Professional Documents
Culture Documents
Chapter No.
A2 Equilibrium of Force Systems. Truss Structures. Externally Braced Wings. Landing Gear.
A3 Properties of Sections - Centroids. Moments of Inertia, etc.
j A4 General Loads on Aircraft.
A5 Beams - Shear and Moments. Beam - Column Moments.
A6 Torsion - Stresses and Deflections.
! ;, A7 Deflections of Structures. Castigliano's Theorem. Virtual Work. Matrix Methods.
r j,
AS Statically Indeterminate Structures. Theorem of Least Work. Virtual Work. Matrix Methods.
A9 Bending Moments in Frames and Rings by Elastic Center Method.
A1Q Column Analogy Method.
All Continuous Structures' Moment Distribution Method.
A12 Slope Deflection Method.
Chapter No.
R_ .
of Plates. • • • . . . . . . . AlB. 12 Torsion - Effect of End UnsymmetriCal Frame • . . A9.2
Streamline Tubing - Strength. C4. 12 Restraint . • • . . . . • A8.16 Unsymmetrical Frames or
Strength Checking and Torsion - Non-circular AIO.4
Design - Problems • . . . C4.5 Sections • . . • . • • • A6.3 Unsymmetrical Frames using
Stren~··_ ".: Round Tubes Torsion Open sections AS. 4 PrinCi~ Axes. • • . •. M.tS
_ ..nder Combined Loadings . C4.22 Torsion of Thin-Wailed .'Tnsymmeirical Structures A9.13
sn-ess Analysis Formulas Cll. 15 Cylinder having Closed Type .' :''"it
" , ':-·"'·A
Stress Analysis of Thin Skin - StiHeners . . . . . . . . . • AS. IS \t, . 'Jy - Load Factor
Multiple stringer Cantilever Torsion Thin Walled Sections. A6.S .....
" • • • .("r- A4.7 •
Wing • . . . . . . . . . . • • A19.10 Torsional Moments - Beams • A5.9
Stress Concentration Factors. C13.10 Torsional Modulus of Rupture. C4.1'7 Wagner Equations .. Cl1.4
Stress Distribution & Angle Torsional Shear Flow in Web Bending &: Shear Stresses CtO.5
of Twist for 2-Cell Thin 4
Multiple Cell Beams by Web Design • . • . • . • • • Cll.18
Wall Closed Section . . A.6.7 Method of Successive Web Splices • • . • . . • • . CIO. 10
Stress-Strain Curve • . • . • B1.7 Corrections . • • • . . A6.10 Web Strength. Stable Webs. CtO.5
Stress-Strain Relations .•• A24.6 Torsional Shear Stresses in Webs with Round Lightening.
Stresses around Panel Cutout. A22.1 Multtple 4Cell
Thin-Wall Holes • • • • • . . • • ClO.17
Stresses in Uprights Cll.17 Closed Section - Distribution AS.7 Wing Analysis Problems A19.2
Stringer Systems in Diagonal Torsional Strength of Round Wtng Arrangements. • • Al9.1
Tension . . . • . • . • • • . Cll.32 Tubes •.•.•..•.•• C4.17 Wing Effective Sectton A19.12
Structural Design Philosophy. ci, 6 Torsional Stresses 1n Wing Internal Stresses • A23.14
Structural Fittings • • • • . . A2.2 Muitiple Cell Thin-Walled
4
Wing Shear and Bending
Structural Skin Panel Details. D3.12 Tubes ••....•... A6.S AnaLysis • • • . . . • • • A19.14
Structures with Curved Transmission of Power by Wing Shear and Bending
Members • . • . • . . . All. 29 Cylindrical Shaft. . . . • A6.2 Moments • . • • . • . . A5.9
Successive Approximation TriaXial Stresses . • • • . Cl.5 Wing - Sbear Lag . • • • A19.25
Method for Multiple Cell Truss Deflection by Method Wing Shears and Moments AS. 10
Beams . . . . . • . . AlS.24 of Elastic Weights • A7.33 Wing Stiffness Matrix.. • • A23.11
Symbols for Reacting Truss Structures A2.9 Wing Strength ReqUirements A19.5
Fitting Units . . . . • A2.3 Trusses with Double Wing Stress Analyl'lis Methods .'1.19.5
Symmetrica sections - Redundancy. • . . AS. 10 Wing - Ultimate Strength . A19.11
External Shear Loads A14.2 Work of Structures Group. Al.2
(1) Wind Tunnel and Fluid Mechanic5 ~est The final results of t.he work of this
Labs. group are formal reports glv~n~ complete a~plied
(2) Structural Test Labs. load design criteria, with ~ny graphs ~nd swu-
(3) Propulsi~n Test Labs. mary tables. The final results ~y 61v8 com-
(4) Electronics ~est Labs. plete shear, moment and no~l forc~s =e~er=~d
(5) Electro-Mechanical Test Labs. to a convenient set of :CY2 axes for major air-
(6) Weapons and Controls Test Labs. c ra.r t units such as the Wing, rus eIage , e t c .
(7) Ar~log and Digital Computer Labs.
THE WORK OF STRESS ANALYSIS ~\m S~R~~GTH GROUP
v. Flight Test Section.
Essentially the primary job of :he stress
VI. Engineering Field Service Section. group is to help specify or deter.nine the kind
of material to use and the :h:c~~ess, size and
SinCe this textbook deals with the subject cross-sectional shape Jt every struct~l ~eQ
of structures, it seems appropriate to discuss ber or unit on the airplane or ~issile, and
in some detail the work of the Structures Group. also to assist in the deSign of all jOints and
For the detailed discussion of the other grou~s, connections for such ~embers. safety with ~ight
the student should refer to the various air- weight are the paramount str~ctural jesl~ re-
craft company publications. quirements. ~he stress group ~ust consta~tly
work closely with the Structural DeSign Sect:Gn
At. 3 The Work of the Structures Group in order to evolve the best structural over-all
arrangement. Such factors as ~ower ~lants,
The structures group, relative to number of bUilt in fuel tanks, landing gear retracting
engineers, is one of the largest of the ~any wells, and other large cut-outs can d~ctate the
groups ot engineers trat make up Section II, type of wing structure, as for example, a two
the technical analySis section. The structures spar single cell wing, or a multiple spar
group is primarily responsible for the ~ultiple cell wing.
structural integrity (safety) ot the airplane.
safety may depend on sufficient strength or To expedite the initial struct~r~l ~esign
sufficient rigidity. This structural integrity studies, the stress group ~ust s~~ply initial
must be accompanied with lightest pOSSible structural sizes based on approximate loads.
weight, because any excess weight has detri- The fi~l results of the work by the stress
mental etfect upon the perfo~ce of aircraft. group are recorded in elaborate reports which
For example, in a large, long range missile, show how the stresses were calculated and hew
one pound of '~ecessary structural weight may the reqUired member sizes were obtained to carry
add mora than 200 Ibs. to the overall weight of these stresses efficiently. The r:nal size of
the missile. a member may be dictated by one or more rae tors
such as elastic action, tne Ias t t c action, ele-
The structures group is usually divided vated temperatures, fatigue, etc. To insure
'into sUb-groups as tollows:- the accuracy of theoretical calculations, the
stresS group must have the assistance of the
(1) Applied Loads Calculation Group structures test laboratory in order to obtain
(2) Stress AnalySiS an~ Strength Group information on which to base allowable design
(3) Dynamics AnalYSiS Group stresses.
(4) Special Projects and Research Group
THE WORK OF THE DYNAMICS A~LYSIS GROUP
THE '"ORK OF THE APPLIED LOADS GROUP
The Dyna~ics AnalysiS Grou) has rapidly
Before any part ot the structure can be expanded in recent years ~elative to number of
finally proportioned relative to strength or engineers required because supersonic airplanes,
rigidity, the true external loads on the air- missiles and vertical riSing ai~craft have pre-
craft must be determined. Since critical loads sented many new and complex problems in the
came tTom many sources, the Loads Group must general field of dynamics. In some airc~tt
analyze loads fram aerOdynamiC forces, as well companies the dynamiCS group 1s set up as a
as those forces from power plants, aircraft separate group outside the Structures Group.
inertia; control system actuators; launching,
landing and recovery gear; a~ent, etc. The 7he engineers in the dynamiCS group are
etrects of the aerOdynamic forces are initially responSible for the investigation ot Vibration
calcUlated on the assumption that the airplane and shOCk, aircraft flutter and the establish-
structure 1s a rigid bOdy. Atts: the aircraft ~ent of desig~ requirements or c~2nges for its
structure is Obtained, its true rigidity can control or correction. Aircraft contain dozens
be used to obtain dynamic effects. Results of of mechanical installations. Vicration of ~~y
wind t~~el model tests are usually necessary part of these installations or systems ~y be
in the application of aerodynamic principles to of such character as to cause faulty operation
load and pressure analYSiS. • or danger of failure and therefore the dynamic
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A1.3
characteristics must be changed or modified in groups have a speCial SUb-group which are work-
order to insure reliable and safe operation. ing on deSign problems that Nill be encountered
in the near 1r distant future as aviation pro-
The major structural units of aircraft such gresses. For example, in the r.t r-uc tur-ea Group,
as the wing and fuselage are not rigid bodies. this sub-group might be studying such problems
7hus when a Sharp air gust strikes a fleXible as: (1) how to calculate the thermal stresses
wing in high speed flight, we have a dynamic in the wing structure at super-sonic speedS;
load situation and the wing Nill vibrate. The (2) how to stress analyze a new type of wing
dynamicist must determine whether this vibration structure; (3) what type of body str~cture is
1s serious relative to induced stresses on the best for future space travel and what kind of
wir~ structure. The dynamics group is also materials will be needed, etc.
responsible for the determination of the
stability and performance of miSSile and flight Chart 1 illustrates in general a typical
vehicle guidar.ce and control systems. The ~ke-up of the Structures Section of a large
dynamics group must work constantly with the aerospace company. Chart 2 lists the many
various test laboratories in order to obtain items which the structures engineer must be
reliable values of certain factors that are concerned with in insuring the structural
necessary in many theoretical calculations. integrity of the flight vehicle. Both Charts
1 and 2 are from Chance-Vought Structures
THE ',jaRK OF THE SPEC IAL PROJECTS GROUP DeSign Manual and are reproduced with their
permi saton.
In general, all the various technical
~EIlOE'-'\STlc:'···4
U..f ,.
11:'
' ""
lAKlIlATo'" I
~~u,
I. "'",ru,,,
'liSI U,,"T
1
J i
, -.c...."I.IC
'NO "Owu
' ..... 1 ""ST
: UN"
, .....c:.. 'H~
COMPUIATICH
~'''G'' GIlOU'
~~,
-:3
e.13W m
= r ETT? 5ZWF
MATERIALS OF
CONSTRUCTION
FA.STENERS
WELDING
BONDING
?LATE~ND BAR
FORGINGS
STIFFNESS CASTINGS
CRITERIA €XTRUSIONS STRESS
SHEET METAL
FLUTTER SANDWICH
ANALYSIS
CONTROL SYSTEM STABILITY ?tASTIC l,OMINAIT SKIN PANElS
BEARINGS BEAM ANALYSIS
PANEL fLIJTTER-SKIN CONTOURS STRAIN COMPATIBILITY
CONTROL SYSTEM DEfLECTIONS STRAIN CONCENTRATION
JOINT ANALYSIS
THERMAL EFFECTS BEARI~G ANALYSIS
MtCHANICAL VIBRATIONS BULKHEAD ANALYSI S
fITTING ANALYSIS
ROLL POW£R-O IVERGENCE i1'ERMAL STRESS
AEROtWtAMIC CENTER SHIFT ,'oIECHAN1CAL COMPONENTS
D:PERIMENTAL STRESS ANALYSIS
DYNAMIC RESPONSE
LOADS AND
ENVIROMENT MATERIALS AND
FLIGHT LOAD CRITERIA
GROUND LOAD eRITER IA QUALITY CONTROL
FLIGHT LOAD DY~M'CS DUCTILITY
LAUNCHING DYNAMICS
LANDING DYNAMICS STRESS-STRAIN
DYNAMIC RESPONSE HOMOGENEOUS MATERIAL
aECOVERY DYNAMICS
,UGHT LOAD DISTRIBUTIONS RESIDUAL STRESS
INERTIAL LOAD DISTRIBUTIONS COMPONENT ALLOWABLES
HEAT TREAT CONTROL
Fl..D:IBILITY EFFECTS YIHDING
GROUND LOAD DISTRIBUTIONS
ANALYSIS STRESS CORROSION
FRACTURE
REPEATEO LOAD SPECTRUMS UNIT SOLUTIONS STABILITY AT TE~P£RATURE
TEMPERATURE DISTRIBUTIONS °ATlGUE
INOffiRMINATE STRUCTURES SPECIFICATION CONFORMANCE
LOADS FROM TllERMAL WEAR, BRINELLING
DEFORMATIONS WING ANALYSIS BLUE PRINT CONFORMANCE
PRESSURES-IMPACT CREEP
TAIL ,l,NAlYS1S
DEFliCTIONS
FUSElJ.GE SHElL A""lYSIS
Tl'ERMAl O"FECTS
THERMAL ANALYSIS
STIFFNESS
DfFlICTION ANALYSIS
COMBINED LOADINGS
STIFFNESS
3UCKlING
Chart 2
From Chance- Vought Structures Destgn Manual
CHAPTER A2
EQUILIBRIUM OF FORCE SYSTEMS. TRUSS STRUCTURES
A2.1 Introduction. The equations of static force system pass through a cammon point. The
equilibrium must constantly be used by the reSUltant, if any, must therefore be a force
stress analyst and structural designer 1~ ob- and not a moment and thus only 3 equations are
taining unknown forces and reactions or unkno~n necessary to completely define the condition
internal stresses. They are necessary whether that the resultant must be zero. The equat10ns
the structure ..or machine be s tmp.Le or complex. of equilibrium available are therefore:-
The ability to apply these equations 1s no
doubt best developed by solving ~ny problems. l:F x =a m, = o
This Chapter l~ltlates the application of these ZFy = a or m, = o } - - - - -(2.2)
important phySical laws to tt.e force and stress
ar~lysls of structures. It 1s assumed that a
l:Fz =0 m, = o
student has completed the usual college course A combir~tion of force and moment equations
in engineering mecrznlcs called statics. to make a total of not more than 3 can be used.
For the moment equations, axes through the point
•
A2.2 Equations of Static Equilibrium. of concurrency cannot be used since all forces
of ~he system pass through this point. The
To completely Cefl~e a force, we must know moment axes need not be the same direction as
its ~agnltude, direction and ~olnt of ann1ica- the directirns used in the force equations but
tion. These facts regaTding the ~arce are of course. they could be.
generally refer~ed to as the characteristics of
the ~orce. So~etimes the more ~eneral te~ of EQUILIBRIUM OF S?AC::; ?J..RALLEL FORCE SYSTlli
line Of act~on or lecation is used as a force
cr~racteristic in Place of paint of application
In a parallel force system the direction of
designation. all forces is known, but the magnitUde and
location of each is unknown. Thus to determine
A force acting in space is completely
~gnitude, one equation Is required and for
defir-ed :: we %now its components in three location twa equations are necessary since the
directions and its ~oments about 3 axes, as for force is not confined to one plane. In general
example FX J F~, Fz ~nd ~x, tly and Xz • ~or the 3 equations commonly used to make the re-
equilibrium o~ a force system there can be no
sultant zero for this type of ~orce system are
resultant force and thus the equations of
equllibri'4n are obtained ~y equating the force one force equation and two moment equatiOns.
For example, for a space parallel force system
and moment cCill~onents to zero. The equations acting in the y direction, the equations of
~f static equilibrium for the various types of
force systems 'Nill now be suear-tzeo . equilibrium would be:
.AC
A2.2 EQUILIBRIUM OF FORCE SYSTEMS. TRUSS STRUCTURES.
n1 z =0
or
4I'1z Ii =a } -----2.6
~'l
(The moment centers 1 and 2 cannot be on the
same y axis)
-
II p
EQUILIBRITJr1 OF COLINEAR FORCE SYSTEM
A collnear torce system 1s one where all For any force such as P and Q acting i~ the
forces act along the same line or in other xy plane, the line of action of such a ~~rce
words, the direction and location of the forces ~ust pass through the ,in center since t~e
is known but their nagnt tudes are unknown, thus fitting unit cannot resist a ~cme~t about a z
only magnitude needs to be found to define the axlo through the pin center. Therefore, fJr
resultant of a collnear force system. Thus forces acting in the xy plane, th~ direct:cn
only one equation ot equilibrium is a'ffiilable, ~~d line of action are established by th8 pin
namely joint as illustrated in the f:gure. Sir.ce a
Single pin fitting can resist :noments atou: axes
ZF =a or ZI1 1 = O ----2.7 perpendicular to the ~in axis, the cirectlon and
line of action of out of ,lane :orces is ~tere
where moment center 1 is not on the line at fore not established by s tng l e pfn ~it-:lng umt.s .
action of the force system A ,B
A2.3 Structural Fitting Units for Establishing the Force
Characteristics of Direction and Point of Application.
-<@o========:j):Jl_
If a bar AB has si~Gle pin f~t~ings at
To completely define a force in space re- eac~ end, then any :orce P lying in the xy
qUires 5 equations and 3 equations if t~e force plane and anplied to end B ~ust have a direction
is limited to one ?lane. In ~eneral a structure and line of action co tnc tc t ng ....ith a line jo tn-.
is loaded by ~own forces ar.d these !orces are i~g the pin centers at end ~it~1ngs A ar.d 3,
transferred tPIough the struct~re in some since the :lttings cannot resist a ~oment about
~nner of internal stress distribution and then the 3 axis.
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A2.3
Double Pin - Universal Joint Fittings Another general fitting type that is used
to establish the direction of a force or reaction
z Q is illustrated in the figure at the bottom of the
L. § /-/11 P
first column. Any reacting force at joint (A)
~ust be horizontal since the support at (A) is
so designed to provide no vertical resistance.
Fh" liillirn
L B
-~~-
fitting support or attachment units are to be
used or are assumed to be used in the tinal
design. The following sketch symbols are com-
~onlY used for coplanar tares systems.
t t
Pin
In order to pe~it structures to move at ioi
~
",4", namely through point (b). Instead of using
I direction as an unknown, it is more convenient
Lubricated Slot ! ~
b'RI=Jn-'
',~
Double RoUer
to replace the resultant reaction by two com-
ponents at right angles to each other as indi-
cated in the sketches.
/1.
~,
A2.4 EQUILIBRIUM OF FORCE SYSTEMS. TRUSS STRUCTURES.
!
Rollers FA Roller.
brium. A structure can be statically indeter-
minate with respect to exter~al reac~ions alone
or to In:ernal stresses alene or to 60th.
I I
The additional eq~ations :hat are needed
to solve a statically indeternlnate structure
are obtained by conSidering the distortion of
The above fitting units using rollers fix the structure. This means that the size of all
the direction of the reaction as normal to the members, the ~~terial ~rom which members are
roller bed since the fitting unit cannot resist ~de must be known since dist~rtions ~ust be
a horizontal force through point (b). Hence calculated. In a statically cete~lnate
the direction and point of application of t~e structure this information on sizes and ~terial
reaction are established and only magnitude is 1s not required but only the configuration of
unknown. the structure as a whole. Thus deSign analYSiS
for statically deter,ninate structure is stra~ght
forward whereas a general t~ial and error pro-
fixed
(bJM., cedure is required for design analYSiS of
rrrrrt '" R X - I I I \ \ \ \ statically indete~inate structures.
~In \~S::
internal stresses can be found by statics arter
haVing found the external reactions. Obviously,
-, the internal stresses will be affected by the
internal reactions at C and 0, so we draw a free
,He fRB .J.VC body of the super-st~cture as illustrated in
Fig. A2.4 ?ig. 2.6 and consider the internal forces that
eXisted at e and 0 as exterr~l reactions. In
~ig. 2.4 shows a straight ~ember 1-2 carrying a the actual structure the members are rigidly
known load system ? and supported by 5 struts attached together at point C such as a Nelded or
J 1 'L.-
I
' _ '\{4x5(
the strJcture Nould be still stable and all ~ine the axial loads in the members and the re-
~ember axial stresses could be found by the actions on the spar.
equations of static equilibrium without regard
to their size of cross-section or the kind at Solution: The first thing to decide is whether
material. Adding the second diagonal member the structure is statically determinate. From
in each panel would necessitate knOWing the the figure it is observed that the wing spar is
size of all truss members and the kind of supported by five struts. Due to the pins at
material used before member stresSes could be each end of all struts, we r~ve five unknowns,
found, as the additional equations needed must ~~elYJ the magnitude of the load in each strut.
come from a consideration involving distortion DirectIon and location at each strut load is
of the truss. Assume for exa~ple, that one known because of the pin at each end at the
diagonal in the upper panel was lett out. We struts. We have 3 equations of equilibrium for
would then be able to find the stresses in the the wing spar as a single unit support'ed by the
~embers of the_upper ?anel by statics but the 5 struts, thus two ~ore equations are necessary
lower panel Would still be statically inde- it the 5 unknown strut loads are to be found.
terminate to 1 degree because or the double It is noticed that the wing spar includes 2 in-
diagonal system and thus one additional equation ternal single pin connections at points a and 0'.
is necessary and would involve a consideration This establishes the fact that the moment of all
of truss distortion. (The solution ot static- forces located to one side of the pin must be
ally indeterminate trusses is covered in equal to zero since the single pin fitting can-
Chapter A.B.) not resist a moment. Thus we obtain two addi-
tional equations because of the ~NO internal pin
A2.7 Example Problem Solutions of Statically Determinate fittings and thus we have 5 equations to :ind 5
Coplanar Structures and Coplanar Loadings. unknowns.
Although a student has taken a course in
sta:ics ~efore taking a beginning course in Fig. 2.15 shows a tree body of the wing
aircraft structures, it is felt that a limited spar to the right of hinge fitting at O.
~eview of )roblems involving the application
1013",(30+ 15)45
of the equations of static equilibrium is quite
j~stlfied, part~cularly 1: the prOblems are I 20~ 2
posSibly somewhat more difficult than ~ost or
~he problems in the usual begir~ing course in
stat~cs. Since one ~ust use the equations or
Fig. A2.15
static equilibrium as ?art ot the necessary
equations in solving statically indeterminate
st~uctures and since statically indeterminate In order to take moments, the distributed
structures are covered in rather complete detail load on the spar bas been replaced by the re-
Ln other c napt ers at :::115 book, only limited sultant load on each spar portion, namely, the
s?ace will be given to ~roblems involVing total load on the portion acting through the
statics in this chapter. centroid of the distributed load system. The
strut .r-eact.t on EA at A tas been shown in phantom
~a~ple Problem 8. as it is ~ore convenient to deal with its com-
ponents YA and XA' The reaction at 0 15 un-
Fig. A2.14 shows a much SI~plitied wing known tn ~gnltude and direction and for con-
structure, cor~lsting of a wing spar supported venience we will deal with its components xo
'oy lift and cabane struts wmch tie the wing and YO. The sense assumed is indicated on the
spar to the fuselage structure. The distributed figure.
air load on the wing spar is unsymmetrical about
the center line of the air!r~e. The wing spar The sense of a force Is represented
is ~ade in t~ree units, readily disassembled by graphically by an arrow head on ~he end of a
using ern fittings at points 0 and 0 1 • All vector. The correct sense is obtained from the
su~porting wing struts have Single pin fitting solution ot the equations of equilibrium since,
~l~ts a: eac~ end. The problem is to deter- a force or ~oment must be given a plus or minus
Sign in writing the equa~ions. Since the sense
201ll/in. 401ll/in.
of a force or moment is unknown, 1t is assumed,
..I...,....,--r;i t • j " " ' . i
~Oi/in.
Ill'
l5#/in.
t f ! ,--, and if the algebraic solution of the equilibrium
equations gives a ~lus value to the magnitude
-r i- 45"~ 82" ---iZO'!_60" ---! Hinge
o' , ';0 then the true sense is as ass~ed, and oppOSite
A 8' B 0" A PIn to that assumed it the solution gives a minus
Lilt Caban , -+- sign. If the unknown forces are axial loads in
Struts truts C~ c 36" ~embers it is camoan practice to call tensile
Fuselage -' "30":.......i..- stress plus and compressive stress minus, thus
E' E Fig. A2.l4 if we ass~.e the sense of an unknown axial ~oad
'i 8ym. 'i ~s tenSion. the solution of the equilibr!um
A2.8 EQUILIBRIUM OF FORCE SYSTEMS. TRUSS STRUCTURES.
equations will give a plus value for the magni- To find strut load g'C I ~ake m~ents about
tude of the unknown if the true stress is point C.
tension and a ~inus sign will indicate the
assumed tension stresses should be reversed or L~ ~ 1325 x 65 + 2000 x 40 + (58eO - 44JO)
compression, thus giving a consistency of signs. 30 - 1500 x 10 - 993 x 35 - 30 (3'8')
30/33.6 =0
To find the unknown YA we take ~cments
about point 0 and equate to zero for equilibrium whence, BtCI = 6000 lb. with sense as sho~TI.
- 2460 x 41 - 1013 x 102 • 82YA =0 To find load in member BIC use equation
ZFy '::: 0 =. 1325 + 2000 + 1500 + 993 - 6000
Hence YA = 204000/82 = 2480 lb.. The plus sign
means that the sense as assumed in the figure (30/33.6) - 2720 (30/33.6) - B'e (30/54)
1s correct. By geometry XA = 2480 x 117/66 = =0
4400 lb. and the load in strut ~ equals
v4400 3 + 2480 3 = 5050 lb. tension or as whence, BIC =-
3535 lb. The ~inus sign
assumed in the :lgure. means it acts opposite to that sho'Nn in fi~~e
or is compression instead of tension.
To find Xo we use the equilibrium equation
ZFX = 0 = Xc - 4400 = 0, whence Xc = 4400 lb. The reactions on the spar can now be
deter.nined and shears, ber.ding moments and
To find YO we use, axial loads on the spar could be round. The
numerical results should ~e checked for eqUili-
ZFy = 0 = 2460 + 1013 - 2480 - YO = 0, whence brium or the spar as a whole by taking moments
YO = 993 lb. ot all forces about a dif:erent ~oment center
to see i t the result "is zero.
To check our results for eqUilibrium we
will take moments at all torces about A to see Example Problem 9.
it they equal zero.
l:11A = 2460 x 41 - 1013 x 20 - 993 x 82 = 0 check
On the spar portion OIA' J the reactions -r-r-'
are obViOUSly equal to 40/30 times those :ound
for portion OA since the external loading is 40
I
strut i ~ 12 12"
as compared to 30.
i\~ I ~ Brace Strut I
Hence AIEl. 6750, XOI = 5880, YO' = 1325 24lJ~ \, E' ---+--
i
12 ~ I
Fig. 2.16 shows a tree body ot the center 12"
spar portion with'the reactions at 0 and at as
found preViously. The unknown loads in the
struts have been assumed tens10n as shown by
the arrows.
Fig.A2.17 I
F--
Shock
Strut
+
12"
1500",50x30
A -L
r 300
" 30
'':In,"
15
, " Fig. 2.17 shows a Simplified airplane
C"9----~ landing gear unit with all members and loads
I Flg.A2.16 confined to one plane. The brace struts ~e
I I pinned at each end and the support at C is of
L __ J the roller type, thus no vertical reaction can
be produced by the support fitting at point C.
To find the load in strut Be take moments The member at C can rotate on the roller but
about s' horizontal movement is ~revented. A knovm load
ot 10,000 lb. 1s applied to axle ~nit at A. The
!MB' = 1325 x 20 - 2000 x 5 - 1500 x 55 problem is to f1nd the load in the brace struts
and the reaction at C.
- 993 x 80 • 60 (BC) 30/33.6 = a
Solution:
Whence, Be • 2720 lb. with sense as Due to the Single pin fitting ~t each end
assumed. of the brace struts, the ~eact1ons ~t a and D
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A2.9
are collnear with the strut axis, thus direction the aluminum alloy tubular truss. Trussed type
and point of application are known for reaction beams composed of closed and open type sections
RB and RD leaving only the magnitude of each as are also frequently used in Wing beam construc-
unknown. The roller type fitting at C fixes t1on. The stresses or loads in the members ot
the direction and point of application of the a truss are commonly referred to as nprimaryW
reaction Re, leaving magnitude as the only and wsecondarJn stresses. The stresses which
unknown, Thus there are 3 unknowns Re, He and are found under the following assumptions are
RO and with 3 equations of static equilibrium referred to as primary stresses.
available, the structure is statically deter.ni-
nate with respect to external reactions. The (1) The members of the truss are straight,
sense of each of the 3 unknown reactions has weightless and lie in one plane.
been assumed as indicated by the vector. (2) The members of a truss meeting at a'
point are considered as jOined together by a
To find Rn take moments about point B:-
common frictionless pin and all member axes in-
ZME = - 10000 sin 30° x 36 - 10000 cos 300 x 12 tersect at the pin center.
- RO (12/17) 2~ = 0
(3) All external loads are applied to the
whence, RO ~ - 16750 lb. Since the result truss only at the jOints and in the plane ot
comes out with a minus Sign, the reaction RO the truss. Thus all loads or stresses produced
has a sense OPPosite to that shown by the in members are either axial tension or compres-
vector in Fig. 2.17. Since the reaction RO is sion without bending or torsion.
colinear with the line DE because of the pin
endS, the load in the brace strut DE is 16750 Those trusses produced in the truss mem-
lb. compression. In the above moment equation bers due to the non-fulfillment or the above
about B, the reaction RO was resolved into assumptiOns are referred to as secondary
vertical and rorizontal components at point D, stresses. Most steel tubular trusses are welded
and thus only the vertical component which together at their ends and in other truss types,
equals (12/17) RD enters into the equation the ~embers are riveted or bolted together.
since the horizontal component has a line of This restraint at the joints may cause second-
action through point B and therefore no moment. ary stresses in some members greater than the
He does not enter in equation as it has zerO primary stresses. Likewise it is common in
moment about B. actual practical deSign to apply torces to the
truss members between their ends by supporting
To find RE take ZFv = 0 many equipment installations on these truss
members. However, regardless of the magnitude
ZFv = 10000 x cos 30° + (- 16750)(12/17) + HE at these so-called secondary loadS, it is
(24,/26.8) = 0 cammon practice to first find the priwary
Whence, HB = 3540 lb. Since Sign comes stresses under the assumption outlined above.
out plus, the sense is the same as assumed in
the figure. The strut load BF is therefore GENERAL CRITERIA FOR DETERl'lINING WHEI'JlER
3540 lb. tension, since reaction HB is collnear TRUSS STRUCTURES ARE STATICALLY DEI'ERMINATE
wi th line SF. WITlf RESPECT TO INl'ERNAL STRFBSES.
I .
, ~.
/'
A2.10 EQUILIBRIUM OF FORCE SYSTEMS. TRUSS STRUCTURES.
It the truss has more members than indi- 30" -+- 30" -----.j
cated by equation (2.8) the trusS is considered i i i
redund~~t and statically lnGeterminate since ',1000
the member loads cannot be found in all the
members by the laws of statics. Such redundar.t
(1
L~~ ~_:)oo
, -
structures it the members are properly ?laced Fig. A2. 19 \ - ,
are stable and will support loads of any u,~
arrangement.
or tension. In equation (0) the load of 1250
ANALYTICAL IlE:rHODS fOR DE:l'SRtrINI'IG in L~L:;l was substituted as a minus value since
PRIMARY STRESSES IN TRUSS STRUCT1JRES it was foune to act opposite to ~hat sho~m in
Fig. A2.l9. Possibly a tetter p r-ccedur e would
In general there are three rather distinct be to change the sense of the ar~ow i~ ~he :ree
methods or procedures in applying the equations bodY diagram for any solved ~embers Jefore writ-
of static equilibrium to finding the primary ing further equilibriUW equati~ns. ~e ~ust
stresses in truss type structures. They are proceed to joint L:::I instead c r joint U2, as
otten referred to as the method of joints, three ~~~own nembers still exist at jOint U2
moments, and shears. Whereas only ~NO at jo~nt La. Fig. A2.20 shows
f~ee body of jOi~t La cut ou~ by section 2-2
A2. 9 Method of Joints. (see ~lg. AS.IS). The sense of the urJL~own
It the truss as a whole is i~ equilibrium
member s~ress L:;lU:;l has Jeen assw~ec as COID-
pressic~ (pUShing toward jOint) as ~t is ob-
then each member or joint in the tr~ss must
likewise be in equilibrium. The forces in the Viously act~ng this way to Jalance :he 500 lb.
members at a truss jOint intersect in a common load.
point, thus the forces on each joint form a 500
concurrent-coplanar force system. The ~ethod
~, ~H
of joints consists in cutting out or isolating
L. 250
a joint as a free bOdy and applying the laws ot u.,
equilibrium for a Concurrent force system.
Since only two independent equations are avail- tu. Fig. A2. 20
able for this type of system only two unknowns
can exist at any jOint. ThUS the procedure is For equilibrium of jOint La, ZH and ZV 0 =
to start at the jOint Where only two unknowns
exist and continue ,rogressively throughout the
ZV =-500 + LaUa = 0, whence J LaUa = 500 lb.
Since the Sign came out plus, the assumed sense
truss joint by joint. To illustrate the method in Fig. A2.20 was correct or compreSSion.
consider the cantilever truss of Fig. A2.18.
From obse~lation there are only two members ZH = 250 - LaLl : 0, wr.ence LaLl =250 IJ.
With internal stresses unknown at jOint L~.
Fig. A2.19 shows a free body of jOint L~. The Next consider joint Ua as a :ree jody c~t
stresses in the members L~ La and L~ U~ have out by section 3-3 in Fig. A2.18 and erawn as
been assumed as tension, as indicated by the Fig. A2.2l. The known member s~resses are showr.
arrows pulling away from the jOint L~. with their true sense as ~reviouslY found. The
two ~own member stresses U2L l and UaU l have
The static equations or equilibrium for Jeen assumed as tension.
the forces acting on joint L~ are ~~ and ZV =o.
!V =- 1000 _ L,U. (40/50) =a - - - - - - -fe )
whence, L ~U:;l =-
1250 1'0. Since the Sign Fig. A2. 21
came out minus the stress is opPOSite to that
aSStooed in Fig. A2.l9 or compression.
Eli =- 500 - (- 1250)(30/50) - L,L, =0 - -(b) ZV = -500 - 1250 (40/50) + UaLl (40/50) = 0
Whence, LaL~ = 250 lb. Since Sign comes whence, UaLl = 1875 is. (tens 1on as
out plus, sense is s~e as assumed in figure asaumed ,.)
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A2.11
ZH = (-1250) (30/50) - 1875 (30/50) - U.U. =0 The algebraic sign of all unknowns came out
whence, U..U a = - 1875 lb. or opposite in
positive, thus the assumed direction as shown
sense to tr2t assum~c ana therefore compression. an Fig. A2.22 was correct.
Note: The student should continue with succeed- Check results by taking &~B = 0
ing joints. In this example involving a canti- ~B = 1400 x 150 + 500 x 30 - 500 x 120 - 500 x
lever truss it was not necessary to ~lnd the 30 - 1000 x 90 - 1000 x 60 = 0 (Check)
reactions, as it was ~oSSlble to select joint
L 3 as a joint involving only VNO unknowns. In To determine the stress in member Fl , Fa and F"
trusses such as illustrated in Fig. A2.22 i t 1s we cut the section 1-1 thru the truss (Fig.
necessary to first find reactions R:l. or R a wni cn A2.22. Fig. A2.23 shows a free bOdy diagram of
then provides a joint at the reaction point in- the portion of the truss to the left of this
volving only two ~cwn forces. section.
f
, 'f~ _
Fig.A2.22
1400
a
o ,
500 1
500
~:
~; •
1400 1
Fig. A2.25 the ferce F 3 is resolved into its the other two ~Jcnowns F~ and ~s lles at i~flni
component F 3V and F3H at point 0'. 7hen taking ty. Thus for conditions where two of the 3 c~t
members are parallel Ne have a ~ethod of solving
~or the web ~ember of the t~uss ccmr.,only re-
ferred to as the ~ethod cf shea~s, or the sum-
mation of all the forces nO~ial to the tNO
parallel ~{nown chord members ~us~ equal zero.
Fig. A2. 2S
Since the parallel chord ~embers have no com-
ponent in a direction no~al to their line of
action, they do not enter the above equatton of
equilibrium.
1000lii
!
Fig. A2. 26
50'"'
1400lii
Fig. A2. 27
/IF,
500 ...."::,,,---¥--7"- F ,
500 3
moments about point 0 as before:- 11400
Fig. A2. 28
members are horizontal and thus have zero verti- since no external vertical load exists at joint
cal component. E. Similarly, by the same reasoning for LH ~ 0,
load in DE = O. The load in the diagonal FO
(2) The truss verticals in general resist equals the value on the diagonal of the panel
the vertical component of the diagonals plus index triangle or 167 lb. It is tenSion by
any exter~l loads applied to the end joints ot observation since the shear in the panel to the
the vertical. right is up and the vertical component of the
diagonal FD must pull down for eqUilibrium.
(3) The load in the chord members is due
to the horizontal components ot the diagonal ConSidering Joint F. ZH = - FG - FOR = 0,
members ana in general equals the summation of which means that the horizontal component of the
these horizontal components. load in the diagonal OF equals the load in FG,
or is equal to the value of the horizontal side
To illustrate the simplicity at determining in the index triangle or - 133 lb. It is nega-
stresses in tITe members of a parallel chord tive because the horizontal component of DF
truss, consider the cantilever truss of Fig. pUlls on Joint F and therefore Fa must pUSh
A2.29 with supporting reactions at points A and against the joint for equilibrium.
J.
Considering Joint D:-
150
ZV = DFv + DO = O. But DFv = 100 (vertical side
of index triangle)
• ", DG = - 100
LH = DE + DFR - DC =0, but DE=° and DFa =
133 (from. index triangle)
DC = 133
550 36"-+- 40"
Considering Joint G:-
Fig. A2.29
ZH=-GH - GF - GCR = O. But OF =- 133, and GCR
= 300 from index triangle in the second panel.
~lrst, compute the length triangles in Hence OH =-
433 lb. Proceeding in this manner,
each panel of the truss as shown by the dashed we obtain the stress in all the members as shown
triangles in each panel. The other triangles in Fig. AZ.29. All the eqUilibrium equations
in each ~anel are referred to as load ar index can be solved mentally and with the calculations
triangles and their sides are directly pro- being done on the slide rule, all member loads
portional to the length triangles. can be written directly on the truss diagram.
The shear load in each panel is first writ- Observation at the results at Fig. AZ.29
ten on the vertical side at each index triangle. show tr~t the loads in the truss verticals equal
~hus, in panel EFGD, considering forces to the the values of the vertical sides or the index
right of a vertical section cut thru the panel, load triangle, and the loads in the truss di-
the shear is 100 lb., which is recorded on the agonals equal the values of the index triangle
vertical side of the index triangle. diagonal side and in general the loads in the
top and bottom horizontal trJSs members equal
For the second panel from the tree end, the the summation of the values ot the horizontal
shea~ is 100 + 150 ~ 250 and for the third panel sides of the index triangles.
100 + 150 + 150 = 400 Ib , , and in like manner
550 ~or fourth panel. The reactions at A and J are found when
the above general procedure reaches joints A
The loads in the diagonals as well as their and J. As a c~eck on the work the reactions
horizontal components are directly proportional should be determined treating the truss as a
to the lengths of the diagonal and horizontal whole.
side at the length triangles. 'Thus the load in
diagonal member DF = 100 (50/30) = 167 and ~or Fig. A2.3C shows the solution for the
member CO = 250 (46.8/30) = 390. The hori- stresses in the members of a Simply supported
zontal compor.ent of the load in OF = 100 (40/30) Pratt Tr~ss, symmetrically loaded. Since all
= 133 and :or CG = 250 (36/30) =
300. These panels have the same width and height, only one
values are shown on the index triangles tor length triangle is drawn as shown. Due to
each truss panel as shown in Fig. A2.29. We symmetry, the index triangles are drawn for
start our analysiS for the loads in the ~embers panels to only one side at the truss center
8f the tr~ss by considering joint E tirst. line. First, the vertical shear in each panel
is written on the vertical side of each index
USing LV = 0 gives EF = 0 by obser~ffition. triangle. Due to the symmetry of the truss and
~ ,/
.,/;c
A2.14 EQUILIBRIUM OF FORCE SYSTEMS. TRUSS STRUCTURES.
~
5" Triangle
30" which must be balanced ,:,y - 562 tr; member' U aU 3 '
33" 50 50 50 50
The reac:icn R~ equals t~e val~e on the
-499.5 U -562 lu, vertical side of our i~cex triang:e in the 91.j
j .9;> 0 panel, or 375. This should oe c~ecked usi~g
;.= 1'.9 .s '1 the truss as a whole and ta~lng ~c~ents ~jout
';~~ R, •
312 312 ~L~ 499.5 p..3 jL. If a truss is lcacec unsJ~etr~cally, t~e
i!R, 100 100 100 100 100 reactions should be ceter.r.ined ~irst, atter
6 Panels @ 25" = 150" which the index triar.~:es can be dra'HTI, sta:-t-
ing with the end ~anels, since the ;anel shear
1s then readily calculated.
center panel = (100 + 50) 1/2 = 75. The verti-
cal shear in panel U:I,UgL1L a equals 75 pl~s the A2.12 Aircraft Wing Structure. Truss Type with Fabric
external loads at Ua and La or a total of 225 or Plastic Cover
and Similarly for the end panel shear = 225 +
50 + 100 = 375. With these values kTIQwn, the
TI:e metal covered carrt t Lever' wing 'N1t;'1 i ts
other two sides of the index triangles are di- better overall aerodyna~lc e~:iciency and s~f
ficien~ torsion~l ri 6 ! : l t y has ~ractlcal1y re-
rectly proportio~41 to the sides of the length
triangles tor each panel, and the results are as placed t~e externally braced wing excep~ for lew
sho~m in Fig. AZ.30.
speed cornmerc~al or ?rivate )i:ot aircraft as
illustrated by the aircra:t 1~ tigs. A2.31 a~d
32. The wing covering is usually fabric and
The general procedure fram this paint is to
find the loads in the diagor21s, then in the therefore a drag tr~ss inside the wi~g is
verticalS, a~d finally in the horizontal chord necessarJ to resist loads in the drag t7USs
member's •
direction. Figs. A2.33 and 34 shows the gen-
eral structural layout a: such wings. The two
The loads in the diagonals are equal to the spars or beams are ~etal or wood. Instead of
va.Iues on the hypotenuse 0 f the index triangles.
USing double wires in each d~g truss bay, a
The sense, whether tension or campression, is Single diagonal strut capable ot t~<ing either
deter.nlned by inspection by cutting ~ental tension or compr-ass tve loads could be -rsec ,
sections thru the truss and noting the direction The external brace s t r-ut s are stream j rne tuocs .
ot the external shear load which ~st be bal-
anced by the vertical camponent of the diagon-
als.
Wing Tip (End) Bow in all ~enbers of the lift anc drag tr~sses will
11e / be dete~,ine~. A Si~plified air loac:ng Nill be
Fig. A2. 33 ,; ~ (0., Plywood Tip Fairing ass~~ed, as the ?ur~ose of this problem is to
;iv8 the st~dent practice in solving sta:ically
, Aileron determinate space t~uss str~ctur8S.
~
, t: ~\ Aileron Spar
Plywood Leading
Edge Fairing --l........;'1
I
I
,
/
X
\.,r:=r
.»"----.ll
~'I Compression Rib
Drag wtre
ASSL~D AIR LCADING:-
-
Fitting ~
Airstream ~--,...-,-/i L ift
Fig. A2. 36
Drag
Drag Truss
Front
Beam
,
Center of
Pressure Rear
Beam
=:i"':
-'"
;; Wlng Chor-d e 72"
Cr
lO~'+- 36" ---1
F.Bl
e p
. •
R.B.
~
I
- ' ,1..
I
A2 16 EQUILIBRIUM OF FORCE SYSTEMS TRUSS STRUCTURES
'\.1/
X /11
"'-I
Drag Truss
Table .\2.1
Taking moments about pof nt (2) lIember gym.: V D , t, I V;1. I D;1. I ';1.
114.5R .. - 114.5 x 30.26 x 114.5/2 - 15.13 x Front Be" FB
I 5.99 o 114.34 114.501.05231 0 1.9986
70.5 x 149.75 - 15.13 x 35.25 x 138 = O. Rear Beall. RB , ~.99 o 114.34 114.501.05231 0 1.9986 I
ZV = 0 where V direction
V ... vertical direction,
Take 1s taken normal to
beam I D • drag direction,
ZV =- R~
- 3770 + 30.26 x 114.5 +
S ~ side direction,
1
(30.26 + 15. 3 ) 70.5 =0 L ~ Vv2+ D2 + 32
2
hence R~ =
1295 lb. We stare the solution of joints jy sta~i~g
(The student should always check results by w~th joint (1). Free bOdy sketches cf joint (I)
taking moments about pOint (1) to see it ZM .. are sketchel below. All members are considered
=0) two-f or-c e r.ember-s or havtng pf ns at each end,
thus magnt .ude is the cn.Iy unknown character-
Rear Beam istic of ~~ch member load. The jrag truss ~em
bers coming in to jofnt (l) are replaced :,y a
single reaction called Jl.' Ar t er- Dl. is r ounc ,
(W'" 14. 7#/in. ~
its influence in causlr.g loads in drag truss
(4) 1I ! Ilftltll;,,/2
114. 5" ija 70. 5,,--1
~embers can then be found when the drag :russ as
a whole is treated. In :he joint solution, the
R. R, drag truss has been assumed parallel to drag
direction which is nct quire true frc~ Fig.
The rear beam has the same span dimensions but A2.35, but the error on member loads is negli-
the loading is 14.74 Ib/in. Hence beam re- gible.
actions R~ and R3 will be 14.74/30.25 = .4875
tilnes those for front beam. JOINT 1 (Equations of ~quilibri~~)
hence R s = .4875 x 3770 = 1838 lb.
~
770
R~ = .4875 x 1295 = 631 lb.
(1) D,
The next step in the solution is ~he B ~v
solving tor the axial loads in all the members. SF 4+D
We will use the ~ethod of joints and consider
zs == - 3770 x .0523--
the structure made up of trIee truss systems zv 3770 x .9986 .0523 FE- .4501 S" = 0- ( i
as illustrated at the top of the next col~. .;966 F& - .36'7'8 SF = 0- ( i
namely, a front li!t truss, a rear lift truss
and a drag truss. The beams are co~on to ~oth
ZD = - .0854 SF + 0, = 0 ----- - ----! , I
Joint (3) (Equations of equt Lrbr-i ua) points (2) and (4). In the design of the beam
and fittings at this point, the effect of ~he
(drag truss actual conditions of eccentricity should of
t 1838
~
1 83 8 reaction on
RL.-J(3l (3) ---D, ptn (3\)
course be considered.
~ RB
R V-Splane SR V-Dplane Combined Loads on Drag Truss
:V = 1638 x .9986 - .0523 RB - .4486 SR = 0 -(4) Adding the two load systems of Figs. A2.38
:5 = -1838 x .0523 - .9986 RB -.8930 SR = 0 -(5) and A2.39, the total drag truss loading is ob-
ZD = D, + 0 = 0 - - - - - - - - - - - - - - -(6) tained as shown in Fig. A2.40. The resulting
member axial stresses are then solved for by the
Solving equations 4, 5 and 6, we obtain method of index stresses (Art. AZ.9). The
RB = - 4189 lb. (compression) values are indicated on the truss diagram. It
SR = 4579 lb. itension) is customary to ~ake one of the fittings attach-
03 = 0 ing wing to fuselage incapable of tranSferring
drag reaction to fUselage, so that the entire
Fig. A2.38 shows the reactions of the lift drag reaction from wing panel on fuselage 1s
struts on the drag truss at Joints (1) and (3) definitely ccnrtnec to one paint. In this ex-
as found above. ample paint (2) has been assumed as point where
drag is resisted. Those drag wires which would
,J><1Xt~X
be in compreSSion are assumed out of action.
~~3G---f1-36~--i
39.5 37.5 58.5
118.5 231 225 281.5 254
Fig. A2.38 798 1191 114\ I j 4189 j
4579
4205
2055
7520
"8
0
-1110 -
8290
4090
.00
The wing bea~ ~ue to the distribu~ed l~~t Lengths k Directional Components of Cabane Struts
al~ loads acting upon them, are also s~bj~cted
to bending loads in add! t t on to the axf.a ; Loads . Member Sym. V D I s L VIL DIL 'IL
The wing be~s thus act as bea~-cJll~s. ~he Front Cabane CF 30 10 12 7 1, 4 1 . 59 1 .721 .240 .648
sUbject of beaT.-col~lh~ action is treated in Diagonal Caba,ne CD 30 30 i27150.17 .597 .597 .538
another chapter of this book. Strl.\t
C~ D1 ~D
/L:.. S
CR C
ZV - 1650 - .7)1 CR = a
(1)(3)
whence
Fig. A2. 41
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A2.19
rig. A2.43 shows the V loads on the center front 1157 568 568
beaw and the resulting V component of the cabane
reaction at joint (7).
:;:,
Solution of force syste~ at Joint 7
-17308 -1730B -1502
~
2535
(11)
CFB 7 Fig. A2.45
Reactions
-~I-/1\I~ As a c~eck on the ~ork the ~~selage re-
13m.~/L\J~7
actions will be checked against t:'e applied
loadS. ~ajle A2.3 ~i7es the '1, D a~d S com-
~on2nts of t~e ~~sela~9 rec~t10~S.
t I 122IT 2.rr- (211
190a' n,
26341 Fig. A2.. 44
'R,
2634#
1190a
I !
,-
A2.20 EQUILIBRIUM OF FORCE SYSTEMS. TRUSS STRUCTURES.
Table A2.3
36
Poiut lIeaber c=, V D S •
sl
9 Protlt 5tr\,\t
C.
_105
-" -'5 _68
,. Rear Strut CR
D1a. Strut CD
2260
4368
1650
2610
335
_2610
1510
2356 72
I, 24
e Front
S.
L1~t Strut 9333 4205 798 8290 i
E
e Rear Lift Strut 4579 2055 o 4090
Total.
'. 10444 _1502 16178
F
G
Applied Air Loads
Fig. A2.46
V component = 7523 (oute::- ~anel) + 65 x 45
= 10448 (check)
D component = - 1110 (outer pane L) 65 x -
6 -- 1500 (er::-or 2 lb.)
\1
,
T
/
I
-
\
'\
'c!
/
l.s
II;';
*'1
"'
I
I~Ac,
)'"1/ •
have ZS = 16178. From Fig. A2.45 the load in
L V / \/ (Q~
the front beam at b of airplane equals - 17308
and 568 for rear beam. The horizontal component ABC
ABC
D
D
E
E
,=
~
at the diagonal drag strut at joints 11 equals
216 x 45/57.6 = 169 lb.
,.g 'JLI!
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A2.21
my :: (6 + 66 + 72 + 72 + 36 )36 - 36 Oz :: 0
To check the equi,mlence of the derived F1g. A2.49 shows a d1a~ of this spar
joint load system wl~h ~he original air load with its joint external loading. The axial
system, the magnitude and ~oments of each loads oroduced by this loading are written on
system must be the same. Addi~g up the total the trUss members. (The S~Jdent should check
joint loads as shown in Fig. A2.46 gives a total these ~ember loads.)
of 756 lb. which checks the original air load.
The moment of the total air load about an x 72 144
, 144 ,
24
~i~-"23,,,3,,6"""==-i-'-""7<?"'---;-:='-:n,,",C"'""",,--"--i
~xis at left end of structure equals 756 x 42 ::
t
-11841 -1184 0 I
31752 in. lb. The ~oment of the joint load 2336
':» I "0 i~
system in Fig. A2.46 equals (66 x 12) + (72 x <\"~ :.?o~~
36) + (72 x 60) + (56 x 84) + l?4 (24 + 48) +
, IN
(120 x 72) + (24 x 84) = 31752 in. lb. or a BOO BOO 224 224 32
check. The moment of the total air load about
line AE equals 756 x 12 = 9072 in. lb. The
Fig. A2.49
moment of the distributed joint loads equals TRI.ANGUUR TRUSS SYSTE:1
(6 + 66 + 72 + 72 + 36)36 = 9072 or a check.
The load system along t~e trailing edge as
Calculation of Reactions causes stresses in both the spar truss ~~d the
diagcr~l truss system. ~he support fitting at
The structure is supported by Single pin paint a provides a reaction in the Z direction
fittings at pOi~ts A, Nand 0, with pin axes but no reacting ~oment about the x axis. Since
parallel to x axis. It will be assumed that the loads on the trailing edge 11e on a y axis
the fitting at N takes off the spar load in throu~~ 0, it is obvious that all these loads
Z direction. Fig. A2.46 shows the reactions flow to ?oint O. Since the bending strength of
Oy, 0z, Ay, Ny, Nz• To find Oz take moments the tra~llng edge member ~s negligible, the
about y axis along spar AEFN.
A2 22 EQUILIBRIUM OF FORCE SYSTEMS TRUSS STRUCTURES
load of 36 lb. at Jol~t S in order to be trans- reaction on the s;ar but does produce ~ couple
fer~ed to pOint 0 through the dlago~al truss force on the spar <~ the v dir~ctlcn ~r.lch ~ro
system must follow the path SDRC~PAO. In like duces compression in the ~0P chord of t~e s~ar
~nner the load of 72 at R to reach a must take truss and tension ~n the bottom c~ord.
the path RC~~PAO, etc.
Consider Joint R
Calculation of Loads in Dlaeonal Tr~ss Members:-
The load to be transferred to truss ~CJR
Member , y x L z/L y/L x/L is equal to the 72 lb. at R ~lus the 36 1J. at
S wn.tcb CCII'.es to jot nt R from tr-cs s DRG.
All Diagonal 4.5 12 36 38.2 .118 .314 .943
Truss Members Hence load ~n Re = (~2 + 36)0.5 x (1/.118)
4.5 0 36 37.5 .120 a .960
457 lb.
AO, NO
Tdhence RJ = 457, C~ = 457 and JQ =- 457
Consider Joint S Joint Q
The truss SEF carL~ot assist in
trla~Jlar
any portion of the 36 lb. lead at Load to be transferred to truss ~BL = 72 ~
transterrlr~
S because the reaction of this truss at SF 72 + 36 =
180 lb.
would put torsion on the spar and the spar has Hence load in ~ = (180 x G.5)(1/.118) =
no appreciable torsional resistance. - 762.
X~~:-
as a free body and writing
the equilibrium equations; Joint P
whence Tz =
0, which ~eans the diagonal Likewise at Joint J, Ty = 287
truss produces no Z reaction or shear load on
spar truss at D. At Joint s:-
ZFy =- .314 x 159 - .314 x 159 - Ty =0 Ty =- (762 + 762) .314 =- 479
, T
whence Ty =- 100 lb. Likewise at Joint c, y =479
If joint G is investigated in the same manner, At Joint A:-
the results will show that Tz a and Ty 100.= = m
'y =- (1040 x .314) =- 326
The results at joint D shows that the rear Likewise at Joint N, Ty = 326
diagor~l truss system produces no shear lead
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES
A2.23
Fig.A2.51
McDonnell Aircraft
(Military Airplane)
Douglas DC-8
Jet Airliner
Piper-Apache
Piper Tri-Pacer
Navy F4-J
North American Aviation Co.
J6" !
wheel and the other with a gear involving two ;0 cJi:~~ <;
wheels.
. "
Example Problem 13
> ED~'b
-r-E = - 10063 x 14 + 6500 x 8 + 11109 x 8
,, I"u
I ...
,,>'
=- 140882 + 52000 + 88882 O(check) =
24" 1 <> = 5690 x 8 - 433 x 8 - 3004 x 14
I I 31 = 45520 - 3464 - 42056 = 0 (cheek)
~ _ ·C--CGO=31 a e G
I "I"'- •
The next step in the solution will be the
26" I L. calculation of the forces on the oleo strut
unit. Fig. A2.58 shows a free body at the oleo-
lel-3u9* 0
strut-axle unit. The brace members EI and CG
are two force members due to the pin at each
end, and thus magnitude is the only unknown re-
action characteristic at points Band C. The
fitting at point E between the oleo strut and
the top cross member FH is designed in such a
TE=24952 manner as to resist torsional moments about the
Fv
I--12"-
I By oleo strut axis and to provide 0, V and S force
reactions but no moment reactions about 0 and S
'.JI
FS-S :E~s=3920LI~- axes. The unknowns are therefore BI, CG, Es,
~ " •
ro"rEV=13332
14"--l...f
:;:@-BIS=3920
BIV=7840
EV, ED and TE or a total of 6 and therefore
statically determinate. The torsional moment
TE 1s represented in Fig. A2.S8 by a vector
with a double arrow. The vector direction
represents the moment axis and the sense of
FS--S
to
1@)_3920
TE 39,20. HO
----~
rotation of the moment is given by the right
hand rule, namely, with the thumb of the right
F. H
hand painting in the same direction as the
arrows, the curled fingers give the sense of
i . E~.2571 I rotation.
~. +- 16" ---t Fig.A2.59
T~ fInd the resisting torsional moment T~
take moments about V axis through E. .....
To find Fv, take moments about a D axis
through point H. ZME(V) = - 3119 x 8 + TE = 0, hence TE
= 24952 in. lb.
l:l'lH(D) = 16 14672 x 8 - 22 FV = 0
Gv +
= 16 x 6500 + 14672 x 8 - 22 Fv = 0 To find CG take moments about Saxis
tnr ougn E.
:~ence FV = 10063 lb. with sense as as-
sumed. ZME(S) ~ 3119 x 50 - (24/31.8) CG x 3 -
24 (21/31.8) CG ~ 0
To find Ho, take moments about V axis Tdhence, CG = 8610 lb.
tihr-ough F.
ZMF(V) =- 6 On - 22 Hn + 3119 x 14 = 0 This checks the value previously obtained
when the reaction at G was found to be 8610.
- - 6 x 5690 - 22 HD + 3119 x 14
=0 The 0 and V components or co thus equal,
~bence, HD = 433 lb. CGD = 8610 (21/31.8) ~ 5690 lb.
-~"'-~,~.~
.~'-"
A2.28 EQUILIBRIUM OF FORCE SYSTEMS, TRUSS STRUCTURES.
To find Eo take ZD =0
ZD = 5690 - 3119 - ED = 0, hence ED = 2571
To find EV take LV = 0
LV = - EV + 14672 - 7840 + 6500 : 0, hence
Ev = 13332 lb.
Fig. AZ.59 shows a tree body of the top
member F1i. The unknowns are Fy, FO, FS' Hyand
HD. The loads or reactions as found tram the
analysiS of the oleo strut unit are also re- Lower
Cylinder
corded on the figure. The equations of
equilibrium for this tree bOdy are:- Fig.A2.60
Torque
(l )~;~O::::iAxle
1:3 =0 3920 + 3920 + FS = 0, or FS =0 LinJ< ~
Whence, HV =
11110 lb. This check value
obtained previously, and therefore is a check
on our work.
The oleo strut consists of ~NO telescoping An airplane level landing condition with
tubes and some means ~ust be provided to trans- unsymmetrical wheel loadii~ has been assumed as
mit torslor.al moment becNeen the two tubes and shown in Fig. A2.6l.
still permit the lower cylinder to move upward
into the upper cylinder. The most cammon way 80LUTION
of providing this torque transfer 1s to use a
double-cantilever-nut cracker type ot structure. The gear is attached to supporting struc-
Fig. AZ.60 illustrates ~ow such a torque length ture at points A, Band C. The reactions at
could be applied to the oleo strut in our these pOints will be calculated :irst, treating
prOblem. the entire gear as a free body. Fig. A2.62
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A2.29
"
zv =- 78070 + 60000 + 40000 + 66666 - VA
~oDo Fig. A2. 61
=a
whence, Ay = 88596 lb.
I tV
L. 5
To find BD take moments about Y axis through
-0
po trrt (A).
36"
ZMA(V) =57142 x 19 - 15000 x 9 - 10000 x
15000
Whee11 __
I 29-38BD a=
=
whence, So 17386 lb.
10000 I
ZD = 0
Wheel
40000 To find An take
6000
ZD =-57142 + 15000 + 10000 + 17386 +
AD = a
whence An 14756 lb. =
To check the results take moments about Y
and D axes through pOint o.
ZMo(V) =5 x 10000 + 14756 x 19 - 17386 x
19 = a (check)
&~O(D) = 20000 x 10 - 88596 x 19 + 78070 x
19 =a (check)
ITE
r 17"'-1- i7"'
ITE
-1
E
.:l;G_ - -
Fig.A2.62
EO-
I 00
E _ES-J:-
' r
Yi
~
'\! 9
~
sense as
"0" -l
25000 M'o"
assumed in Fig. A2.62. 100,000 lb. 200,OO@ -, 100,000 lb.
c
t 50000"lb. t50,OOO"lb.
The reaction at C must rave
a line of action along the line ~~,
CD since ~ember CD 1s pinned at 28G the axle centerllnes have been transferred to
each end, thus the crag compon- 24 0 point (0). Thus the total V load at (0) equals
ent and the load in the strut 60000 + 40000 = 100000 and the total D load
2D tollow as a matter of geometry. Hence, equals 15000 + 10000 = 25000. The moment of
=
Co ~ 66666 (24/25) 57142 lb. these forces about V and D axes through (0) are
CD = 66666 (36.93/28) =
87900 lb. tension Mo(V) = (15000 - 10000) 10 50000 in.lb. and =
Mo(D) = (60000 - 40000) 10 200000 in.lb. =
To find BV take moments about a drag axis These ~oments are indicated in Fig. A2.63 by
through point (A). the vectors with double arrows. The sense of
the ~cment 1s determined by the right hand thumb
A2.30 EQUILIBRIUM OF FORCE SYSTEMS. TRUSS STRUCTURES.
*
Fig. AZ.S4 shows a tree body at member AB ~-10 I I (6)
~
10tr: (5)
with the known applied forces as found from I A (4)
A ; t B
the previous reactions on the oleo strut. A 10 10
A
'*'
B
,
1000*
I
...J. jl0 r O
t10 (16)
15"~'
T 30"
10
12 4 (
24
124"1
....J..
36" -+- 36" -+- 36" ~ 45"------1
B B'
1'000 30 dihedr
~'~:£::
CF
CD
CR
...L _800 10 .i- D,D'
I l- I 1 30'....
I
1-30" -+
2000 (IB) I . I I
15"+15'..-.l
:~tL Fuselage
% A.A'
Flg.21
(4) Determine the axial loads in the :nembers
of the structure in Fig. 19. The member-a (7 )
are pinned to supports at A, Band C.
I-- 5 Bays @ 30" --.j
rl C D
T ~-DragTrUSSLS
3I
At t t t l t t t t t f t t tt t 25 lb/in.
Pin Reactions
v
t D
" '" B "'0.
C
~"'.
B
at A, B, C '-D 60" Fuse-
.L lage
~v."'~
::,.... '5
fL
t.-s
L:
I
D
(8 ) (9 )
The r i tt tng at In the
6"
,
....,.-_-,A B C
20"
points A and B ~or
the Landfng gear
structure in ?ig. 23
prOVides resistance
E F
t
L22"----';
C F
landing gear-
of :'1.>;.
the br-ace
member's
2~,
BC
I to VJ D and S :-8- and SF are
24" actions and moments B f,'IW r or-ce
-13000 v member's • The
lb. about D and V axes. '0" LD .::'itt::-:1g at E:
Fig. 23 Find the reactions at
Wheeo wheel wheel
~
CD~
10000 A and a and the load cr-evices ::-e-
10000 lb.
in member CD for . .J500' sistance tc
333 A V, 0 ar:.d S
given wheel loading. wheel 2
"113'
res.c1:1.ons but
15000 10000 only momentFig. 24
resistance
about 1J axis.
Find reactions at ~ and loads in :::embers SF and
BC under given wheel loading.
y TABLE 1
I .c:": Section Properties of Areas
d --1- x ....,
I >dA Rectangle Area = bd
fe. g. Fig. A3.2
Y =
d
2
I I1._1. = bd 3 / 12
I _
a a = Oct'
2dA.
3"
The first term,! x represents the ac-
ment of inertia of the body about its centroldal f ,-, = .289 d
axis y-y and will be given the symbol I. The f------------:.-------------l
second term 1s zero because~xdA 1s zero since Area = bh
TJ 1s the centroldal axis of the body. The 2
last tern, d2~dA = Ad 2
or, area of body tl~es Y = h/3
the square of the distance between axes yy and
,~3
11._1. = bh 3
Y1.Y1." ~ Triangle "V"
vO
Thus in general,
I=I+Ad a
1 ---r- - - I 1,;1_a = bh 3
This expression states that the amount ot 12
inertia of an area with respect to any axis in 2 1 _ 'l 2
3 =
bh'
the plane of the area 1s equal to the moment ot
inertia ot the area with respect to a parallel
3
f---- ---i
"4
centroldal axis plus the product of the area and = .236 h f ,-,
the square ot the distance bec#sen the two axes.r-----------------~-------------------------_1
Area = bh
Parallel Axis Theorem For Masses. If·instead of '2
area the mass at the body is considered, the
parallel axis can be written: x = 2/3 h
h~
I = I + Md a , where M reters to the mass at 11.-1. = 4,8
.3
the body.
12 W
20
deb +a)3 _
= (dt-t') (b'-bt)
zaec -
Ix - I y
and I:a- II see equa-
given for angle.
6ab:ae
Tl1
l::
a
~al1
x Circular
Fillet
Area ::: .215 a a
X ::: .223 a
~
u-sectton Xl ..L
-jtJI- a~r.o-
Area A ::: 2dt 1 + at
-
-r-
.x ==rr I ,>,
r:n.n
lIT:±[b1
¥
y ::: d a t 1 ~ .5t a a
A
~
1
I'
'I_it- Parabolic
Fillet
x ::: 3
4:
y = .3 Y.
Xl
r-b~=r 3
Area = nr'
Circle
= nr'
2--;-~-2 1 1- 1
T y ::: 2 y.
" 'i
.... -; la_Ii = 5 n r-
l\~l 4: TABLE 2
e1-l = r- Properties of SolidS
2
Ring Area ::: n (roll - rill) Solid Cire. Cyl.
Vol. - n r-'l-L,. (r - radius)
~ M = W (Total wt.)
.. '/~
',I
1 ~ ~ -----:..-.l- 1
"
.~
ro '
Jo. l .'
11. -1 ::: n (ro" - rl")
4:
,
)5:5'
'" .-
~"
'\ .'
11 -1
s
= ~.r.a
"2
I,~ = M [r a + (L f31])'4
Area _ n r "
Serm-crrcfe 2 Hollow Circ. Cyl. ~ Vol. =- IT L{r1 a - ra~
~ y::: .424r
\~ . 11. -1 ::: M(r1 a + r aa)/2
~~
4-
II' j (r1 :J -ra 3 )
xf'x =a r 1 ,. radius of sphere
~~
'2 d1a.
=
2Mr, '
"""""3""
13.4 CENTROIDS, CENTER OF GRAVITY, MOMENTS OF INERTIA
TABLE 2 - Co~t~nued
I -- r-e t ~-
sIn 20
- (l-cosa 0) •i
NAx 2 4
Ring with Circular
Section Y M= w = mass per unit volume
J;tf'x y
g of body.
1:0: =mnaRaa[R' + (Sa a/4)J
a
I yy =! ml1 a Raa (4R + 3a )
2
1I
TABLE 4
Centroids ot ':'rap1zaidal Areas
Lc1JT
b I J.
TABLE 3 ll-y -!-x I-
Section Propert1es of Lines
Distal1Ce x Distance y
(t 1s small in comparison to radius) Ratio a
b
0.40992 0.5008
Circular Arc Area = 2 11 rt 1.01
0.4098-4. 0.5016
1.02
1. 03 0.40976 0.50240
11._1. = 11 rat 1.04 0.4988 0.5032
1.05 0.40960 0.5040
I
pol ar = 2 11 r"t 1 06 0.40952 I 0.5048
0.5056
1.07 0.40944
1.08 0.40936 0.5064
E'x = .7(f7r 1.09 0.4928 0.5072
1.10 0.4920 0.5080
E' polar = r- loll 0.40912
0.40905
0.5088
0.5095
1. 12
1.13 0.40898 0.5102
seeu-ctrcte Arc A = n rt 1.140 0.4890 0.5ll0
1.15 0.4883 0.5ll7
y = .6366 r 0.4877 0.5123
2~2
1.16
1.17 0.40870 0.5130
= 11 rat 1.18 0.4862 0.5138
I.::L-__L.LI -2-
t:~~ o:;~~
0.5145
~r4-.--'-1
t l 1.3< 0.40758 0.5242
1.36 0.4746 0.5254
y = .6366 r 1.38 0.40733 0.5267
1---;2 1.<0 o 4721 0.5279
= 11 rat 1.405 0.4693 0.5301
-4- 1, 50 0.4667 0.5333
I 1 • .5.5 0.4641 0.5359
2 I 1. 80 0.4616 0.5384
I._iii = .149 r"t 1.6.5 0.40.592 0.5408
I
1. 70 0.4568 0.5432
1.75 0.4545 0.5455
1.80 0.4523 0.5477
CIRCULAR ARC 1." 0.4502 0.5498
1.90 0.4482 0.51H8
1.95 0.4462 0.5538
2 00 0.4443 0.55.57
2.10 0.4409 0.5591
2.20 0.40375 0.5625
2.30 0.4343 0.5657
2.40 0.4312 0.5688
0.4284 0.5716
;.~~ o 4259 0.5741
2.70 0.40233 0.5767
2.80
2.90
3.00
0.4209
0.4188
0.4168
- 0.5791
0.5812
0.5832
3.20 0.4128 0.5872
J 40 0.4090 0.5910
~-
3..60 0.4060
J.80 0.4030 0.5970
Area = art a in Radians 4.00
4.20
0.4000
0.3975
0.6000
0.6025
0.3950 0.6050
i =r sin a
4.40
4.60 0.3928 , 0.6072
a ' 4.'~
'.00 ~:~:; ~~~~
6:0.6152
5.50 0.3848
(a + sin 2(1) ~. ~810 0.6190
2 ~.~ o 778 0.6222
7.00 0.3750 0.6250
7.50 0.3725 0.8275
8.00 0.3702 0.6298
9.00 0.3668 0.6332
10.00 0.3636 0.63640
Y =r (1-008
c 0) ,l1xx=.W- = rlilt (j-cos a)
- r"t (0
I xx - sin 2(1)
2 - 2
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A3.5
Table"
HorizOIItal Vertical
I"
where Ix, I y, and I z are generally referred to
Totals 3150
as the rolling, pitching and yawing moments of tia about the horizontal centroidal axis for the
inertia of the airplane. area shawn 1n Fig. A3.4
w = weight ot the items in the airplane Solution. we rrrst :C1nd t he moment of inertia
x, y and z equal the distances from the about a horizontal reference axis. In this so-
axes thru the center of gravity ot the airplane lution, this arbitrary axis has been taken ~s
and the weights w. The last term in each equa- axis x'x' thru the baae as shown. Having this
tion is the summation ot the moments of inertia moment ot inertia, a transfer to the centraidal
of the various items about their own X, Y and Z axis can be made. Table 5 gives the detailed
centroidal axes. calculations for the moment ot inertia about
axis x'x'. For simpliCity, the cross-section
It w is expressed in pounds and the distan-has been divided into the tive parts, namely, A,
ces in inches, the moment or inertia is express-
ed in units of pound-inches squared, which can B,C,D, and E~
be converted into slug feet squared by multiply- lex 1s moment at inertia about centroidal
x axis at the particular part being conSidered.
ing by 1/32.16 x 144. Distance tram axis x'x' to centroidal horizontal
Example Problem 1. Determine the gross weight Y
axis = = :Av = 17.97 = 2.9l w
center ot gravity of the airplane shown in Fig. ZA 6.182
A3.3. The airplane weight has been broken down
mt c- the 10 items or weight groups, with their BY parallel axis theorem, we transter the
individual e.g. locations denoted by the symbol moment at inertia from axis x'x' to centroidal
+. axis xx,
Solution. The airplane center at gravity will be Ixx=Ix'x' -Ajlt=79.47-6.18x2.9P.=
located with respect to ~NO rectangular axes. In
27.2 in"
this example, a vertical axis t~~ the center-
line at the propeller will be selected as a ref-
erence axis tor horizontal distances, and the
Radius of Gyration, Pxx = xx =lj27 .2 = 2.1" VI
thrust line as a reference axis for vertical dis- A 6.18
tances. The general expressions to be solved
are:-
x = Zwx
distance to airplane cvg , from
ref. axis -S-i5
Zw
y = Zwy distance to airplane c vg , from Centroidal
ZW ref. axis X-X
Table 4 gives the necessary calculations,
whenee
x = 417180 = 133.3" aft ott propeller
3150 Fig. A3.4 Fig. A3.5
Example Problem #3. Determine the moment at in-
y 5480 -1.74" (below thrust line) ertia of the stringer cross section shown in
3150 Fig. A3.5 about the horizontal centroidal axis.
11)\ I
Solution. A horizontal reference axis x~xP is
assumed as shown. The moment of inertia is
first calculated about this axis and then
.......,F=""-,---- - - - transferred to the centroidal axis xx. See
7 • Table 6.
Fig. A3.3
, Ref. line
Z
.3 • ,
CENTROIDS , CENTER OF GRAVITY , MOMENTS OF INERTIA
·-" •·•"
e
•
•
- ,
• .
", • _.ent of inertia of
portlon about centroldal
lI: uia.
•,
! ,
I
I appr ox ,
.~72 = .547"
y = .375 + 2 :OJ. = .375 + .172 =547"
A 0.'" 2.7li 15.13 1/12 x .5 • 13 .. .0< 15.17 n
Al 0.50
'"
a. , 2.75 15,1:3 .. . .0< 15.17 lex= [.1098(.0841+ .0625) + .04x .54-
,• 1.00 '.25 5.2S 27.50 1/12 x 2 z .53 • .ca 27.52 .283x .0841 x .0625 x .04] = .002375
'0 2.12 2.875 6.10 17,56 1(12 x .5 x 4.25 3 • 3.19 20.75 .54
0 0.281 1.00 0.28 0.28 bb 3/36 .- .75 4(36.0.01 0.29 appr-ox , lex = .s s .04x .27"= .002365 In~
D1 0.281 1.00 0.28 0.28 .. .. 0.01 0.29
Problem #4. Determine the ::J,oment of inertia of
• 1.50 0.375 0.56 0.21 1/12%2x .7:i 3 • 0.07 0.28
the flywheel In Fig. A3.5a about axis of rotatlc n.
i5 \1116 . 182 17,97
I 179.47111.4
The material 15 aluminum alley :::ast1ng (weight =
• 1 lb • per cu. Inch. )
Table 6 AI Sec. AA
PartiOD Area , A, A,' lex Ix 'x' .. lex... Ay2 V/// / ~-------,
r/
u~
I + l' .0. 0 0 o .00281 .00281
, .06~4 .706 .04617 .0326 .00168 .03428
f;i
I l,.,.,f/ ""'~
3 + 3' .0679 _.547 _.0371 .0203 .000475 .02078
I cb •.. 5" i 1" I I
I
2" 10"
w~
s~ .1933 .009 • O~787
\ T I I I
,
I
I
y = l'.Av = .009 = .0465" -'--
l'.A :T933 r/////il 1" i
. -i--'
.
1= = lx~x'
_ Aj3
= .05787 - .1933 X IA Fig. A3.5a
:--3'l
( .0465)' = .05745 in~
Solution: ~he spokes ~y be treated as slender
Radius or gyration i.ox-x
= V· 05745 = .55"
.1933
round rods and the r-rn and hub as hollow cyl Ind-.
ers , (Refer to Table 2)
Detailed exptanat i on of Table 6:- Rim,
Portion 1 ... l' weight ," == n(5"'-4.l1)
n(R 3_r a)x3x
.1
~2' 75')
bd' .3=8.48 lb.
lex = 1 x 2
= (-04 ) = .5x8.48 (5
2"'r 2 2
12 I = .5W(R "' 4 2 ) =
173.7 lb. In'
= .00281 in- •
Hub,
Portion 2 (ref. table 1) = n(la_ .5 2 ) x 3 x .1= .707#
y = .375 + 4 Rlil ... Rr -+ r-« = .375 + 4 'I" = .5x.707 (l a .... 5:1 .44 lb. in~r=
3n R+r 3n Spokes,
(.5 3
.... 5x .54 .... 54 2 ) = .375 .. 331 = .706" Length of spoke = 3"
( .0 + .54) '""eight of spoke = 3 x .25 2 X .1 =
.588#
By approx. method see Table 1. I or one spoke = ',.rL:a/12+·>"j'r~ .588 x
y = .375 + g rl, = .375 + z x (.52) = .706" 3'1/12+ .588x2.S a =
n 4.10 lb. 1n 3
rt
I for 4 spOkes = 4x 4.10 = 16.40
Icx = .10ge x (R a + r-") t (R + r) - .283 Ra + r.llt
Total I of wheel = 16.40 + ;44, + 173.7 = 190.54
R + r
10 • In'
• . 1098(.2916+ .25)x .04x1.04- I = 190.54/32.16 x 144- = .0411 slug ft~
.283 x .2916 x .25 x .04 ~ample Problem #5.Moment or Inertia of an
1.04 Airplane.
Approx.
= .002475 - .000793 = .001682 in ~.
lex = .3trl," = .3 x .04 x.52 -. = .001688 in".
':'0 calculate the :nO:r:lent of inertia or an
airplane about the coordinate 9.xes through the
gross weight cen.ter of graVity, a breai< down or
the air?1ane weight and its distribution 15
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A3.7
necessary, which is available in the weight and axes, the prOduct of i~ertia of the weight about
balance esti=ate of the airplane. the reference axes 1s necessary. Column (14)
gives the values about the reference axes. To
Table 6a shows the complete calc~lation of
transfer the product of inertia to the c.g. axes
t~e moments of inertia of an airplane. This of the airplane, we make use of the parallel
table is reproduced. 'from :i.A.O.A. 'I'ecnnt ca L no te axis theorem. Thus
#575, ~sstimation of ~oments at inertia of air-
planes from Design Data.~ Z'NXZ C• g • -=Z'NXZ(Rer. axes) -Zwxz =48,857,589-
Explanation of Table 5325.3xl15.9x77.8=B39,253 lb. in.a
Fig. A3.5b shows the reference of planes To reduce all values to slug ft.a multiply 1
and axes which were selected. After the moments 32:l7
of inertia have been determined relative to x.J:....
these axes the values about parallel axes 144
through the center af gravity of the airplane Hence Ix -= 3061, I y = 6680, Iz -= 9096, Ixz -= 181
are found by use of the parallel axis theorem.
Having the inertia properties about the co-
Column (1) of Table 6a gives breakdown of air- ordinate c.g. axes, the moments of inertia bout
plane units or items. the principal axes are determined in a ~anner as
Column (2) gives the weight of each item. explained for areas. (See A3.13).
Columns (3), (4) and (5) give the distance or
the c.g. of the items from the references planes The angle 0 between the X and Z axes and the
or axes. principal axes is given by,
Columns (6) and (7) give the first moments of tan 2 0 = 2I xz = 2 x 181 -= .05998 hence >:1 =
the item weights about the Y' and X' reference Iz- Ix 9096-3061
axes. 10 43"
columns (8), (9) &id (10) give the moment of
inertia of the item weights about the ~eference
axes.
Columns (11), (12) and (13) give the moments ot
inertia of each ltem about its own centroida1
axis parallel to the reference axes. Such items
as the fuselage skeleton, wing panels and engine l
.
, '
have relatively large values for their centroid-
al ~oments of inertia.
/ ". " ' \ ~ .
The last values in columns (3) and (5) give
\ \
, '. .
....
the distances from the reference planes to the ~ \
. " , ,
center of gravity of the airplane.
Xc •g . -=Z'~·'X-=617.024 (col. 6) ==115.9 in.
~';~
l
.
Zn 5325.3 (col. 2)
ze ~
'0'
-= ~wz-=
_
414.848 (col. 7) -= 77.8 in.
. \
Zw 5325.3
The last values i~ columns (8), (9) and
(10) were obtal~ed by ~se of the parallel axis i
:
\\ , (
theorem, as follows:-
Zwxa about e.g. of airplane -= 97,391,595 - E325.3 (
x 115.9 2 -= 26,691,595.
, ',0--_.-- .
(
\---:...-,'-- -.---'1-}' ~
The tmru :'rom the last value in columns
(11), (12) and (13) give tr..e .noment.s of inertia
or the airplane about the x, y ar.d z axes \\t-"'; /~ ~~
through the e.g. of airplane. T~e values are \'~ ! .: :
obtained as follows: /
1y =Zwxa + ZWZli + Z<lI y -= 26,691,595 + 999,035 +
3,120,384 -= 30,304,014 lb. in s
Jx = Zwy~ + Z'Nz. 3 + Zl;I x -= 10,287,522 + 992,023 +
L-'-
.
~
\
\
'\
-"'y,
"\' ~ l
:
!
~
,,-
i . ,.
elillt
,.
•
..
Y
S.
• "
••
"
,. .,,'
COMPU!ATIOHS 0' MOMXiTS OF IBERTIA
. ••
"Y' ..'
10 .. 11.
."
12 •
t
lJ •
.,,
H.
m
C.ll~e:r ,eCl't1oD. 106.6 102 - " 11,098 6,20a 1,131,955 - 353,4;;1 361.339
·'
- 251,229 532,563
.....
no,......C1bly
Clln~ll:r oectloQ
• tc .
....
204.6 121
H.
- "
· ee
.
24,751
141.462
11,662
4,631
2.995,549
1.844,317
0 664,745
254,705
491,24.5
aoZ,164
-
33,680
4091,245
235,844
1,411,126
685,388
...... ....
Center .ecUon 0
rlbl, eee .
Flap ".0 ,.0 - .553
lsa
3,960 1,166 71a,8OO - 61,796 48,598 -- 48,596 209,880
Outer p&l1el ~
Outer paDfll
104.6 105
155.5 120 lsa '5
10,983
18.672
6,799
10,114
1,153,215
2,240,640
2,545,546
3,186,682
441,935
657,4ol0
164,514
274,478 0
I 184,514
374.,4078
713.895
1,213,680
Outu p&l1.1 rib' 139 ISS 54 12,462 5,747 1,735,Oae 2,185,373 367,621 158,407 17,601 175,008 798,861
Al1nona 31.4 172 lsa sa 5,401 1,947 928,938 764,150 120,702 55,390 0
55,390 334,850
rUI.lac· oklle- 3H.0 1.,. - .1. 55,4&' .35,434 9,726,484 - 2.060,154 69.394 1.5713,594 1,5'ro,OOO 4,4075,384
to.
Eng1Il..
ruztl.~k
_ _t
( fa.1:rll'11)
40.5
".5 ...
'0 0
-
·ec
.0
.0 '.""
141,319
3,4140
a.aac
145.800
3.139,026
-
0
- 259,200
310,400
5,184
15,181,
5,184
57,861
5,164
56,648
194,400
985,520
f1l'n&ll
Step.
11.0 '0
'.0 170
.0
'0
'70
:l4O
••0
140
53.iOO
51,800 600
70,400
'.800 - .3,200
0
1,100
0
1,100 61,600
23.800
..
C.. bln &lid. 66.5 10. 7,18a l,417,5l4t 0
77S.e56 2,394 106,400 108,794 l,048,S7<l
_1lll11.bi ..111
foo-c tI'O~U
floor, ,,~
'.0
'.S 210
•• 0
1M
1,995
I,.
62'
11,858
418,950 0
SO ·9,248
4l,3S2
0
'0.
0
1.Z69 ·
1.976
10,.72
131,670
.. ·
"'tllg flllOlU 18.:1 142 20 58 .,Sa7 1.073 373.034 '.400 62,234 18,944 18,944 152,366
8o'Cto.. eow11Jl4 27,0 140 11 '5 3,180 2,02:1 529,200 3.267 151,875 0
24,300 24,300 Z8:3, .500
~ 0111e fram••
,,., "'. sa 104,853 5,160
• s 23,250
Al'l'el-C1JWJ ;;1001' 1.3
"
0
-
0 0
"'"
3,302
a,8S8
1.394
11.3
1,040
••3
347,633
49,005
83S.708
1,111.688
0
0
5.se8 114,308
7,144
83,200
62,108
1,088
720
".394
120
28,288
·
15,509
--
av.acc
720
15,509
120
199,342
18.111
z64,l60
::>S2,763
.... .. ••
Co_Unr, ata- 12.0 110 0
se 1,320 1.140 145,200 0
108,300 1.200 1.200 :1...35.400
dOI18 1-3
CIla,e1. (1'_ 332.4 115 Sl. 26.726 11,6:12 3.073,490 e77,678 604.472 0
23,240 23,240 1,383.0<36
.. ....
trected)
Retr..cting "ocb-
""
- "
no as
sa
3,146 1.916
S._
see.ceo 17,875 128,JS5 0
- 0
210.7841
_.e '.-
Wb••l., 91.0 lU 54 U.83l 1,809,171 285.356 385,378 0 0 0
718,536
1'atl _.bee.l ".0 "0 1.924 3.369,600 0
142.376 ee 0
ee 692.640
l:J:1g1n. acc••-
aorl ••
Eng1A. control.
PTOp.ller
1049.0
90.' ,
,.
11.0 1.03
aaa.c
37.0 sa
J3
52
'.3
0
10
0
.0
.3
SO
M,617
1.133
a.065
83,920
'.71.1. '.520
.,.
17.760
1.142,~1
244.9941
118.699
19,201
·-
I,l.OO
·-
6, '7l3.600
ea4,143
83,538
1,420,300
253.658
9,060
-
177,600
a.
3:13,858
·
0
ea.sce
253.858
'.060
0
69,300
..,
2.769,3&0
391.030
88,109
185,168
S'ar1:1~ .ye-c_ .5 .3,0'73 3,145 116,032
- 361,325
'" 116,120
0
,.
Ll,lbr1ca'Uq aye· ae,c
•• 0
sa I._ .3.132 U3,798 l'U.834 1,374 0 1,.374 147,J.08
fuel .,.,to_
InetZ'WIIl!'l::l'.
841.0 12.
38.0 10. ·
0
.0
sa
10,496
3,8'16
'.sao
3.498
1.343,488
395.3541 ·- 534,300
331.6341
.32,008
.3,4341
20,008
SO. 600
35,668
63,233
839,680
356.592
81U'face COA- 81.5 160 0
'1. 13.040 5,187 .3.088,400 0
4010.842 U13,ll84 130,4.QO 254,:562 9415,840
trol.
F\lnl1a.bio.p. 160.0 lsa - SO .34.960 13,800 3,893,160 - 1.0M,OOO 108,160 400.000 508,1150 1,996.600
Slectr1aa.l
.qu1~t
Koht 011c.g
UJU!'L.,t.U
. IKP'n'
100.7 128
---'.0 US
3867 •• 106.6
IS
0
0
SI
10.
---
",""
,go
----81.'
8,157
7i,35O -
1,849,869
82.424
74.8 41.41,199 .2a9,l38 67,342.573 10,283,443 aa,a63.738
32,556 560,593
·
.3,881.572
ea.
0
·
2,890.«14
·
...
4.980,508 31,090,714
l,04.,O!8
70,380
.,....
Pllot 200.0 10' - go 21,000 1.8,000 .3.205,000 1,6ao,OOO 33.600 28.800 '.000 1,890,000
- •• -
0
Ob.e1"W'.r 200.0 205 41,000 17,800 8.405,000 1,584,.300 33,800 38,600 '.000 3.649,000
Fuel
OIl
Vlry P1eto1 , ..
780.0 132
75.0 Yl
195
0
·U
85.S 1041.990
.5
"
5,3a5
751.
",.
66,690 13.590.720
6,375
., 378.075
148,296
0
..
,t ••
;',701,996
541,875
21,938
141,180
3,615
-
172,380
0
157,560
3,675
--
8.603,Oao
452,625
57,038
... -, ."
0
se
.....
3.:)118 eaDd.l.,
.,.
'.0 IS' 14 13:1,424 39.864 0 0
70,6sa
.....,.
Flo.. t l1rttta '.0 190
148.7 l7S
14 72
.5
I, 7l.O
25.«11
,
324,900
12.130 4,5.31,307
1.7&4
0
46,656
1.031,008
0
5.U7 ·
0
·5,13'1 143,120
3,159,051.
ese ·
·- ,as
Cbart board.,
Or1!t 11ght '"
1.6
'0
I,.
23.680
m
J3 32,693 0 0
.37,824
Tlre-C a.1d.
Life :raIt
'.0 I.,
34.0 1.'"
14
10
0
54
101.
'55
"0
4,S.34 3,434
78.854
108,900
628,864 - '00
U • .390
12,996
3046,834 --
0 0
· see
aQ,837
37,540
461,024
U8U1JL LOAO 1457.9 140.5 0 88.7 204.828 1.26.450' 30.549,023 4 ,079 ~88 ,299 311,898 229,980 1'16,878 17,786.87:1
fOULS 5335,31 817,0.24 414,848 97,asl,595 10,.387,saa ~S.2,035 2,899,470 3,120,384 5,157,188 48, 857,58a
COflR.u::,.IO. 53a5.3 1+5.9 0 77.8
,,
r71 200,000 Or~O 000
IE.t\9i,5il5 10.Z67.5aa l~a,oj5
48, 0 1 33t1
a~ ,aS3 1
1
,
, , 10.287,5412 as. 691. see,26,691,595 ,
,,
,
,
,, I ,
99. S'~ 36,804,014 4,136,303
.
i 292 035 1~ 287 52.3
,
,,
," ..
,
I • ,
,, I 1",179, , a ,, ,,
,
,
,
, , I .3,662
,, 51.8 , 76.1
5,784
, 7,912 I
, , ,
J:c.g. a ~--"'~'i" ,,
,I
1'x
e.g. I a.g.J
, : :_~ :'~·---~'3 ..,,< .... '.~y ' ". • ~. "~':;, >.... ,..,..r"_.~"'-~~"_. t, -'): ~",. <:», ·,...:<··: . . . >:';.,~··~~··-~;-"·'-::': ...~·:·~:~i~~ . . .;;.
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A3.9
The principal moments of inertia are given by will be given the symbol Ixy, hence
following equation.
Ixy = I xydA - - - - - - - -(1)
I xp = 11: cos" £I + I z sin" 0 - Jxz sin 2 £I. (see The unit, like that of moment of inertia, 1s ex-
Art. A3.lll pressed as inches or feet to the 4th power.
Since x and y may be either positive or negative,
I yp = I y
the term Ixy may be zero or either positive or
I z = Ix sin" 0 + I z cos" .0 + I xz sin 2 .0 negative.
SUbstituting ~oduct of Inertia at a Solid~ The product or
inertia or a solid is the sum or the prOducts
Ixp = 3061 x (0.9996)" + 9096 x (0.0300)' - 181 x obtained by multiplying the weight or each small
.0599 = 3056 portion in which it may be assumed to be diVided
Izp = 3061 x (0.0300)' + 9096 x (0.9996" + 181 x by the product ot its distances trom two or the
. 0599 = 9102 three coordinate planes through a given point •
Iyp =6680 Thus With respect to planes X and Y
Ixy = I xy dW
A,3.7b Problems
Jxz = I xz dW
n r Zi ""i I yz =I yz dW
Z
1
S
T I
' 1-'"
1<.'-1"
+ uct or inertia at the area with respect to a pair
at parallel centrold.al axes plus the product at
the area and the distances at the centroid at the
Fig. A3.9 total area tram the given pair.ot axes", Or, ex-
A3.S pressed as an equation,
Ixy F Ixy + AiY- - - - - - - - - -(2)
(3) Dete~ine the moment or inertia about This equation is readily derivable by re-
the horizontal centroidal axis tor the section terring to Fig. A3.l0. yy and XX are centroidal
shown in Fig. A3.8. axes for a given area. YY and XX are par-a.lLeL
(4) In the beam cross-section at Fig. A3.9 axes passing through point O.
assume that the tour corner members are the only The product or inertia about axes YY and XX
effective material. Calculate the centroidal is I XY = I(x + x)(y ~ Y) dA
moments of inertia about the vertical an~ hori- = IxydA + 'is I dA + x I y dA + Y I xdA
zontal axes.
The last two integralS are each equal to
A3..8 Product of Inertia zero, since !ydA and IXdA rerer to centroldal
axes. Hence, I xy = IXYdA + i j IdA, which can be
In various engineering problema, particu- 'Nritten in the tarm or equation (2).
larly those involving the calculation of the
~oments ot inertia at unsymmetrical sections,
the expression I xy dA is used. This expression
is referred to as the product of inertia at the
area with respect to the rectap~~lar axes x and
y. The term, product ot inertia of an area,
A3.10 CENTROIDS, CENTER OF GRAVITY, MOMENTS OF INERTIA
In Fig. A3.ll
·_-+-x I X1. Y:l, =!X1Yl dA=!(xcOs~+:;sin,2j)(ycos.z;
xsln0)dA
= (cos s 0 - atn- ~ / xyG.A'" cos Zl sin 0
y Fig. A3.10 (y:a..x a ) d.A
Port
.. ,
'ox '0, ~
A~' 'x
Ax_ Iy !
A x r A, Ax A" Ax'
1 .3705
." ,1" ,0469 ,281 ,OO~8 ,211 ,03~1 fi:z 1.5U~3 e , 0019 1 Jl:1xl 53 •• 070 ,0077 ,281 !
.
12 4 '
a .500 ,1205 1.205 ,625 .06205 ,7800 ,0078 .0781 1 1 3
12 :z,,:Z2 ,167 ~2X2J:.2S3 •• 0026 ,9470 ,010
i
,6 19 .J.oli35 ,7858 .2188 ,1132 ,95.1 .291
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A3 11
-- -
"i
5,
I
6' Ti:!--f~
r f XRe f.
lows: - x
.---=-- - ---L... -,-. x
~~I
0
,!
tan 2 0~2 Ix" ~2(-.150 =-.30 =1.06 x
12 13
I y _ Ix .157- .440 - .283 .r ~4 ,
15 I~
T ,
2 0 ~ 46 0 - 40' 0 = 23 0 20'
-
I I
Calculate moments or inertia about centroidal
Fig. A3.13
I. I
principal axes as follows:- Yp Y y
x .918 m'
·,• ..• :::':,·..., '"" I.....
~
.. . '
.3965 504 Area
" I
-
I Y1l ~
-
Ix sin 2 .0 + -I y cos 2 0 + 2I- xy sin 0 cos el , ", • lx,
"
~f.
"<
o~
<
" ~.,
,••r ." .,
q:-:'!- •o-t • 17 7.37 1.253 9.234 -21.18 _3.600 18.205 _26.53
T' ? t~
:.ra .~
.0< .U
.>0
1.50
~.288 9.68 I-~~' 60
1. 215 9.563 -12.60
-
_2.8::
_2.142 ~:::: :i~: ~~
I fL a .ra
.0<
· 0<
.U
.U
1.30
6.90
1.241 9.0:19 8.60
1.111 8. ceo _ 4.00
_1.462
_0.680
12.51 _10.61
2.12 _ 4.69
." ."
." ."
s .J<
~t
2f ,\1' ';""-"
,"'~i
6.50 1. 885 12.252 - 0.35 _0.100 0.04 -00.65
io .U -a.nc _0.561 1. 8.51 _33.2:1 _5.6:13 181.96 111.611
U .ur
." .U -4.90 _0.83 4.0112 _29.28 _4.918 1411.16 24.39
ra .U ." ."
ra .La _5.911 _1.6661 9.913 -24.8.5 _6.9581112.90 41. 40
, ~-~; ""ie .n
." .n
.n
.30
_1.40
_8.13
_2.2941 16.916 _LS.10
_2. S2~ 20.481 _12.42
_1l.191 108.40 42.90
_2.850 4.1.82 31.30
I-L -
•
'+-,(
, ...L
.n
.J<
ata s
.n
.n ."
a.
.30
-
_8.62
-8.81
...
_2.612 22.432 _ 6.10
-3.094 21.444 -00.35
.0<
_1. 891
_0.122
-,
11.54 16. ;10
• .,
O. il4 1. 08
- a. s
-I y p = -Ix e-
s In e 0+ 'y c os e el+2 I xy s t n 0 A3.14 Section Properties of Typical Aircraft Structural
Sections.
cos 0= 186.46 X .:~822 + :;)1.7 X .2695 2 + Table A3. :0 'thr-ough A3.15 and Char-t ......' 0 ; v.~
2 -36.41 x ,9::95 x (-,1"32) =437 in ... si'le the section pr-oper t t es of a few st:-uct.. . ral
shapes G8rrnOn to arrcra rt . Use of tr-ees tab Ie s
will 08 :TIade in later cha;Jters of th:!.s jook.
, ,
f--e-J ,~
n"TI
---.-.z~1
'I
I R
£=T ,il
-'-T
TaDI• ..1.3. 10 TB I I 'I
- r+ ~T Table ..1.3.11
x-' 'i
X PfDlI"rlies of
zee Secl'Q/\S
X ! I !
!
X
Properties ot
Channel ~ectlons
, ,::,/-R AS<T I V-R I R--) . )
---..'-
I J ..J,_ __Ai2==:jI
~a-- -T
, -T
I
I
I I
y
A
"""
U
A B
IDcMa Square
Inches
Puc
Inches
"""
U
A I B
InChes
r I >-
:>q..are
InChU
Inch
'n
Incnes'
I Iyv I '"
I
"0'< I ~v,'
Inche.
". on
=t:
-+ . '00
.eas · 050
· 06J
.050
· ".
·
· 125
· 0852
· 0977
· 0035
.00<12
00"
· 00<.
.0062
· 0122
· 203
.197
""
'"
rn
· 353
1-+
I-}
· 375
· 500 '"
· 050 ".
.0127
· 0852
· 106
· 153
· 0009
· 0020
.0060
· 0075
I .111
'" '""
~96
· 125
· 1231
0 o
, 0071
· 0146
· 0069
20'
· 255
.346
'" H:
H-
.750
'" ~
· 063
· 155
,n .207
JI5
· U045
.oi rr
.010S
.0!-l7
· 190
275 ,,.
· 296
j+ .625
'"
so •
• 115
0 0070
· <lOS3
cia '"
· 259
· 268
'"m 1-+
· 875
· 315
· 094
",
' 125 .aao
· 0790
.327
· 0995
0159
0009
· J193
· 00 7
,;;
106 ,
290
H- 1.000
.063
00,
.125
· 1~5
· 131
.163
0084
.0109
0148
.0367
.253 · 336
H- .500
.050 .094 .09\6 .146 .0020 .csro .148
, 347
I i1 . 625
.050
'"
·.cso
.094
,
· 125
.0977
· 123
.0090 .00l>S
.0082
;0 •
.298 ~ .625 ~
.125 .131 '"
· 196 · ,}047 015"
'" 347
· Olla
.no
· 259
pt .076
, · 158 · 201 0055 0183 1ST
, · HI
~
.750 .110 .010 .0122 ·~ 3
"
.750
,a ." ra us 0148 · 304 .n
i-#-
.875
.750
--*l-
· '0 4 .11.
.147
245
.247
0065
0079
0151
.OIS4 ,
2JT
aaa
'"
0
1. 000
,,, ".
.125 · IT .0165 .0367 .312 · +64 .or .17 • · ~53 .0094 0214 ",
Ri:te
;0
.050
r*-" .S2S
"', ".
· 125
·, 1~5
· 104
· 131
· 111
0130
· 0158
· 01 1
00"
.0122
ce
· 35J
· 347
.360
· 251
· 251
aaa
I. 000 '~
us
· 125
.328
aie · 362
.319
· 1~0
· 0~13
.03ll4
OG26
· (;276
0376
· .14
.304
.;H
;)58
aaa
304
"*"* 4*
18 .750 ,n61
1
Hl-- .87S .078'" .125
.147
.177
.0164
0214
O14B
.0175
254
.348
· ;)17
.314
.-g.
.500
". .H5 0029 .0211 HI 377
H¥-- ". ~
.139 187 0049
'" · 187
, 394
... *
.050
'" ~.
· 0172 · 0197
'"
· 365
--li'-
. '" '"
· 193 · 0058
Hl--
f--#-
.875
~
.076
.153
· 197
.caro
,0245
· ,0241
.0287
.359
.353
.385
-+.\- ~ · 155 .229 0083
· 0252
.0250 · 231 '"
· .02
~
ceo .0285
.0123
'"
~
.625
· '"
.U 00' .400
, ."" ". ~152
,"
· 294 .0335 ~03
· 125 · 139 oe .875
• ., " .244
4 · 094 · 244
." '00 · 01,6 OJ83 '159
t
397
.'w
'" , -*ae -1E- · 2. · 041
",0643 .417
.. m
1.000
~ .170 .0285 .0242 · ~1J9 ~ '00 ~
· 131 134 02
" "
"ae H\!- .158 · 139 .0031 · 0281
.,.,
· HO · -122
;::ll:: .875
~ .125 .206 .0335 .0287 · ~03 · 373 1.12..5 .125
,,.
·". ." ;86
.0363 .0332 .397 -+.\- ~
· 141 .178 · 805l nB5 H'
f--#- 1.00< ~ ao .036 .415 .444
aa .078 .171
, · 0060 03:H lH
'"H
*
01(12
I : ~94
232 · _3
"
as sas
· 076 .226
'"
0 sea • .750
2n .237 .0117 .0444 , oa225
~
'" .050
· ".
· 125
'"
• 142
0265
, 0304
0083
· 0197
..
.441
· 464
· 237
· 374 *
:j
as
~
~
· 181
.216
.278
.283
.0135
0156
0319
0441
.2,.
27\ ~52
,"
~
,875 .176 .0374 .0241 .368 .875
s.t""*"" ~
.256 .289 .0183 .0507 :68 .. 5
., ·, , ~
.216 .0441 .0287 m ,364
.. .
1.115
0< a .445
L 125 · 125 .125 .328 ase · 0225 · '-1616 sea .33
.125
! .,.
Pi!:
· 236 · 238 0232 .0495 .214 I 4Je
=ll!:
.210 .0467 .0 30 , .72 .503
1. 12.5
." ". .0637 · 464 · 500
L 000 · 279 · 341
asa
0269
,""
.0569 .310 .52
"*"*"
.125 .359 0695 303
, '"
· .303 .0632 .0T44 · 457 · .96 m · ,66
';0
." ." ~~5.
· 508 0661 0757
~ .150 ~ .," · (1421
.04 7
,0148
.01.
1. 375
I .
,. · .53
,,
SI9 .0832 I 0930 · -123 · .53
, ,
'sae
"
! '" sa
1. 2~O LOOO ~ .202 · u~32 o · 514 --11- · 220 0071 0493
·· ~57 "
.
.125 .267
.094
~
.291
".
.0729 0511 · 501
'" Hi-
~
,875 ~
.125 .344
.278
288
.0190
.0234
.0650
'J79 · .51 481 '"
I. 'W ". ,
· 0765 .0887 · 519 · 558
,..g. 1.250 .265 355
~
.338 0886 · 1039 512 · 125 · 377 · 0334 0698 5n
· 555
..... ". ·
I. 125 .314 .383 .0389 0607 ,07
=iso
· l52
*=1t ."
0
" · 543
12 csa · 'J484 · (>994 .345 -ss
...
·,500
:jf ::$.: · 210 0662 .0367 .562
·, '" ,
..
1. 500 .560
· 25~ 07 4 · 125 · 1106 1292 · .70 · J08
,2,
,, "
1. 375 1.000 .125 · .15
~
.197 · 169 .00 3 0540 .179
"*-*4-
· .103 .0911 .0511 .549 .HI .525
.094 .232 · 115 .00B · i)621 171 · 517
~ ·" o.
.
I. 250 us ~ · ~68
.2.16 .263 0169 0705 547
· ,53
. ., s
1365 · 1308 · 549 .537 -fr
t
aae
."
. ~
· 8'15 279 ~68 .0198 · 0818 · :66 543
," .07 5 .:; 2
~
1. 375 · 125 ' .359 · 278 .0242 .1002 ' · 359 S2S
~ · 265 .0961 0439 .602 ,407 · 125
I. 500 000
· .115
,,
· .114 .111 0511 .596 .403
, ~ · 326 · J71 .0403 1010 I .352 557
=!t" I. 750
I. 2 0
,c
L .00
.094
,
". ras '"
.284
,
.134
· 11 4
1377
,
.1040
.0439
.0\1
u
.096
a
· 393
H}
ss
1.125
L sas
sae ~
~
.,
· ;56
..
· ~22
.513
.244
· 31l'1
.390
."
.0504
059\
, \710
.0076
1246 I .346
.1451
· 2035
.J76
.339
.177
SH
,
.532
,52
f:i!:
L 'SO .094 · 365 · 1925 .1039 707 .520 · :66 · ;0 I
ns
:::g: 2.000 L .00 .094 361
.12.
2194 .0511 " · 779 ate ~ \.;00 I avs ~ .125 m .269
-a 0090
.0249
0939
· 1236
:7
· 258
HB
;H
"
3852
,. 2346
"
.029
" · 338 · 380 · 0414 · 1233 · 350 004
=H: 2,500 m ~
'" · I2S
· J23
· 385
.2890
· .1383
_0288
0331 """ ·'" ,298
=t 1.000 ~
"
00.
·;
.3723
.4707
0512
"00
.955
.941
"
.354
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A3 13
w
Dimensions
, R
Area
~.ID. I
Axis XX or YY
P d I
Axis ZZ
P , ~:
,x
'I'
3/.
3/'
3/32
1/16
3/32
1/'
1/'
1/'
.111
.089
· 132
0.004
0.004
0.006
0.183
0.220
0.219
0.181 .0015
0.199 .0018
0.214 .0026
0.111
O. 142
0.141
'"'"
",
1.-
I
y
, A~~
II' II' .171 0.217 0.227 .0034
1
3/'
1/16 1/16 .122
0.008
0.012 0.311 0.211 .0048
0.141
0.199 '"
",
,,0
a
•
NOMINAL DIMENSIONS
R AREA
SECTION EL.£MENTS
I
'" "'xx i
1 3/32 II' · 178 0.016 0.301 0.276 .0066 O. 19:J A a Y '------- i i ~ t"yy
1 1/' II' .23' 0.021 0.298 0.290 .0085 0.191 ", "
w
0 Incbe.s
Square
1 3/16 II' 0.029 0.293 0.314 • 0124 0.192 "
,
Inchel
"''' lJIChell
"'''
1~ 1/4 3/32 3/32
· 339
.230 0.033 0.38 0.34 .014 0.24 '"", .oea
, .,
", " · "'''
· 625 · 500 .050 · 0535 .126 · 187 .OOIl .H6
1-1/4
1_1/4
141/4
II'
3/16
1/.
3/16
3/16
3/16
.30
.43
.56
0.042
0.059
0.074
0.37
0.37
0.36
0.35
0.37
0.40
.017
.025
.032
0.24
0.24
0.24 '"'" H=
f-t .751l
·,
· 625
s
· ""
a .083
· 0818
aa
.
· 162
· 150
· 222
'"
·· 250
• · 0027
· 0029
" ., ,
· 01144
· 0069
"
·· 221
i· 274
~
...."
· 875 .US .100 0049
'" · 0073 · 270
..
1_1/2 .28 0.46 0.40 0.30
1-1/2
3/32
II'
3/16
3/16 .36
0.058
0.074 0.45 0.41
.024
.031 0.29 '" q
",
· 750
.625
~
.184
.<0
• 211
.,
I . l40
· 272
.320 .,
.00 2 .0123
, .,., .,.,.
.211 .258
,a
3 II'
3/16
II'
1/4
.49
.,.
.n
0.18
0.27
0.61
0,61
0.53
0.56
.08
.11
0.40
0,39
'45'
"
~
f#
1.250
.125
~
.125 .231
.139
.181
.239
.427
"I
.0092
.0124
.0346
.0217
, .291
00
~,
.381
, ".
~
1.000 .202 .250 .0\71
1/4 1/' 0.34 0.60 0.58 .14 0.39 fil-
ia ra .262 · es
· 373 .0306 .389
~ ~
.15 .0077 .0463 .194
.750
.125 .167 .534 .0096 .0584 .l91 .465 "
..... ~
.......
.228 .228 .472 . 0182 .0512 .282 • 474
,.:
fielf 1. sec '.300 -:.m-
· 156
· 156
'"
·.361 · 239
. 249
• 48~ .0226
.0265
.0647
.0712
.276 · ~68
.271 , csa
c#
~
1.250 ~
~
.251
· J27
.310
.321
.433 .0347
.0436
.0550
.0698
.312
.365 ·.462 '"
1. 00
· 156 .401l
, .331 .455 .0518 .0834 .360
'"
s
~
.125 .591 .02 6
1.250 .125 , · 3~8 .300 "" ',"
".
'" · 0465 · 1078 · 360 '
'~W
1.750 ~.m .389 .383 .506 .0775
ti&- I. 500 .156 .478 asa · 517 · 0928
.1144
1377
· ~47 ' .542
· «1 I
,
'"
y
,
--'--
,
~0il~
i
, 'r
I , n -11
A T
Table A3.1S
X...Li ~ik-R C
X I'ro,",rllell of
l"Y
I , T ..ble A3.I3
PrOpertin of
6~
1.8e<:tionll
I ~ ~ ~
:7 ,~!a-j ,
'----"
"T
I,, ,_A~oT
•a
SOMINAL DIMENSIONS SECTION ELEMENTS NOMINAL DIMENSIONS SECTION ELEMENTS
C
z
• , Y I R AREA
, ", "'V< a Y I R AR<A
", 'u '"
"
,~
j Ip'IY A ~
5q1JaI"e W Squate
0 Inches Inch [nches Inches 0 incheS 1JIches"' Inches
Inchell
" " Inches
-+ '"
R=j
L
· 625
· 7sa
, eo 0
· 050
.063
.0'
.063
,125
.
". , · 094
,
· l27
,
.H
.228
,'" · 0028
0055
.0046
.Ol31
· 0011
.0098
or
.0129
· 173
· 209
.209
.296 ,
· 287
ie
ia
-"
, ,H=, .750 L
L '00 '" ~
078
· 07
· 125 '"
· 250
· 289
-,
· 0206
· on9
0284
· 0193
.0235
OHI
. 118
.310
.314
'"
.307
.318
-
~1.
· bO
~
0
'"ec ~
I. 375
375 L '00
r. 00'
on
· 'J94
ias
'"
ne ,
,:H2
aas
.0182
0<, · 0158
O1B8
· 292
.290 ,
294 ats
L eas 094
m
,aa ,
· 0~2
,•
0375
; '"• •.u · 338
H=,
094 240 .327 .0345 .0 ISS · 379
L 2~0
,;0 .063 .lH
, ," .141 .0052 .0169 .192
" I. 000
L 370
l. 625 "
.094 · 125
.341
.388 0676
.0375
0626
HI
$:1.
., ;00
I. 000
t. 000
t , 250
,078
'J94
I. 250 ; .094
. 125 '"
.228
ua
.252
.225
.3H
,
,0 60
.232 I · J186
· 0301
Q354
'
0200
.0245
.0245
,
,
286
· 371
ate
· 321
.328
· 312
tz-
I::±tJ
~l.
II
13
250
l. J "
l. 02
L B7:;
2. 125
· ,)94
.094
.oec
· 125
H
",
458
50S
.11
1285
1442
cI
.0911
H .,H
516
"530
.321
as
· 4~0
r*- 120 .ca n, .0449 .371 ra
!
, • .b
•
· 531
I.l~
"
"
~
, tea .180 II
, J2ij H' 't I. :50 .125 .5H 2268 .1023
0
.61; "
H;
H!- ! "
azs oJI :' : 1.500 ia 'OS a ,
'" 0407 · 234
"." Ht-
2. COO .025
f.li.-
Ht- I. 62~
I. l25 ,
,
· l25
.878 '"no
.212 .254
314
.0228
· ,; 39
.0263
· o~oa
na
322
rsa [7 , 2. 500 us · 781
,,'
3106
· 1543
305~ 636
· .65
626
L 125
" '" , " H¥s- 3922 1183
..,44 , ,.'"
· ~63
! .094 2,5 .3H .047; 0314 H5 .338 2.000 2.000 · l25 719 .4946
1. J75 ';0 464
, ,
L. 3,
L.1 ,
· l25
"•,
a
.n,
a
,,.0 0,
,
,)40~
sa , '"
0";
.Hl ,
,
' , , 240 . 17 013l 234 · 40~
"
· 1546
· II 4 L z '"
I~L
'--"-' 2. ~OO
f--iIT- , · 156
; 23 I .67>
,
· 125
894 I
.31l .191 .01M , 0512 230 ,," , aa
3.000
\ira
· ~51 L 262
, a
l5~1 '
,. "
L aa ."
24 , I L 125
'"H"
252- · 0273 J393 .322
"" I. SOO ;H1 I,
~'
, ias , 1,8 , ,- .314 I · .167 '"
740
'0 , · iOO
<0'
~ ".
L .0412 062 · 125 I. l88 · 1184 • L
750 L 315 , ". ,os -, . 333 04 4 0393 HI , .370 2. 000 2, ~24 · l547 i t. se
" ,
L .175
,,25
156
')9.
.45 .3,8
'"
· on: , · 0627 · .03
.500
371
, a " 4.000 3. coo
""
Hm .188 , l. 52 4. 136
L a
S5d6 I L ;;
, ss
550
7'"
,
, , ,
.310
'" , · 'J394
)5l3
· J56
se 3.500
2.00 5. 521
• 368 ,
,,
L 794
"'"
12:; 4.14 .0995 2.60 t , 63d .794
L 625
625 l56
· 405
Table A3. 16
SXC'1'IOI'f PROPERTIES OF TYPICAL .URCRAFT RX'I'RUDiD SEC1'IOI'fS
I
'Sect. Dimensions ATOll 'rop.,u•••bM x-x' src•• rU••• bou~
i xo. , A 8 I c 't R in. AlI:1s AlI:l.S
I ! I I
'1
I
'2 !
IlI:X ,
I
e lI:X e I I rY e rr e'
, 1 17/32 13/16 I 1/' ,040 ,050 1/16 .068 .00686 ' .317 .319 .00150 .148 ,112
"
,
- ...C;..-..- 2 I liZ 1 ,
, 7/32 ,050 ,050 .050 .113 .01515 ,3665 ,510 ; ,OO15~ .117 ,110
~::-[:
,
, 3 lI2 7/' 1 3/16 1/16 1/16 I 3/32 .1025 i .0100 i ,313 ,
.383 ,00162 .125 ,116
• ,
,
9/16 7/. I 3/16 .060 .060 3/32 .1031 , .0102 ,313 ,383 .00215 I .114 ,130
-, ,I
, , ,
_e_
..l. 'T.
,
-*
5
e I
9/16
3/4
7/'
1-l/8
3/16
3/16
,050
.0751.075
,078 3/32
1/16
.1061 !
.1514 I .0239
.0120
I
.329
.397
, .345
.435
!
I .00550
.00302 .165
,190
,
: .182
,133 ,
I
"
,
,
'-- A---- 7 3/4 1_5/16 9/32 3/32 3/32 9/64 I .230 I .0507 , ,471 ! .574 , .0080 .181 .171
,
, I
,•
7/8 3/8 i
,
1/16 1/16 I 1/16 i .095 , ,0101 .326
,
,4315: .00111 .108 ,
,
,
,108
7/8 3/' . 3/32 3/32 I 3/32 .1375 .0136 .314 .4375: .00148 .104 .118
-r-F I
!
! 10
. '-'l-'
1 ,I lI2 13/32 3/32 ; 3/32 i .1736 i .0240 ; .37l
,
.500 .00376 .147 I .158 I
II 1-1/4 11/16 3/32 3/32 3/32 I ,2320; .0536 I .480 .625 .00992 ,206 ,210
~ I 1'0 ,_'t\ I
,
~
12 1_13/16 11 ,
3/1' 3/16 ·3/32 ."4
,
.2920 I .677 .906 : .0555 .295 .326
_L-L., ,
, .338 ,I
-. 4
C
13 I 2_1/2 1/. I 1/8 illS .581 .525 ! .950 1.25 I, .0662 .323
l -
11 ,
,
14 : 2_1/4 1.1/8 l/S 3/16 ' 1/8 ! .666 ' .524
,
.888 1.125
,
I .0812 .362 I .375 ,
15 1_3/8 , 11/16 1/. 1/16 1/16 I 1/16 : .131 i .00425 .176 .133 .0178 ~ ,360 ,I .687
rr6
r
IS 1_5/8
I i ! .168 I .812
I~r-
3/4 5/32 1/16 1/16 ! 1/16 ,
.00580 1
,
.186 ,133 , .0335 .446
,ee
1 17 1~5/8 3/4 5/32 i 1/16 3/32 ! 5/64 I .206 .00589 I ,170 ,121 I .0458 i .474 i .812
ir+-r- §...-/C IS I 2 1- 1/ 1 6 ,
3/1' ! 3/32 ! 3/32 i 3/32 .306 : .0252 , .287 .201 i ,0191 I .509 1• 0
i~ .L.
1
1
18 ,
2 i 1_1/4
,
3/16 1/8 I 3/32 I ' 3/32 .359 .0475 .364 .300 : ,0821 i .479 ,1. 0
~--"
;-~ -; 2.
I 2 1-5/32 1/4 3/32 I 1/8 i 1/' ~ .378 ! ,0291 : .279 .217 i .1095 i .541 i1. 0 I
22 I 1_1/2 i 1-l/8
,
1/16 1/16 I ; 1/16 , ,161 .0186 ,
.342 .255 ; ,0171 .326
I .750
j"-:-J:' 23 1-1/2 2 I l/S I .384 .1562 .641 .658 ,0346 .302 ,750
Tf"'-f~T
3/32 i 1/8 I i ,
,. ,
I 1_3/4
1
2
,
9/32 1/. ; 1/8 .760 .315
I .643 .757
,
I ,0599 .281
, ,875
I'a
! 1 I
;
I as
'C
I, 1-3/4 2_9/16 : 1/4 .5/32 I ; 5(32 .918
,
.600 I .809 .959 , .0730 .282 , .815 I
I
i.,.e.'J, ,,
26
!2 i 2_1/2 ! 5/32 1/. 5/32 .605 ! .385
; .800 .793 .159 .514 i 1. 00 i
27 11/16 1 l/S .050 .050 ; .050 3/32 .135 .0218 , .404 ,500 .0163 .348 • 6625!
,..A-! I i
28 I 13/16 1 1_ 1/ . 3/16 .060: .060 i .050 3/32 .180 I, .0315 ! .456 .562 .0295 .405 .78251
i--r~
!
, .595
I'8 - - ' -e - T 28
! 5/S 1~~/2 13/16
! 1/4
.050 1/16 1/16 ; 3/32 .166 , .0603
,, .793
.75 .0151 .298 .600
I ·'r·''.,-J
,
3. 3/4 2 .050 1/16 : 1/16 3/32 .214 .1382 1.00 .0281 .358 ,725
I..L- I ~ ,L 'I 31 i 3/4 2 i 1/4 1/16 1/16 : 1/16 3/32 I .2378 i .14.49 i .773 i 1.00 ; .0282 .341 .719
't-e'..I 132 , 3/4 ,2 11/4 1 1/ 1 6 3/32 ! 3/32 3/32 .2855 ! .1820 i .795 1.00 .oaes .356 .719 i
: 33 I '/1' : 7/8 I .050 .050 i ,1/16 I .099 ! .0120 I .349 : .437 , .0056 .238 .531
...
:
'--A ;:I
.~,
1.,,10-
" e
I 34
35
38
I
1/2
9/16
9/16
I
I1
,1 1/4
7/8 : 1/16
1/16
.ove
1/16
.080
.075
I
i
,
1/16
' 3/32
3/321.170
.115
.151
,, .0239 ,!, .397
:,
I
.0135
.0384
.341
, .475
:
,
.437
.500
.625
I .00446
I
I
,00828
.00876
.197
.234
I .227
.469
.531 I
.525 1
a ...- -~ 31 11/16 1
1 _ 1/ 4 1/18 .100 3/32 : .213 : • 05~0 I .508 .625 I .0194 .302 .655
I
L'" e
LII~J-
I
I ~
38
3.
,, .825
3/4
1.1/2
2
i
i
.075
.072
.120 I
i .012 ,
3/32 : .302
1/' : .249
I .110
.1446
,
: .604
,, .757 i
.750
1.00'
I .0391
I .0180
.360
.267
I
,787
.714
I
,,
i ' 4. , 3/4 2 , .102i· l 0 2 i 3/32 .330 .1826 .750 1.00 I .0228 .265 .700
CHAHT A3. 1 PHOPEHTIES OF FOHMED CHANNEL SECTIONS
1.0
.. ., ,.
w JIi I'
~ .1
11:1Mo.3O ..
,:1: i I 1, ,.r-
Z
.. , ! II
1
...
~ ., II ., ~
~
.'_, i: , . , I'I
:'1' ;':. :1
"" ~
~ '+, • • i: 'i~
I .:: I T1 ::1. ,i 1
,.
Z
0.25 F.itt0'l!I: i1ii !!II:1 !l~i :II!: ill~~' ;:11 ~ -1" I-!- I~; Iiii1: ill:i : IilliIiII: 1111~1 illIlllllli! ~ '1II: i III Ll!: I!il;:~
.. , I
tl
tl
'"
~
~
030 i
~
~
~
~
~
"z
.,
0
.,
0
Ii
~
0
Iiw -.,"
r-
0,15 ~ ~
0
~ '"
<:
'"
'"
~
0
r-
0.10 1";'I'~!I!"!Im!I'!'T"!lIIWI'III'II'I~j0P;IPI'':¥'IU;;ni'W!UUwnUI'~p!!1
till r._. 1m .. , .m itT. !I. I:-p. Itt! ....... " ", ." .... ','.". "r. " I " . . . " "" . " . "
'".,
I~
'0c"t
.,
0.05 1II111l1\l!llHllllllllml!l!!Yf111 IY!!'III1i~:iIIlIJl!lIiHIM1'I!l"J!ml!!! 'III ::1 !I,'I II! II i!!!l1i 111:11 ·',IIII! I: !I .,
"
I~
~
.'-c.
'.
~.
1.0 1.0 2.0 2.5 3.0
Mom. ollnarUa.:-
I
,
I;
II - IIEIGIiT OF CHANNEL .025
U
II; HEIGIIT OF CHANNEL
3.5
•
V~
\1 3
A3.16 CENTROIDS, CENTER OF GRAVITY, MOMENTS OF INERTIA
-ll-/.,
l:~;; l.- 4-- 8 ..;... ...+
Ir..
I '
...
IS I. ,
Il"
' I
10
J I·'.T L
'i)t. "
0·8
., i
Ii /0_ I zo·'
~~t•
I-- z" -.J ~ Fig. A3.19
,,..
I
o.? Od~
Fig. A3.14 Fig. A3.15 1'- t +• 15
1"8 i
(1) For the section of Fig. A3.14 determine
the moment of inertia about each of the or-mer- (6) F~~. A3.19 shows a wing 0e~~ section
pal axes Xp and Zp. with a cut-out on the lcwe~ surface. Jeter~ine
(2) Calculate the moment of inertia of the the ~oments of inertia about the ~r1~:i~~1 axes
section in Fig. A3.l5 about the ?rincipal axes. assuming the eight stringers are the cn:y ef~2c
t ive :naterial.
It ~ l : D~
I 5 spaces @5'" 35" I F,g. A3.20
Fig. A3. 17
(7) Fi~. A3.20 shews a 3 cell ~ulti,le
(3) Fig. A3.l6 illustrates a box type beam flan?e -beam. !he 7 flan~e T.embers on t~e upper
section with six longitudirral stringers. De- face of beams have ~~ area of .3 sq. in. each
termine the ~oment of inertia of the beam sec- and those on the bottom skin 0.2 sq. in. each.
tion about the principal axes tor the follow- The bottom skin is .03 inches in thickness.
ing assumptions:- Compute the moments of inertia about the
(a) Assume the beam is bending upward principle axes assuming that the flange :nembers
putting the top portion in compression and the and the bottom skin comprise tr.e effecti~e
lower portion in tension. Therefore, neglect material.
sheet on the top side since it has very little
resistance to compressive stresses. The sheet
on the bottom side is effective since it is in
tension. For simplicity neglect the ver~ical
webs in the calculations.
(b) Reverse the conditions in (a) thus \
placing top side in tension and lower side in \_Cutout [or door
compression.
(4) For the three stringer single cell boX
beam sectien in Fig. A3.17, calculate the mo- Fig. A3.21
ments of inertia about the principal axes. As-
sume all 'Neb or wall material ineffective.
>4
A4.2 GENERAL LOADS ON AIRCRAFT
l
sk1 type). Physics, the accelerating ~orce F = ~, where M
(2 ) Landing.Loads Landing on ~~ter. is the ~ss of the bOdy or ~/g. In Fig. A4.1
the unbalanced force system F causes the rigid
Arresting. {Landing on Air- body to accelerate to the rl~ht. Fig. A4.2 shews
craft Carriers). the effect of this unbalanced ferce in ;roduclng
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A4.3
- -- - -
\
tw where w = angular velOCity at the paint A.
a angular acceleration at paint A.
Fig. A4. 1 Fig. A4.2 r radius at Cllrrnture at :light path
at point A.
The inertia forces are equal and OPPOSite
a force on each mass particle of mLa, m~a, etc., to these etfective forces as indicated in Fig.
thus the total effective force is zma = Ma. I f A4.3. These inertia forces can then be con-
these effective forces are reversed they are re- sidered as part at the total force system on the
terred to as inertia torces. The external airplane which is in equilibrium.
forces and tha inertia forces therefore torm a It the velocity or the airplane along the
force system in equilibrium. path 15 constant, then at = zero and thus the
From basic Physics, we have the following
inertia force F = 0, leaVing only the normal
relationships tor a motion of pure translation t
it the acceleration is constant:- inertia torce Fn•
(1) It the angular acceleration Is constant,
v - Va= at - -
lf
the following relationships hold.
s ve t -+- iat - - (2 )
v" _ va If = 2as W - WQ= at - - (6)
(3 )
Q = wot ... ~a.tlf (7 )
where, wa _ W o If = ZaQ (8)
5 = distance
moved in time t ,
where Q = angle of rotation in time t.
Vo = initial velocity w = initial angular velocity in rad/sec.
v = tinal velocity atter time t. w = angular velocity after tune t.
In Fig. A4.3 the moment To or the inertia
Inertia Forces on Rotating Rigid 80dies. forces about the center of rotation (0) equals
A common airplane maneuver is a motion I'tra(r)= MTaa. The term MT a is the mass moment
along a curved path in a plane parallel to the pf inertia of the airplane about point (0).
XZ plane of the airplane, and generally referred Since an airplane has considerable pitching
to as the pitching plane. A pull up trom steady moment of inertia about its own center ot gravity
flight or a pullout from a dive causes an air- axiS, it should be included. Thus by the
plane to follow a curved path. Fig. A4.3 shows parallel axis
an airplane following a :urved path. It at
point A the velOCity is increasing along its To = loa -+- I e.g a - - - - - - - - - - - - - (9)
path, the airplane 1s being subjected to two where I = Mfa and I e.g = moment or inertia of
Q
accelerations, n~elY, at' tangential to the
curve at point A and equal in magnitude to airplane about Y axis through e.g. at airplane.
Inertia F0rces For Pitching Rotation of Airplane
Center of Curvature about Y Axis Through e.g. airplane.
01 Flight
Path
r In flight, an air ~~st may strike the hori-
I zontal tail prodUCing a tail torc~ which has a
I moment about the airplane c.g. In some landing
an=rwlf=Vlf / ? conditions the ground or water forces do not
pass t~xough the airplane e.g., thus producing
=E2~~-MTa a moment about the airplane e.g. These moments
AI '.g. cause the airplane to rotate about the Y axis
M? w lf = Mv a / ? through the e.g.
Therefore for this etfect alone the center
Fig. A4.3 or rotation in Fig. A4.3 is net at (0) but at
A4.4 GENERAL LOADS ON AIRCRAFT
the e.g. at airplane, or r o. Thus F and F air fOrC(3 on the wing. For example, consider
n t the two air pressure intensity diagr~~ in ?lgs.
equal zero and thus the only inertia ~~rce tor A4.6 and A4.7. ~hese distributed ~orce systems
the pure rotation is I g C, (a couple) and can be replaced by their resultant (R), which
c. .
thus the moment of this inertia couple about the of course must be known in magnitude, direction
e.g. = Tc. g• = Ic.g.~' and location. ~~e location is specified by a
term called the center of pressure which is ~he
As explained before if the inertia forces
are included with all other applied forces on point where the resultant R intersects the a1r-
the airplane, then the airplane is in static foil chord line. As the angle at attack is
equilibrium and the problem is handled by the Changed the resultant air force Changes in mag-
static equations for equilibrium. nitUde, direction and center of pressure
location.
A4.5 Air Forces on Wing.
R
The wing at an airplane carries the major
portion of the air forces. In level steady
flight the vertical upward torce of the air on
the wing, practically equals the weight of the
airplane. The term airfoil is used when re-
terring to the Shape of the cross-section at a
wing. Figs. A4.4 and A4.S illustrate the air
pressure intensity diagram due to an air-
Fig. A4.6 Fig. A4.7
'UGHT DlllltCTIOlI
'!'?5?:>
FLIGHT DllIECTIOIII
a positive and negative angle of attack. The
shape and intensity at this diagram is in-
fluenced by many factors, such as the Shape of Fig. A4.8 Fig. A4.9
the airfoil itself, as the thiokness to chord
ratio, the camber of the top and bottom sur-
faces etc. A normal wing is attached to a Aerodynamic Center (a.c. ). Since an air-
fuselage and it may support external power plane tlies at ~y different angles of attack,
plants, wing tip tanks etc. Furthermore the it means that the center ot pressure Changes for
normal wing is usually tapered in planform the many flight design conditions. It so hap-
and thickness and may possess leading and pens, that there is one paint on the airfoil
trailing slots and flaps to prOduce high lift that the moment due to the Lift and Drag torces
or control etfects. The airflow around the is constant ~or any angle ot attack. This
wing is affected by such factors as listed point 1s called the aerodynamic center (a.c.)
above and thus wind tunnel tests are usually and its apprOXimate location 1s at the 2S percent
necessary to obtain a true picture of the air of chord measured from the leading edge. ~hus
torces on a Wing relative to their chordwise the resultant R can be replaced by a 11ft and
spanwtse distribution. drag force at the aerOdynamic center plus a wing
moment Ma.c. as illustrated :n Fig. A4.9.
Resultant Air Force. Center of Pressure.
A4.8 Forces on Airplane in Flight.
It is convenient when dealing with the Fig. A4.l0 illustrates in general the main
balancing or equilibrium of the airplane as a forces on the airplane in an accelerated flight
whole, to deal with the reSUltant of the total condition.
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A4.5
Fig. A4.1O • 0
T_
Fig. A4.11
W
T =engine thrust. forces in steady horizontal flight. L repre-
L = total wing lift plus fuselage lift.
sents the total airplane lift (Wing plus tail).
D = total airplane drag.
=
Ma moment of Land D with reference to wing Therefore L = W. Now assume the airplane is ac-
a.c. (aerodynamic center) celerated upward along the Z axis. Fig. A4.12
shows the additional inertia torce wag/g acting
W = weight of airplane.
I L = inertia force normal to flight path. downward, or opposite to the direction ot
accaleration. The total airplane lift L tor the
In inertia force parallel to flight path. Z
I
1
m rotation inertia moment. ",L
E tail load normal to flight path.
For a hortzontal Constant velocity flight
condition, the inertia forces I
L, 1D, and 1m "...- e.g.• - 0
T
would be zero. For an accelerated flight con-
dition involving translation but not angular w
Fig. A4.12
acceleration about its own e.g. axis, the
inertia moment 1 would be zero, but I and 1 hg.•,
m L D
unaccelerated condition tn Fig. A4.ll must be
would have values.
mUltiplied by a load factor n to produce statiC
a
Equations of Equilibrium For Steady Flight. equilibrium in the ~ direction.
From Fig. A4.10 we can wrlte:-
Thus, n L - W - ~ e., ::; 0
a g a
l:F D + W sin Q-T cos ~
x 0,
° Since L =W
ZF ~.= 0, L
- '" cos Q+T sin ~ - E =° Hence n ::; 1 + a~
~
ZMy = 0, -l1a- La - Db + Tc cos ~ + te = a g
weight W of the airplane J hence a defin~te ~ax~~w~ r:~;~: ,eloc~ty :ar ~om
mer-c Ia l airplanes the ve Loc i t y is imi t.ec to 2.
'~
n W ax (See Fig. Al4.l3) reasonable ~lide speed whic~ ~s su ficient to
x g take care of ~easo~a8le flis~t opera~icns.
~.
ities issue requirements whic~ specify the From ?!g. (b) for small angles, 0" KU~l
design conditions for ~he various class1ficat1on and from Fig. (a) J 6C = m 6.0., wnere cn the
of airplanes. Generally speaking J any air~lane ZA
flight altitude can be defined by stating the Slope of the airplane no~~l force curve (C 7
existing values of load factors (acceleration) ~A
and the airspeed (or mere properly the d}~ic per r-anian) .
pressure) •
The accelerations en the airplane are pro-
duced fram ~NO causes J nanelYJ ~~euvers and air
gusts. The accelerations due to maneuvers are
subject to the control of the pilot who can
manipUlate the controls so as not to exceed a KU~
certain acceleration. In hi&~ly maneuverable
military airplanes J an accelerometer is in-
cluded in the cockpit instruments as a ~ide to V
limit the acceleration factor. ror commercial Fig. b Fig. a
airplanes the maneuver factors are ~de high
enough to safely take care or any maneuvers tr~t Tr4e load factor increment jue to the gust
would be reqUired in the necessary fl~ght opera- KU can then be expressed
tions of the particular type of airplane. ~hese
limiting maneuver factors are based on years of KUVS
operating experience and have given satisfactory m (A)
results fram a sarety standpoint without pen-' 575 ',.,[
alizing the airplane from a weight deSign con-
sideration.
The accelerations due to the airplane U gust velOCity in tt./sec.
striking an air gust are not under the control
of the pilot since it depends on the direction K*= gust correction factor depending on
and velocity of the air gust. From much ac- wing loading (Cur7es for K are prOVided
cumulated data obtained by instal:ing accelero- by Civil Aeronautics Authorities).
meters in commercial and military aircraft and V indicated air speed in miles per hour.
flying them in all types of weather and loca-
tions. it has been found that a gust velocity or S wing area in sq. ft.
30 ft. per second appears sufficient. W gross weight of airpla~e.
The speed or velocity of the airplane like-
wise effects the loads on the ai~plane. The • NACA Technical Note 2964 (June 1953), proposes that the
higher the velOCity the higher the aerodyna~ic alleviation factor K should be replaced by a gust factor,
Wing ~oment. Furthermore the gust acceler- Kg :: 0.88 Mg/(5. 3 + Mg}. In this expression Mg is the
airplane mass ratio or mass parameter, 2 W!o.pcgs. In
ations increase with airplane velocitYJ thus it which c is the mean geometric chord in feet andg the ac-
is custcmary to limit the partic~lar ai~plane to celeration due to gravity.
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES
A4,7
It U is taken as 30 ft./sec. and ~ as the would gl ve an acceleration less than the lim-
c~ange in C wit~ respect to angle of attack ited values given by l1nes AB and CD.
zA The positive ,and negative gust accelera-
in absolute ~~its per degree, equation (A) re- tions due to a 30 ft./sec. gust normal to f:i~ht
duces to the following path are shown on Fig. A4.l4. In this example
diagram, a positive !list is not critical within
on • ~V _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ (E) the restrictAd velOCity of the air?lane since
the 6Ust lineS intersect the line BD below the
line AB. For a negative gust, the zust load
Therefore the gust load factor n when air- tactor becomes critical at velocities between F
plane is flying in horizontal altitude equals and D with a maximum acceleration as given by
paint E.
n ;: 1 ... 3KmV (e) For airplanes 'Nhich have a relatively low
- I,.J/S required maneuver factor the IUst accelerations
may be critical for both pOSitive and negative
and when airplane 1s in a vertical altitude accelerations. ~ination of the 3Ust equation
indicates that the most lightly loaded condition
3KmV (smallest gross weight) prOduces the highest
n •• - - - - - - - - - - - - - - - - - (D)
- ',,/s gust lead factor, thus inVOlving only partial
pay load, fuel, etc.
A4.10 illustration of Main Flight. Conditions. On the dia~am, ~he points A and 3 corre-
Velocity-Load Factor Diagram. spond in general to what is referred to as high
As indicated before the main design flight angle ot attack (H,A,A,) and low angle of attack
conditions for an airplane can be given by (L.A.A.) respectively, and points C and 0 the
stating the limiting values of the acceleration inverted (H.A.A.) anrt (L.A.A.+ conditions re-
and speed ~nd in addition the maximum value of spectively.
the applied gust velocity. As an illustration, Generally speaking, if the airplane is de-
the design loading requirements for a certain Signed for the air lOBds produced by the veloc-
airplane could be stated as follows: ~The ity and acceleration conditions ~t paints A, 8,
proposed airplane shall be designed !or applied E, F, and C, it should be sate from a structural
positive and negative accelerations of + 5.0g strength standpoint if flown Within the speCified
and -3.5g respectively at all speedS tram that limits regarding velocity and acceleration.
corresponding to CL up ~o 1.4 times the Basically, the flight condition reqUire-
max• ments of the Civil Aeronautics Authority, Army,
maximum level flight speed. Further.TIore, the and Navy are based on consideration ot specified
airplane shall Nlthstand any applied leads due velocities and accelerations and a consideration
to a 30 ~t./sec. gust actl~g in any direction ot gusts, thus a stUdent understanding the basic
up to the restricted speed of 1.4 times the discussion above should have no diffiCUlty un-
maximum level flight speed. A design factor of derstanding the deSign reqUirements of these
safety of 1.5 shall be used on these applied three government agenCies.
loads" . For stress analySiS ~urposes, all speeds
In graphical form these design require- are expressed as indicated air speeds. The
ments can be represented by plotting load fac- "indicated~ air speed 1S detined as the speed
tor and velocity to obtain a diagram which is which would be indicated by a ~erfect air-speed
generally referred to as the Velocity-accelera- indicator, that is, one that would indicate
tion diagram. The results of the above speci- true air speed at sea level under standard at-
fication would be similar to that of Fig. A4.14. mospheric conditions. The relation between the
Thus, the lines AS and CD represent the re- actual air speed Va and the indicated air speed
stricted pOSitive and negative ~euver load
factors which are limited to speeds inside line Vi is given by the equation
BD "Nhlch is taken as 1.4 times the maxtmum
level fli~~t speed in this illustration. These
restricted maneuver lines are ~er.TIinated at VI .~ Va
points A and C by their intersecticn With the
:nax1m.um C values of the a ir-cjcne • At speeds where
L
beCNeen A and B, the pilot ~ust be careful not
to exceed the ~neuver accelerations, since in Vi indicated airspeed
general, it would be posSible tor him to ~n
ipu1ate the controls to exceed these values. Va actual airspeed
At speeds beloW A and C, there need be no care
or the pilot as far as loads on the airplane Po = standard air denslty at sea level
are concer~ed since a ~neuver prodUCing C~.
Pa ;: density of air in which Va is attained
A4.8 GENERAL LOADS ON AIRCRAFT
i!~;~~~~;i:;'"lli~~,~;["!
practice to place the airplane under accelerated
conditions of motion into a condition of static
equilibrium by adding the inertia farces to the
applied force system acting on the ai~plane. It
is usually assumed that the airplane is a r~gid
body. Several example problems will be pre-
sented to illustrate this general proceaure.
hence, hence,
R = 19420 lb. R, = 29800 lb. (up)
~o find the distance (d) take moments about c.g.
at airplane,
ZF~ = 29800 - 9000 + R1 =0
hence, R, = - 20800 lb. (acting down)
ill
c.g.
19420 x 24 - 42700 d = 0
hence, The velocity at end ot catapult track can
d = 10.9 In.
be found trom. the rof towtng equation
Landing velocity Vo ~ 60 M.P.H. = 88 tt/sec. V·_Voll~2as
Example Problem 3
hence stopping distance s = 34.4 ft. Assume that the transport aIrplane as il-
lustrated In Fig. A4.l7 has Just touched down in
Example Problem 2 landing and that a brakmg torce at 35000 lb. on
An airplane equipped w1th float is cata- the rear wheels is beIng applied to bring the
pulted into the air tram a Navy Cruiser as il- airplane to rest. The landing horizontal vetoo-
lustrated in F1g, A4.l6. The catapulting torce 1ty Is 85 M.P'.H. (125 rn/sec) , Neglecting air
P gives the airplane a constant horizontal ac-
forces on the aIrplane and assuming the propeller
celeration of 3g{9S.6 tt/sec·). The gross forces are zero, what are the ground r-eac t i ons
weight ot airplane 9000 lb. and the catapult R1 and R.. What is the landing run distance With
track is 35 ft. long. F1nd the catapu rt ing the constant braking forcs?
torce P and the reactions R1 and R. tram the
catapult car. The engine thrust 1s 900 lb.
:~at is airplane velocity at end at track run? W ,. 100,000 lb.
Lx
9000 lb.
Max.....--c.g.- T--
-l5",J~,"""===ThrU8tLine _ _ -<
c·g:--Ma
A
I 78"
t,
A Solution: -
R R,
~85"-j The airplane is being decelerated horizon-
Fig. A4.16
tally hence the inertia torce through the air-
plane e.g. acts toward the tront ot the airplane.
Solution: - Since the braking torce is given we can solVe
tor the deceleration factor by the equilibrium
The torces will be determined just atter equation,
the beginning of the catapUlt run, where the
car velocity 1s small, and thus the lift on the l:Fx = 35000 - Ma
x 0
airplane wing and the airplane drag can be
henes,
neglected. Max = 35000
Horizontal inertia force acting toward the air- or
plane tail equals, HIIg a
X= 35000
whence
/"", = (9000) 3.0g = 27000 lb. 35000
g (100000) 32.2 = 11.27 tt/sec'
From statics: -
To find landing run (s),
ZFx = -900 - P + 27000 ~ a, hence P = 26100 lb.
v· - Vo • = 2 a x s
To find R~ take ~oments about point A,
o - I25' = 2 (-11.27) s
2:1: = 9000 x 55 + 27000 x 78 - 900 x 83 - 85R li hence,
A =0 s = 695 ft.
A4.10 GENERAL LOADS ON AIRCRAFT
To find Re take moments about point (A) Airplane Load Factor = AirPl~e Lift
ZM
A
= 100,000 x 21 35000 x 9 + 38 Ra o
111800 - 4100 7.7
Re = 47000 lb. (2 wheels) 14000
93700 lb.
T
,
L
210" J ..
tIll. 800 lb.
I
210'
- D c :. . • Engine Thrust
z
t"'so'""",
L. w 14000
Solution: - Solution: -
•
V' 732 Fig. A4.19 shows a tree body at the air-
Acceleration aa = r 2500 = 214.5 tt/sac a plane With the lift and inertia forces as found
or 214.5/32.2 = 6.67g (uj>'IaI"d). in Problem 4.
The inertia force normal to the flight path The additional inertia torce due to the
and acting down equals angular accelerat10n a = 4 rad/sec a • equals,
Calculations of resultant load on pilot: - are for deSign loadS, which in general are 1.5
times the applied loads. It would not be correct
x = 372" to say·that the Wing deflections under the ap-
plied loads for these two High Angle of attack
c.g. of pilot condltior~ would be 2/3 the deflections sh0'Nn in
the photograph since under the deSign loads a
iL _ considerable portion of the wing would be stressed
/' beyonu the elastic limit of the mater!al or into
a c s g , air- the plastic range where the st i rrnass modulus rs
= 0.93 plane
Fig. A4.21
- 180
Fx = Mza = 32.2 x 12 x 40 x 0.93 17 lb.
Up until World ~ar II practically all air- for applying the load on the wing when striking
planes were assumed as rigid bodies for struc- the air gust.
tural design purposes. During the war failure NACA Technical Note 2424 reports the flight
of aircraft occured under load conditions which test results on a twin-engine Martin transport
the conventional design procedure based on rigid airplane. Strain gages were placed at various
body analysis, indicated satisfactory or safe points on the wing structure, and strains were
stresses. The failures were no doubt due to read. for various gust conditions for Which the
dynamic overstress because the airplane 1s not normal airplane accelerations were also recorded.
a rigid body. Then slow pUll-Up maneuvers were run to give
Furthermore, airplane design progress has similar airplane normal accelerations. The wing
resulted in thin wings and relatively large had a natural frequency of 3.8 cps and the air-
wing spans, and in many cases these wings carry plane speed 'HaS 250 M.P.H. Two of the con-
concentrated ~sses, such as, power plants, clUSions given in this report are: - (1) The
bombs, wing tip fuel tanks etc,. Thus the bending strains per unit normal acceleration
flexibl1i ty Of_ wings have increased which means under air gusts were approximately 20 percent
the natural bending frequencies have decreased. higher than those of Slow pull-ups for all mea-
This fact together with the fact that airplane suring POSitions and flight conditions of the
speedS have greatly increased and thus cause tests, and (2) The dynamic component ot the wing
air gust loads to be applied more rapidly, or bending strains appeared to be due primarily to
the loading is becoming more dynamic in char- eXcitation of the tundamental wing bending mode.
acter and thus the overall load effect on the These results thus indicate that air gusts
wing structure is appreciable and cannot be apPly a air load more rapidly to a wing than a
neglected in the strength design of the wing. maneuver load gl vtng the same airplane normal
acceleration for a commercial transport type at
General Dynamic Effect of Air Forces on airplane, and thus the dynamiC strain effect on
'..ling Loads. the Wing is more pronounced for gust conditions.
Figs. A4.23, 24 and 2S show results of dy-
The critical alrlaads on an airplane are namic effect at air gusts on a large wing as de-
caused by maneuvering the airplane by the pilot
termined by Bisplinghotr·. The results in these
or in striking a transverse air gust. A trans- tigures show that dynamic effects tend to con-
port airplane does not have to be designed tor Siderably increase wing forces on same portions
Sharp maneuvers producing high airplane accel- of the wing and decrease it on other portions.
erations in its job at transporting passengers,
thus the time of applying the maneuver loads is Fig. A4.23
conSiderably more than a fighter type airplane -000
I Comparative Shear
pUlling up sharply from high speeds. Dtstribution
Fig. A4.22 shows the result of a pUll-Up :; -400 , 1\ ,
~neuver on the Douglas D.C.3 airplane at 180 •
t5
< Dynamic Analysis
M.P.H. relative to load factor versus time of -200
- - - Rigid Airplane
application at load. As indicated the peak <, Analysts
load of load factor 3.25 was obtained at the i'..
end of one second of time. o .2 .4 .6 .a 1.0
4
~ 3 / ' -, Fig. A4.24
.
<;
•2
1
./
<,
-, Fig. A4.22
c
e -20
-30
,
-,
Comparative Bending
Moment Distribution
0
1
I'
.5 1 1.5 2 ~ -10 I" I
Pull-up of DC-3 Airplane at 180 mph. .; i
The author estimates the natural frequency of
the 0.:.3 Wing to be around 10 to 15 cycles per
o .2 .6
<, ,
-
.6 l.0
secor.d, thus a loading time of 1 second against Fig. A4.25
a tDne of 1/10 or 1/15 for halt a wing deflec- !
tion cycle indicates that dynamic overstress -3 ,
I I
~ c~~~:a~~v;o~~~~i-
should not be appreCiable. In general, it can
be said that cynamic over-stress under maneu-
-2
•g.-1 , -. I
!:Ill
c
Abou.. Elaatic Ax1s.
The higher the air gust velocity and the • Report on an Investigation on stresses in Aircraft Struc.
higher the airplane velocity, the less the time tures under Dynamic Loading. M.l. T. Publication.
A4.14 GENERAL LOADS ON AIRCRAFT
The advent of the turbo-jet and the rocket 90" ---I Fig. A4.26
type engines has opened up a range of possible
airplane airspeeds hardly dreamed of only a few zontal deceleration and the stopping distance
years ago, and already trans-sonic and super- for the airplane?
sonic speed airplanes are a cammon development.
From. an aer-odynamtc standpoint such speeds have (3). The flying patrol boat in Fig. A4.27
dictated a thin airfoil section which has thus makes a water landing with the resultant bottom
promoted a high density wtng. Thus tor air- water pressure of 250,000 lb. as shown in the
planes With appreciable Wing spans like Mili- figure. Assume lift and tal~ loads as shown.
tary bombers and near future jet commercial The pitching moment of inertia at the airplane
transports, which usually carry large concen- is 10 million lb. sec.~ in. Determine the air-
trated masses on the wing such as engines, fuel plane pitching acceleration. what is the total
tanks etc ,; the assumption that the airplane is load on the crew member who weighS 200 lb. and
a rigid body is not sufficiently accurate enough is located in a seat at the rear end of the
because the dynamiC stresses are appreciaDle. r.ull?
The calculation of the dynamiC loading on 1000 lb.
the Wing requires that the mass and st 1tfness Fig. A4. 27 25000 lb.
20"~ L .
distribution or the wing structure be known.
Since these factors are not known when the g. of Crew
structural design of a wing is started, the • ~ Member
general procedure in des tgn would be to first 30"- / __ __ 0"
base the design on the assumption that the wing '1 c. g, 'LJ'
is a rigid body plus correction factors based on r--'oo" --rr-
W: 50000
past deSign experience or available research in-
formation to approx1mately take care of the in-
fluence of the elastic wing on the airplane
aerodynam1C characteristics and the build up
dynwmic inertia forces. ~lth the wing thus in- (4). The jet-plane in Fig. A4.28 is diving at
itially deSigned by this procedure, it then can a speed at 600 X.P.H. when ~ilot starts a 8g
be checked by a complete dynamiC analysiS and pUll-out. Weight of airplane is 16,000 lb.
modif~ed as the results dictate and then re- Assume that engine thrust and total airplane
calculated for the modified elastic Wing. This drag are equal, opposite and colinear.
procedure is now practical because at the avail-
ability of high speed computors. (a) Find radius of flight path at start of
pUll-out.
A4. 15 PROBLEMS. (b) Find inertia force in Z direction.
(c) Find 11ft L and tail load T.
(1). The airplane in Fig. A4.26 is being
launched from the deck ot an aircraft car~ier by
the cable pull T which gives the airplane a for-
ward acceleration of 3.25g. The gross ~eight of
the airplane 1s 15,000 lb. T
;,
~
(a) Find the tension load T in the launching ~
launching time t?
(2). Assume the airplane of Pig. A4.26 is
landing at 75 M.P.H. on a runway and brakes are Fig. A4.28
applied to the rear wheels equal to 04 of the
vertical rear wheel reaction. ~hat is the hori-
CHAPTER A;
BEAMS SHEAR AND MOMENTS
AS.l Introduction.
In general, a structural ~ember that sup- 30"
~orts loads perpendicular to its longltudlonal .1
axis is ~6ferred to as a beam. The structure of
aircraft provides excellent examples of beam
units, such as the wing and fuselage. Very .1----15"'
25 7S lb.
seldom do bending forces act alone on a ~jor
aircraft structural unit, but are accompanied by Fig. AS. 1
g
axial and tors1onal forces. However, the bend- . I OO . P
ing forces and the resulting beam stresses due 10 -1
to bending of the beam are usually of primary :i
~
importance in the design ot the beam structure. N
"
AS.2 statically Determinate and statically Indeterminate >1
Fig. AS. 2 a Fig. AS. 3
Beams. 75 75
A beam can be considered as SUbjected to
known applied loads and unknown supporting re-
100 ... 1
actions. If the distribution of the applied
~
I OO ' P
tclM
known loads to the supporting reactions can be :i
determined trom the conditions of static equil- C .....
H
ibrium alone, namely, the summation of forces d >-
and ~oments equal zero, then the beam is con- T
sidered as a statically d~terminate beam. How- ::; I. 75 I. 75
ever, if the distribution of the known applied Fig. AS. .o4 Flg. AS. S
loads to the supporting beam reactions is in-
fluenced by the behavior of the beam material right side portion as a free body in eqUilibrium
during the loading, then the supporting reactions as shown in Fig. A5.2. For static equilibrium,
cannot be found by the statical equilibrium ~V, ~H and ZM rouEt equal zero for all forces and
equations alone, and the be~ is classified as a moments acting on this beam portione Consider-
statically indeterminate beam. To solve such a Ing ZV = 0 In Fig. A5.3: -
beam, other conditions of tact based on the
beam deta~tions ~ust be used in combination ZV = 75 - 100 = - 25 lb. - - - - - - (1)
With the static equilibrium equations.
thus, under the forces shown, the force system
AS. 3 Shear and Bending Moment. is unbalanced in the V direction, and therefore
A given beam is subjected ta a certain ap- an internal resisting force Vi equal to 25 lb.
plied known loading. The beam reactions to hold must have existed on section a-a to produce
the beam in static equilibrium are then calcu- equilibrium of forces in the V direction. Fig.
lated by the necessary equations of static equi-
librium, namely: -
A5.3 shows the resisting shear force, Vi =
25 Ib, which must exist for equilibrium.
ZV = 0, or the algebraic summation af all verti- °
Considering ZM = in Fig. A5.3, take
moments about seme point a on section a-a,
cal forces equal zero.
ZH = 0, or the algebraic summation of all hori-
zontal forces equal zero.
ZM, =- 75 x 15 + 100 x 5 =- 625In.lb.(2)
ZM = 0, or the algebraic summation af all the or an unbalanced moment of - 625 tends to ro-
moments equal zero. tate the portion of the beam about section a-a.
A counteracting reststing ~oment M • 625 must
~ith the entire beam in static equilibrium,
exist an section a-a to provide equilibrium.
it follows that every portion at ~he beam ~ust Fig. A5.4 shows the free body with the Vi and
likewise be in static equilibrium. Now consider !'Ii acting.
the beam in Fig. A5.1. The known applied load Now ZH must equal zero. The exterr.al
of P ::ill 100 lb. is held in equt Itbr-tua by the two forces as well as the internal reSisting shear
reactions of 25 and 75 Ibs. as shown and are Vi have no horizontal components. Therefore,
calculated from simple statics. (Beam weight is
the internal farces prodUCing the resisting
neglected in this problem). New consider the
beam as cut at section a-a and consider the moment Ml must be such as to have no horizontal
AS.I
A5.2 BEAMS -- SHEAR AND MOMENTS
-+- 5'~5'
to make ZH • O. I' 1 2 '3 6 '1'7 B'I
The tendency of the loads and reactions 1"--4' -----L.... 3' .-!
acting on a beam to shear or move one portion of RA = 1110 lb. RBi", 690 lb.:
a beam up or down relative to the adjacent por-
tion ot the beam is called the External Vertical Fig. A5.6
Shear, or commonly referred to as the beam Vert-
leal Shear and is represented by the te~ V. Calculations for Shear Diagram: -
From equation (1), the Vertical Shear at We start at ~he lett end of the beam.
any section at a beam can be defined as the al- ConSidering a section just to the right of the
gebraic sum of all the forces and reactions 500 lb. load, or section 1-1, and ~onsldering
acting to one side of the section at which the the portion to the left of the section, t~e
shear is desired. It the portion of the beam to Vertical Shear at 1-1 = ZV =_SOO (negative,
the lett of the section tends to move up rela- down on left.)
tive to the right portion, the Sign of the + 610 lb. + 610 lb.
Vertical Shear is taken as positive shear and
negative if the tendency is opposite. Or in .,~ ~b ·3001b,
~
other wordS, if the algebraic sum of the forces
is up on the lett or down on the right Sids,
then the Vertical Shear is positive, and nega- _ 500 lb. _ 500 lb. - 390 lb. - 390 lb.
tive tor down on the left and up on the right.
Fig. A5.7 (Shear Diagram)
From equation (2), the Bending Moment at
any section at a beam can be detined as the al- ~in.lb.
gebraic sum of the moments of all the forces o ~O
acting to either side of the section about the
section. It this bending moment tends to pro- ~ -900 in. lb.
duce compreSSion (shortening) ot the upper fib-
ers and tension (stretching) ot the lower' fibers' -2000 in. lb. Fig. AS. 8 (Bending Moment Diagram)
or the beam, the bending moment 1s classed as a Next, consider section 2-2, just to left of
positive bending moment, and negative tor the reaction RA'
reverse condition.
ZV - - sao, or same as at section 1-1.
AS.4 Shear. and Moment Diagrams.
Next, consider section 3-3, just to ri&~t of
In aircrart design, a large proportion ot RA'
the beams are tapered in depth and section, and
also carry a variable distributed load. ThUS, ZV =-
500 + 1110 =610 (pOSitive, up on
to design or check the various sections of such left side at section).
beams, it is necessary to have a complete piC-
Next, consider section 4-4, just to left ot
ture as to the value ot the vertical shear and 1000 load.
bending moment at all sections along the beam.
It these values are plotted as ordinates trom a ZV =-
500 + 1110 = 610 (same as at
base line, the resulting curves are referred to section 3-3).
as Shear and Moment diagrams. A tew example Section S-S, to right of 1000 load:
Shear and Moment diagrams will be plotted, to ZV =-
500 + 1110 - 1000 390 (down =-
retresh the' students kncw.l edge regarding these on lett).
diagrams.
Check this shear at section S-5 by using the
Example Problem 1. portion of the beam to the right of 5-5 as a
Draw a shear and bending moment diagram tor free body.
the beam shown in Fig. AS.6. Neglect the weight ZV =-
300 + 690 = 390, which checks
at the beam. (Sign of shear is minus, because ZV is up on
In general, the tirst step is to determine right). Section 6-6, use the portion to right
the reactions. as a free body:
To find RB, take moments about point A. ZV = - 300 + 690 = 390 (minus Shear).
ZMA = - 4 x 500 + 1000 x 5 + 300 x 13 - lORS = a
Section 7-7:
hence RB • 690 lb.
ZV =- 500 + RA - 1000 - 300 + 690 = 0 ZV =- 300 (positive shear, down on right)
hence RA = 1110 lb.
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A5.3
RA and consider the forces on the left side only: RA=270 RB=210
Fig. AS.9
ZM = - 500 x 4 = - 2000 in. lb. (Negative 270 lb.
moment, because of tension in the top fibers). 'n-,.,._Vcs180 lb.
V =60 lb. Fig. A5.l0
Consider the section under the 1000 in. lb. load:
A;~ 1--6"
ZM to left = - 500 x 9 + 1110 x 5 = 1050 in.
lb. (positive moment, compressing the top ~ 15" Shear Diagram -210 lb.
fibers). Fig.A5.~.~
~
Check by considering the forces to the right:
ZM right = 300 x S - 690 x 5 = - 1050 in. lb. Moment Diagram
Next, consider a section over RB:
ZM right = 300 x 3 = 900 in. lb. (Negative The vertical shear be~Neen points D and B, when
moment, tension in top fibers). x is the distance of any section between D and
B trom A:
At Section 8-8:
ZM right = 300 x 0 =0 VDB =270 - 120 - lOX - - - - - - - - - - (1)
~~~.at~~~~~t~~al
Resistance at Point
~
tI
I
I
II
0
--I·I
15000 lb.
16"
16"
l
since the right hand portion of equation (5) is "0"
equal to the shear.
Hence, dM = Vdx - - - (6 ) SOLUTION: -
Which means that the difference dM be~Neen To find Vc take moments about point 5,
the bending ~oments at ~NO sections that are a 2MB - - 15000 x 0.5 x 42 + 15000 x 0.866 x 5.77
distance dx apart, is equal to the area Vdx + 20.77 Vc = O.
under the shear curve between the two sections. hence, Vc ~ 21550 lb.
Thus, for two sections Xl and Xu The axial load in member CE therefore equals
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES AS.'
11550/eos 30° or 13330 lb .•'. He = 11550 x 15/26 Bracket _____
= 6660 .' ~[;,""''l--500
To find HA take moments about point D, 1 00 lb. E ,:, II
Bracket
ZMD = 11550 x 15 - 6660 x 10 - 15000 x 0.5 x 32
... 10 HA = O.
hence, HA = 13340 lb.
To find VB take ZFV =0,
ZFv • 15000 cos 30° ... 11550 - VB - O. Fig. AS. 14
hence, VB = 24550 lb.
The axial load in member ED therefore equals
24550/cos 30° = 28360 lb. (compression). The SOLlJrION: -
reaction HB therefore equals 28360 x sin 30
14180 lb.
Calculations of reactions at A and B: -
Fig. A5.13 shows the oleo strut as a free To find VB take moments about paint A,
body with the reactions at A, D and E as calcu- :MA =-500 x 7 - 500 x 6 ... 1000 x 20 ... 1000 x
lated. Fig. A5.l3 also shows the axial load, sin 45° x 10 + 1000 cos 45° x 2 - 22 VB = O.
vertical shear and bending moment diagrams. hence, VB • 999.3 lb. (up).
The bending moments due to applied loads
without regard to bending'deformation of the To r Ind VA take ZV = 0,
beam are usually referred to as the prtmary zv = 999.3 - 1000 - 1000 sin 45 Q 500 + VA = O.
bending moments. If a member carries axial hence, VA = 1207.8 lb. (up).
loads additional bending moments will be pro-
duced due to the axial loads ttmes the lateral To r inc HB take ZH = 0,
deflection of the beam, and these oe, :ing mo- ZH = - 500 ... 1000 cos 45° - HB 0, hence =
ments are usually referred to as seC)ndary bend- HB • 207.l.
ing aomerrts , (Arts. A23-30 covers t.he calcula- With the exception of the 1000 lb. load at 45°,
tion of secondary moments). all loads are applied to brackets which in
turn are fastened to the beam. Therefore the
13340
next step is to find the reaction of the loaded
brackets at the beam centerline support points.
6660
F7/?/?/'?//hi'T,;==:z:z:J~~~:am
133~O
e-e.Compz-eaefon
,o-rl ,'.
~:84b
-7500 lb.
~~'Moment
Diagram
Bending
,. !I
2'j
50
8"
L MlY'40ci~P"'500
Fig. b
...
1000 5"
HF"'500
VF"'I000
F1g. AS.15
FIg. c
000
L~~'O'.1.
~~.,
E-
Fig. d
,,-:.
,,<;:)
~
The load of 1000 at 45° and applied at point E" have rraximum peak moments without the Vertical
will be referred to point E the centerline of Shear paSSing through zero. To illustra:e
beam. Fig. d shows the reaction at E due to the this tact, consider the bean of ?~g. A5.20,
load at E'. The reaction at B should also be namely, a Simple supported beam with an ex-
referred to the beam centerline. Fig. A5.16 ternally applied couple ~o~ent of 10 1n. Ij.
shows the beam with the applied loads at points magnitude at paint C the center point of the
C 0 E' F and 8'. Figs. AS.17, 18 and 19 show beam. The shear and bending moment dlagr~~
the axial load, vertical shear and bending mo- are as ind~cated and a maximum bending ~ome~t
ment diagrams under the beam loading of Fig. occurs at C but the shear diag~am does not pass
A5.l6. through zero.
500 707.1
,....-..,MC: 1O" -II'
t,
3000 4000 'e '
e -0 I
'R:l
10"
1207.8
-a l I Shear Dia.
±'"
Fig. AS.16
Bending
-500 lb. -707.1 Axial Load Dia.
Fig. AS.20
~"-II' Moment Dia.
Fig.A'.lS
d -
-----
707. S,-,lb",,-. _
bl ==~O. 7
[I-
Shear Dia.
A couple is two equal and opposite forces
not in the same straight lir.e. Let it be as-
sumed that the 10 in. lb. couple is made up of
torces equal to 100 lb. each and an a~ Jetween
-~ -999.3----' them at 0.1 inch as illustrated in Fig. A5.21.
4285 5788.7
3000
Bending 1~ ~O 1"
Moment Dia. e
I
-207.1
'1 + lIb.
Fig. AS. 19 11
Fig. AS.21
U99 Shear Dia.
........---
500
Final Moment Dia.
-- ceterainet;e because there
are only 3 unknown elements,
namely the magnitude and di-
rection ot the reaction at A
Pa ::rIO
!!. =10
Fig. A5.-22 Fig. AS. 23 and the magnitude ot the re- P..",10
action at B. For conven-
ience the reaction at A is
support at the right end. The final bending
resolved into two magnitudes Pin~HA"'20 B
moment curve for the true given beam then equals
the sum of these separate diagrams as illustra- as H and V components. The
reactions VA' HA and VB can ""A"'10 B::r20
ted in the figures. Fig. AS. 27
then be found by statics and
STATIC MOMENT CURVES IN SOLVING STATICALLY the results are shown on Fig. A5.27. Fig. A5.28
shows the bending moment diagram on this frame
INDErERMlNATS STRUCTURES due to each load or reaction acting separately,
The usual procedure in solving a statically starting at A and going clockwise to B. Fig.
Indete~1nate structure 1s to first make the A5.29 shows the true bending moment diagram as
structure statically determinate by removing the the summation at the separate diagrams.
necessary redundant or unknown reactions and -=::;::::::I-50-Due to P a
then calculating the deflection of this assumed - 60 -60
o ?'--'"'
statically determinate structure as one step in .....1J.e v t->" Due to V ..J~I00
y A
the overall solution of the problem (See Chapter
AS). In the solution of such structures it is
likewise convenient to treat the bending moment
diagram as made up of parts. To illustrate,
Fig. A5.24 shows a loaded rectangular frame
fixed at paints A and B. The reactions at both
Due to P li
Due to P ,
--=::::J -50
-601 \-60 Fig. A5. 28
j~nu"~
~"r-_,,'O
'i
.I'"
180 30 in. lb.
..1 B Final Bending
-60 Fig. AS. 24 Moment Diagram
-110 Free 60 - 50 90 (Tension on inside
-6 .110 of frame is posi-
-80 FIg. AS. 25 tive moment)
A B
Fig. AS. 29
5" A5.8 Forces at a section in Terms of Forces at a
100 ...L
Previous Station.
STRAIGHT BEA11S
paints A and B are unknown in magnitude, di- Aircraft structures present many beams
rection and location, or each reaction has 3 un- which carry a varying distributed load. tlinl-
known elements or a total at 6 unknowns for the mum structural weight is at paramount importance
two reactions. With 3 static equilibrium in aircraft structural design thus it is de-
equations available, the structure is statically sirable to have the complete bending moment
indeterminate to the third degree. Fig. A5.25 diagram for the structure so that each portion
illustrates one ~er in which the structure of the structure can be proportioned effic-
can be made statically dete~nate, by freeing iently. To decrease the amount of numerical
the end A to make a bent cantilever beam fixed work required in obtaining the complete shear
'( i:t",
~.
A5.8 BEAMS -- SHEAR AND MOMENTS
MlC
(1)
Lr~v"OM" (2)
Fig. AS. 32 Fig. AS. 33
Fig. AS.31
Then from Fig. A5.33 we can 'Nrite for the
resultant torces and moment at point (2) at
station 2: -;
at the beam portion 8evNeen stations land 2.
The Vertical Shear V1 at stat~on 1 equals the
summation of the forces to the lett of station 1
and M1 the bending moment at station 1 equals
the algebraic sum of the moments at all forces
lying to lett of station 1 about station 1.
Now conSidering station 2: - The Vertical
Shear V" = Vl + F 1 _ " , or stated in wordS, the
Shear V" equals the Shear at the previous sta- Having the resultant forces and moments for a
tion 1 plus the algebraic sum of all forces F given point on a given station, it is usually
lying between stations 1 and 2. Again consider- necessarJ in :inding beam stresses to resolve
ing Fig. A5.31, the bending moment M" at station the forces ~nto components no~~l and parallel
2 can be written, MOl = Ml + Vld + Fl_"a, or to the beam cross-section and also tr~sfer
stated in words, the bending moment tl" at sta- their location to a point on the neutral axis
tion 2 is equal to the bending moment Ml at a of the beam cross-section.
previous station 1, plus the Shear V at the For examp ;e Fig. A5.34 shews the resultant
pr-evious station. 1 t1.mes the arm c , the dist-
ance between stations 1 and 2 plus the moments
at all forces lying be~Neen stations 1 and 2
about station 2.
AS.9 Equat.1an.s for Curved Beams.
Many structural beams carry both longitud-
ional and transverse loads and also the beams
may be made of straight elements to r orn a frame
or all beam elements may be curved to form a
sP W v
Fig. AS. 34
s
Fig. AS. 3S
" "J6 ~~
Mo.:Ml wNe
s
Fig. AS. 36
curved frame or ring. For example the airplane
tuselage ring is a curved beam SUbjected to
torces of varying magnitude and direction along
its boundary due to the action or the fuselage
skin torces on the frame. Since the complete
bending moment diagram 1s usually desirable, it forces and moment at point 1 of a oe~ cross-
is desirable to minimize the amount of numerical section. They can be resolved into a normal
work in obtaining the complete shear and bending terce N and a shear for S ~lus a moment Ml as
moment values. Fig. A5.32 shows a curved beam shown in Fig. A5.35 where,
loaded with a number of different vertical loads
P and horizontal loads Q. rig. A5.33 shows the N = H cos a + V sin a
beam portion 1-2 cut out as a treo bOdy. Hl S = V cos a - H sin a
represents the resultant horizontal force at
statton I and equals the algebraic summation ot Later on when the beam section is being de-
all the Q forces to the left of station 1. Vl s igned it my be round that the neutral axis
represents the resultant vertical torce at 11es at point 0 instead of point 1. Fig.
station 1 and equals the sum of all F :orces to A5.36 shows the forces and moments referred to
lett of station 1, and Ml equals the bending point 0, With Mo being equal to Ml - Ne.
moment about point (1) on station I due to the
moments of all forces lying to the left of
point 1.
• s
to carry out a certain Job or duty and to do The above unit load conditions are for con-
that Job requires a certain ~lmum airplane ditions ot acceleration in translation of the
A5.I0 BEAMS -- SHEAR AND MOMENTS
airplane as a rigid body. Unit load analyses ratio or the spanwise variation of the lift
are also made for angular accelerations of the coeff~cient C
L in terms ot a uniform dIstrIbu-
airplane which can also occur in flight and tion CL' In Ehis example we have ~aken thIs
landing maneuvers. ratIo as unity sInce we have no wind tunnel or
The SUbject of the calculation of loads on aerOdynamic calculatIons for this wing relative
the airplane is far too large to cover in a to the spanwise distributIon of the lift force
structures book. This subject is usually cover- coeffIcient. In an actual problem involvIng an
ed in a separate course in most aeronautical airplane a curve such as that given in FIg.
curricula after a student has had initial A5.36 would be available and the values to
courses in aerodynamics and structures. To il- place in Column 3 of Table AS.l would be read
lustrate the type at problem that Is encountered from such a curve. Column (2) gives the Wing
in the calculation of the applied loads on the chord length at each station. Column (4)
airplane, simplified problems concerning the gIves the wing running load per inch of span
wing and fuselage will be given. at each station Doint. Since a total unit
load of 17760 lb. was assumed acting on the
AS.12 Ezample Problem 01 Calculating Wing Shears and half wing and since the wIng area is 17760 sq.
Moments for One Unit Load. Condition. in., the running load per inch at any station
Fig. A5.37 shows the half wing plantorm of equals the Wing chord length at that station.
a cantilever wing. Fig. AS.38 shows a wing In order to find shears and moments at the
section at station O. The reference Y axis has various station points, the distributed load is
been taken as the 40 percent chord line which now broken down into concentrated loads which
happens to he a straight line In this particular are equal to the dIstrIbuted load on a strIp
wing layout. and this concentrated strip load is taken as
acting through the center of gravity at this
distributed strip load. Columns 5, 6, and 7
show the calculations for determining the
(~Pz) strip loads. Column 8 shows the lo-
cation of the ~Pz load which is at the centroId
ot a troplzcidal dIstrIbuted load whose end
Yalues are given in Column (4). In determin-
ing these centroid locations it is convenient
to use Table A3.4 of Chapter A3.
The values of the shear Vz and the mo-
ment Mx at each station are calculated by the
method explained in Art. A5.S. _Columns 9, 10,
11 and 12 of Table AS.l gIve the calculations.
For example, the value of Mx = 9884 in Col-
umn (12) for statIon 220 Bquals 2436, the Mx
xL.
moment at the preViOUS station in Column (12)
plus 4908 in Column (10) which Is the shear
aer~mic J at the preVIous statIon (230) tImes the dist-
center a.c" 0' ...... Ref.
ance 10 Inches plus the moment 2540 in Column
I i 96" I Fig.A5.38 (9) due to the strip load be~Neen statIons
"? Ref. Axls 230 and 220, which gives a total of 9884 the
value in Column (12).
The strip loads 6Pz act through the
The total wing area Is 17760 sq. In. For
convenience a total unit distributed load of aerodynamic center (a.c.) ot each airfoil strip.
17760 1hs. wIll be assumed acting on the halt Column (13) and (14) give the x arms which is
Wing and acting upward in the Z direction and the distance trcm the a.c. to the reference Y
through the airfoil aerodynamic center. The axis. (See Fig. A5.38). Column 15 gives the
spanwise distributIon at this load will be ac- My moment for each strip load and Column 16
cording to the (CL) 11ft coefficient spanwise the My moment at the various stations which
distributIon. For simplicity in this example equals the summatIon of the strip moments as
it will be assumed constant. one progresses fram station 240 to zero.
Table AS.l shows the calculations in table Fig. A5.39 shows the results at station
form tor determinIng the (Vz) the wing shear in (0) as taken from Table A5.1.
the Z direction, the bending moment Mx or mo-
ment about the X axis and My the moment about
the Y axis tor a number of stations between the
wIng tip statIon 240 and the centerline station
O.
Column 1 of the table shows the number ot
stations selected. Column 2 shows the CLIC
L Fig. A5. 39
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A5.11
TABLE A5.1
=
. .•""
C
L/_
CL
• · i ..-
~
.->
Ratio,
Assumed
~L
Unity.
a o
II
~;:
g.
Q,,"''S
~g
o 7.2
.u
>-
~g
.
..·· . :;>-:8
.-
~
~.
~S...;
. s-:
o
'40
"
-49.0 1.0
"
49.09
-48.54 >.0 <2. 2.49
242.7 o 60s 605 7.36
7.28
1787 1767
49.63 .0 48.1 2.49 7."
230 50.18 1.0 50.18 -490.8 1213 618 2438 7.53 1846 3613
51.27 10.00 512.7 4.97 7.89
220 52.36 1.0 52.36 1003.5 4908 2540 9884 7.85 '940 7553
54.00 1 .00 810.0 7.4 11. 09
205 55.64 1.0 55.6'" 1813.5 15052 6020 309S8 8.34 65'iO 14093
57 27 8 9
7 "
190 58.92 1.0 58.92 2672.5 27020 6490 64488 8.85 7390 21483
60.56 15.00 908.4 7.43 9.09
175 62.20 1.0 62.20 3580. 9 40087 6760 111313 9.33 8260 29743
.84 1.00 .4J
160 6 ., 1.0 65. '" 4538.5 53 100 17 1 9170 38913
67.13 1 .00 1007.0 7.43 10.06
14' 68.78 1.0 68.78 5545.5 68077 7480 247683 10.31 10100 49013
70.39 1.00 1055.5 7. -43 10.55
"0 72.00 1.0 72.00 6601.0 83782 7840 338705 10. 80 11100 50113
73. 15.00 1104. 7. "'3 11.04
75.23 1.0 75.23 7705.3 99015 8200 445920 11. 28 11480 71593
76.89 15.00 1153.4 7.43 11.54
I 100 78.55 1.0 78.55 115579 8570 570060 11.80 13270 84863
80.18 15.00 1202.7 7. i3 12.04
as 81. 81 1.0 81. 81 10061 132880 8940 702949 12. 28 14480 99343
83."'8 1.00 1252.2 7.43 12.52
70 85.10 1.0 85.10 11313 150915 9310 863174 12.76 15600 114943
1 .00 1300.9 .4J 13.01
ss 88.35 1.0 88.35 12614 169700 9660 1042534 13.27 16930 131873
89.98 15.00 1349.7 7.42 13.49
91. 62 1.0 91. 62 139 4 189210 10000 1241744 13.71 18200 150073
9 • 1 .00 1 9. a 14.05
o .00 1.0 . 0 ie , 9 1 00 1539 4 14.40 26350 176423
96.00 .00 1920.0 10.00 14.40
98.00 1.0 96.00 17760 318808 19200 1875630 14.40 27848 214011
Sum : 17760 Cheeks Total Limit Load. Assumed all HaJ! Wing.
l~en the time comes to deSign the structural !orces are the major forces. For the body load
make-up of a cross-section to withstand these analySiS the direct air pressures are secondary
applied shears and moments, the structural de- the major forces being of a concentrated nature
Signer may wish to refer the forces to another in the iorm of loads or reactions from units
Y axis as for ex~ple one that passes through attached to the bOdy. as the power plant, wing,
the shear center of the given section. This landing gear, tail, etC. In addition, since the
transfer of a force system with reference to an- body usually serves as the load carrying medium
other set of axes presents no difficulty. Unportant forces are produced on the body in re
sisting the inertia forces of the weight of the
SHEARS AND ~OMENTS eN AIRPLANE BODY interior equipment, installations, pay load etc
As in the case of the wing, a large part
AS. 13 Introduction.
of the load analysiS can be made without much
The body of an airplane acts essentially as consideration as to the structural analysis of
a beam and in some conditions of flight or land- the bOdy. The load analysis of an airplane bOdy
ing as a beam column which may be also subjected involves a large amount of calculation, and
to twisting or torsional forces. Thus to deSign thus the treatment in this chapter must be ~t a
an airplane body requires a complete picture of slmpli:led nature, and is presented chiefly fer
the shearing, bending, twisting and axial ~orces the purpose of Showing the stUdent in general
which may ~e encountered in flight or landi~g. how the problem of load analysis for an air-
Tn the load analysis for Wings, the direct air plane bOdy 1s approached.
I _
..,(
A5.12 BEAMS -- SHEAR AND MOMENTS
A5.14 Design Conditions and Design Weights. AS.i5 Body Weight and Balance Distribution.
The airplane body ~ust be designed to with- The resisting inertia ferces Que to the
stand all loads ~rom specified ~light co~ditions dead weight of the body ~nd its contents plays
for both maneuver and gust conditions. Since an important part in the load analysis for the
accelerations due to air gusts vary inversely airplane body. ~~en the initial aerodyna~ic
as the airplane weight, it customary to analyze and general layout and arrangement of t~e air-
or check the body for a light load condition for plane is ~de, it is necessary that a complete
flight conditions. :n general, the design weight and balance estinate of the air?lane be
weights are specified oy the government agen- made. This esti~te is usually ~ade by an en-
cies. For landing conditions, hcwever, the gineer from the weight control ~ectlon of ~he
normal gross Hei~~t is used since it would be engfneer tng cepar tmerrt who has had experience
more critical than a lightly loaded condition. in estimatin~ :he wei~ht and distribution of
The general design conditions which are airplane units. 7:'i8 estimate which is pre-
usually investigated in the design of the body sented in report fo~, gives the wei;hts and
are as follows: (e.g.) locations of all major air;lane units
Flight Conditions: or instal:ations as well as for many of the
minor ~~its which nake up these ~jor airplane
H.A.A. (High angle of attack) assemblies or installations. ~~is weight and
L.A.A. (Low angle of attack) bal~~ce report forms the basis for the dead
I.L.A.A. (Inverted low angle of attack) weight inertia load a~alysis which forms an
LH.A.A. (Inverted high angle of attack) important part ~n the load analysis of the air-
The above conditions generally assume only plane body. The use of this weight and balance
translational acceleration. In addition, it 1s estimate will be illustrated in the ex~~ple
sometimes specified that the forces due to a problem to follow later.
certain angular acceleration of the airplane
AS.16 Load Analysis. Unit Analysis.
about the air~lane e.g. ~ust be considered.
The bOdy is usually required to withstand Due to the many design conditions such as
speCial tail loads both symmetrical and unsym- those listed in Art. A5.14, the 5eneral pro-
~etrical which may be produced by air ~~sts,
cedure in the load analysis of an airplane body
engine forces, etc. Also, the body should be is to oase it on a series of unit analJ~es.
checked for forces due to unsymmetrical air The loads fo~ any part1cular deSign condition
loads on the wing. then :cllows as a certain combination of the
unit results with the proper ~ultlplylng fac-
Landing Conditions: tors. A Simplified example problem follows
In general, the body is investigated ~or the which illustrates this unit method 8f approach.
following landing conditions. 'The detailed re-
AS.17 Example Problem lllustrating the Calculation of
quirements for each condition are given in the
Shears and Moments on Fuselage I:e to Unit
government specifications for both ~illtary and
Load Conditions.
commercial airplanes.
Fig. A5.40 and A5.4l shows a layout of the
LandpLanes : Level landing. airplane body to be USed in this example er-e-
Level landing with side load. Lem, It happens to be the body of an actua...
Three point landing. airplane and the wing used in the preViOUS ex-
Three point landing with ample problem Nas the wing that Nent with the
ground loop. airplane.
Nose over or turn over II
condition.
Arresting. (Usually for only
Navy Carrier based air-
planes) .
($'-1\"1~=r=~~~I-1 hrust Line
• _. - -- - _ £.jBo!_.!..xi.»... - - -
Seaplanes or Boats: ~c sc. G. (groSS wt. ) I 11'2"
Step landing With and without ~
~
= I I
I
angular acceleration.
:1
I '
i
TABLE A5.2
a
~
'.m
1
3
N=.
PawerpiaDt grOlq)
Fuselage groull
w."",
y
1100
-
Boriz.
,
19
113.5
Bori:l:.
""'m.
~
30900
39700
Vert.
.um
,
o
Vert.
""'m.
~
Reference
Vert. (Z) arms Ill_red from thrUst line
(+ II up)
; ,• Tail group
'50
11' 38'
121
31550
"
1
'"
3690
,•
-1190
'30
, , " 118
10700 -10 - 350
'.m
No. N=. w. I
wt. it
Boriz.
.um
Bortz.
Moment
Vert.
.um
Vert.
Mom,
"~
~
10
Furn1ab.iDgs
"'"'"
Wela:ht em~ty •
'"
136
d55
191
25520
22600
1t197oo
10
~
1100
19
~
20900
(Z)
n
~
o
'50
~
0
. 11
12
Plio<
"""~,
GroSl weigllt " Z
200
200
es
151
"Z
33200
19800
19700
••
Z
'00
SO,
a Fueelaie Gro~ 35' 113.5 311700 1
"'"
3
•s
WiD( Grou-p
Tail Grou~
SDrface Controls
'50
11'
ae
" 38'
72750
31550
10800
-18
"•
-1:ui00
'04'
_ 1190
Emllty ~ z 169700 • 78.7"
""
Z.~.2..00"
", '""
-l<
e 133 , 7930 2~~~ ". ~
- '"
Electrical System With \18efUlload. i • • BG.O" • 2.32"
t Chu;ais Front 73 16450 -52 -12200
,a Tall Wheel Gro~ 35
'" 10700
'50"
-10 35'
5 I 1100
10
FuruilItw1ga
Radio '"
'" 131
11'
22600 I 10 "50
11
ra
13
Weipt emllty •
PUot
Student
Fuel Syllt.em
,,
Grosl weight ·1
,
3150
20'
'00
'"
..
151
88 ,
"8900
'0200 [
196OO! "
67600 ) _27
·21380
• ."
." _20500
weight distribution should study the inboard
prOfile drawing of the airplane which shows the
general arrangement of all the installations and
4300 37~00 ~40280
equtpment , Furthermore, he should study the
overall structural arrangement as to its possi-
.CalcuJ.ation of C, G. locations:
GroSI wt. x z 376500/4300 z 98,S" a,ft of Re! • .Mia ble influence on fuselage weight distribution.
i z. z -40280/4300 • 9.4" below line thrust
The whole process involves considerable common
sense if a good apprOXimation to the wei~~t d1s
SOLUTION: trlbutlon is to be obtained. Fortunately the
large dead weight loadS, such as the power
WEIGHT At'IITJ BALANCE OF BODY ITEJ"!S. plant, tail, etc. are definitely located, thus
~~IGHT DISTRIBUTION. small errors in the distribution of the minor
distributed weights does not change the over-
Table A5.3 gives the weight and balance all shears and moments an appreciable amount.
calculations for all items attached to fuselage In order to obtain reasonable accuracy, the
or carried in t~e fuselage, except the wing and fuselage or body is diVided into a series of
items attached to the wing as the front landing stations or sections. In Fig. A5.42, the Sec-
gear and the fuel. tions selected are deSigned as stations which
In order to obtain a close approXimation to represent the distance from the Z reference
the true shears and ~oments on the fuselage due axis. The general problem 1s to distribute the
to the dead weight inertia loadS, it is neces- concentrated loads as shown in Fig. A5.43 into
sary t~ distribute the weights of the various an equivalent system acting at the 7arious
items as given in Table A5.3. Fig. A5.42 shows fuselage station pOints.
a side view of the airplane with the center of Obviously, if a weight item trom Table A5.3,
graVity locations of the wei~~t items of Table represents a concentrated load such as a pilot,
A5.3 indicated by the (+) Signs. In the various stUdent, radio, etc., the weight can 08 dis-
deSign conditions, the direction of the weight tributed to adjacent station points inversely
inertia forces changes, thus it is convenient as the distance of the weight (c.g.) fram these
and customary to resolve the inertia forCes into adjacent stations. However, for a weight item
X and Z components. ThUS, in Fig. AS.43, the such as the fuselage struct~e (Item 2 of Table
weights as given in Table A5.3 are assumed act- AS.3) whOSe e.g. location causes it to tall be-
ing in the Z direction through their (c.g.) lo- ~Neen stations 80 and !20 of Fig. A5.43, it
cations. The loads as shown would not give a would obViously be wrong ~o distribute this
true picture as to the shears and ~oments along weight only to the two adjaCent stations since
the fuselage, thus these loads should je dis- the wel~~t of 350= 1s for ~he entire fuselage.
tributed in a ~nner which should si~ulate t~e This weight item of 350~ shou1 4 thus Je 11s-
actual weight distribution. In ~ost weight and tributed to all station points. The contrOlling
balance reports, the weight items are broken require~ent on this distribution is tr~t the
down into considerable ~ore detail than that moment of the distributed system about the ref-
shown in Table A5.3, which ~kes the wel&~t dis- erence axes must equal the moment of the orig-
tribution ~ore evident. The ~erson maKi~g the inal Height about the s~e ~~es. Fig. A5.44
A5.14 BEAMS _. SHEAR AND MOMENTS
STATION
so 80 120 170 200 220 260 290 315 0 11 50 80 120 170 200 230 250 290 315
I
i 1 I I I I I
I I I
I I 110 I
!I Fig. AS.42. Location of weight items of Table AS. 3. Fig. AS. 46. Weight items from Table AS. 3 acting in X direction.
Z.
011 50 80 120 170 200 220 260 290 011 50 80 120 170 200 230 260 290 315
I
I I I 1 I I I 1
I
I I
I I I 1
I I
'" I
~
1- ~ --1-r1--ig- T
I -
,,I
.... ~
.... '" ~
~M
I
l~
<:cl N
I
-
'"
N
~; i
, ,
I I I
Fig. AS.43. Weight items of Table AS. 3 acting in Z direction. Fig. AS. 47. Vertical distribution of fuselage dead weight.
011 SO 80 120 170 200 220 260 290 011 so 80 120 170 200 230 260 290 315
I 1
I
I
I I i
I I
I I
I I' i
I I
I I
" I I I I
190
--II
80< 780 6511
j -
----i!-- j 30# 100
I
rpl 90
90 .-r-80~
3:0# 2':34#
I I
I I
Fig. AS. 44. Results of fuselage weight distribution to stations. Fig. AS.48. Fuselage weight referred to X axis plus couples.
SO 80 120 170 200 220 260 290 215 o 11 50 80 120 170 200 230 260 290 315
I I I I I I I
I
I ' I I I I I I I
I I I I I I I
I I I I' I
',9'
409*
I
311*
I
7~#
I
100 11191
11i93
8 758 882 1850 '90 10 '274 - 2
-(_288(;-307 -(-'09-C-311G"C- 10 (_21 'tf: 12e
I I
Fig. AS.45. Final weight distribution to station points. Fig. AS.49. Final weight distribution in X direction referred
to X axis plus proper couples.
ANALySIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A5 15
shows how the dead weight of 350i: was distribu- aanner as to have the same resultant e.g. 10-
ted to the various station points considering cation as the original we Ight system which [s
the weights to be acting [n the Z direction. illustrated in Fig. A5.46. Fig. A5.47 illus-
Table A5.4 shows the results of this sta- trates how the fuselage weight distributed
tion point weIght distribution rcr the weight system as shown in F[g. A5.44 [s distributed
item.s of Table A5.3. The values in the hori- in the vertical direction at the various
zontal rows opposite each weight item shows the station points so that the moment at this sys-
distribution to the various fuselage stations. tern about the X axis [5 equal to that of the
The summation at the weights [n each ver-t tca.L original fuselage weight or 350#. For con-
column at each station point as given in the venience, these distributed fuselage weights
third horizontal row rz-cn the bottom or the can be transferred to the X axis plus a moment
table gives the final station point weight. as shown in Fig. A5.48.
These weights are shown in Fig. A5.45 for Table AS.4 shows the vertical distribution
weights acting in the Z direction. The moment at the various items at the various station
at each total_station load about the Z axiS [s points. The bottom horizontal row gives the
given in the second horizontal row tram the moment about the X axes at the loads at each
bottom of Table A5.4. The summation at the station point, which equals the indiVidual
moments in this row aust equal the total WX loads times their Z distances. The summation
moments of Table A5.3 or 219700"*. This check at the values in this horizontal row must equal
[s shown in the last vertical column of Table the total wz aoment ot Table AS.3. This check
A5.4. is shown at the bottom of the last vertical
The distributed system must also be distri- column. Fig. A5.49 shows the results as gtven
buted [n the Z or vertical direction [n such a in Table AS.4 for the weight distribution in
the X direction.
TABLE ~ ••
,,
roUD ,
, , - ,- , ""
urlace contro
, ."
II
, "
io
I ElllCtrical s !item
Radio
Furni •
raUl)
0
12$ i ,
• •
,
to ..
rs rc •
,
,
,
, -" , , .10 I
,
.W
".""
"", ,
•
...
, , ,
U i Pilot .OO
"",
." "'.
ra ""'.m ,
• 2.47 , " • au I te 10 1.0 11 uml ,'oo
"'''"
2555 I 893 .02
ToW ..
Horlo;o"l::I,I MomlMlt • :. ""'. 9820 r
3881
\9390 "" 24560 I
."
491001
1.67
52800
5. J
isace
6.45
"""I
1181 19.1
134200 I
.",
A5.18 Unit Analysis for Fuselage Shears and Moments. (1) Unit acceleration or load factor tn Z
Since there are many flight and landing direction and acting up.
conditions. considerable tl.:ne can be saved i t a (2 ) Unit acceleration or load factor [n X
\mit analysis is made fer the fuselage shears.
direction and acting rorwarc .
axial and bending forces. The design values in
general then folloW as a sunnnation of the values (3 ) Unit tail load normal to X axis acting
in the \m[t anaLys t s t imes a ~roper multiplica- down.
t~on factor.
The loads on the fuselage [n general con- Unit analyses are also usua.Ll.y carried out
sists ot tan loads. eng:!.ne loads, Wing re- for engine thrust and engine tor-que , side load
actions, landing gear reactions If attached to on ~ll and angular acceleration, but to keep
ruse rage and tner-t i.a fore es due to the airplane the example calculations from becoming too
acceleration which ~y be due to botn trrans Ia-. lengthY only the above 3 unit conditions will
tional and an~ular acceleration of the airplane. be carried out in detail. The others will be
For s 1.:nplic1t.y, these loads can be resolved into discussed in detail [n later paragra phs •
cGIt::onents parallel to the Z and X axes.
To illustrate the unit analysis procedure. Solution for Unit Load Factor in Z Direction.
a unit analYSiS tor our example problem will be
carried out for the following unt t conditions: Fig. A5.50 shows the dead weight loads
A:) 16 BEAMS -- SHEAR AND MOMENTS
•
ZF z . - 2555 + £iF + RR • 0 - -- - --- -- (3 )
290
- -
0
118
22
140
I
30
! - 550
I
- 550 j
VN
I I
I I
I i
I I I I I I I I
4
I
I
I
I
~
I
I
I I i
I :
'
I
I
I
22
' 116 •
-
0
77' 1I
967
192 i
36J
- 3668 I, -53888
I
II
,I ,
* ·-
I
893 388 307 409 311 76 19 21 ~ 0 192
. - --.--.: .-t--. --t-->: - - t - . +- •- +- . .- 80
- 307 499 , i
,
- 6912 1
-60800
I 17" 20" I
.I '--- ~--- --f'-7::. •- 0 499
7
I
I
i:i' RW O .~.43"~,R",775 I
73 1780 -1281
I
- 3493 I -64~90
Ry=1780 Fig. A5.50
"" I 23
-•
STA.73 STA.1l6 388 -1281
'0 -29463 -34827
0 - 893 I
39
Solution for Unit Load Factor in X Direction. II •
-
893
0
- 693
a
-348~
Fig. AS.51 shows the panel point dead (3) Up on left and down on r-ight side of a section IS I
weight distribution for loads acting in the X pcemve shear.
direction and att, as taken from Table A5.4 or (8) Tension in upper fuselage portion is negative bending
Fig. A5.40. To place the fuselage in equlli- moment.
brlum the wi::lg reactions at points (A) and (8 ) (I) •
refers to aft side of station.
- refers to forward side of station.
will be calculated.
• M = M at previous station in col. 6 plus 1M in col. 5•
lF x = 2555 - RH = 0, hence RH = 2555 lb. (forward)
Take momentis about point (A)
II 80 120 173 200 230 260 290 31'
0 I 50 I I
5920 - 4, RR = 0, I I I I
lI1A ,. 2555 x 17 + I I I
i I I I
I
i I i i
hence RR - 1147.8 (up) I I I ;",
I
I
I
r
i
I
'\
I
I
(5920 equals the sum of the couples from Table 19 I,'
~93t ..(-388(-307 (-409
758 662 1650 490 10 274 2249
(-311(-760-10 (- 21 (.....],[3.<:-2 2
A5.4.
1. R':=t~~ ___.B
I
220
I -4' ---- 43"
lF z = 1147.8 - RF = 0, I
hence Rr = 1147.8 lb. (down) I Rp1l47.8 iRR"'1147.8 Fig. AS.51
STA.73 STA.116
Table A5.6 gives the calculations for the shears,
moments and axial loads for the loading of Flg.
,- ,
.<>..:>.5....
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A5 17
TABLE AS.S 1
TABLE A5.7
fUSELAGE SHEARS, MOMENTS" AXIAL LOADS FOR
FUSELAGE SHEARS &; MOMENTS FOR UNIT
ONE LOAD .-ACTOR IN X DIRECTION
I
, a , a I
lnerua Loada Acting Alt
s • ; a , s
HORIZONTAL TAn.. LOAD IN Z DIRECTION
...
,
, (Load Acting Down)
i w" : :w::iP":.~w" v • z w: "'Mt = r c x • I.::.M>l·- I
mX ,load. ~
'"" •
Couple i D,S!..
I
So. i 'wa.l SlIear Moment Ibetween vt:,,: i
1 2 3 4 5 6
Dlr. !ZDlr. LGad ;htloD. I
, , Load or V , shear ~X :: di st,
o: Moment
" u
,,-- ~M
c
-j ,I· c
1 315 •
! . 22.! '1 "aa 220 !
1
1
I
22g I sta. Reaction
w lbs.
: Z w
Ibs.
between
stations , V~X in. tbs.
as
t asc : I 11~ I cI ,<0 u • 22411 1
o' sac
- 2029
,I
2l\(J • i
-, " ,I ,<0
0
I
- 37~ I
30
1
I c 1_ 2029
,- 2303
315 -- 0
0
0
0
0
0
0
2li
,., '" 1 30
25
230 : i
,
o:
10 i
"
161
m c 1
1
-
c !
10 !
30
c :-
2303
• _ 2313 1 290 ·1
- I
0
0
0
0
0
0
0
L
,
12.5
200
-.! 01 rn c 1- 49~
, , c ~_23131
,'.38031, 0 0 0
,,
" ! u1
'" 272.5 0
170 : I
,I
311 : oI ...
'" i
1
o 1
,
· l6Sg L
30
su
!
I
,
01.28031
:- 445 3 1
=1
- I
-100
0
100
100
17.5
- 1750 -- 1750
a
I 'I 358
· 68~ I
c i [- 44531 260
- 0 100 1750
'I
120 _ 40IIj
o o !_ 3135 1
'"
0
4 , 30
• - 3000 - 4750
,
~ :! '" ,
,
i-' - SUS 0 100 4750
-
115
: I
I
1147.6!
'"sat , -IH7.6
_1141.8
, 1
ae
Q
5135
1 230
- 0 100
30
"" ~:~:; i,
60 •. Q 1 41320
- 3000 -- 7750
,
- rse !
:i
3071
c
1274
1214
_1147.8
I
-1147. a i
_4343~ ~- a035
Q
,
43462 i
200 -- 0
0
100
100
7750
I
73 2555'-1147.al " -1281 u 7
o 27 ! 30
, " • 0 100 ·10750
1
50 ::
3861
o1 01
-t281
- a93 !
o I :
'
I
I
c ~I
170
- 0
I 100 I -
3000 -10750
.- a9~ I
,
I I~ " i
50
: "1" •
!, u - 893 o ,I c 0 100 -15750
"I o 1
I
I
120
- 0 100 I- 5000
-15750
1 • :' 17 x (2555) : 4343:5
•
1 Col.
,..
• refers to alt aide 01 staJ:ion
1 _ refers to forward side of station
te pJ,u lor ten!lion in -fllHlage.
116 -- 0
475.6
100
-375.5
36
I! - 400
-16150
-16150
i ICoL S) M • M iU ~rl!Vlou. station In CoL. 9. ll1u .; Ml of col. 6 ~l...
-- .- 2629
:'M2 ill 001. a I 0 -37!5.6 i
80 13521 2629
0 -375.6
7
Solution for Unit Horizontal Tail Load Acting Down. a
73 -- 0
-375.6
-375.6
0
23
2629
0
The fuselage shears and :noments will be
C8mputed for a unit tail load of 100 lb. on the 50 -I
-,
0 0 0
tail acting in the Z direction, with balanc Ing , 0 0
39
0
reactions at tna wing attachment pOints. The
-i-I
0 0 0
center of ;Jressure en the hor:!.zontal tail is at 11
0 I 0 I J
station 277.5. Fig. AS.52 shows the fuselage
loading. To r ind wing reactions at (A) and (B) : (Col. 8) , M= M at previous station in Col. a plus
6M in Col. 5.
ZMA = 100 x (277.5 - 73) - 43 RR = 0,
hence RR = 475.6'" (up)
ZF z .
- 100 + '::'75.6 - RF 0, . CALCtru-TION
MOMENTS ANIl
OF APPLED FUSELAGE SHEARS,
AXIAL LOADS FOR A SPECIFIC
hence Rr = 375.6# (down)
FLIGilT CONDITION.
Table A5.7 gives the detailed calculations
for the shears and moments at the various sta- Using the results in Tables AS.5, AS.6, an d
tion :;)oints. .- 7 the applied shears and ~oments for a g:!.'l8 n
,,"0., J
0 11 50 80 120 170 200 230 260 290 315 flight condition follow as a m.tter of pr'cpor'-.
I I I I I I 127~. , i tion and addition. To illustrate, the applied
I, ! I I values ~or one flight condition 'Hill be given.
,I
1 1
1 1 ! 1
I I 1-'100* , It will be assumed that the aerodynamic
I
! I , ,
1
i calculations for this air?lane for the (H,A.A.)
! 1
hfgh angle of attack condition ;ave the follow-
+---------------- - - - - tng r-esu Lt s , which the student will aave to
; ~ RH-~~4 --2~-_ accept wt -nout knowrenge of how they were
I 43" -- -- cotameo •
R F =375. a RR=475.6 Fig. A5.52
STA.73 srA.116
A5.I8 BEAMS -- SHEAR AND MOMENTS
TABLE A.5.a
'" .i "
• 6850
: 173 0 _ 86855 I
17J2.0 ·286650 1 :290
'"
'" factor ot 7 would equal 7 t1mes the values in
columns (3) of Table AS.S to obtain shear ~.d 7
',:;i~:g: ~~:~
116 • - 6850 ' 17770
_ 6850 1 17770 I
",., 2910 _313690 I 1290
times column 6 to obtain bendi~g moment.
I
, .,
80 : 47250 mo ,_314640 : 1700 Likewise the farces due to the 2.98 load
73 • I H900 0 40 I 1700
factor in X direction would equal (-2.98) ti~es
• ,.,
'_385704 1_1710
" _208937 ;-1190
0
the values in col~s (4)7 (5) and (9) of
-1190
o
Table AS.S to obtain axial loads, shears and
bending moments respectively.
The final or ~r~e for~es woult be the
Col.. 1 _ 6 :l valu•• In '01""",,, 3 at Talli. A.5.5.
Col. 3 • 1.333 '" valu'" ill .00umn 5 01 Tabl .. .405.B. algebraic sum of these results.
Col. 4 _ -1.10 '" valu" ill .00""",n 3 01 Ubi. A.5.7.
Col. 5 • eol""",,, (1) • oolWIIII (2) •• olumll (3).
Col. B • 8 '" valUH III .01"""," B af 1':lQI. A.5.5. Landing with .~~gular Acceleration
Col. 7 _ 1. 3;13 :l VlI.1ues In .0IW1111 9 of T:lQI.. -'5.6.
Col. 8 _ to 10 '" vaiUH ill .01"""," 6 at TaDl. AS.7.
Col. 9 _ col""", ... {B) • (7) • (8). In a level UL~ding condition. i t 1s some-
Col.10 _ 1.333 '" ..uuu In .01 ....... 14) of Tabl. AS.6. times specified that the horizontal component
ot the ground reaction must be a certain pro-
AS.19 Example of Fuselage Shears and Moments for portion of the vertical componeht, which causes
Landing Conditions. the line of action of the grO~~d reaction ~ in
Fig. AS.53 illustrates the airplane in a rig. AS.53 to not pass tr~ough t~g e.g. of the
level landing condition. The ground reaction is airplane, which creates an external pitching
assumed to pass the center at landing gear wheel moment on ~he airplane. This ~oment is us-
and c.g. of airplane. The fuselage Shears, mo- ually balanced by the inertia forces cue to
ments and axial loads are r-ecurr-ed when the the angular acceleration produced by the ..in-.
vertical ultimate load factor is 7. (Gross balanced moment about the e.g. The shears and
weight = 430ot). moments on the fuselage cue to this externa;
moment could be :aund as ex~la1ned in p~t.
SOLtlrION: AS.20.
The vertical or Z component of the ground AS.20 Inertia Loads Due to Angular Acceleration.
reaction R is specified as 7 load factors which In some of the flying conditions, it is
equals 7 x 4300 = 3010oi. One halt ot this is samet~es specified that the a1rplane ~ust be
acting on each Wheel. SUbjected to an angular acceleration as well as
The horizontal or X comDonent of R is 30100 translational acceleration. ~is angular ac-
tan 23° 425 x 30100= 12800# and acting aft. = cele~ation of the airplane produces :nertia
The horizontal load factor on airplane forces which must be calculated if the air-
equals 12800/4300 = 2.98. plane is to be treated as a body in statiC
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A5.19
13j 3 1 4 $ 5 1 I • 10
!
about e.g. of airplane.
= pitching moment of inertia of air- II ~~.
~.
!~!I Arml
I"
Arml
Z :XI
Arm
z.,
i Arm
I ~ 1
". ~;:~I~;:;'I'
!x:.J. (~.~l""
I.. IFlI:
a
plane about airplane e.g. = Zwr 1.~1I 51. 3.1 I
i· ~.nn.s!
- 14.3l<u.e I
The tangential inertia force F for a mass 8:1.$ -1I.:l
wig due to an angular acceleration a equals J 110
200
IJlll
,711
~.~!170:
[4.701
5.4'1.00;
IS. il IU.SO e: n,sl 1011 \
1S50
2180000
~UOOO
130 : 10
: l 1 O ' 21
I 1. 00 1 .so i
10.40 142.50
13.0GI 250:
2.1.ta 173.$0
I 20460 i
30:nO I
204100
535000
i.1y
F = :!!..
g r-c , but a = 1y g 290 ,UBI 19. 10 I. '"
I 28.
501 :W3.50 JlGU
I
, 3110000
141.3 1_20.11
hence
'"
W'l..!lI
rroutC18IUi..12351_~'
Fual.
1:r.2. <10.0 • 315 - 0.1 3:17.50
1S01_[., 911_ I.G;
i
11.50
10;11-41.5 i.<11.50
710 1 - :n
i
I n <17.51 1.50
51100
1M
Zl20
313 i~
u"",",
133000
4il5ooo
::: l'"~:: I
• :5.5
1.1
51.3
U.3
My I
F = Ty 'N ~, wher-e r 1s the distance from I ,I,. : 11091100 1 •
1:12.0".
Planform Fig. AS. 62
10L
Fig. A5.54 _
1----- 200" --.....;
A5.22 Problems
~ i 10'~soo 150"
I--- 20'~
o~i=:I:::J'=C=IIII::t[Il
30" (55) (56)
100 00 400 SO
1.
.L
t 1 ]lj'kS·3,0
I
Relative Span-
wise Distribution
1 1-2,..,' --I10j- }-2H I """'""1:20"1--
A AI L (56a)
~1O"f-
Fig. A5.63 shows plan ror-a of a cantilever 'Iring.
60 --j 10"
The total distributed air loa~ nc~al to sur~ace
.Cable is 10000 lb. The relative spanwise ~istriJut1cr.
is shown. Take center of ~ressure at 24 percent
400 (57) o~ chord fro~ leading edge. Divi~e wing i~to
'\.-.600 a" 10 inch width strips and calculate VZ 1 Mx and
~=;r===±i:==~' -L My, ane plot curves for same.
01" r't-12" 9"
AI
l' ~2 ,
IV. F:g. A5,64 shows an externally Jraced
monoplane wt ng , Take an aver-age 'Nir.g 11ft
500 lb. load of 90 l'J./sq.tt. norma I to wing with
center of pressure at 27 percent o~ the
chord from leading edge of wing and calcu-
late anc draw the tron0 and rear beam prl-
~ry shears and bencing ~c~ent d1a~rams.
100
M
I
Ii
60c
4 T
2.0"
,e..L
c
F.B. R.B.
(II) Draw bending ~oment diagram for structures
and loading in Fig. AS.51, abc •
4~~5°
\ 10';-
100f
,j I~O
/ 100
_ _ ::\450
iii .L (b) 1-12""",
1--16.. --l (c)
Fig. AS. 61
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A5.21
-, ------~----------'~~~-
AS.22 BEAM -- COLUMNS
M where
t(w) is a ter::n which does not include the
sin L
load P or the end moments l11. and M:a. The
axial
J expressions tor r(w), C:\. and CII depend on the
To find the location of the maximum moment, dit- type of the transverse lead.
terent1ate equation (AS.3) and equate to zero. Table AS.I gives the ~alue of these 3
terms tor types or transverse loading on a Single
d11 _ c, X Ca sin '!:
dX - 0 = T cos J - J J span which are frequently encountered in airplane
structureS. The Table also gives equations for
whence the pcfrrt of maximum bending moment; and its mag-
nitUde.
L
x C D. - D1 cos - Table AS.II 15 a table or slnes, cosines,
tan - = --" = --,--"J IA5.5) and tangents tor L/.1 in radians which is more
j c, L
D1 sin 1 convenient to use than the usual type at trigo-
nometric tables. This table is based on values
The value or x must tall within x = 0 to x = L, given in Appendix ~ of Air Corps Information
otherwise I11 or Ma is the maximum value. Circular #493. The A difference have been added
The value of the maximum span moment can be to facilitate rapid use of the tables.
found by substituting the value trom equation For single span beams, the critical value
(A5.51 in (A5.41, which gf vee
of L/j is fi; that is, it the axial compressive
load is such that the term L/j = fi, the center
n.ax = D__ + wJ' - - - - - - - - - (A5.5) region of the beam will tend to deflect until the
cos x combined stresses equal the tailing stress of the
J material.
The moment M at any point x along the span can
also be written:
AS. 27 Moments for Combinations of the Various Load
I1=D:l. [(tan ~ . sin ~rcos ~]+Wjll - (AS.7) Systems as Given in Table AS. I, Margins of
J J J Safety. Accuracy of Calculations.
where Xm refers to the value at x where the span
moment is maximUm, or equation (AS.S). Since it The principle or superposition does not ap-
is customary to locate the point of maximum span ply to a beam-column, because the sum of the
bending moment and its value before investigat- bending momenta due to the transverse loads and
ing other S~l points, the value at tan !m is the axial loads acting separately are not the
J same as the moments when they act simultaneously.
known from equation (AS.S) and thus is available In combining several transverse load systems
to use in equation (A5.7) tor tinding moments at with their accompanying axial loads, the principle
other points along the span. ot superpOSition can be said to apply it each
If the equation tor the beam detlection is transverse loading is used with the total axial
desired, it can be found by SUbstituting the load for the systems which are being combined.
value of M trom equation (A5.3) in equation ThUS, in Ta~le A5.I, to tind the :noments tor
(AS.l), which gives: several combined loadings, add the values ot C:\.,
ell and t(w) for the several loadings and use
---_:.
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES AS.23
Table AS. I
P+~~j====I~P
o
sin L M 1 sin..!::.
I- ~L . T j
x <: a, .Wjsinb
T o o
sin L
T
x o a, + Wj sin a - Wj sin.!. o
1"
tan L
)
T
Triangular Loading. No (NOTE A)
End Moments ..l. o To obtain Max1mum Moment, compute
p__ -c:rT r 1! t ~JI
w~in. moment at 3 or. " points in span. Draw a smooth
curve thru plotted results.
t--X~ L
Triangular Loading. No End
U ro-::::r
~ Momems
w./p'"iJTi I f
'Fx L
Couple Loading (Clockwise) x < a, - m cos b o o (See Note A)
T
sin L
T
X) a. - m cos..!.. m cos.!.. o
) J
tan L
T
TABLE AS. II
3.02 O. 2129
.00 92
-0.99262
O. 0127
-0.12219 · 10 6
0.00993 0.00118 0.01013
3.03 O. 1136 -0.99378 -0.11206
0.00994 0.00108 0.01011
3.04 0.10142 -0 99484 -0 10195
0.00996 0.00097 0.01010
3.05 0.09146 -0.99581 -0.09185
0.00996 0.00086 0.01008
3.06 0.08150 -0.99667 -0. 817
0.00997 0.00077 0.01006
3.07 0.07153 -0.99744 -0.0717
0.00998 0.00066 0.01004
3.08 0.06155 -0.99810 -0.08167
3.09 0.05156
0.00999
-0.99867
O. 0057
-0.05184 · 003
0.00997 0.00046 O~
3.10 0.04159 -0.99913 -0.04162
0.01000 0.00037 O.
3.11 0.03159 -0.99950 -0.0318
0.00999 0.00027 0.01001
3.12 0.02160 -0.99977 -0.02180
0.01000 0.00018 0.01000
3.13 .0 16 . 9 3
o 00007
-.
3.14 0.00160 -1.00000 -0.00160
0.01001 0.00003 0.01001
3.15 -0.00841 -0.99997 0.00841
0.0100e 0.00014 O.
3.16 -0.01841 -11.99984 00.01841
0 9 3
3.17 -0.02840 -0.99960 00.02841
,
3.18 -0.03840
0.01000
-0.99926
0.00034
0.03843 i · 2
3.19 -0.04839
0.00999
-0.99883
0.00043
0.04845 I ·
0.00999 0.00083 0.01003
3.20 -0.05838 -0.99830 0.05848
0.00998 0.00064 0.01004
3.21 -0.06836 -0.99766 0.06852
0.00997 0.00073 0.01005
3.22 -0.07833 -0.99693 0.07857
0.00996 o 00084 0.01007
3.23 -0.08829 -0.99809 0.08884
0.00996 0.00093 0.01009
3.24 -0.09825 -0.99516 0.09873
0.00995 0.00103 0.01010
3.25 -0.10820 -0.99413 0.10883 I
c
65
these values in the general expression tor M as of the bea~, or 4420 x .75 -= 3315"# positive be-
given at the top of the Table. cause it procuces compression in the top tibers.
In a beam-column member, the bending mom- The moment at (2) due to the cantilever overhang
ents do not vary directly as the load 1s i~crea equals (20 + 10) 36 x 16 = 8640"#. Fl§;. A5.:58
ed. Thus, the student should realize that marg- 2
ins or safety based on direct ?roportton of mom- shows the beam ?ortion between points (1) and (2)
ents to loads are incorrect and lie on the un- as a free body.
safe side. From Art. AS.Z5, we have the follOWing pre-
It is recommended that four significant cise equations for a beam carrying a tr-ansverse
figures be used in computations, making use ot uniform distributed load with end compressive
the so-called precise equations, since the re- loads.
sults in many cases involve small differences x L
between large numbers. tan J D~ - Dl cos J - - - - - - - - -(A)
P-';"~I~~;b~g;~g~~~~~t'~~B~~.J...
.A. Wood Sea 2
Wj
D1.
"= 20 x 2941
M1 - wja
= 58820
= 3315 58820 55505
~...<;e lSOQ.IP=Reaction
DII = M.. - wjll 8640 58820 50160
'S:'~
,0' from Strut to
Lower Wing
L 100
J = 54.23 = 1. 844
'\..!~'\.
From Table A5.II sin J = .96290 and cos J -= .... 26981
...... w= 2Of/in.
D1 sin J
50180 - (-55505 x - :26981) -65156
1.2192
-5~505 x .96290 -53441
x tan-' 1.2192 = .88383
J
Fig. AS. 68
Hence, x = .88383 x 54.23 = 48 ft , which equals the
distance trom the lett end at the beam to the
hinge point at the lett end. point at maximum span moment.
l:l'lb 2000 x 50 - 540 x 116
- 1500 x 100 + 70.75 Th = 0
= ..lL:-
cos x
+ «r , cos ~ = .63419 from Table
j
A5.II
hence J
Th = 442Qj1
Hence 55505
The axial compressive load induced by the lift
strut at pOint (2) then equals - 4420#. ~ - :63419 + 58820 28,700"#
Taking ZH = 0 tor the load system or Fig. To obtain an idea as to the magnitUde at the
AS.S? gives P = - Th -= 4420#. The end moment on secondarY bending moment, that is, the ~oment due
the beam at (1) equals the end load timeB the to the axial load times the lateral beam deflec-
eccentricity or the hinge :ram the neutral axis tion, the primary bending moment at a point 48~
-,,;; . r
tram the left end w111 be computed. substituting values of C1 and Cm and f(w) :rom
Table A5. I in the above equa.t i ons ;
M..II := 3315 + 48x20x 24- 940:<48:= -18765"#
Thus the secondary bending moment equals I1= (M .. -!11. cos L/j) sin x/j +M cos x/j
sin L/j 1.
- 28700 + 18765 = - 9935"# which 1s a large per-
centage of the primary moment. The transverse
deflection of the beam at the po1nt of max. span 20, OQQ!...lF==={~2='=!=2="=0=6=3=8=,,="=1=T=U=be=J~02~~000t
moment then equals - 9935 := 2.25 inches upward. . L: 41. 762 ,(
- 4420
Bending Moment at any Point Along Span Fig. A5.70
(j) ®
Let the moment at a pcfrrt 10" from paint
(2) be required. In this case, x = 100 - 10 = 90 But, M:a. = 0 in our problem, henes,
M:= Dl.~tan ~ . sin })+ cos
L .
11
+ wja (Ref. Eq. I1 = Masin x/j
A5.?) sin L!j
x 90 x
- - = 1. 6596, sin j- .99605 =" {Ei:~ /"'2"'9"'X'l"07''''X-."746' =' ~ = 25 826
J 54.2:3 VP - V 20000 V66'l •
58820 = - 3664'"
.
M =-55505 [(1.2192 x .99605) +-.08867] ...
suost i tutlng in the above equations tor 11 /'
E:x:ample Problem #2 M = 36000 x .72327 = 26066""
Fig. A5.59 shows a simplified landing gear .99892
structure carrying a vertical load of 12000# on This compares with a prtmarY moment of
the axle. M.ember ABC is continuous thru B and 36000/2 ~ 18000"#. The deflection at the mid-
ptnned at C. Let it be recurred to determine the point or BC := 26066 - 18000 = .403 In.
bending moment at the midpoint or member BC and 20000
its lateral deflection due to the 12000# verti- The maximum moment is gl ven by the equation:
cal design load.
~ =s~n L and it occurs at x =n2j (See
~ 1 26" ~ Ie J Tabla A5.I)
~D ~
"0:"------;;:86:::2"."'" "--------1- A5. 29 Stresses Above Proportional Limit Stress of
Material.
12
The equations as presented in this chapter
4 assume that E is constant or in other words the
stresses are Within the elastic range. In air-
crart structural destgn the applied or limit
21/2-083. loads must be taken without suffering permanent
Round
Steel 10 deformation, hence E 1s constant under such
loads. However the aircraft structure must
Tub" I take the deSign loads which equal the limit
loads times a factor at safety (usually 1.5)
Fig. A5.69 without failure. In many cases structural
failure will occur under stresses in the plastiC
Line of action of DB goes through E
AxlJL range where the material stiffness Is less and
BiE,,( A not constant.
"""'"3 2000*
A good approximation for ~~ effective
Solution:- modulus E' Is obtained as :ollows:-
Solving for reactions at C by staticS, we
obtaIn :he axial load in Be = -20000. The (1) Compute Fc = PIA for the given number.
bending momen~ at 8 due to 3" eccen~ricity of
the 'wheel load = 3 x 12000 := 36000"#. (2) With this value of Fc enter the basic
Fig. A5.70 shows a r~ee body of portion BC column curve diagram for the given material (for
of member ABC. From Table A5.1 end fixity C = 1) and find value of L'/p cor-
M = C~ sin xj + ,v~ cos x
} + '()
~ w responding to the stress Fc'
~~--~-------~~
AS. 30 BEAM -~ COLUMNS
(3) USing these values of L' /p and FC J (4) Determine the bending ~oment at the
compute centerline of the beam-c01~~s shown in Fig.
Assume ~I =
64 J OOO, 000 lb. in. sq.
"' -""Tti"
to
_Fe (L')'
P
A5.74.
~
50" ----1 18000"-IF
(3) Determine the maximum. bending moment The secondary moments in a partic~lar
for the wood wing beam and loading of Fig. member due to beam-column~action also effect or
influence the deflections in adjacent members
A5.73. I of beam section = 17 in.· E 1.3 x = of a continuous structure. This rat~er involved
10".
problem can be handled quite simply and rapidly
by the moment distribution method as explained
and illustrated in Arts. All.l2 to 15 of
I cnap'ter All.
t f ! t
p-r* V
10001 20007
s.
"000..
.
t-<."+-:""--.
#rlll
:-
.
I I I tIl I
""-j
tyeoo IO<I/in.
tt I tl l!
0'"
1~ I 100001
200" - - - - . - . "
Flg. AS. 74
CHAPTER A6
TORSION.. STRESSES AND DEFLECTIONS
(3) Diameters remain straight atter twisting of The ~oment of the shearing stress on the
section. circular strip dA about 0 the axis of the bar
(4') Material rs homogeneous, isotropic and 1s equal to,
o'ig}dA
elastic. dM .. -r pdA
c = '--L- and thus the total in-
(5) The applied loads lie in a plane or planes
perpendicular to the axis of the shaft or ternal torSional resisting moment is,
cylinder. p
ao"QdA
\nt.
,
\
L
B'
=
r--L-
\
For equilibrium, the internal resisting
A moment equals the external t or-ai onaj moment T,
- i
and since GQ/L is a constant, we can write,
,,
J
Fig. A6.1
,
p
-- .T = Mint. =
GQ
L Jr
0
2
p ctA = T
GQJ (2)
C-,.'
vb
AB.2 TORSION
T = H.P. x 396000
2nN
63025 H.P.
N
(7) 5" L
I
24" i 24" -15" L
Fig. AB.4
EXAMPLE PRCBLillS.
Problem l. SOLUTION:
Fig. A6.3 shows a conventional central
stick-torque tube operating unit. For a side airload on the sur~ace ~er.ds to rotate
~he
the ailer8n ~round the tC~Gue t~be. ju~ nove-
load of 150 Ibs. on stick grip, determine the
shearing stress on aileron torque tube and the ment is prevented or crea~ed 2y a control rod
angle of ~Nist between points A and B. attached to the torque t~be over the center
supporting oracket.
SOLUTION: The total load 0n a 5t~ip of aileron cne
lucre wide = 40(15 x 1/:'44) ~ 4.16 lb.
Torsional moment on tube AB due to slde Let w equal intensity of loading per inch
stick force of 150# ~ 150 x 26 = 3900 in. lb. of aileron span at tha.,leading edge point cr the
The resistance to this torque is provided by the aileron surface. (see ~ressure diagram in
Fig. A6.41.
150*
Then 3. + (0.5 w)12 = 4.16
J24 ST At Alloy
,...control Stick l G ::: 3,800,000 psi
hence w = 0.463 lb.
The total load forwarG of the center-
p~
varies cirectly as the cistance :rom the end of twists of non-circular cross-sections is given
the ailerons, the angle of twist g can be CQm- in Art. A6.5.
?uted by using the average torque as acting on
entire length of the tube to one side of horn or A6.5 Elastic Membrane Analogy.
a distant J.... = 29", :tence The \Shape of a~watD~a cross-section of a
non-circular cross-section in torsion is
261 x 29 needed in the analysiS by the theory of elas-
2 x 3800000 x 0.06678 ticity, and as a rasult only a few Shapes stich
as rectangles, ;eJ:,lJ2.-seS, triangles, etc., have
AB.4 Torsion of Members with Non-Circular erose- been solved by the theoretical approach. How-
sections. ever, a close approximation can be made ex-
The for-~las derived in Art. A6.2 cannot be perimentally for a Iznos't any shape of cross-
~sed for non-c1rcular Shapes since the assump- section by the use of the membrane analogy.
tions made do not hold. In a circular shaft It was pointed out by Prandtl that the
subjected to pure torsion, the shearing stress equation of torsion of a bar and the equation
distribution is as indicated in Fig. A6.5, for the deflection of a ~embrane subjected to
namely, The maxi~um Shearing stress is located uniform pressure have the same form. Thus if
at the most remote fiber from the centerline an elastic membrane is stretched over an open-
axis of the bar and is perpendicular to the ing which has the same shape as the cross-
radians to the stressed paint. At a given dist- section of the bar being considered and then if
ance from the axis of rotation the shear stress the membrane 1s deflected by subjecting it to a
slight difference of pressure on the two Sides,
the reSulting deflected shape of the membrane
provides certain quantities Which can be mea-
Sured experi~entallY and then used in the
theoretical equations. However, pOSSibly the
main advantage ot the membrane theory is, tr~t
Fig. AB.5 it provides a method of visualizing to a
considerable degree of accuracy haw the stress
conditions vary over a complicated cress-section
Sf
A6 4 TORSION
to take the shape of the bar boundary as the From Table A6.l it is ~otised that :or
boundary 1s approaChed. ~lg. A6.Sa shows a sec- large values 0:' bit, tne values 0-:: t he con-
tion through the contour lines or the ~erlected stants 1s 1/3, and t~us for such narrow rec-
membrane along the lines 1-1. 2-2 and 3-3 of Fl~. tangles, equations (6) and (7) reduce to,
AS.9. It is obvious that the slopes of the de-
tlected surface along line 1-1 will be greater _ _ 3 T
bt:l (5 )
than along lines 2-2 or 3-3. From this we can "MAX -
conclude that the shear stress at any point on
line 1-1 will be greater than the shear stress _ 3 T
g - bt 3 G (9 )
tor corresponding pOints on lines 2-2 and 3-3.
The maxlmu~ slope and therefore the maximum
, , •
,
,
,:>
,
Although equations (8) and (9) have been
derived for a narrow rectangular shape, they
can be applied to an approx~mate analysis of
I ,
~
~. __I ._.
\ \
shapes made up of thin rectangular members
such as illustrated in Fig. A6.10. The ~cre
generous the fillet or corner radius, the
I smaller the stress concentration at these jun-
ctions and therefore the more accuracy of these
,,! ,,
/
approximate formulas. Thus for a section made
3
I .J up of a continuous plate such as illustrated in
•
Fig. A6. a Fig. A6.9
stress will occur at the ends of line 1-1. The .r-: b " --! ..J::t "
slope of the detlected membrane will be zero at
the center ot the membrane and at the four ~ET
b.
corners, and thus the shear stress ~t these
points will be zero.
...
fo-bl.~~tl.
A6.6 Torsion at. Open Sections Composed of Thin Plates. Fig. A6.10
Members having cross-sections made up of
narrow or thin rectangular elements are some- Fig. (a ) of Fig. A6,.lO, the width b can be taken
times used in aircratt structures to carry tor- as the total lteiigthof the cress-section. For
sional loads such as the angle, channel, and Tee sections such as the tee and H section in Fig.
Shapes. A6.l0, the polar moment of inertia J can be
_;"'~'Ir-For a bar of rectangular cross-section or
taken as Z bt~/ •. Thus for the tee section of
width b and thickness t a mathematical elastiCity Fig. A6.10:
analysis gives the following equations tor maxi-
mum shearing stress and the angle of twist per 9
m
=-'- = G 3Z Tbt 3
un1 t length. OJ
= T a (6 ) = 3 T
a b t G (blt~ + bat~)
EXAMPL" PROBLEM SHOWING TOPBIONAL STIFFNESS OF AG.7 Torsion of Solid Non-Circular Shapes and Thick-
Walled Tubular Shapes.
CLOSED THIN ',.JALLED TUBE COMPARED
Table A6.3 summarizes the for.nulas for
TO OPEN OR SLOTTED TuBE. torsional deflection and stress for a few
Fig. A6.lla shows a 1 inch diameter tube Shapes. These for~ulas are ,based on the as-
with .035 wall thickness, and Fig. A6.11b shows sumption that the cross-sections are free to
the same tube but with a cut in the wall making warp (no end restraints). Material is homo-
it an open section. geneous and stresses are within the elastic
range.
For the round tube J. = 0.02474 in~.
A6.8 Torsion of Thin-Walled Closed Sections.
The structure of aircraft wings, fuselages
For open tube 3 2 = ~ x 3.14 x .035~ = 0.000045 and control surfaces are essentially thin-walled
tubes of one or more cells. Flight and landing
loads often produce torsional forces on these
major structural units, thus the determination
at the torsional stress and deformation of such
structures plays an important part in aircraft
structural analysis and design.
Fig. A6.12 shows a portion of a thln-
walled cylindrical tube which is under a pure
Fig. AG.lla Fig. A6.l1b torsional moment. There are no end restraints
on the tube or in other words the tube ends and
Let Q1 equal twist of closed tube and Q 2 tube cross-sections are free to warp out of
equal twist of open tube. The VNist, will then their plane.
be inversely proportional to J since Q = G~J'
Therefore the closed tube is Jl/J a = 0.02474/
0.000045 = 550 times as stiff as the open tube.
This result shows why open sections are not ef-
ficient torsional members relative to
torsional deflection.
TABLE Ae.3
~qds
T
FORMULAS FOR TORSIONAL DEFl.ECTION A.ND ,rrRESS
FORMUloA FOR
SECTION K SHEAR STRESS
SO
SOLID ELLlPITICAL
'T
'tMAX :: ft'iijr (1Lt elida
of minOr ana).
(b) •
Let ~ be the shear force intenSity at point
(a) on the cross-section and qb that at point
Now consider the segment a a ~ b of the tube
wall as shown in Fig. A6.l2 as a free body. The
applied shear force intensity along the segment
edges parallel to the y axis will be given the
SOLID SQUARE.
values ~y and qby as shown in Fig. A6.l2. For
T
D1
SOLID RECTANGLE.
K tO.141a 4 't MAX :: O.208 aJ (at
m1dp::lint of each 'ide).
a plate in pure shear the shearing stress at a
point in one plane equalS the stress in a plane
at right angles to the first plane, hence
qa = qa and qb = qb .
Y Y
ao3[~ 3.3e~(1 = T(3a Since the tube sections are free to warp
CJI K t _ • 'tM.\X - 1. 8bl
8a.2t)%
mtdpo1Dt of 10ll(side. there can be no longitudional stresses on the
2a--., tUbe wall. Considering the eqUilibrium of the
~
~] segnent -in the Y direction,
SOLID TRIANGLE. ~Fy = 0 = Qa~qb;"
- =
0, hence Qa y qby and =
6~ therefore Ga = qb or in other words the shear
torce intenSity around the tube wall is con-
stant. The shear stress at any point T = qjt.
It the wall thickness t Changes the shear stress
For an extensive U,t of formula. for many ~, both .soUd and bollo..,
refer to book, "Formulas For StreBB and stl"ll.in" by Roark, 19!14 Ed.itiOQ.
A6.6 TORSION
changes but the shea~ force q dOes not change, T~e elas~lc strain energy dU stored in :his
or element therefore equals,
ds
We will now take moments of the shear flow
q on the tube cross-section about some point (0).
In Fig. A6.l3 the force dF on the wall element
=qds , Its arm from the assumed ooment cen- or U =
.- cs , 'ehe mt egraI y
f 8Jl:""Gt
ter (0) is h. Thus the moment of dF about (0)
is q cs h. However, ds times h 1s twice the area is the line integral ar8und the periphery of
of the shaded triangle in Fig. A6.13. the tube. From Chapter A7 from Castigliano's
Hence the torsional moment dT of the force theorem.
on the element ds equals, au _ ~ cs
dT • qnds • 2qdA
f"I
'7
_
- aT - r 4A,.T OIGt ~ - - - - - - - - - - - (17)
where A is the enclosed area of the mean periph- AG.9 Expressi.on for Torsional Moment in Terms of
ery of the tube walL Internal Shear Flow Systems for Multiple Cell
The shear stress ~ at any point on the tube Closed Sections.
'NaIl is equal to q, the shear force per inch ot Fig. A6".l6 shows the internal shear f Low
wall divided by the area ot this one inch length pattern for a 2-cell thln-walled tube. when
orlxtor the t~be is subjected to an external torque.
q~, qa and 43 represent the shear load per inch
(16) on the three different portions of the cell
walls.
For equilibrium of shear ~orces at the
TUBE TWIST junction point of the interior web With the out-
Consider a small element cut from the tube Side wall, we know that
wall and treated as a free body in Fig. AS.14, ------(21)
With ds in the plane at t~e tube cross-section m
and a unit length parallel to the tube axis.
Under the Shearing strains the plate element
T ,
<Is '0----1
to- 1 -0\
1. t
I
>0- 1 --<
qds~D: t ,
t
\
----I Fig. AS. 16
Fig.A6.14 Fig. AS. 15 n
detorma as illustrated in Fig. AS.l5, that is, Choose any ~oment axis such as paint (0).
the tace a-a moves with respect to face 2-2 a Referring back to Fig. A6.l3, ~e found that the
distance O. The torce on edge a-a equals q ds moment of a constant shear force q acting along
and it moves thrOU~~ a distance O. a wall length ds about a pOint (a} was e4~1 in
_ _ _ _ _ _ _ _.........··iII'itMTlllil'i
magnitude to twice the area ot the geometrical inch in the web 1-2 and (qe - q~) ~or web 2-3.
Shape formed by radii from the moment center to For eqUilibrium, the torsional moment af
the ends of the wall element d5 times the shear the internal shear system must equal the ex-
flow q. ternal torque on the tube at this particular
Let To • moment ot shear flow about point section. Thus. from the conclusions of article
(0). Then from Fig. A6.16, A6.9, we can write:
(24)
2GQ=.s.i~ (25)
A j t
Hence, To • 2q,A, + 2q,A, - --- - - - - - - (23)
ThUS, for each cell of a ~ultiple cell struc-
tf
where A1 = area ot cell (1) and A3 = area of
cell (2). Therefore, the moment of the internal ture an expression ~s can be written and
shear system of a ~ultlple cell tube carrying equated to the constant value 2OQ. Let a l Q•
pur~ torsional shear stresses 1s equal to the
sum of twice the ~nClosed area of each cell
times the shear load per inch Which exists in
represent a line integral f ~s for cell wall
1-0, and ala, a~o, a2~ and a~Q the line in-
t~e outside wall or tr~t cell. (Note: The web
mn 18 referred to as an inside wall of either
cell) •
tegrals f
~s for the other outside wall and
interior web portions of the 3-cell tube. Let
AS.10 Distribution of Torsional Shear Stresses 1n a clockwise direction of wall shear stresses in
Multiple-Cell Thin-Walled Closed Section. any cell be pOSitive in sign. Now. substituting
Angle of Twist. in Equation (25), we have:
=2GQ - - - - - - - - - - - - - - - - - - (27)
Fig. AG.17
Cell 2
----(29)
_ 1 [
2GQ-367.4 -
1 :: 25. 25"
.040"
External
g =GTJ - - - - - - - - - - - - - - - - - - (32) Torque
83450"*
Fig. A6.19 shows a typical 2-cell tubular 83450 - 2 X 105.8 q, - 2 x 387.4 q. =0 - (36)
section as formed by a conventional airfoil
shape, and having one interior web. An external Solving equations (35) and (36), q1. = 55.S#/in.
applied torque T of 83450 in. lb. is assumed and q. = 9Z.St/in. Since results come out
acting as shown. The internal shear res ist tng ?OSitiv6 1 the assumed direction of coun~er
pattern is reqUired. clockwise was correct for q~ and q~ or true
signs are q1. ~ - 55.6 and qa - - ~2.5.
Calculation at Cell Constants
Cell areas: - A, = 105.8 sq. In.
q1.~ =- 55.6 + 92.5· = 36.9#/in. (as Viewed
from Cell 1).
A. = 387.4 sq. In.
A = 493.2 sq. In. Fig. A6.20 shows the resulting shear ,at-
Line integrals a = ~s : f -
tern. The angular twist of the complete cell
can be found by SUbstituting values of q1. and
~~ in either equations (33) or (34), since twist
ot each cell ~ust be the same and equal to
=26.9
•025
= 1075; au =~
.04 = 335 twist of tube as a whole .
aao = ---:o:;r
25.25 + ~ + 25.3 = -''''35
•05 •032 (
I
Solution by equating angular ~Nist of each cell.
General eQuat~on 2GQ ~ ~s. Clockwise tf
tlow at q is pOSitive.
Cell 1 Subt. in general equation
2Gg = 1
105.8
q1.
r- x 1075
qa
+
•• 002456 T
au = .~~ = 333.3 10
a" = 2nxx .025 = 629 au = 10
.05
= 200
- 20
altO -:03 .04 = 916.7
+..l2... _ 20
=.l2...=
a a e - .03 = 667 a ••
.03
333
c:=:ITO 0,---,---,---1,---
Fig. AB.24
Subat I t.ut tng r
With one unknaNn shear :low q tor each cell,
the solution by the previous eq~tions becomes
roe = l~O [200 qlil - 200 q , + 667 c, + 3:33 q , - Quite labor! ous ,
The method of successive approximations
provides a simple, rapid method :or fl~dlng the
- - - - - - - - - - - - - - - - - (39)
shear flow in multiple cells under pure torsion.
234. H/in.
Therefore assuming GQ~ = 1 fer cell (1) of
Fig. a, we can write ~rom equation (41): _
Fig. A6.23 shows the resulting internal
shear flow pattern. ~he angle of twist, if de- • Based on Paper, "Numerical Transformation Procedures
sired, can be found by substituting values of for Shear Flow Calculations" by S. U. Benscoter.
shear flows in any of the equations (38) to (40). Journal, Institute of Aeronautical Sciences, Aug. 1946.
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES
A6.1l
=-
2 X 89.3
25.7
--
.04
10,
.05 .-- 0.212 lb./in.
(1-2) is SUbjected to a shear flow qJ1l
lin. (counterclockwise with respect to cell (1)
and therefore negative), in addition to the
shear flow q1. = .212 of cell (1). The negative
.109#
r-:---
resulting value tor G91. = 0.875 instead of 1.0
. I
,.-
212
I~
~
q '" . 109
as started with.
Thus in order to make 091. = 1 again, we
j (11 I I (2) will have to add a constant shear flow q'l to
cell (1) which will cancel the negative ~Nist
~ j due to q~ acting on web (1-2). Since we are
I considering only cell (1) we can compare cell
Fig. b Fig. c wall strains instead of cell twist since in
equation (41), the term 2A is constant.
In a similar manner assume cell (2) sub- Thus adding a constant shear flow qi to
jected to a shear flow qa to make GQ 3 = 1. Then cancel influences at qJ1l on web l-?, we can ve-tta:
l.Q.
.05
2 x 39.3
+ 15.7
.03
=
78.6
723 :. 0.lOS#/1n. hence
qi (z n cell (l) - q, (~) 'Neb 1-2 = 0
II \
I.
C.O.F. (1 to 2)
= (-¥-) web (2-1) = 200 = .277
qT .237 x .00717 = .0017~/in.
(z ~) cell (2)
723
q~ = .277 x .0139 = .00385#/in.
Fig. e shows the constant shear flow q~ and Fig. g shows the result s .
qt that were added to make GQ = 1 for each cell.
However these cor- The final or re- :;....-_ _ q'~ '" . 00385
rective shear flows q'~ e , 0587 sulting cell shear 0i,~=.OOi71j
were added assuming flows then equal the
the cells were again
independent ot each
original shear flows
plus all corrective
\ (I) t!(21)
other or did not cell shear flOWS, or ~
--- »
have the cammon web Fig. g
Fig. e
(1-2). ThUS in
bringing the cells
together again the Interior web Is SUbjected to
be resultant shear flow of q~ - q~. In other q = q. + q'. + q". + qm.
wordS it we were to add the shear flows of Fig. e • (r inat )
to those of Fig. d, we would not have GQ 1 and GQ~ Fig. h shows
equal to 1. The resulting values would be closer :;;..- 1 r . 1787
the final results.
to 1.0 than were r OUIld tor the shear flow system
or Fig. d.
To check the
final twist at each
I'q,= .2534 fll I
Considering Fig. e, we will now add a second
set ot correctIve shear flows q~ and to cells q: cell the value GQ
will be computed for
I~ (1) ll(~
(1) and (2) respectively to ~ke GQ, and GQ~ = 1 each cell using the ~--
tor cells acting independently. q values in Fig. h. Fig. h
Considering cell (1), and proceeding with
same reasoning as betore~ cen (1)
1 [~534 25.7
'-' ..... .- ~ x -04
x "9::; • + .0747 x .~~J '" .997
Hence
Cdl 1 Cell 2
(~) web 1-2
10
.05
C.O. (1-2) = = = .40
1
2
C.O. FactoX'
AssUIIled
.277
a a
237 z (t) cell 2 ..lQ. + ...2Q.
.05 0.1
~ ~~
J CO 02~8
•
m
C.O. 3
. 00
• web 2-1 10
T • 2A <15. zo·" 14 1".
C.O. (2-1) = = .~5 = .25
(~l
0," T .3""
Z cell 1 :05
.Explanation of Table 1
Line 2 gives the shear flow q in each cell,
Line 1 gives the carryover factors for when it Is assumed each cell is acting separ-
each cell, computed as explained before. Line 2 ately and is subjected to a unit rate at twist
gives the necessary constant shear flow q in or GQ - 1.
each cell to give unit rate of twist to each The calculations for the q values are as
cell acting independently. (GQ = I). Line 3 follows: -
gives the first set of constant corrective shear
flows to add to each cell. The corrective q re- For cell (1 )
ferred to as the carryover q or C.O.q in the
table consists of the q in the adjacent cell q. • 2A. . 2 x 100
• •25
times the C.O. factor of that cell. Z (~) cell 1
40
.05
Thus .237 x .109 = .0258 is carryover from
cell (2) to cell (1) and .277 x .212 = .0587 is
For cell (2 )
carryover from cell (1) to cell (2).
Line 4 gives the second set of corrective
carryover shear flows, namely .277 x .0258 =
.00717 to cell (2) and .237 x .0587 - .0139 to
q. •
Z (~)
2A.
cell 2
- 2 x 100 • • 4
10 + ~
.05 0.1
cell (1). Line 5 repeats the corrective carry
over process once more. Line 6 gives the final A. and A. equal cell area of cells (1) and (2)
q values which equal the original q plus all respectively. L - length of wall element and t
carryover q values. its corresponding thickness.
Example Problem 1 (2 cells) In order not to start out with dec1lllal values
Determine the ~nternal shear flaN system of q, the values above will arbitrarily be
for the ~NO cell tube in Fig. A6.25 when sub- multiplied by 100 to make q. = 25#/in. and
jected to a torque of 20,000 in. lbs. q2 :; 4Oi/in. Since we want only relative
I-- 10" -----.J.-- 10" -----I values of terms this is permiSSible. These
values are shown in line 2 of Table (2). The
r
10" .05
.05
Cell
1 .05
0.10
Cell
2 0.10
corrective carryover process proper is started
in 11ne 3 of table (2). In cell (1) the amount
carried over of q. = 25 to cell (2) equals the
C.O. factor times 25 or 0.4 x 25 = 10 Which is
written along the vertical line under cell (2).
1 .05
Fig. A6.25
0.10
Likewise in cell (2) the amount of q. - 40 that
Is carried over to cell (1) equals the C.O.
factor t !mes q. or .25 x 40 = 10, ·.mich is
written along the vertical line under cell (1).
OPERlTIOMS TA8U 2
The second set of corrective carryover
constant shear flows are given in line 4 ot
Cell 1 Cell 2 Table (2), thus, .4 of the q. :: 10 :; 4.0 is
1
,
CO. l"ac'toJ:"
l!IU11ed
0.'
0
O.2~
.0 0
carried over to cell (2) and .25 x value of
q. = 10 = 2.5 is carried over to cell (1). Line
J C.O 10 10
• C.O•
•0.
2. S
• 5 repeats the process namely 0.4 x 2.5 = I to
a cell (2) and .25 x 4 = 1.0 to cell (1).
7 . . 0 10
, This process of carrying over 'lalues of q
>n.
2A
•
For cell (1) T = 2 x 100 x 38.85 = 7770 in. lb. OPERATIONS TABLE 3
For cell (2) T = 2 x 100 x 55.5 = 11100 in. lb. Cell 1 Cell 2 Call 3 I
,
Line 10 gives the sum at th9 above two val~as 1 co Factor 23 .21 .'88 .288
a Assumed 25.30 18.45
which equals 18870 in. lb. 3 CO 3 88 2 3. o.
The original requirement of ~he problem was •5 0
C 0 °
1 "
505
° 394 1 51
0.430 0.13
2.53
0.685
the shear flow system ~or a torque of 20,000"#. , 6 C.O. 0.244 0.116 O. .33
co.
, ,.
Therefore the required q values follow by direct 7
, 0.072 0.056
.0
proportion, whence 9200 6020 2215
10
11
"
T Total ~ 19000 + 6020 + 2215 ~ 1 43S'"1lI
q _ to;,: T=100,000 183 '"0 112.5
20000
=I88'iO x 38.85 = 41.2#/ln.
20000 Explanation of sol~tion as given in Table 3: -
q~ = I8870 x 55.5 = 58.9i/ln.
Cell (1) Cell (2 ) Cell (3 )
These values are shown in line 11 of Table
2. Check on twist of cells under fir~l q values.
A~ .: 144 = 96
A. A. = 56. 5
The relative total strain around each cell
1
"i = 1140
L
"t= 1040
L
Zt::; 1055
boundary is given by the term. A z q tL for the
Shear flow q for GQ • 1 for each cell acting in
cell.
Thus tor cell (1) dependently: -
Cell (1) q = 2A, = 288 =
"~ 1140 .253
t
, C.O. =
m web 2-1
;:rw
240
= .21
I - 12"
I
I
8" ---I,
(2-1)
!(~)cell 1
T .04 .032
i I Cell (2 ) to (3 )
.041 \
12" .04 Cell 1 .0' Cell 2 'Cell 3 I (~) web 2-3
!y
_ 300
C. O.
(2-3) = - 1055 = .234
z(~)ce113
I
i
...i... i .04 0032
I
Fig. A6.26
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A6.15
8.90. Pne car-r-y over q to cell (1) is then .21 For the open type stiffener as illustrated in
x 8.90 = 1.87 and to cell (3) is .224 x 8.90 = Fig. A6.28, the torsional rigidity of the in-
2.53. dividual stiffeners as compared t 0 the torsional
In line 8, the final q in cell (2) equals rigidity of the thin-wal!ed cell 1s so small
the original q of 18.45 plus all carryover q to be negligible. However a closed type
values from each adjacent cell. stiffener is essentially a small sized tube and
Line 10 in Table (3) shows the total torque its stiffness 15 much greater than an open
developed by t~e resultant internal shear flow section of similar size. Thus a cell with
is 17435"~. Since the problem was to find the closed type stiffeners attached to its outer
shear flow system for a torque of - 100,000 in. walls could be handled as a multiple cell struc-
lb., the values of q in line 8 must ~e multipli- ture, With each st_tfener acting as a cell with
ed by the factor 100,000/17435. a common wall With the outside surrounding cell.
Line ~l shows the final q values. Since in general the st11'fness provided by the
stiffeners 1s comparatively small compared to
Example Problem 3. Four cells. the over-all cell, the approximate Simplified
Determine the ir.ternal shear flow system procedure as given in NAeA T.N. 542 by Kuhn can
for the four cell structure in Fig. A6.27 when be used to usually give sufficient accuracy. In
SUbjected to a torsional moment of - 100,000 tn. this appraximate method, the thin-walled tube
lb. and closed stiffeners are converted or trans-
40 formed into a Single thin-walled tube by modi-
fying the closed stiffeners by either one of the
(1600) follOWing procedures: -
.025
16" (1) Replace each closed stiffener by a doubler
14001 plate having an effective thickness
(1600) te=tSTs!d, and calculate ¢ dslt with these
(500) (625) doubler plates in place. The enclosed area
cen i Cell 2 Cell 3 Cell 4 of the torsion tube still re~lns (A) or
the same. See Fig. A6.30.
Fig. A6.27
,-
, 12
, Transiormation Procedure (2)
a
9
10
", .
7620
47.1
31560
65.3
, 22550
49.9
8900
29.41
by Procedure (1)
I -,- d ----oot ,.tsK
by
I •
~ .. - ,.
~- . ~. , ... ~ ~. .
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A6.17
Note: The deflection of a cantilever ~e~ with 7he shaft rotates at constant Soeec. The dif-
a load F at its end equals FL~/3EI, and I the ference in be!t pull on two sides of a p~lley
moment of inertia of a rectangle about Its c~n are shown. an tne figure. Calculate the ::2...'(1~~
ter axis tb~/12. torsional shearing stress in the 1-3/4 inch
shaft bet~een pulleys (1) and (2) and between
hence (2) and (3).
(2) rl. 1/2 HP. motor operating at lOCO RR~
Now let Tt be the ?ortlon ot the tctal torque rotateS a 3/4-.035 al~~lnum alloy torque tuee
carried by the ~e~ber in pure torsion. T~e ap- 30 inches long which drives t~e gear =ectan~~
proximate sol~tion for open sections co~posed of tor ~p&ra~ln; a Wing flap. Dete~ine the ~~<:
rectang~lar ele~ents as given in Art. A6.6, ~~ torslcnal st~ess in the torque sha:t ~cer
equation (12) will be used. fUll ~~wer and RPM. Find ~he ang~lar de:lec~ian
of shaft in the 30 lr.ch ~ength. ?olar =c~ar.~ ot
= 3 Tr L inertia of tube = .01 in. ~odulus ot r:gidity
9t Gt'(b + b w) G = 3300,000 psi.
Tni.n.lb.
Equating GS to tit we can ''iT' i te
3 E h
2
b3 I I
8L 2Ge(2b
+ b.n ) / I
/
,/ I
Substituting values when L : 10 inches. 'f--- I- "
T
I I
3 x 10.5 X 10 4 X 3.4 x 1.75 I /
/
io- X 3.7 X 10 4 X 0.1 11(2 X 1.75 + 3.3}
= 9.65
Fig. AS, 36
'rner-ercre , percent of total t cr-s tona L T -15" :-
taken :y :e:'.d.ir.g of flanges e~~als,
'!'c ':'8
.... '1' (100) =,,~:-+"='...-(lCC): 90.5 percent.
Ide t.ertafne t:r.etne crs ona ;
(3) I ." . .
~,
t
__ ..1."" r::
__ " "," e ..,..'!!: • •"\._
"01'1""'''' """' '{:: • .;.0
snea r :lo'iJ :::. :-es:~'::::g
t
~ -ee
.J_ • C
.T
3 . -t -& I tne ext er-na L t orque 0: scccc :n.lb. -,.;,:2: ::.:.:.
wall t'u cxnees ar-e gi '''e:: or: tno r:;,;u.:-e. As "'\.::",e
the tube 1s 100 In. long and :in~ :~e :c::::~3.1
It we ccnsid.er the section 4Q inches :~c~ de f t i cn , za cer 15 <:ll:"''"ll.L'1.u::l.
L e c (J-=
i a t a.I L o y ,
Fi.g. AS. 37
of 120,000 In.lbs. USing the ~ethod of 10" .04 04 .035 ~035 .03 .0' .03 I (b]
successive approximation calculate reststing
shear flow pattern.
1 .03 .03 I .03 .03
1-==::::::::===
All interior wece e . 05 thickness
I i i I'"
Top skin .084" thickness
Bottom skin .064" thickness
Fig. A6.38
'. ,
,.
The big helicopters of the future will be used in ~ny important industrial and military opera-
t t ons , The helicopter presents many challenging problems for the structures engineer.
(Sketches from United Aircraft Corp. Publication "BEE~HIVE", Jan. 1958. Sikorsky Helicopters)
•
CHAPTER Ai
DEFLECTIONS OF STRUCTURES
ALFRED F. SCHMITT'
A7. 1 Introduction.
Calculations of structural deflections are p
i~portant for tNO reasons: - ,
(1) A knowledge of the load-defo~ation Work =- L: Pi
I~l
6 Oi
characteristics of the air?lane 1s of prllnary in the limit as
ireportance in st~dles of the influence of
structural flexibility upon airplane perform- 60-0
ance.
Work =Lt'J P dO
(2) Calculations of deflections are neces-
sary in solving for tt.e In~ernal load distri-
butions of complex redundant structures.
Fig. A7.1
The elastic deflection of a structure
under load i s the cumuLat;'... ''j result of the curve ls a straight line whose equation 1s
strain deformation of the individual ~lements P = k6 and the strain ener~J is readily com-
composing the structure. AS SUCh, one method puted to be
of solution for the t~tal deflection ~ight in- p'
,o~ve a vectorial addition of these individual U equally, U = 2k'
:ontriJutions. ~he involved geometry of most
?ract~cal str~ctures ~kes such an approach
A7.3 Strain Energy EXl)ressions for Various Loadings.
prohiJitively ditf~c~lt.
For complex structures the more popular STRAIN SNERGY OF !ENS I ON
tec~~lques ~re analytical rather than vectorial.
7hey::ieal dt r sc t Ly 'Nith quarrt t t Les which are not A tens 11e load S acting at ~he end of a
themselves ceflections but from which deflec- aniform bar of length L, cross sectional area
tions ~y be obtained by SUitable operations. A and elastic ~odulus E causes a deformation
~he nethods employed herein tor deflection cal- a = SL/AE• Hence S AE a and
L
culations are analytical in nature.
A7.2 Work and Strain Energy. U • ~ Sd5 • AE 5' (1)
'Q L 2
~ork as defined in ~echanlcs 1s the prod-
uct of rcr-ce t ines cr stence . It the rorce Alternately.
v~aries over the distance then the work 1s com-
puted by the integral calculus. Thus the work S'L
U' (2 )
cone by a varying force P in deforming a body 2AE - - -
an aaount 6 is
',.Jork =.rPd.a and is represented
by the area ander the load deformation (P-6)
Squations (1) and (2) are equivalent ex-
pressions ~Jr the strain energy in a Qni~o~ bar,
curve as shown in Pig. A7.1. the farner expressl~g U in terms of the de-
If the tefor.ned body 1s per~ectly elastic fo~~tior. and the latter expressing U in~e~
the energy stored i~ the body nay oe co~pletely of load. 7he secand form of expression is ~ore
r-ec over-ed , the bocy un Loac tng a Lcng the same 80nvenlent for ~eneral usage and succeeding
?-6 c'Jrve :ollowee for increasing load. T~is strain energy :ornulas will be ~ut in this form.
energy is ca l Ied the elastic ~ energy q£ ?cr a tensile bar haVing non-uni:orm prop-
GeforT.aticn (hersa~ter ~he strain energy, for erties (varying AE). or for which the axial load
brevi~y) and is denoted by ~~e s~bol U. Thus S varies, the strain energy is computed by the
U = t Pdc. Should the cody '_be .Li near Iy
calculus. Thus the energy in a dir:erentlal
element of length dx is atven by eq , (2) as
elastic (as are ~ost bodles jealt w1~h in
structural analYSiS) t~en the load-defo~tion s-ex
dU :: ;;'2AE Where S and AE
(1) Design gpec iarist, Convair-Astronautics are average ';alues over the ~ength of element.
A7.1
A7.2 DEFLeCTIONS OF STHUCTUReS
~WT
total strain energy one has
1. • 40·
2
.to • 2 in U = j dU = ~ jl1'dx
q(~ m1
40 (5 )
J: _ lO::do 6 pd 2 EI
" 'to. 2' lb/in
Exa~ple Problem 3
I. Flg~ A7.3
For~
the beam of Fig. A7.4 derive the strain
energy expression as a function of Mo.
Solution:
= q , 2L
n sin 2L mI' The bending ~oment diagram (Fig. A7-4a) was
found and an analytiC expression 'Nrltten ~or v
•
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A7.3
L
~=GJT -------------- (6)
Then
or
(7 )
1. 01"":~- __
Fig. A7.5a
Fig. A7.5
Inspection of the diagram revealed that o !_ _ ~ __ ~_~ __ ~_~_
~:-~-:.i.?1:
.
;,::~ ~.~<~--~
---~'. ':";\':,,,.
,
A7 • DEFLECTIONS OF STRUCTURES
V'dx
_ - - _ + 15 f
4- n-z
+ 3f
n-L
+ ~f )
4- n
U=
J-2AG - - - - - - - - - - - - - - - - (10,
Sxa:nple Problem 6
Determine the strain energy of shear in
TABLE A7.1 the Jeam of Example Problem ~.
(T)'
STA.
7
T (T)~ GJ
ill
Coe£!. m~ Coeff. Sc Iut t on :
a 1.0 1.0 1.0 1.0 . 0667 .0667 The shear load diagram was drawn first .
.98 .98 .78 1. 26 .267 .336 L L
•• .9' .88 .54 1. 63 .133 · 216 ~ 2 1 2 =-jP o
.8
.8
.07
• 2• ."
. 084
. 37
.24
1. 22
.35
. 250
' . 200
· 305
· 070 Po Ee" . . . . . . AG • CONSTANT
1.0 a a .14 a . 0833 0.0 V (xli I! iii I i_!J~L! i ! ! Ii i ii iii j
z= · 994
Then
Therefore the strain energy was ,L/2
2 x J 0 Poa dx
u = LTR'
_.- x .994
GJR ::: Po aL
2AG
STRA IN ENERGY OF SHEAR
Eq , (9) may be used to comput.e the shear
A rectan~~lar element "dx" by "dy" of
thickness "t" into the page under the action of strain ener 5J ~n a thin sheet. The elemen~
uniform shear stress ~ (psi) is shown in Fig. dx x dy Is visualized as but one of many in the
A7.6a. sheet and the total ener~J is oJtained by
summing. Thus
From eIementar-y strength
ot materials the angle of
shear strain J is pro- U =~ SS7't gx dy - - - - - - - - - - - - (11)
portional to the &hear
stress -r as Here a double integration 1s required, t~e
summation being carried out in both directions
~ = ~ where G is the over the sheet. In dealing with ~hin sheet the
material elastic modulus use or the shear flow q Tt 1s often convenient
in shear. For the dis- Fig. A7.6a so that eq , (11) rewrites
placements as shown in
the sketch only the down U = 2~ Sf q'dxGt dy (l1a)
load on the lett hand side does any work. (In
general all four sides ~ove, but if the mot~on A very important special case occurs when a
is referred to axes lying along ~NO adjacent homogenous sheet or constant thickness 1s
sides of the element, as was done here, the de- analyzed assuming q 1s constant everywhere. In
rivation is Simplified). This load is equal to this case one has
't' x dy x t and moves an amount a' x ex, Then _ q II _ qas
U - 2Gt JJ dxdy - 2Gt - - - - - - - - - - - (12)
dU =~ 7Et dx dy
where S = J5 dxdy is the surface area of the
1 7' sheet.
=2 G tdxdy - (9 )
j-~7
p - ---------
6 U= PM
~u
Limit flo .. 0,
dU = Pda
T" f.-b--l "------...JL.j- a
,
Fig. A7.8
Fig. A7.9a
Solution: Thus
q=.1.= T In words,
2A Zbe ItThe ~ 9S chang;e ~ ~ energy ~
respect to deflection is equal t9 the associated
constant around the ?erlmeter of any section. load" .
Then ---- Sq. (l4) and the above quotation are state_
= q"S ments of the Theorem of Virtual t.or-k. The
U 1: 2Gt reader may find this theorem sta~quite dif-
ferently in the literature on rigid body me-
T'L(b+c) chanics but should be able to satisfy himSelf
4Gb lie it that the expressions are nevertheless compatible.
A useful restat~ment of the above theorem
THE TOTAL ELASTIC STRAIN is obtained by rewriting eq. (14) as
~GY OF A STRUCTURE
dU - Pda =0
The strain energy by its ~efinition is
always a positive quantity. I~ also is a It is next argued that if the Change in dis-
scalar quantity (one having TSgllltude but not placement dO 1s sufficiently small the load P
direction) and hence the total energy of a re~Alns senSibly constant and hence
composite structure, having a variety of ele-
ments under various loadings, is readily found dU - d(PO) =0
as a simple sum.
diU-po) =0 - - - - - - - (15)
The quantity U - PO is called the total
potent:'..al or' the system and eq, (15), r-esemb Ltng
as it 10es the ~th~~atical condition for the
+ J'{"d:<
ZAG
+Ii q",j.xc.y
2Gt
- - - - - - - - - - - - (13) ~inimum value of a function, is said to be a
statement of the Theorem of Minimum Potential.
From the foregoing i~is clear that the
The integral symbols and comacn use of "XiI Theorem of Mini~um Potential 1s a ~estatemen~ Of
an an index of integration should not be taken the 1':leorem or 'iirtual .vor-x .
too 11terally. It is probably best to read In structural analysiS the ~ost Dnportant
these terms as "the sum of so and so over the ~ses of these theorems are T~de in problems con-
str~c~ure" ra)her than "the integral of It, for cerning buckling instability ~d other non-
quite often the te~ are formed as simple linear1ties. No a?plications will be ~de at
sums without resort to the 8clculus. ~he 8al- this poInt .
culus is only ~sed as an aid in Some ~pp1ica
tions. A7.5 The Theorem of Complementary Energy and
It is seldom that all ~he terms o~ eq. casngnaoo'e Theorem.
(13) need Je employed :n a calculation. ~any
of the loacings, i~ actually ~resent, ~y be of Again in t~e case of an elastic bOdy, ex-
a localized or o~ a secondary nat~e and their aoinatior. of the arsa above the load-deformation
energy contribution ~ay be neglected. cu:ve shows that increments in this area (called
(',- ,.-
the complementary energy,U·), are related to the Force {Ibs .") Translation (inches)
load and deformation by (see Fig. A7.G(b). Moment (in. Ibs , ) Rotation (radians;
Torque (in. Ibs.) Rotation (r~dians)
Pressure (los/in"') vojuae (Ln ")
L' r Ou- Shear Flow (lbs/in) Area (in"')
TI;/'
III I/U:,
u- dU· ::6dP Any generalizations of t~e ~eanings of
. "P ,
,
' ~force~ and ~deflection~ are possible only so
, long as the units ~re such that their product
,, yields the units of strain errergy (in. IbS).
Once again !or emphasis it is repeated that,
while the complementary ~ature ot eqs. (14) and
(17) are clearly eVident, the use of eq. (17)
dU' (Castigliano's T~eorem) is restricted to linearlY
iiP =6 - - - - - - - - - - - - (16) elastic structures. A brief exa~ple will serve
tor illustration at the pOSSible pitfalls.
This is the Theorem or Complementary STIergy. The strain energy stored in an initiallY
Now tor the linearly elastic body a very straight uniform column under an axial load P
tmportant theorem follows since (Fig. A7.9c) when deflected into a hal! sine wave is
dU = dU· so that
P~U'
, ,
~ a. u
. rt x
Y :: Y oSlDT; M::PY
y Consider Flexural Energy Only
;.., 6 U .fM""" • P"Yd L
2EI 4EI
=0 Theretore
1 (17a)
P"L"
nilEI = P
The partial derl~tive sign in eq. (17a)
or
_ n 2 EI
indicates that the increment in strain energy is
due to a small change in the particular load P ,
P - r::r
all other loads held constant. i (Euler formula for uniform co nem ) .
Note that by "load~ and "detlection" may correct r-esuj.t ,
be meant: Application at Castigliano's Theorem.
eq. (17), leads to the erroneous result:
Load ASsOciated
Detlection dU = 0
dP
dU 2P6L~
dP = nAEI
• The proof of the theorem for the case of multiple ioads is
generally formulated more rigorously. appeal to a simple
diagram such as Fig. A7. 9c being less effective. se e, •• See Art. A17.6, Chapter A-I7 for detailed
for example, "Theory of ElaSticitY" by S. Timoshenko. derivation of this equation.
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A7.7
(incorrect) Then
'r1-i-1.-
~"~'~:
r
~~
.-$:' p 60'
I_
u = J:..
2GJ'
t q ads
t
=~ ~ cis
8AG~ t
Solution:
USE OF FICTITIOUS LOADS
The strain energy considered here was that
due to axial lo~ding in each of the four mem- In the follOWing example the desired detlec
bers. The load distribution was obtained fram tion is at the free end of the bar where no load
statics and the ener 5 J calculation was made in is applied. A fictitious load will be added for
tabular form a~ follows: purposes of the calculation. The rate of change
of strain ener~J with respect to this fictitious
load will be found atter which the load will be
TABLE A7.2 set equal to zero. This technique gives the de-
si~ed result in as much as the deflection 15
MEMBER S LaS
AE X iO"'5 s.aL x 10" equal to the rate £f change at stratn energy
LBS AE with respect to the load and such a rate exists
I OA -1. 50 P 40.0 136 0.66 P Ii even though the load itself be zero.
AS 2.50 P 50.0 11.8 26.48 P
AC 1.58 P 63.0 11.8 13.33 P Example Problem 10
DC -2. 12 P 4,5 13 2.79 P Compute the axial motion at the free end at
I;:; - 43.26 P Ii
the tapered bar of Fig. A7.12.
A7.8 DEFLECTIONS OF STRUCTURES
q - -flA'"
constant J
~
Ao u- 2L
x) q
A.o '" 40.,/in.
= 4 in. 4
L = 40 in.
erally are easier to
:Q:a'nple ?roblem 11
eval~ate.
Solution:
--- -:-L
L
3
: L
3 3"
Fig. A7.13
After addition of the ~istitious end load
R the axial load frem statics was found to Je
Solution:
S(x) = p. -I- q (L - x)
A fictitious load 21. ·NaS added at point B
Hence, since loadings other than tensile are of and the bending moment dia;ram was 1rawn in VNO
a secondary nature parts.
U = !
2
t0
S'dx
AE Mp
0<X<:L/3
I~
1 [Po + q IL - x)]'dx 0<Y<L/3
2 AoE (1 - 2
XL)
0< ;;<L/ 3
Po'
2A,E
I L
o
dx p.q JL 1L - x ) dx
(1 _ -"-)+ A,E
2L
- 1(_ X-)
0-2L
Fig. A7.13a
r/'
to be differentiated with respect to P, and the
subsequent setting of R = 0 would drop out both ,
the first and last terms. Hence only the second 1 ~ P + P, (L +
term was evaluated. + 2EI , [gIZL - 3Y) 3YIJ dy
9
L~
r
II
u = X + 2L Rq (1 _ 2) -I- X ,
A,E
+ .z.
2E!
3 [IP 2P1.) (L- 30)J do.
-I-
9
Then c
1 rL
/3 T:!1.us P
+ E:I :1), (2L - 3y) (L + 3y) dy Fig. A7.14b
1 _ PR)lI s in" Q Rd g
U = 2EI
Fig. A7.14a shows Putting P J the fictitious load equal to zero and
the vector resolution of Plan View integrating gave
ene applied torque To on
beam elements. TL (Q) =
ToGas Q and the moment au = ToR
li
4 - IT
( ~-EI
IT )
OVERT = IP 4
nl,{Q) = To sin 'g. AP-
plication of a fictitious E
vertical load'? (dOwn) at Since J = 21 and G ~ 12.6 the deflection was
the point of desired 1e- negative (up).
flection gave the loadings Fig. A7.14a
shown in Fig. A7.14b.
A7.7 Calculation of Structural Deflections by the Method of Dummy-Unit Loads (Method of Virtual Loads).
The strict application of the calculus to Castlg11ano Js theorem as in Art. A7.6 J leads to a
number of clli~bersame techniques ill-suitee to the solution of large complex structures. A more
flexiJle approach, readily adapted to improved ~book keeping~ techniques is the ~ 2f ~
Unit Loads developed lr.dependently JY J. C. ~arNell (lB64) and O. Z. Mohr (1674).
--- That the eQuations for the Method of Dummy-Unit Loads may be derived in a number of 'Nays is
attested to by the great variety of names applied to ~~is nethod in the literature ~. Presented
below are t~o deri'lations of the equations stemmir.g ~rom different viewpOints. One derivation ob-
tains the equations by a reinterpretation of the S~bols of Castigliano's theorem - essentially an
a?peal to the concepts of s~~ain er.er~J. The other derivation uses the ~rinciples of rigid ~ody
mechanics. Based as they are upon a commoe set of consistent assumptions, all the ~ethods must, of
course, yield the sa~e result.
Derivation of Squations for the Method of Dummy-Unit Loads
(Virtual Loads)
~ Variously called the Maxwell-Mohr Method, Method of Virtual Velocities, Method of Virtual Work, Method of Auxilliary
Loads, Dummy Loads, Method of Work, etc.
A7.10 DEFLECTIONS OF STRUCTURES
I Cont'd. IT Cont'd.
U = ~
2
J S "dX + ~
AE 2
J!'l"u
:::1 + ~ J T~f + - - - etc. small ~ount withcut dis~urbing the equl:1br1~
balar.ce, the work done is zero - obv1ously,
differentiate under the integral sign with since zero res41tant torse ~ovlng through a
respect to Pi to obtain distance does zero work.
6
1
= IS~dX
~
i
+ J M~dX
1
,1
+ - - - - - - - etc.
Consider now the set of equilibrium force~
applied to the truss of Fig. A7.15(b). The set
may be divided into ~~o parts: the "external
system", conSisting of the ~~it load applied at
Consider the symbolS the point whose deflection is ~eslred ane the
~S ~M ~T two reactions fiXing the line of referer.ce. ~d
<lP '"'JP'1'P'"" - - -- - etc. the "internal system" consisting of the axial
1 i i loads acting on the truss members to produce
Each of these 1s the "rate of change of so equilibrium. T~ese latter are denoted by the
-and-so with respect to Pi" or "how much so- synbol ~u". T~is set o~ forceS 15 considered
and-so changes when Pi changes a unit a~ount" OR small enough so as not to.affect t~e actual be-
EQUALLY, "the so-and-so loadln~ for a unit load havior of the structure during subsequent ap-
PI' plication of a real set of ~Jor loads. This
Thus, to compute these partial derivative unit load set is present solely fer ~thematical
terms one need only compute the internal load- reasons and is called a "virtual loading" or
ings due to a unit load (the virtual load) ap- "dummy loading".
plied at the point of desired deflection. For Assume now that t~e struct~e undergoes a
example, the term elM/elPt, could be computed in deformation due to application of a set of real
either ot the ~HO ways shown in Fig. A7.15a. loadS, the virtual leads ftgoing along for t~
ride n • Each ~ember of the structure suffers a
RATE METHOD UNIT METHOD deformation denoted by ~ ~ • The Virtual load-
ing system, bei~g in equl11b~1~ (zero reSUltant;
by hypo'thes t s ,10es work (I'virtual work") equal
to zero. Or. conSidering the s~bdivision of the
virtual loading system, t~e work done by the
external Virtual load must equal that absorbed
Fig. A7.15a by the internal vrr-tuaj loads. The work done by
the external virtual forces is equal to one
pound times the deflection at joint C, the re-
actions Rl. and Ra not moving. !~at is
I1 = Pl.X m = dummy loading
Exter~al Virt~l Work = l' x 6 c
(=~~J
oM
»l. = x x
The
=
internal Virtual Nark is the sum over the
structure of the products of th~ ~ember vir~ual
Likewise, d s~ Pc, where Pc is a load (real loads u by the member distortions~. That is,
or fictitious) applied at jOint c ot Fig. A7.15b, Internal Virtual ~ork = ~ u.~
is given by the loadings for the unit load ap-
plied as shown. Then equatin~ these works,
In practice the use at the WIlt load is
most convenient. Using the notation 1 x 0c = Z u·d .
_ oM If the deformations 6 are the result of
,m =~, elastic strains due to real member loads S then
1 Do = 8L/AE for each member and one has
v a
Pi'
q = aq
- ~
:V 6 = Z u 8L
c AE
for the unit loadings, the deflection equation The ar~~ent given above may be extended
becomes QUickly to include the internal virtual work of
6 =
1
Sud:< +
AE
IlmdX +
EI
~
GJ
f J J
virtual bending moments (TIl), t or s i on loads (t),
shear loads (v), and shear flows (q) doing work
during defo~tions due to real moments (M),
+ J If
~ + q qGfdY - - - - - - - - - (IS) torques (T), shear loads {Vr;-ind shear flows
(q). The general expression becomes
-1.0 -1.0 6 f Sud=< ~ J Ttc.x
1
=
AE
+ ;
EI
...
GJ
+J AG +J;q q
VvdX dXdy
Gt - - - - - (18
Fig'. A7.1Sb ~ Note that the deformations are not restricted to those due
to ~lastic strains only. They may be the result of elastic
"un loads due to a unit (virtual) load. or tnetesuc strains. temperature strains or misalign_
ment corrections.
. '.?:'"B.'P!
• ., II.
.....,. """""
~s 10· ~JCIO.
iv. Summation and/or Integration.
Here again note the general nature of the
AS .. Ibo
4.1•
AI ""Ib
8.2'1'0 10,500 1.' a a.. " a
..
CD ...." 3.074
3.0T4
g.T5'
g. '1'1
2,~0
2,250
a•
• • •
... "
a a
a
dummy-unit loads to the determination of ab-
solute and relative deflections, both rotation
and translatioo.
,., .. 5.385
5.385
5.5111
5.1111
_6,~
-5,2llO .0. T5 1
1 22.01
12.01
_21.~
_21.31
aB 3." 1.821
• c 1 a •
Example Problem 13
The pin-jointed truss of Fig. A7.l6 is
•• 50 lO.a 4.1128 _8, 'I'1lO 1-1.25
• 53."
•
acted upon by the external loading system shown.
BQ 50 3.4' 14.3. _.000 1." • a 111.80
•a
The ~ember loads are given on the fi~~re. Mem-
ber properties are given ~n Table A7.3. Find
00
SF ....
50 5.3G
3.0'1'4
, 11.320
13.012
.3, 'liO
3,000
•
•
•
• a a
the vertical movement of joint G and the hori-
zontal ~ovement of jOint H.
co
DB .. 3.074
3.0'1'4
13.012
13.012
_l,OOO
3.000
•
•
• •
• a
•
a
.
A 10
1000"
sods 2250
1000*
Ie 2250
1000'*
10
fj~gl~(?'bgl~~~ l
Answer-s : oG ::: 0.286"
VER
~"
0
-;o:"g40"
'-5250 .5250 (l N 1 O~OR ::: -.0587" (the negative Sign means
\~30" t30"'+ 30"1 the joint moves to the LEFT since the unit load
was drawn to the RIGHT in Fig. A7.16b).
2000t 2000t
Example Problem 14
Fig. A7.16 For the truss of Fig. A7.16 rlnd the fol-
Solution: lowing relative displacements of joints:
c) the movement of joint c in the direction
only the energy of axial loadings in the of a diagonal line joining c and F.
members was considered. Unit (virtual) loads
were applied successively at jOints G and H as d) the movement of joint G relative to a line
shown in Figs. A7.15a and A7.16b. All Sand u joining points F and H.
loadings were entered in Table A7.3 and the
~elatlve deflections are determined by
applying unit (Virtual) loads at the points
calculat~on ~or 0 = J Sudx
AE
was carried out JY where the deflections are desired and by support
ing such ~nlt load systems at the reference
forming the sum ot SuL/AE terms fer the ~embers poir.ts of the motion. ThUS, ror solution to
,art (c) a unit load system was applied as
UL
, e.
of the truss, .. 6 = ~"
'" SAE
_ shown in Fig. A7.16c and for the solution of
A7 12 DEfLECTIONS OF STRUCTURES
part (d) the system of l~it loads of r1g. A7.l6d Rotations, both absolute anQ "elative a~~
.cas used. Table A7.4 completes tne solution, dete~ined by a~ply1ng unit (virt~al) couples
the ~eal loads and member :leXibilities to the member or port~cr. of struc~ure ~hose
rotation is desired. The unit ccuple is ~esis~
(A~) being the s~e as for ex~ple problem 13. ed by reactior.s placed en t~e line of reference
for the rotation. Thus Figs. A7.16e ~nd A7.l6f
o -.6 roO -.375 -.375 show the unit (virtual) loadings fer par t.s (e)
o I~,,;>°t"
and (f) respectively. T~ble A7.5 completes the
/\/0
1/50*, 1/50# R=1/40'"
"~~?@?1
"us" loads "u ..." loads
AS 8.210 10,500 0 - 0 0 0
BE
eo
4.926
14.368
-8,750
'.000 ! 1.0
0 o
.e"
o
71.84 ....
0 AE
Be
6.a70
9. 7S~
10,500
a. 450
. 0.5
. D25
0
_. D.5
1."
.55
0
.0.55
DO
BF
.9.320
13. 012
-3,~0
2.000 -
-
....
0
·.50
.'20 0
·20.82
-21.8
·13. a
CD
EF
9.
~.591
7~9 2,250
-5,2:10
.020
·.025
0
0
.55
•. 73
0
0
CG 13. on ·1.000 0 .10.41 0 FG 5.591 ·5,2:10 ·.025 i 0 •. 73 0
DB 13.012 2.000 0 _.50 u ·13. a GB e.eu 0 0 0 0 0
z = 65.87 Z =·19.38 BE 4.926 -8, 750 0 0 0 0
BF 13.012 2,000 0 0 0 0
,
CG 13. 012 -1,000 0
I
0 , 0 0
Therefore the movement of joint c towards
, DB
joint F was 0 = .06587 inches and the motion of 13.012 2,000 0 0 0 0
I!l. = 0, far x cb
-1 (x - b), for x)b
,±r '°'1
a Fig. A7.18
,
C
, Fig. A7.17 J E1
l== L
c
4p =F:r
1 k 50
15 x~ (-.5X + 10)dX + ;;'1
1 h 80
15 x"
1' It
a Fig. A7.17a .... .1. 20 " 50
(.5x-40jdx
c r- B
)' .,l un1 t.
Fig. A7.17b
150x,l50 + l. [7.5 x' _ 600 x' l80
ooupl_
3 J20 S1 4 3 jso
ec,
=JMmdx
2;1 = ISO; [_ 11.72 + 6.25 + .300 - .400 +
For vi~tua1 loading See Fig. A7.11b 76.8 - 102.4 - 11. 72 + 25.0]
a = 0, x cb , - I, x>b
17.9 X lOll
EI
Hence !1mdX = (-?xi (-Ii ex ?x ex
Therefore deflection of point 0 relative to line
joining ~E is down because result comes out nega-
J Mmdx_PfL
SI - ~ b xcx 2EI
o
tive and therefore opposite to directicn of Vir-
tual load.
0' , Example Problem 18
If b = 0, ~ = 2EI x~
j:~~lr,1
1'''' lb/rt
'; II i i i
a.~ple Problem 17
Tor tne um r orm.Ly loaded cantilever beam
of Fig. A7.18, f~nd the deflection of point 0
relat~ve to the line joining points C and ~ on
the elast~c curve of the beam. This is repre-
sentative of a yr~ctlcal ?rOblem in aBronauttcs,
in that A3 :n.1.~ht represent a rear wing beam and
tn ~L-
(a)
•
",to
'2
L
2
(b)
Fig. A7.19
x-'1
m diag'l
wx e
• u
'oiL'
'2
( c)
R=l-1i' ~ '\
R=l/L unit
couple
1
Fig. A7.20
R=l/L
A7.14 DEFLECTIONS Ol" STRUCTURES
hence
where V 15 the shear at any section due ~o g~ven
L(WLx WX')
2 - T h d x =r,(WLx'_'/IX') 1!. JL
loads. v = shear at any sect10n due to unit
•f EI
~ 4 6. EI
c
hypothetical load at 'tae point where tne aer iec-,
tion is wanted and acting in the desired direction
of the deflection. The reactions to the hypo-
1 whV' thetical unit load fix the line of reference for
• 12 EI the deflection.
A is the cross-sectional area and Es the
To tind angular deflection at C apply a unit
imaginary couple at C with reactions at C and D. modulus of rigidity. Equation (a) is slightly
F.ig. A7.20 shows the virtual m diagram. in error as the Shear:ng stress is not ~~i~orn
over the cross-section. e.g. being ~raballc ~cr
a rectan~~lar section. However. ~he average
So =JL~EI
1.
EI
= fL (WLx _
2
WX') ~ dx
2L J;I,
= ijL
shearing stress gives close results.
For a unito~ load of w per unit le~gth, the
• , L center deflection on a s~ply supported beam 1s: -
(W~' - w:: ) dx = ,;\
1
= 24 Ef
[~' - w:~ J
wL;'
c
°center = 2f~~=2
L AE s
,-
~
wL'
r
e
(wL
2
wx)
AE s
1
~
ax
"'~dLdy
seldom. greater than 12
~--
6
-f-5-' ----ft
T,.
Thus the Shearing deflection in percent of
the bending deflection equals 4.1~ for a ~ ratiO
Fig. A7.21
t '+ lO'4
'O
A
IF==='
Unit.
Load.
acting in the plane of the desired de-
flection. (inch Ibs/inch lb)
Example Problem
Fig. A7.22
Example Problem 20
Find the vertical deflection of free end A Fig. A7.Z3 shows a cantilever landing gear
due to shear deformation for beam of Fig. A7.22 strut-axle unit ABC lying in XY plane. A load
assuming shearing stress uniform over cross- of 1000* is applied to axle at point A normal to
section, and AE constant. XY plane. Find the deflection of point A no~l
s to XY plane. Assume strut and axle are tubular
"' _ I Vvdx and of constant section.
vA., - -
AE-s Solution:
V = 100' for x = 0 to 10
V 150 for x 10 to 20 The loading shown causes both bending and
twisting of the strut axle unit. First find
v = 1 for x = a to 20 bending and torsional moments on axle and strut
hence due to 1000* load.
°A 1
AEs to ...L
0 100·1 dx + AE
s
fO
10
150·1 dx
I
Y
10 20
1
= IE:s [lOOX] + ~.[150X] 2500 I / 36" 3°1
0
1/ \\·~~3'i s:
AE s
0 10 ,.
Method_of Virtual Work Applied to Torsion of Cylindrical Bars. B I
B A
The angle of twist of a circular shaft due Fig. A7.23 Fig. A7.24
to a torsional moment may be found by similar
reasoning as used in previous articles for find- Ttdx
ing deflection due to bending or shear forces.
E J
The resulting expressions are: - s
o = JTEts~ - - - - - - - - - - - - - (A)
Member AS M = 1000 x, (for x
T = 0
o to 3)
6 = J I'!mdx -+-
81
JTtdx
E J
A7.27 ~s an explOded vi~N of the beam shaNing
the unit (Virtual) loads.
s
= tr J: 1
1000 X . X dx • Er f: 6
K1000 x • 1026)
(x + 1.026)J dx -+-
1
E-J
f36 (2820) (2.82) dx
s 0
rL I
20"
t=.0
32
1l t=.032 1
A=.08
t=.032
A=.08
A-.05"
TIS"
--..L
panels were made using the shear flows en the
(nearly) horizontal sides as average values,
assumed constant over the panel. with this
E
~ 20" ----..+- -20"
rF I G
-+- 20"
H+ P.j80"*
-4 Simplification
Gt
= qAV 55 qAV q,w :ixdy
Fig. A7.2S
where S 15 the panel area.
Solution: The calculation was completad in Table
A7.6.
It was assumed that the webs did not buckle
and carried shear only. Fig. A7.26 is an ex- c==.:===;::::~~~====
•
ploded view of the beam Showing the internal
real loads carried as determined by statics:
3. OP_ 2. 184P 1. 201P : :.:.,. p, , ".'"
~ID·
-··*~i ;~~
·'.00 P ,_L,H P, ·1.00,_:.011 .un I .,.,., ,
.'.'M "'-'-:10""'.01' 'u. ".0 P
.04
3.0P...".. 2:"I84P
.O!~~~~. W I 1.201p P ...:,:" ,-',W")
... ~.r'._.·
",,"'''1 i •
.,.\< ""I
i.,'. • ",.i,~......5. •".,
I : • 171.0 P."
C·D-<l-HI._P' : .1
• The equations of statics for tapered beam webs are
derived in Art. AIS.18, Ch. A-IS.
ANALYSIS AND DESIGN OF FLIGHT VEH LE TR C RE A7.17
L - x
Solution: u
U h
(~ - "L)
The thermal coefficient of expansion of the
rod material was u. Hence a rod element of T~en ~he ceflection equation was
length ax experienced a thermal defo~tlon
a = a . T • ax. Application of a unit load at
the bar end gave u = 1. T~eretore ~J~
o n
• a T
T . dx
SXa.m.ple Pr::;ble:n 24
~he ~irst step in computing the the~l
stresses i~ a closed ring (3 times indeterminate)
involves cutting the ring to ~ke i~ statically
23
2.X8.!l:p19 Proj19m determinate and r1~d1ng the relative movement of
~heidealized ~No-flange cantilever beam the ~NO cut faces.
of ?ig. A7.2Sa undergoes rapid ~eatlng of the Fig. A7.3Oa s~cws a unifo~ circular ring
upper flange to a ~emperature T, uni:orm span- whose insi~e surface is heat~d to a temperature
Nise, above that of the lower flange. Deter- T above the outside surface. The temperature is
xlne the resulting displaeemer.t of the free end. constant ar8und the c~rcum!erer.ce and is assumed
A7.l8 DEFLECTIONS OF STRUCTURES
Remarks:
In the three elementary examples given
above no stresse~ were developed inasmuch as the
idealizations yielded statically dete~inate
structures which, with no loads applied, can
have no stresses. Indeterminate str~ctures are
treated in Chapter A.a.
A7.9 Matrix Methods in Deflection Calculations.
• b Introduction. There is much to recommend the 'lse of
Flg. A7.30 natr rx methads'~ for the hand Ling of 'the quarrt Lty
of data ariSing in the solutions ot stress and
Solution: deflection calculations ot complex structures:
~he data is presented in a form suitable for use
An element of the beam of length Rd¢ is in the routine calculatory procedures of hi~~
shawn in Fig. A7.30ce Due to the linear tem- speed digital computersj a fleXibility of opera-
perature variation an angular change dQ = tion is present which permits the solution of ad-
RaT ditional related proble~ by a simple expansion
~ d$ occurred in the element. The Change in
of the program; The notation itsel: suggests new
length of the midline (centroid) ot the section and improved methods both of theoretical ap-
was 4 = R~T d~. Unit redundant loads were ap- proach and work division.
The ~ethods and notations employed here ar.d
plied at· the cut surface as shown in Fig. A7.30b later are essentially those presented by Wehle
giving the following unit loadings around the and Lansing@ in adapting the Method of D1..1L1!IlY-
ring. Unit Loads to matriX notation. Other appropriate
From. unit redundant couple (X)
references are listed in the bibliography.
m
a
-R sin $
ua - sin $
The deflection equation by the dummy-unit load
method is (.) (b)
4Z
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A7.l9
~ ql)", JI
~qg
\ h
q1.
wise readilY written, With proper account taken
q, (b) (d) tor the cross influence at one :orce upon an-
other (the na a3 qa q~~ term).
M = q1.X O<x<L1. q. q'f No'te that
qtj I)
M;Q3+qaY O<y<L. (e)
= = "jl - - - - - - - - (20)
Fig. A7.32. Some possible choices
of generalized forces.
Hence
~ •• (Maxwell's Reciprocal Theorem)
-("Relative displacements in the individual member") ,.
A7.20 DEFLECTIONS OF STRUCTURES
,,
+Q"q1. (l"l + q"q" (l"3 ... - - - _ - _ - - - _I
,
,I
+qfoQ1. au .
,,
- - - - - - - - - - - - - - -1
,, Fig. A7.32f
,,
,
I
I I
+ qNql aN' + - - - - - - -- - - + qN qN (lNN
In matrix notation
or, more concisely, The matrix [Jim] 1s called the ~1t load
dlstr1butlon~ inasmuch as any one col'~ at
A7.21
L J r,L':rJ1J)1:1 i,I
plication itself yields a summation to complete
2U p
m
P - - - - - - - - - (25) the calculation.
?rcm this last discussion it is clear that
eq , (26a) also may be derived by formulating the
Sq. (25) expresses the strain energy as a Dummy-Unit Load equations (Art. A7.7) in matrix
function of the external ~pplied loads. notation. -
In the ~pecific example being used To illustrate the application of the ~trix
methods presented thus far, a brief and element-
ary example 1s worked with those tools already
~ rLa1 1L'J
10
developed.
Example Problem 25
Determine the influence coefficient matrix
rorxransverae forces to be applied to the uni-
DEFLECTtONS BY CASTIGLIANO'S THEORE!1 form cantilever ot Fig. A7.33 at the three points
Application of Castigllano's theorem to indicated.
eq , (25),
,@ ,lID ,®
(25)
t b I.
3"""" 3 -e- 3
b, b J
c:::::::=::=J.~ q1.
=:J~)q"
yields q.
="q
q. '
I~ ) (26 )
I ~p n .f Fig. A7.33
Solution:
The steps in passing to eq. (26) may be
demonstrated readily by writing out sq. (25) T~e choice and numbering at genera11zed
for, say, a set of three loads (a, n = 1,2,3), forces are shawn on the figure. These forces
differentiating successively with respect to were placed so that pr~vlously derived express-
Pl.. ,'Pa and P" and then re-collecting in -catrr tx ions for the a's could be used. The following
rorn, member flexibility coefficients were computed.
The matrix ~mnJ gives the deflection at Note that the only non-zero coefficients ot
mixed subscripts (i not equal to j) are those
the ext ernal points "a" ror urn t values of the for loads common to an element.
loads? and 1s therefore, by definition, the
n
l
of influence coefficients.
~trix
---- L/~.dx L'
COt1PARISON ·.ITH DUMt'lY-UNIT LOADS SQUATIONS
al,l = Q ~ = BIEr = aa::a = a.....
It is instr~ctive to Nrtte eq. (26) out as This expression was adopted from that de-
veloped for a transverse shear force on a canti-
lever beam segment in the preceding illustrative
example.
ing to
Ai:
L...,. T~e ma tr tx pr'oduct l'J:l] :Pn (
This expression is for the cross influence
;ives the :nember load distrlb~t~ons due to the af a couple ~~d a shear load an a cantilever
real applied loads, hence these are the "B" segment. Collecting in matrix form,
leads. Finally the operation of matrix :nulti-
A7.22 DEFLECTIONS OF STRUCTURES
as per eq • (26).
1
27 0 0 0 0 T'he matrix G!Il11] is seen to be svmetc-rc
0
1 1
0 0 about the main diagonal as it ~ust be: from
27 6L reciprocal theorsm A
~~Nell's .c.'
nm (see
[a !] l 3EI
L'
0
6L
1 1
V 0 0
eq. (20)). mn
~
~
1 L/3 1 2L/j 0 1..
27 1..
6L
o 0 1 1 a 1 j 0
Fig. A7.3-4a
-- d•q/"
'U _ 1 JLx , dx
- AE =
L'
3AE
" 0
~'U = AE
= oqi)qj IJLxdx 0 = 2AE
L'
•
.Y!
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A7.23
S = surface area
a11 =..§...
Fig. A7. 34c::::::!:d Gt
~e~ v~tes lin~rlY :J: ~
Since by statics qj =
=
- h, Fig. A7.34g
Ql -.-, one could use Q as
a. j
an alternate choice of generalized torce and
BEAm
The energy in the uniform beam 0: Fig. ajj -
_(h.)' S.
h. lit
A7.34d is given by
TORSION BAR
qj r1---====<'
1 1 q,
A uniform shaft under torque qi has strain
L energy
L
Then Fig. A7.34d Then 0.11 = GJ •
7-£
A7 24 DEFLECTIONS OF STRUCTURES
Con.tant Axial
Load; 'l.i-'l.j 'K 1 2 3 4
-=~
1 0 1.0 0 1.0
!ben
"'1.- e(i1'" ~jj'" 0 0
2 0 0
OC1j+<:o.j1
.. L!AE 3 0 -1.0 C 0
~i~
Fig. A7.35 Fig. A7.35a 4 0 1.25 0 1.25
=
deflections at points E and r. 5 0 -1.03 0 -1.03
Solution: 6 -1.0 - .75 0 - .75
The member flexibility coefficient for a 7 1.13 .848 1.13 .848
unitorm bar under constant axial load is
L/AE. Fig. A7.36a gives the ~umberlng scheme 8 -.825 -.165 -.825 -1.55
applied to the members and the q's (these being
one and the same, since q is constant in a given 9 .20 0.40 .20 0.40
member). Fig. A7.36b shows the numbering scheme
adopted for the external loading points.
Then the matrix triple product
plo Pa
~Jn]
tsl3J3 DJsL
1 2
t=] = Emi] [OiJ]
6 a a a a a 185 a a a
could have been achieved by leaving out the
7 a 0 0 a 0 a 155 a a
8
9
a
a
a
a
0
a
0
a
0
a
a
a
a
a
55.7
a
a
229
first two raws of @",i] 0he transpose of @bl])
The matrix fo~ of equation above ~s useful
in organiZing the computation of deflections for
In the case of a Pin jointed truss, where a number of different loading conditions. Thus
only a single generalized force is required to should there be several different sets of ex-
describe the strain energy per member, the ternal loads P corresponding to various load-
matrix or member fleXibility coeffic18nts 15 a n,
diagonal matrix as above. ing conditions, each set is placed in column
fo~ giving the loads as the rectan~~lar matriX
Unit lead distributions were obtained by
placing unit loads succeSSively at external
loading points 1, 2, 3 and 4 (Fig. A7.36b). ~nk] , k different numerical :ubscr'1pts for the
The results were collected in matrix form as
load conditions. The matrix product
,
tip'
A7.25
LAB~J
L'
~llJlJ
iL-':c-J<.......
."- = -"
. 'nn
P
nk
(26b) au a"
AB;SEI
a••
L;~
r1SW 21'1e5 t.ne 'deflections at each point (:r,; for
t~e ~arlous load conditions (k). a" a .. a" = L8~
3EI GJ
~a.T:J.ple ?roblem 27
= a,. = La~
Deflections at points 1, 2, 3 and 4 of the
truss of ?lg. A7.35 are Geslred for the follow- a,. BO a •• a .. = L8~
ing loading conditions: 2EI EI
~'"'J
1 8 0 0 0 0 0 sae 0 15.500
;; 257 252 257 252 800 -2100 -1200
389 789 252 789 450 -1750 -1100
Exa."'11ple Problem 29
~he following ~e~ber flexibility coef::cients T~e bea~ at example problem 21 is to be re-
'Nere oaterainec solved by the T~trlx wethoas ?resented ~ereln.
•
A7.26 DEFLECTIONS" OF STRUCTURES
.... =0'
1
17 8
•
'-'.
....
..... 2500
8.89
®
Unit load Yalues were obtained as in Fig. Stringer Areas
A7.27, considered to be external loading num- Constant", . 33 in a 20" 1:-5"
ber "two". S1ml1ar diagrams were drawn for unit '-J.-
, 15.. -J
'
T
loads at points "one" (H) and "three" (Ft(llt'. rig.
A. 7-2"
1 2 3 Fig. A7.39a
1 1.0 0 a Solution:
2 .0667 a a
Fig. A7.39b is an exploded view of the beam
3 1.201 a a showing the placement and numbering of the in-
~~
4 a 1.0 0 ternal generalized forces. Note that only the
upper side of the beam was numbered, the lower
5 .0540 .0600 a side being identical by symmetry.
6 2.184 a Member fleXibility coefficients were com-
11.0:
7 a 1.0 puted by the formulas ot Art. A7.l0 and entered
in ~trix form as below. Note tr2t all entries
8 .0447 .0496 .0545 tor Which there Nere corresponding leads on the
9 3.00 2.00 1.00 lower surface of the beam were dOUbled. ay
this ~eans the total strain ener~J of the beam
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES
A7 27
deflection at tne ccmt (c r . eq , 103). ..~e now In rig. r co~sider Mex ~3 a load on a si~~ly
show tr~t the ceflecticr- expression, uSin: :he 81
above equat~on, is the SaT.9 as the bending supported bea~ a~c dete~ine the benG:n; 200ent
moment expreSSion fer ~ simple beam loaded by at pOints band c G~e ~o ~dx acting at pOl~t c.
--"- ~ Md.x "IT
an elastic weight 21 Mdx L XLdx
In Fig. A7-40, the la~cing at (a) pr ocuces
4EI :2 BEl
the real moments of (0). Consider the ce r jeo- L
tlons ot points 3 and C due to the ar.gular
ch~nge Mdx
~ in a 0 ea~ ~~emen t
",1 a tA" (Fig. A'7-4C':).
"
values of ~~e te~ bending ~o~e~ts at
~hese
points 0 and care ldentica: to :~e deflections
at band c by the virtual ~ork equa:ions. ~~e
noment diagram ~ for a unit ~oad at b a~d c
(Figs. d and e) is numerically pr sc t se Ly tne sane
B
as the influence line for illo~e~t ~t ~oi~ts b
41--, -....--
L I L
4
l.h-Lhj
-1- 4 ' 4
and c.
Therefore deflections of a si~ple bea~ can
I') be determined by cons Ider mg the :1 cur-ve as an
81
~WEI diagram ~ imaginary beam loading. ~he bending ~oment at
any pOint due to this !1 load.ing equals the 1e-
.., !"odx
(h) EI
flection of the beam ili~der the gi7en loads.
Mdx
EIpj Likewise it is easily pr)ved :~zt :he ~n~~
lar change at any section of a sl~ply supported
(c) beam is equal to the shear at that section due to
~ !'! dt azr-am acting as a beam Lcac .
I Sol
-= L~4 ~ A1.13 Example Problems
(d) mb, moment diagram for unit
load acting downward at point B Examnle Problem 31. ~ind :he ~2rt~cal
3 deflection and slope of paints?. and b :or be~
r16 L and loading sho'Nn in Fig. A7.41. T~e lower F~g.
shows the :noment c Iagr-am for load? acting at
f ----
center at a sl~ple beam.
(el me" unit load at point C
r-~1 IP
~-;P
Mdx
lEI
EI Constant
If) a b
3 Mdx
4EI
Fig. A7.40
lMdx
4EI
P f L
2
2
For a unit load at point b, Fig. d shows the m
PL
T Fig. A7.11
diagram. The value of m at the midpOint of dx
(point a) = LIS. Hence -6"-
L L
2
°b = ~~ . ~ ~~ PL'
16 PL'
16
For deflection ot paint c, ,1raw m diagram ror a Deflection at point a 8Guals bending ~cm
unit load at C (see Fig. e). Value or ~ on ele- ent due to M diagram as a load di vtoec by EI.
aerrt dx = L (See lower Fig. at Fig. A7.~1)
16
PL2 L PL2 L) 1 11 PL3
Hence 6a ( 16 4. - 64 12 EI = 768 EI
00
Db = (~2 ~ i~2 ~ )Elr = 1a ~~3
.. For srmoncrty the points A, Band C were placed at the one-
quarter span points. The reader may satisfy himself with The angUlar c~nge ot any point equals the
the general character of the proof by substituting xA, xB and shear due to ~JEI diagram as a load.
XC for the point locations and then following through the
argument once again. = (PL2 _ PL2)2:.. = 3 PL2
Ua 16 64 E1 64 EI
-, 7
A7.29
rection of beam
axis. ) rHear Beam
I Anercn E FIA
A B C
?rcbl,:::n ·32. De t cr-ntne ':;;'a cer Iec t Lcn
~x":.:nJ'"'!
,
of a sixple be~~ loaded unifo~y as shown in ..-- Ercnt Bean-
Fig. A7.42. The bending moment exor ess i on for /
a unlforn loac M = wlx - ,NX2 or ~arabolic as ,
2" 2" Plan View I: 2 Wmg
shown in F'1;. A7.42a. The deflection at mfd-
?oint equals the Jending J10ment due to M dia-
gram as a load.
r 0"1
.,.....Air Load
IiffffHHHflH fJ HJ f OHm!! H Hbj"
25li'iin.
3
rt:
I IO US63 5.S 21000
t~~ee ?oints. In the desi~ of the aileron beam 2 IO 20063 6.5 30900
and similarly for cases of wihg flaps this de- J IO 31063 7.S 41450
flection produces critical bending ~oments. As- •
·5
IO
IO
44580
60580
'.5
9.5
52300
63850
suming that the running load distriblted to the 6 IO 79050 12.0 6.5940
rear beam as the wing bends as a uni~ is as shown ,s
7 10
10
1000.50
123$50
16.0
20.0
62530
61770
i~ the Fig., tind the deflection of pOint 3 with 10 149550 24.0 62320
10 10 118150 28.0 63600
respect to straight line joining ~oints A and C,
0 0 0 0 0 0 0 0 0
which will be the deflection of £ with respect to 0 OJ e- eo ~ en 0 en 0
line jOining D and F it bracket deflection is co eo c- 0> co eo ~ 0
ro
.o
OJ ~
so
"':'J
-o 0'_0
eo OJ
.c l.D
~ ~
OJ
neglected. 7he moment or inertia of the rear
beam between A and C 'varies as indicated in
'" '"
:he Table A7.6
A7.30 DEFLECTIONS OF STRUCTURES
Bending ~oment at paint B due to above elastic A7.14 Deflections of Beams by Moment Area Method·
loading = 7,100,000 .'. de~lection at B ~elatlve For certain types of beam problems the oeth-
to line AC = 7,100,000 = .71 inch ad of moment areas has advantages and this ~eth
E::IO,OOO,COO ad is frequently used in routine analysis.
Example Problem 34 Angular Change PrinCiple. Fig. A7.44 shows a
Fig. A7.43a shows a section of a cantilev- cantilever beam. Let it be required to deter-
er wing sea plane. The wing beams are attached ~ine the angular change of the elastic line be-
to the hull at paints A and B. Due to wing ~eent~pointsA~8~to~givenlMdl~.
loads the wing will deflect vertically relative From the equation of virtual work, we have
to attachment points AB. Thus installations
.,
such as piping, controls, etc., ~ust be so lo-
cated as not to interfere with the wing deflec-
tions between A and B. For illustrative pur-
poses a simplified loading r~ been assumed
as shown in the figure. EI has been assumed as
constant whereas the practical case would in-
volve variable I. For the given loading deter-
~ine the deflection of point C with respect to
the support pOints A and B. Also determine the
vertical deflection of the tip ppints D and E.
0
0
::.
0
0
0
::,
0 0
'\ ,
\
N 0
~r400"
~
300'~1w
, 40" I Fig. A7.43a
D A 0 B EI Constant E
a
Moment Diagram
o!
o
in.'
1000
""
0,
0,
, Fig. A7.43b
\ c
.,
0·1
,
Fig. A7.43c
!
Fig. A7.43d
° fA
~ a;:
~i ~ ,~o ~4
N
Hence Max
r- 576 " -
8 = B This
-EI x r-- 720" ...
eXDression ~epresants the 1st noment of the
M uiagracaoout a vertical t.hru B. 'fhus the Solution:-
~I Due to symmetry ot loading, the tangent to
deflection principle ot the moment area method the deflected elastic l1na at the center line of
can be stated as follows:- "The deflection of a airplane is horizontal. Therefore, we will find
point A on the elastic line at a beam in bending the deflection of points A or B away tram the
normal to the tangent of ~he elastic line at a horizontal tangent ot the deflected beam at pOint
point 3 is equal numerically to the statical mO- C ~hich is equivalent to vertical deflection of
ment or the !1 area between points "An and "B" C with respect to line AB.
. EI Thus to find vertical deflection or A with
about point A". respect to horizontal tangent at C take moments
of the M diagram as a load between points A and
Illustrative Problems
EI
ExaT.ple Problem 35. Determine the Slope C about point A.
and vertical deflection at the rree end B of the
cantilever beam shown in Fig. A7.45. £1 1s con- Whence
stant. (area) (arm)
1~~0 (tangent at C) =E~ (650 ~ 646) 40 x ZO =
1--- L ------i
1
EI (518400) = deflection of C normal to AB.
Fig. A7.45 To tind the vertical deflection of the tip point
~~am D with respect to line AB. tirst rind deflection
of D with ~espect to horizontal tangent at C and
~ PL2 __ .!L--J subtract deflection of a ~ith respect to tangent
2' 3 at C.
Solutlon:- The ~oment diagram ror given load is
triangular as shown in rig. A7.45. Since the ~ (respect to tangent at C) = E; (40000 x 267 +
beam is fixed at A, the elastic line at A is
horizontal or slope is zero. Therefore true 1Z7500 x
576 • 25920 7Z0) = x
(10Z700, 000) ir
slope at 8 equals ar.gular c~~ge becNeen A and 8
(See Fig. A7.46 for areas and arms or M/EI dia-
wm cn equals area of 'nomerrt diagram between A gram). subtracting the deflection of A with ~e
and B diVided by E1.
spect to C as found above we obtain
Hence 1 PL Z
"8 = (-PLo LIZ) £I - 2EI l~gO (respect to Ltne AS) = Ell (102,700,000
The vertical deflection at B is equal to the 1st
518400) = SI (102,180,000)
:no~ent of the :noment diag~am about ~oint S di-
Vided by S1, s~nce tangent to elastic curve at A
1s norlzontal aue to ~ixed support.
A7.32 DEFLECTIONS OF STRUCTURES
A7.15 Beam Fixed End Moments by ~e~hod of Area :1A =- Pab 2 and MB =- ?ta 2 / L " l ' o ' h e r e b:: (L-a)
lIOllIents L'"""
From the rNQ prinCiples of area ~o~ent5 ~s To find the fixed end ~owents ~or a te~~ ""itt
given in Art. A7.14, it is evident that the de- variable moment of Iner-t-Ia use the t""iI crucrana
flection and 510;e of the elastic curve depend in place of :he ~oment clagraos.
on the amount of bending ~oment area and its lo-
2xa~ple Problem 37
cation or its center of graVity. Fig. A7.48 shows a :1x-ended je~~ carry~ng
Fig. A7.47 shows a beam r1xed at the ends
two concentrated loads. finS the f~xed-end
and car~Jlng a single load P as shown. T~e oend
~ng moment shown in (c) can be considered as
moments MA and roB'
made up or two parts, namely that for a load P 200*
acting on a simply supported beam wmcn gives j 100. I t:',
the triangular diagram Hlth value Pa (L-a)/L
tor the moment at the load ~otnt, and secondly
a trapezoidal moment diagra~ of negative sign
4t=
M9"......:.-....12"~9" +.-
__ I ~tB ; (a)
I
(I)
jOints of a trussed structure are required, the
~ ,/1'1\. I
g
method of elastic ~elghts ~y save considerable : d <ltokih
::~e over the method of virtual work used in r-rr : 6 Panels @ p
previous articles of this chapter. !he illethod Bar be (2)
in general consists of finding the illagnitude and
location of the elastic ve i gnt for each member u6L _ 4P6~L.L Shortening
~lection Diagram
of a truss due to a strain from a given truss "F" - ---sr-- .: . Lbc of Member be
loading or condition and applying these elastic o Elastic Load on Imaginary Beam
weights as concentrated loads on an imaginary I.!..
3r
uLb
c
t .::.Lbc
3r (3)
beam. The bending aoment on this imaginary beam
due to this elastic loading equals numerically ~~entCurve (4)
the deflection of the given truss structure. for Elastic Loads
Consider the tr~ss of dia;ram (1) of Fig.
A7.49. Diagram (2) shows the deflection curve Bar ck 6L(d+6p)
for the truss for a ~L shortening at member be,
all ether members considered rigid. This de- /?\~ (5)
flection diagram can be determined by the vir-
tual work ex pression 6 = u~L. Thus for deflec-
~ection Curve
for !J. L Lengthening
tion of jOint O. apply a uhit vertical load act- 6L • djip t
SL, 1M. d+2p of Member ck
ing down at joint O. The stress m in bar hc due f. rl, L"'[':'" "'--Ela~tiC Loads on
to this unit load = 2 2P = 4P. Therefore Rigid Arm t a K .::. Ld Imaginary (6)
:3 r- 3r t:. L! d+6W ~ Beam
6pr~ l:.L(d+6p)
°0
=
3r
4P The deflection at other
lower chord jOints could ~iiiz:
be round in a si~ilar
~anner by placing a unit load at these joints.
Diagram (2) shows the resu~ting deflection curve. Fig. A7.49
7hls diagram is ~lalnly the influence line for
stress in bar be multiplied by ~Lbc' Table A 7.7 Equations for Elastic Weights
Diagram (3) shows an imaglnarJ heam loaded Elastic- Weight for Chord Members
~ith an elastic load 6Lbc acting along a verti- (See Member ab)
r Upper Chord Lower Chord
cal line thru joint O. the moment center for ob- w=~L
taining the stress in bar be. The beam reac- h
tions ~cr this elastic loading are also given.
Diagram (4) shows the bea:n bending moment dia-
gram due to the elastic load at point O. I~ is &1:0""i
tab t
noticed that this moment diagram is identical to
the deflection diagram for the truss as shown in elL
w=-
dt agr-am (2). I r
a b
~\t
The elastic we Lght of a member' is therefore
equal to the =ember defo~ation diVided by the
a~ r to its ~oment center. If this elastic
load is applied to an i~gi~ary beam correspond-
/y\/\r
f 0 j,L tab
w=-
ing to the truss ~ower chord, the bendi~g ::noment r
on this Imagfnar'y beam. ,/.111 equal to the true
tr~ss deflection. b
i/NJ\I~
Diagram 5. 6 and 7 of Fig. A7.49 gives a
si~ilar study and the resul~s for a ~ lengthen-
lng of ::nember CK. The stress moment center :or
.~~'
I aL
this ciagonal ~ember lies at point O~ which 11es ~
,
outs1~e the truss. The elastic Neight ~ at w= Member Axial Deformation J.L
Perpendicular Arm to Moment Center r
rl
point 0 1 can be replaced by an equivalent system The moment center 0 of a chord member is the
at points 0 and ~ on the imaginary beam as shown intersection of the other t ..... o member-s cut by the
in Diagram (6). These elastic loads prOduce a section used in determining the load in that memb-
bending ~oment diagr~ (Diagram 7) identical to er by the method of moments
the deflection dia~am of diagram (5). The Signor the elastic weight w for a. chord
Table A7.7 gives a s~~J of the equations member is pj.ua if it tends to produce downwardde-
for the elastic .vefght.s of truss chord. and 'Neb r Iec t t on of its point of application. Thus for a
~embers together Nith their location and sign.
simple truss compression in top chord or tension
in botton chord pr-oduces downward or positive
elastic weight
• tt rmn=rry
/<'_:'~
- ~/"</\~
I N
a
~( bI CI d~ ...<iaI>...e
wO 6L 000 '2000 2000 '2000 i5000
r ...... __ - ~ ~
1,_-
k:-::.._____ ,i'''l,e, b 4@30" = 120"
,
Id
I a
p. bw=~=2.L
b
Q
FIg. A I
this step are given in the figure, the stresses
Fig. A7.50
~
lations. Table ..1.7:8
I\&i\l\
t \Q i fQ r
Chord Member E1as~ic Weigbts
For a truas diagonal member the elastic AB ,I ao .2172 _5630 _.0268 24,I·Q011171 , I
weights P & Q have opposite signs and are as-
~i,
so
ao
.2172 _7500
.2172 _5630
_.0357
-.0256 I 24 .0014871
24 .001117 ,c I
<
Truss Verticals
(See Member ab)
· · . ,I
•" •
• •• w.d
•
'L •
~ I
.-
il
~I
~!
.,
I
I Q -
II
~
,
:1 " II "
. •• e,
~
'1 '1
•• •"
e,
} I •
,.
A'
28.25 .2421_5880 ._.0238112.75
.146 47101.0314 ....
_.00185
_.00246
a :12.75 . 0018~
:1 ""
.00246
A
,
P,
d
I
~L Bo
"
"
.1461_~·3!>51_.01571
.093 11771 .0123
..
"
_.00123
_.00095.5
•, "
.00123
.000955
•,
PoQ'"
,c " .09311177 .0123 " _.00096.5 C " .00096.5 c
c, " .146 _23.55 _.01.57 " _.00123 " .00123 C
0
"13
0
0
;i
g g
0
r--
~
:3
0
c-
g
;i
OJ
0
0
130 co
~
0
er cut by index section 1-1 used in tlnding 0 0 0 0 0 Fig. A7.5l
stress in ab by method at sections.
1 1 1 1 J' I I'
a
A b c C d D- e
B
.17.17 Solution of Z%aaple PrOble... ~ 8 spaces at 15' _
The method at elastic weights as applied to Ra = .005412 Re = .005412
truss detlection can be best explained by the
solution at several Simple typical trusses. The detlection at any joint equals the bending
Example Problem 38 moment on the imaginary beam of Fig. A7.51.
Fig. A7.50 shows a simply supported t~JSs Den. at J,. = (.005412 + .00185)15 = .007252 x 15 =
symmetrically loaded. Since the axial detorma- .109"
tiona in all the members must be tound, the Den. at b = .109 + (.007262 - .000507)15· .109 +
tirst step is to tind the loads in all the memb-
ars due to the given loading. The results ot Den. at B .
.006755 x 15 = .209"
.209 + (.006755 - .002347)15 = .209 +
.004408 x 15 . .275
,
ANAT·YSIS AND DESIr.N OF Fl.Tr.HT VEHICLE STRUCTURES A7.35
nen. at c = .275+ (.004408- .0027)15= .301" beam loaded with the elastic weights from Table
The slope of the elastic curve at the truss join A7.1l. Table A7.12 gives the calculation for the
points equals the vertical shear at these points jOint deflections.
for the beam of Fig. A7.5l.
Example Problem 39
Find the vertical deflection of the joints
of the Pratt truss as shown in Fig. A7.52. The
member detormationsAL for each member due to the
given loading are written adjacent to each memb-
er. Table A7.l0 gives the calculation of member
elastic weights. Fig. A7.53 shows the imaginary
beam loaded with the elastic weights from Table Fig. A7.54
A7.l0. The detlections are equal numerically to
the bending moments on ..t his beam. Tabb A7.11
0b = .01855 x 25 = .465"
0B = .465 + .053 (~in Bar Bb) = .518" Ilellber 6 L , ... -.ll
, Joint
0c = .01855 x 50 - .00387 x 25 - .833"
....
••
0c = .833 + .031 (~ of ecl = .864 .O82~ 15.0 .00550
BC -.056 17.17 .00320
On = .01855 x 75 - .00387 x 50 - .00623 x
..
U .096 20.0 .00.480 C
I
25 = 1.03"
b ·.083 c •. 091 d -.091 e .083 !
CD
.n
18
_.069
••06
-.0738
.113
21.1-6
~i:i~
20.0
....."
.00322
.00348
.00565
D
J
1
a
•r
~~I~o ~I~~I
BE _.015 17.17 .00437
. '" ~ "0> '=, ~o Q ~e:::,' ~ ,~.,,~lI' g BG .108 15.00 .00120
A
'T
.061
r.-l BCD
.061 .066
6@ 25" • 1S0"
.066'
E
.061
F
.061
G
lfeiRDer 6L
EIa.tie Weight ot Web
.. .... ••
Table A7.10
JD .oraa I ~0.33 .00134 J ~l?~3 _. ~0126
ID .0496 10.33 .00480 1 10.93 -.00454
El 10.80 _.000115 I 11.26 .00891 1
Elzstic Weight Chord Ke~erB
r•
_.0254 9.27
.?~!,5 '.60
_.00274
_. ?~?28 •r 10.60 ,00240
1~.20 .00878
I
•r
lteaber H r
,
... =il Joint G7 _.0585 9.60 _.00610 G 9.60 .00610
AD • 061 3• .00203 b
ac • 061 3• .00203 b
CD
be
• 066
-.063
3•
30
I .0022
.00277
c
C
CO _.091 3. I • 00304
•
.lzstic Weigbt ot Web Me.bers
,
I p Q
I
lfellber
I HI, '1 p .. 6L
"l
Joint r2 Q .~
"
JOint
Fig. A7.55
i Ab
.b .121119.2
.053 dJ
-.00648 19.2 .00648 b Table A7.12
.
00
I be I .128 19.2 -.00667 b 19.2 .00667 C
eC
I-'O~~ i 25
_.00124 I
i
C 25 '
:gg~f~ • hllel PaDel Sher.r Z Shear
.IIloael1t
Shear " de-
12.5 z
Point
CD
ad
, .08g 119.2 _.00417 0 19.2
° ! ileet10D
o , o A
= =
=
~
~ M
N
=
0
~
=
M
N
=
~
= =
~
=
AD
,
• 02158
• .02 58
•• .508
••
.
M M
0 0 0 0 0 0 6
q q 0 q .056 2 .710
ql q CJ .01350
• .8. J
.
~
I I
.
I I I .07181 .900 D
~r
qt1;G
B C D E F IB -. .07429 .930 I
N
~
q Fig. A7.53 -
N
q YG
-.
-.
_.02 04
~
.638
,
• error
Example Problem 40
Find the vertical jOint deflections ror the Example Problem 41
unsymmetrically loaded truss of Fig. A7.54. The
Fig. A7.56 shows a simply supported truss
~ deformations ror all ~embers are given on the
Figure. Table A7.1l gives the calculation of with cantilever overhang on each ena.' This sim-
the elastic weights, their signs a nd points of plified truss is representative of a cantilever
wtng beam the ruseaage attachment points being
application. Fig. A7.55 shows the imaginary
A7.36 DEFLECTIONS OF STRUCTURES
at e and .'.
The ilL deformation In each truss
memoer- due to the given external loading Is giv-
lever over-hang portion of the truss r eIat t ve to
support points e and e l •
en on the figure. The complete truss elastic Since the cantilever :Jor1:iOn is not fixed 2. t
loading 'Nill be determined. vt th the elastic e since the r-es tr-a i rrt Is ceter.nined by the truss
loading known the truss deflections from various between e' and e, this fact must be taken into
reference lines are readily determined. account In loading the cantilever portion. The
reactions on the beam of Fig. A7.S8 represent
< the Slope at e due to the elastic Ioad Ing be twee
A' .M""·.""", ·_~~...n'.M...., .o.a E ·.GSA ;)·DS6C·D6Q8·1170A
n"i
~ " tf},(~
. '01n74.~~ .. ~".
:;.. 'if' '" \~"":"''' ",- <at. ~i;~,'
".
,;~" 0.c:.~
~,("p
\072 '(', . .072 "(,"::,U74\ .U70\"~74\ .:l..-
"",; e and e' . This elastic reaction in acting in the
reverse direction Is therefore applied as a ::"oad
r.---:- ~, d' .e' [ ,,4 c ~~
Q
, ,
"
N
0 01 N
0
a' A' b' 8' c' C- D' e' A a ~.
"
Elutic W"igj'lt Loadinl
0
Fig. A7.57
e'
0'
, ~.
,
e
I
..
.074 20 .0037 0
IlK _.056 20 .0028 0 Thus to tind the cer l ect i on of the truss er
of .012
-.0""8
20
20
.0036
.0024
E (Joint
I ,
a) we tind the bending moment at pOint a
of the imaginary beam or ?lg. A7.59.
....
rl at
, joint
'2
"
Jo1ut 70 + .00333 X 60 .. . 00239 X 50 + .00468 X 40 +
B
8.95
8.95
8.95
-,00":
,
• 00895
.00702
.00838
B
e
upward.
•
the truss
Due
we
to symmetry of 'truss and leading o•
know the slope of the elastic curve
'C _.060 8.95 _.00670
, o;~ 8.95 -.00781 ! C
c 8.95
8.95
,00670
• O~L81
C
,
at the center line ot the truss Is horizontal 0 r
zero. Thus to find the deflection of any potnt
dO -.0605
o.! f .02.
8.905
-. 00 72 8 1 d 8.95 .00728 0
I With reference to Joint f we can make use ot the
"" .076
.09' 12.0
8.95 ~. 008.50
.00792 I 0
D 8.95
12.0
.00850
-.00792
0
II deflection •,
12.0 .0020 12.0
• principle of the moment area method.
~.0020
~I
~ ~ ~ ~ ~ ~ ~
N ~
N N ~ ~ ~ ~
braically the elastic weights tor each jOint t:ro~
~I
0 0 0 0 0 0 0
Table A7.13 the beam elastic loading as shown In
Fig. A7.57 Is obtained. Freete 0
~I
d
0,
"
C
,
0
c
0
'I " 9 , o.
B b A
I, Fixed
ar
.
Let it first be required to determtne the R", .00312
vertical deflection at Joint r relative to the (From Span ee') Fig. A7.59
truSS support pOints at e and e ' .
To detern,ine the deflections of the truss l:l1a = 2.077" (student should make calcula-
between the supports e a d e' It Is only neces- t i ons ) • Previously the deflection of t with r-e
sary to consider the elastic weight loading be- spact to e was found to be -.0586" . Thus de-
tween these paints. Fig. A7.58 shows the portion flection of ~ with respect to paint e ; 2.077 +
of the 1Icaginary beam of Fig. A7.57 be'tNeen these .0586 ; 2.135" which checks value found above.
paints. The deflection at f relative to line ee' Let It be required to f'1nd the deflection of
.
Is equal to the bending ~oment at , for the por- joint c relative to a j rne connec t Ing joints b
tion of the imaginary beam be~Neen joints e and and d. N
e' and simply supported at these points. "N 0=N 0 0N =0 N0 0 0 0 ~ ~ ~ ~ ~ ~
~I~,
~ ~ ~ ~ ~ ~ ~ ~ ~
N
~I
~ ~
Hence deflection at t ; - .00312 x 30 + 0
0
~ ~
For this problem we need only t~ consider (7) For the truss in Fig. A7.66 calculate
the elastic loads betN8en pOints band d as the deflection of jOint C along the direction CEo
loads on a simple beam supported at band d E = 30.000,000 psi.
(See Fig. A7.61) ~he deflection at C with re-
spect to a line bd of the deflected truss
equals the bending moment at point c for the C --.-20,000*
loaded beam of Fig. A7.61. S'
hence
0c = .004743 x 20 - .00239 x 10 =,07 inches B 0 ~
8' 2 S'
A
~
M
=
= ~
M B
L
N ~ N
20' '7'
d ~l ~l ~1 b
A E
~
Fig. A7.67 Fig. A7.68
~=.004743 Rb=.004667 (8) Far the truss in Fig. A7 .67, tind the
vertical and horizontal displacement of jOints C
Fig. A7.61 and D. Take area of all members carrying tension
A1. 18 Prob 1ems as 2 sq. in. each and those carrying compression
as 5 sq. in. each. E = 30,000.000 pSi.
(9) For the truss in Fig. A7.68, determine
A~
~
O"
B the horizontal displacement of points C and B.
E = 28.000,000 psi.
3~'
J
45' \. G 5000 5000 500*
t- - 1000 1000
~
~ 0 A IB Ie 0 ~-r--r.,--r-.,---,
I I i-
to'
H +- t T 10"
C
Fig. A7.62 1>\)(): 3~ A 20" ::i ::'F 2 ,000< 30" T
A B J.. G414~E I e B
40" . 5060 5000 ~ D 4 @20 = 80"
Fig. A7.63 Fig. A7.64
Fig. A7.59 Fig. A7.70
(1) Find vertical and horizontal deflec- (10) For the truss in Fig. A7.69, rind the
tion of jOint B for the structure in Fig. A7.62. vertical deflection of jOint D. Depth at tr~ss =
Area ot AS = 0.2 sq. in. and Be = 0.3. E = 180". l..tIdth ot each panel is 180". The area of
10,000,000 psi. each tr~ss member is indIcated by the number on
(2) For the truss in Fig. A7.63, calculate each bar in the figure. E = 30.000,000 pSi. Al-
the vertical deflection of jOint C. Use AE so calculate the angular rotation at bar DE.
tor each member equal to 2 x 10 7 • (11) For the truss in Fig. A7.70, calculate
(3) For the truss of Fig. A7.64 determine the vertical a.~d horizontal displacement of
the horizontal deflection of jOint E. Area of joints A and B. Assume the cross-sectional area
each truss member = 1 sq. in., E = 10,000,000 tor members in tension as 1 sq. in. each and
psi. those in compression as 2 sq. In. E = 10,300,000
(4) Determine the vertical deflection of pat ,
joint E of the truss in Fig. A7.64 (12) For the truss in Fig. A7.70 calculate
(5) Determine the deflection at joint D the angular rotation at member AS under the given
normal to a line jOining joint CE or the truss truss loading.
in Fig. A7.64. 1000* 5000
I i o
B iii
b
12'
r
" 10
' T
A
10' is'1
c Fig. A7.71 Fig. A7.72
h
g I
(13) For the beam in Fig. A7.71 deter.nlne the
d 1.-
-
e
B C
deflection at points A and B using method of elas-
~a' ---.-a' ~9'
, , tic 'N8ights. Also determine the slopes ot the
I SO,OOQt- elastic curve at these points. Take E = 1,000,000
Fig. A7.65 pSi and I = 1296 In.~
(14) For the beam in Fig. A7.72 find the de-
(6) Calculate the vertical displacement of
flection at points A and E. Also the slope at the
joint C tor the truss in Fig. A7.65 due to the elastic curve at point C. Assume EI equals to
load at joint B. tlembers a, b, c and h have
areas of 20 sq. in. each. Members d , e , t , g 5.000.000 Ib in. sq.
and i have areas ot 2 sq. in. each. E =
30, 000, 000 psi.
A7.38 DEFLECTIONS OF STRUCTURES
500Jl
10"---.1
100,*
A
,--- t=25"
"
-j.<a'
;r.~1
t 2"1
'I
~A
, 5"
5"
50"
lOOOf
Fig. A7.81 Fig. A7.B2
(15) Fig. A7.73 illustrates the airloads on (23) In Fig. A7.81 find the vertical ~ove
a tlap beams ABCDE. The tlap beams is supported ment and the angular rotation of point A. Take
at B and D and a horn load or 500# is applied at EI = 12,000,000.
C. The beam is made tram a 1~-.049 aluminum (24) Determine the vertical deflection or
alloy round tube. 1= .01659 in~; E = 10,300,0 point A for the struc~~e in Fig. A7.82. EI =
psi. Compute the deflection at pOints C and E 14,000,000.
and the slope of the elastic curve at point B. 100#
F ...
B 3"
lOOt 100
"'3" 1"3"
100
3 ., F91t -;- 100
9" + 9 "
100*
j
100
~~1=t=-.~a=58~st=e~el=tU=b~e~~" Fig. A7.83
t 0 1 B A ri:1 20" Y
A
Fig. A7.74 Fig. A7.75 100ill'
(16) For the beam or Fig. A7.74 determine V
c
the dert.ectaons at paints C and D in 'terns or EI
which is constant. Also determine slopes ot the
Fig. A7.B4 L. o r a•
elastic curve at these same paints.
(17) For the cantilever beam ot Fig. A7.75
determine the detlections and slopes or the
elastic curve at points A and B. Take EI as
constant. ExPress results in terms or EI. Front View Side View
IF 11\\ ~
Flg.A7.78 Fig.A7.79
A
tJA RI
I
Fig.A7.80
'11 2
[ aij] {qj} compute the strain energy in the'
truss at Fig. A7.63 (Problem 2). The ~e~ber
flexibility coefficient fer a member under
uniform axial load is L/AE (see Fig. A7.35a).
An". U = 22.4 lb. In.
(27) Using the matrix equation 2U = Lqi I
(20) In Fig. A7.78 EI is constant throughout. (28) Using matrix equation (23) compu~e
Calculate the vertical deflection and the angu- the strain energy in the beam at Fig. A7.71.
lar rotation ot point A. Note: the choice of generalized forces should
(21) For the curved beam In Fig. A7.79 rind be made so as to permit computation at the
the vertical deflection and the angular rota- member fleXibility coef~iclents by ~he equations
tion at pOint A. Take EI as constant. of p . A7.l9. Ans. U =- 3533 lb. in.
(22) For the loaded curved beam at Fig. (29) Re-solve the problem of example
A7.80, determine the vertical detlection and the problem 25 tor a staowed cantilever beam whose
angular rotation at the point A. Take EI as I doubles at point "2 and doubles again at "3".
conatant , (HeaViest sect10n at built-in end.)
_ " . .; . /' : _:;. i' .:.", :,.>.:-:~.::~ -:-:-.. :.~,~ :'::::;:<; ~ ..": <:' ~ -,. -- -
-<";:. ;','" _~ -::!,~. :',." ~2::'':~'' .....<:\-~. ~-: ';0,\, . ._~r '3 :./"-:- "'-~. ~ ,:t ._~ <-.' l: /}",-.f,~~7'f:": N" - ~
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A7.39
T
the influence coefficient ~trix relating
vertical d.eflections due to loads P1.' P 'U P~
and P~ applied as shown. Member areas are
shown on ~he f~gure.
B t:.051"
A 1.0 m 1.0 G : 3.85 x 106 psi
;
1.0 10"
Answer.
(34) Find the influence coefficients re-
lating deflections at points land 2 of the
[" ~.~
44.67 44.67 simply supported be~ of Fig. A7 87. Use
44.67 106.0 99.34 38.0 matrix methods.
[3,nJ. E 1
44.67 99.34 99.90 38.0
33.00 38.00 ~8.00 33.0
TEXT BOOKS wITH MATRIX APPLICATIO~~ ~ehle, L. and Lansing, ~'J A Method for Reducing
the Analysis of Ca~p1ex RedUr..~tructUIes tc
"Elementary ra cr-ices", R. A. Grazer, v : J. ~ Routine Procedure, Jo~n. of Aerc. S:~.
Duncan and A. R. Collar, Cambridge University Vol. i9, 1952.
Press. Langefors, B., ~~alySis of ~las~lc Structur~5
"Introduction to ~he Study of Aircraft Vibration Matrix Transformation with Spec ~al ?,eg-a.r1 ~
and ~lutter", R. Scanlan and R. Rosenbaum, Sere1 Monocoque Structur~ .jour-n , of Aero. Sci.
Mac Millan, New York. 19, 1952.
Langefors, 3., ~trlx r.ethods ~or Recundant
Structures, Journ. of Aero. Sc~ Vol. 2C, :;53.
TECHNICAL PAPERS Falke~~einer, H. Systematic )~alysls o~ Redundant
21astic Struct-:.rres 'Jy :-:ea:;s of ~aL:rix Ca:c'.!lus,
Benscoter, S. U., "The Partitioning of ~atrices .Jour-n , or .Aer-o, 3ci:-: 20, ::''.753.---
!E Structural AnalySTS", .rourn. of Appl , hechs ,
Vol. 15, 1948. Argyris, J. and Kelsey, S., 2nergy Theorems in
Structural Ahalysis, Aircraft ~~gineeri~g,
cct • lS54, e t , seg ,
. .
All.
;..~~
... ~
....-.
• a';~'~ .
"-\'
~. ~12l
gineer employs them often enough so that prob-
lems in which they are used to obtain stress
distributions are often thought of as being
"cetermnace" . c-: c-: Nt-to:
It is frequently the case in aircraft _A D -:'Jlc----'!J!' A D
structural arzlysis that, in view of the re- 20004;' 2000# '"' .707X •
quirements for efficient design, one cannot ob- \.00"* 1200"* hoo,,* 200,,*
tain a determinate problem without sacrificing <a) (b) (e)
necessary accuracy. The Theory of Elasticity
assures the existence of a sufficient number of
auxiliary conditions to permit a solution in Fig. AS.!
such cases. SIngly redundant stress distribution, (a) consisting 01 a stress
This chapter employs extensions of the
methods of Chapter A-7 to effect the solution in static equilibrium with the awlied loads, (b), with one
of typical redundant problems. SpeCial methods
of handling particular structural configurations zero-resultant stress distribution, (c), superposed.
are shown in later chapters.
AS.O The Principle of SU'Pel"?osition.
Only the system (b) is actually required to
The general principle of superposition eqUilibrate the external loads (corresponding to
states that the resultant effect of a group of
X = 0). Note that the system (c) has zero ex-
loadings or causes acting Simultaneously is
equal to the algebraic sum of the effects acting ternal resultant.
separately. The principle is restricted to the [[j Of all the pOSSible stress (foroe) dis-
condition that the resultant effect of the tributions satiSfying static equilibrium the
several loadings or causes varies as a linear one correct solution is that one which results
function. Thus, the principle does not apply rn-kinematically pOSSible strains (displace-
Nhen the member material is stressed above the ments), i.e. retains continuity of the struc-
proportional limit or when the member stresses
are dependent upon member deflections or de- ture.
formations, as, for example, the beam-column, a ThUS, for example, there are an infinite
~ember carrying bending and axial loads at the
number of bending moment distributions satis-
same time. fying static equilibrium in Fig. AS.l (d) since
~ can assume any value. Of these, only one
AB.! The Statically Indeterminate Problem.
will result in the zero deflection of the right
Several characteristics (and interpreta- hand beam tip necessary to maintain structural
tions thereof) of the statically indeterminate continuity witn the support at that point.
prOblem may be pointed out. rhese character is-
AS.!
AB.2 STATICALLY INDETERMINATE STRUCTURES
M. -I===:r-<:'.
M
PL/.
AS. 2. 1 Determination of Redundant Reactions by
Least Work.
Example PrOblem A
Fig. A8.1d
By way of illustration, the proble~ posed
by Fig. A8.2 'NaS carried to completion. The
Singly redundant beam with root bending bending moment was given by (x, YJ ~ measured
tram the left ends of the tr~ee beam dlvls~Qns)
moment Me undetermined by statics.
M
&
....El = constant Differentiating under the lntegra~ 51'S!:, (see
p , A'7.8)
--t---
500-2Rx
L
au
Fig. AS.2 oR x •
A singly redundant beam with one reaction given an ar-
bitrary value (Rx)'
'1
o
~. a - - - - - - - - - - - - - -
aR (1) .. (500 +
x 2
Eq. (I) is true for all redundant reactions
occurlng at fixed su;po~ts. 3ecause it corre_
sponds to the T~the~atlcal condition for t~e
minimum of a function, sq. (l) is said to state
BE.""...._ _IIliIIiiii......... ..
2 a
_
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES AB.3
Rx =- 1500
~ lbs. =- 93.8 lbs., au
ai1= 0
J ~
= [Ll1L+ • + 2PL - l1L X] -Xd y
R 3L 3L
t~e negative Sign indicating that Rx was down. c
Example Problem E
Determine the redundant fixed end moments
for the beam of Fig. A8.2(a).
o =l1L J~ -3~) dx
Fig. AB.2a
~+2PL-'\JL x
A doubly redundant beam with two reactions given + 3L oX(1 - 3L)dx
two arbitrary values.
Solution:
+!lR
2Iy -dY+
'\ + PL - ~ J2~.
-dy
The redundant end moments were designated
as ML and MR for the left and right beam ends
J , 3L 3L
, 3L
~ + 2PL -
respectively and were taken positive as shawn.
The moment equations for the two beam portions
(x trom left end, y from right) were O'l1L
\-dx+ '\ J\.
-r-r dx
3L 3L 3L
~+2PL-,\ Je c
11='\+ x O<x<L
3L
O<Y<2L
Then
4 PL
9
+..l.
2EI 2
~ • - - PL
9
Differentiati~g ander the integral Sign (see AB. 2. 2 Redwldant Stresses by Least Work.
remarks on p. A7-8) The Theorem of Least Work may be applied to
the problem of determining redundant member
forces within a statically indeterminate structure.
ThUS, in-an-fi-tDn8S redundant structure if the
redundant member forces are assigned symbols X,
Y, ~, - - - etc., the values which these forces
~ust assume for continuity of the structure are
such tv~t the displacements associated with these
forces (the discontinuities) must be zero.
Hence, by an argument parallel to that used for
redundant reactions, one writes,
A8.4 STATICALLY INDETE.RMINATE STRUCTURES
au o
50
.~.
ax
au
ay = 0 __ - - - - - - (2)
+
':;=".l.BC j c
(lOOe
\:
+--
1
2EI~e
etc.
In words, "the rate of change of strain eneT~Y 60
~lthrespect to the redundant forces is zero". 1
+-- o (50,000):ld6'
2EICD
Eqs , (2), like eq, (1) I are statements of the
Theorem of Least ~ork. They provide n equations 1
for the n:tlmes redUndant structure. The simul- Obviously there was no need to ccnsider the
taneous solution of these equations yields the energy in CD as its loading did not depend upon
desired solution of the problem. X and hence could not enter the problem. Dif-
ferentiatln~ under the integral sign
Example Problem C
The can~11ever beam and cable system of au 58.3 X
Fig. A8.3(a) 1s singly redundant. Find the
member loadings by use of the Least ~ork ax ~.U3
Theorem. 50
so
v
(1000 - .30_ x\
'_ _...";,,0,,8.:..,,,,_'-)
1x a ex
58.3 EI BC 0
+ -0 )"
_0- o
(b) ( 50.3
Fig. AS.3
A singly redundant structure with one member force given an
arbitrary value (X).
58.3 X ... 11,032 X • 6620 X
Solution: EA trac S; AC
AB <
50 gave
MAC = 58.3 X • Y
X 613 lts.
In eD: #
Then
l""cD = 50.000
~e [1000 ~ 30
vv.,,)
("3)J
0.1. x = 685x
The strain ener~J was therefore
M..',.c
50
52:3 613 .y 520 y
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES AS.5
p p -
R'P
~
j 60'
0 (1 - cos g) sin g d g
0
R' X f90
+ ""-r sin II g d g
"" 6Qo
0
101 Fig. AS.3 .... (d) R'P )90
- 2Ei' sin g d g
60'
60 0
Solution: + ~ locOS g sin g d g
X R Sin g PRIZ
t
EI AE
60'
U = zir E sin Q- PR R (1 - cos Q)J' Rdg
Example Problem E
The portal frame 1. 50
of Fig. A8.3(e) is
three times redundant.
+ zir
90'
2.25 2.25
c
1
50"
+ X sin Q r RdG
are given on the figure.
Solution:
Fig. AS.3e
I".
(l000ll'·!"l. '{a' .. SO '{ ~ ~OO ":':(~C - s': 1S'
LZ5
Fig. AB.3f 50,000 Fig. AS.3g After evaluation of the ~nte~31s the equations
obtained were
-
s s'
v
l 1 ~
• lOOT
direct application by the calculus, as in Art.
A8-2.1 and A8-2.2, is often impractical. ~or t~e
Fig. A8.3h lT Fl •. AB.31
majority of problems the work is facilitated if
carried out by the techniques of the Method of
Dummy-Unit Loads.
The follOWing derivation is for a doubly
redundant tMISS struct~e. The extension te a
more general n-times redundant structure, in
The composite bending moments as funct10ns ot which ether loadings in addition to axial (flex-
5, Q and 5' were
ure, t crst c.i and Shear) are present, is indicated
later.
I1 Consi~er the doubly redundant truss of Fig.
AB=Il+Vs
A8.4(a). It may be made statically dete~inate
MBc 50,000 sin 9 + 11 + 50 V + 50V sin 9 by "cutting" two members such as the diagonals
indicated. Application of the external loads to
- SOT (1 - cos g) this determinate ("cut") structure gives a load
distribution, ~S", computed by satisfying static
MCD = 1000 s' + Il - Vs' + 50 V-100 T equilibrium. At this time discontinuities appear
Then since at the cuts "xn and "yn due to the strains
developed.
U = :2
l J Il
~an
' dS d au
aR=;j1i
au au
aT 0, IP, ,P, ..
one haS,
au
aM = ° f
) 0
so
III
2.50
+ vsl ds M x
S loads
Fig. AS.4a
y
y
ux loads
Fig. AS.4b
Uy loads
Fig. AB.4c
l
each to zero, and rearranging, gives the simul-
Z SuyL taneous equations
AE
~ - z "",C
llL uxuyL
X Z UX + Y Z
u and u are the unit-redundant stress dis- AE AE AE
x y
tributions as indicated in Figs. A8-4b, c. The u 'L
--- - - (4)
X Z UyUx + Y Z -y-= _ Z SuyL
L
subscript wow indicates these relative dlsplace- ICE AE AE
~ents occur in the determinate (Wcut W) struc-
ture with the Wor i gl na l Wstress distribution. Eqs. (4) are two simultaneous equations in the
It is now desired to close up the dis- two unknowns X and Y. Upon solution for X and Y
continuities by application ot redundant loads the true stress distribution may be computed as
X and Y to the x and y cuts, respectively, as
In Fig. A8-4(d). Load X (5)
causes a stress distri-
bution XUx and, likewise,
~ For a structure which is only singly re-
M
Y causes a distribution dundant, eqs. (4) and (5) are applied by setting
YUy. The relative dis- Y = a giVing
placement at cut x due to
redundant load X is given xZ ux aL = _ Z SuxL
by (6 is read wdisplace- Fig. AB.4d AE AE
xx
ment at x due to X W) . or, Simply,
SUxL
Z Ai:
X - - - - - - - - - - - - - - - (4a)
and at cut.y by (read 0yx as wdisplacement at u 'L
Z ...L-
y due to X W ) . AE
0
0
0
_.107
•• 707
_.707
_.707
1.0
0
21210
0
0
0
".
15
15
21.2
.60'
_559
3"
395
3••
following examples. are given at this time. The extension of
the method to more complex structures is immediate and no
further work on de!lections of redundant structures is given
in this chapter (excepting in the case of the matrix methods of
.d 42.4 1.' 14140 1.0 40000 28.3 '55 Arts. AS. 10 et. eeq.}.
• 2
X • true 1~c1 111 redundant _.tier be Find the horizontal ~ovement of :Joint ftd"
of Example Problem 1# 1 under the action ef tne
X •
." hi
A.
~
~ -aUlO
109.S
• _5~9.
load applied there.
t"r Solution: The equation used to find the
deflection is Sq. (18) of Chapter A-7. ',iritten
Example Pr'oo.Iem t 2 for application tv t:-uss deflections it is
(see Example Problem 13, p. A7.l1)
Fig. AS.S shows a Singly redundant
3-member frame. Ftnd the member loadings.
Member areas are shown on' the r rgure , a : z~
AE
--- - - - - - - - - - - - -(A)
~~.
Now for a deflecticn calculation the symbols
T~OV·
100" -' .2
'8' and "u" ;nust be carefully reinterpreted
from their ~eanings in the r-edundant stress
J.2 .4 catcutat i cn.' For a deflection calculat1Cml the
symbols of eq. (A) , above, :nean: "s-Loacs " are
0 the true loads of the redundant structure due
ho"", to application of the real external leading;
Fig. A8.8 "u-Loads " are tae loads due to a dummy-untt
(virtual) load applied at the external ;Joint
where the deflection is desired and in the di-
Solution: Member ex: was selected as the recti on of the desired deflection.
redundant and. was cut in figuring the S-loads,
as in Fig. AB.9. Fig. AS.IO shows the u-load Thus , the '8 loads" for use in ;;q. (A) are
calculation. The table completes the calcu- the true stresses (the sclution] of Example
iaercn, Problem It L
PrOOf of Theorem
Again the symbols "S" and "u ll are to be re- ==
inter?reted for a deflection calculation as To prove the theorem above we return to the
explained above in Example Problem II-A. The virtual work prinCiple and the argument from
"Sc-Lcada" are now the "true Loads" coraout ed in which the dummy-unit loads deflection equation,
Example Problem. #2, above. The lI u_l oadS" are Eq. (18) of Chapter A-7, was derived (refer to
loads due to placir~ a dummy-unit load acting p. A7.l0). It will be remembered that the de-
horizontally on the structrre at pOint O. flection was shown to be equal to the work done
Since this load acts on a redundant structure by the internal virtual loads (u-loads) moving
it would appear that another redundant stress through the distortions (~) due to the real
calculation is required. However, this Is not loads, r . e., 6 Z u e, =
The internal Virtual
necessary. loads are those loads due to a unit load acting
at the paint of desired deflection.
Theorem; For the u-10ads in a de~lection Now for the statically indeterminate
calculation any-s6t of stresses~loads) i~
structure these internal 7ir~ual loads (u-loads
static equllT'5ri'i:iinNith the dum.lily-unit load
may be used , even :~rom the Simplest of "cut;" are, in general, indeterminate since the dummy-
S"trueture5:'" - - --- - - - -- unit load is applied at an external point ot the
structure. However, we recall that,
This theorem says that to get the wu_loaasl i-any stress df s'tr-fbut f o.t in static
for this deflection calculation we may " cut " equilibrium with the lI a pp l i ed load" (for the
any one of the ~hree ~embers and get a satis- ~a::nent now we are thinking of the dummy-unit
factory set of u-loads by Simple statics I load as the "applied load") differs from the
Berore proving the theorem we complete the correct (true) distribution only by a stress
calculation in tabular forn as shown. The distribution having zero external resultant
lI u _I oa ds " were obtained by cu~ting ~ember OC (~ • .'18.1).
and applying a unit load hor~zontallY at O. ii - a zero-resultant stress distribution
moving through a set of displacements does zero
work 4
AS.lO STATICALLY INDETERMINATE STRUCTURES
, ,
~
'";" I -..-- ~A
Member t, A 5
S\I,,1.. ..jL \I""I..! 1'rue Load
" "y
A A S"X\I".Yu.,
I
..,
Be 30. 2~ 0 0 0 0 0 i 43.2 0 .50
Z <I x "sTATIC
CD 40.5 -'000 0 - 0 64,000 0 I 51.2 0
I ·255
........ ,W: , ..
BE 40.1 -300( -.8 -.6,137,000'137,000 36.5 3(;.6 36.6 -1015 I
AB. 5 Trusses With Double Redundancy ED 30.5 -750 0 - 0 27,000 0 21. 6 1 0 -190
....
Trusses with double redundancy are ,
-..
>0,
handled directly by Eqs. (4). By way of AE
'. 0375.000 0 I 0 0 "'0 !
Z
0 0
~62,
0 105.01
I
X Z ~ + Y Z uxUyL - - Z!
SUxL
A A r
So Ivtng ,
X;:: 1550*
Y--932H
A8.5 Trusses With Double Redundancy. cont'd. Solving the two equations for X and Y, one ob-
tains X = 521* and '[ = 416*. The true load in
Example Problem 3 any member = S + XU + Yu which gave the values
x y
Fig. A8.l4 shows a structure composed at in the last :olumn of the table.
tour co-planar members supporting a 2000* load.
With only two equations of statics available for AS.6 Trusses With Multiple Redundancy.
the concurrent force system the structure, rel- By induction; eqs , (4) may be extended for
ative to loads in the members, is redundant to application to trusses which are three or more
the second degree. times redundant. Thus for a triple redundancy,
Solution:
U 'L UxuyL uxUsL SUxL
X Z _x_ _ + Y z~+ bZ~ - Z
Fig. AS.lS shows the assumed statically AE \ AE
determinate structure; the two members CE and
DE were taken as the redundants and were cut at uxuyL U 'L
poin~s x and y as shown. The member stresses X Z + Y Z .:L... + b Z Uyu.,L - Z
SUyL
-(6 )
AE AE ~ AE
tor this structure and loading are recorded on
the members. Figs. AS.16 and AS.17 give the
U and U member stresses due to unit (1*) uxuoeL U UsL uoe aL
x y
tensile loads applied at the cut faces x and y.
X Z
AE +YZ ~ +gZ
AE - - Z SUsL
AE
Table A8.3 gives the complete calculations for
eqs , (4) and (5). The load in the redundant and atter solving for X, Y, b,
member CE was designated X and that in DE as Y.
True Stresses = S + Xu + Yu + bU ---- -(7)
r 4' ..,..' ,2' I"- x Y •
~[ ~/y
AS.7 Redundant Structures With Members Subjected to
Loadlngs in Addition to Axial Forces.
::;qS. (6) are extended readily to cover
E problems in which flexural, torsional. and shear
1200041 +200041 loadings occur. ThUS, for a three times redundant
Fig. A8.14 Fig. AS.I5 structure
Xa + Ya +;;a
~
N j/
~ ~,~:./
xx xy XS
""."..-::3 1*
-~V "T' - - - - - - - - -(8)
ux loads Uy loads
xa + Ya +:!a = - °so
Fig. AS. 16 Fig. AS. 17 sx .y
••
Ux aL uxuyL SUxL where
X Z -A- + Y Z - A - = - Z -A-
".~'''~-.c~:~~;;:?":!.,_
,..',..:;)~~~
A8.12 STATICALLY lNDETERML'lATE STRUCTURES
Solution:
5xO
~ SUXL
--+ J ~xOX +
JTtxox
-"-.- The shear flow in the 2heet panels ~as
AE £1 0" chosen as redundant. Secau2e 0: s~~~et~y :~e
problem was only singly red~da~t. ?1g. AS.IS
+ jj qqxdxdy
Gt
shows the Ux and qx loadings jue to the redand-
ant shear flow X = 1. The real loadir.g in the
dete~inate structure consisted of a ccnstant
load P in the central stringer alone. The
Dye
~ --y-
Su L + J~ OX
-y--+
T\ox equation solved was (ref. eqs. (3)).
J OJ
AE E1
ax dy
x (J Ux ~"ax
...... t..
+ If 0-x' ncct '\ =
Gt
Y
,)
q<l.r
+fJ Gt
_(J SU~E ax + jJ q ~ OXd Y)
Gt
etc. where
Ux ' rnx' t x ' ~ are the unit (virt~l) load- Ux - L - x in s1de stringers
ings due to a unit load at cut x; VIRTUAL
LOADS _ 2(x-L) in cent ra 1 s 'tr-:
r ...nger-s
uy' my, t y ' ~ are for a unit load at cut
{ ~ - 1.0
s, etc. When evaluated, (note that the double inte-
grals Simply reduce to a constant times/the
The redundant force(s) need not be an axial panel area)
force but may be a moment, torque etc.
Atter solution for the redundants, True X C: ~; + 2itL) = - (-P1E')
Axial Forces =
S·+ X u
x
+ y U
Y
+;; u. x• .E.
2L
( 1 1 ME )
+ GtLa
True Bending Moments - - - -- - - - - (9)
Therefore the true stresses were
1 +
...
'!-1o-:-b-t:.lLb-~..l;l
1 l1l J tj "Clamped" I- L .j.. L -'h
Fig. A8.19
x
p
Fig.A8.I8
qx loadings
Fig. A8.19
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES AS. 13
A
~ :: 100
L
g.
L
giving
'+
~ .3716 X ... •1526 Y •• 09011 PL
'<
X=1,·tcu~-1!L B 1 Y=1" *
{ .1526 X + .8121 y. .3616 PL
~L
Ux> m x loadings
P
Uy, my loadings
X .
{ Y· .456
.0645 PL
PL
Note that in Fig. A8.22 the unit redundant True Stresses :: S + X U ... Y U
loading 'NaS applied as a self-equilibrating set x Y
af unit couples. The real and virtual load- rrue Moments = !1 + Xrn. + Ym y
ings were as follows: (member portion 3D J x
haVing no virtual loadings J was omitted. It A8.8 Initial Stresses.
could not enter the calculation.) In a red'.lndant structure In~tlal stresses
I are developed if J upon assemblYJ certain mem-
Member S M Ux "'x uy !lly bers must be forced into place because of lack
of fit. In some situations intentional misfits
AB 2P$ 0 Y2/L 0 -IlL/£" 6/Lv'1l" are employed to obtain more favorable stress
y-L distributlons under load (~prestressing").
BO -2P ?y - IlL L 0 - YI L "If, in Fig. A8.4(a), the redundant member
with the "x cut~ was initially oversize (too
long) an amount 0Xi (an overSize J corresponding
The equations which were solved were (Ref. to a distortion in the positive X direction, is
eq , (8)).
a pOSitive 0Xi)J the modified condition tor
continuity at the x cut would be (compare with
the equations just preceding eqs. (4)).
y ~ Ux,\L ~ SUxL _
After evaluation of the integrals and
~lylng t~~ough by L~ these become
~ulti
X f '_xL
_ _ ...
-- = - °Xi 1
rx:.U .
AE AE AE
- -(10)
L r'L fUA~L
= _ PL ((2 ... 4v:1"1L _ ~\
X y :E =- - °Y1
t
,-, Ai: 621 ) The ~8 loads" of eq. (10) are present because
of applied ext3rnal loads. These mayor may not
be zero depending upon the problem.
AS. 14 STATICALLY INDETERMINATE STRUCTURES
~
making cuts ~x" and "y" to render the structure
determinate, the application ot the temperature B;:rr A ET(~;~
distribution IS visualized. Relative displace- 30" x
ments occur at the cuts,dencted by exT and 0yTo
--l ,
These displacements may be computed by the
Dummy-Unit Load method as shown in Art. A7.8 of
Chap. A? After this calculation Is accomp-
40" C (permits
sliding
Y r-- --l r
Fig. AS. 23 Fig. A8.24
lished, the problem proceeds as for initial vertically)
strains, Art. AS.S. Thus, the continuity con-
dition at the cut gives (compare with the equa- Solution:
tions 1mmediately preceding eqs , (4) and as-
sume tor simpliCity that the external loads are The structure was made determinate by cuts
absent, making axe = aye = 0) x and y as in Fig. A8.24. The unit loadings are
shown in Fig. A8.25.
0xT + On + 0xy = 0 }
0yT + 0
yx
+ 0
-JU
=9 a
r 'L 5 . 008 X - .90 Y T X lOll
x ~ uXuyL
X + y --- x
a Tdx {_ .90 X + 1.50 Y 0
AZ AE
rUL
- - (13) So Lvtng ,
X .!:....!....
AE
+ Y ~U'L
.!-
Ai;;
=
-f uy
a Tdx x = 2.01 a T x 10 II
Y = 1.21 a T x 10 1I
Equations (13) may, of course, be extended
tor application to structures other than True stresses are given in Table AB.4.
~usses. The expressions appropriate to other
loadings have been developed previously (eqs. Exa~ple Problem 10
(8) et seq. in this chapter and other equations upper surface of the built-in beam of
T~e
in Art. A7.3). Fig. A8.26 1s heated to a uniform ta~perature T.
-:
AS.16 STATICALLY INDETERMINATE STRUCTURES
Through the depth of the beam the temperature Then the equations corresponding to eqs. (13)
varies linearly to normal (T = 0) at the lower were written (see also eqs. (8)).
surface. Determine the end moments developed,
neglecting axial constraint and influence of
axial forces.
X(t:~dS J\:dS) y(J u<t Jmx;{s)
+ + +
r- L T I
Solution:
• + O·y + ( -=- =0
"Therefore O·X 1 +-'-
R ' ) ;;
)~ E1
L
X "IT =- TaL/h
Note that from the last of these equations
g = 0,
as it must because of the symmetry of ~~e
X = - TaEI ring. SolVing the first two equations
-h-
8maOLB - (5)
(21)
The reader may satisry himself that the Eq. (21) is a set of simultaneous equations
~crass praduct~ term in the ~ldd1e of the above for the redundant internal forces ~,qs' It
result is correct by Qbserving that, because of :nay be compared With sq. (6) of Art. ,1.8.6, to
the symmetry at [u1J
], which it corresponds. Eq. (21) ~y be solved
directly from the form there displayed or its
solution may be obtained by computing ~rs:JJ
the inverse of the matriX of coefficients, giVing
(~~ - ~iJ ~s[] ~sJ ) {Pm} The moments q~ and q. were taken as the redund-
ants. With these set equal to zero, the internal
force distributions due to application of unit
= (~iJ+ ~iJ l:rmJ ){Pm
} values of P l and P a were determined, giving
r I" 'J
tribution. The beam is loaded by couples over
the supports as shown.
[SJ . [!r!J ~ti] ~J !J
..!
:.1 0 0 -liL 1 1 0
~
f--- L L
1
----1
('I 0
0
0
0
L-/3
L/,
L:, ~L -~L
~~
q2( II
q, } (hi
L
: 6E:.1
q,( Ii
q,
Fig. AB.27
rrnJ • [!r!J rti] ~J;]
-r. J["
I' .:J
So Lut Lbn: L/, 0
£! 0 0i _lit.
." 1 Liz 1 0
The choice of internal ~eneralized forces
is shown in Fig. A2.Z7 (b) .. T~e appropriate . 0 0 1.-/3 Liz 0
~ember fleXibility coefficients were arranged in
0 0 L/2 1 0
~trix :or~ as (ref. Art. A7.l0 :or coefficient
~ ~
expressions) . L 1 2
..
: mOl ... -
~
AB.20 STATICALLY INDETERML.,*ATE STRUCTURES
The inverse of Cars] was fOillld (ref. though the ma'tr-tx of member- aexi'Ji2.1 ty cce-
appendix) fflclents was expanded to a 6 x 6, the
coefficients for q, and q, were zero. ThUS,
~rs ~
II _J. _ 5EI
- 7[ C -J
2
-1 4 VI/3
L/2
L/2
1
a
0
0
a
0
c
0
0
_ L
~lLJ
Next, the unit redundant load distribution
was found (eq. 22). - ;;:1 a a La/3 Liz 0 0
0 a L/2 , 0 0
1
"
- '7 0 0 0 r! 0 0
0 0 0 0 0 0
L· J
1 0 _1/ 0 286 -.42
L 0 0
0 0 1 0 .143 -.286 0 0
= 1 •
L 0 1 I/L - I/L
and, with PJ. and P a zero, successive applica-
0 0 0 0 tions of unit redundant forces q$ ~id q~ gave
-1 0
0'\ 1
1
1.286
2
.428 -I
L
-I
0
1
=L 2 - .286L -.428L ~lJ = 2L L
3 - .429 .858 1 0
4- .143L -.286L La I
(The tabular form at presentation of the The~, multiplying out per eqs. (17) and (18)
matrix Gim, above, is used here only to indicate
~ J
clearly the functioning of the subscript nota- L'
tional scheme. In general, it should be unnec- ["r;;] 2E1
essary to callout the subscripts in this
fashion excepting for the larger matrices, tor
the handling of WhiCh, the tabul~r form may
prove helpful.)
Example Frob lem 13a
~r~
L'
6Ei r 6
~5
51
sl
The redundant beam problem of Fig. A8.27 is The inverse was found:
to be re-solved using the redundant reactions as
the unknowns.
6EI -5]
Solution: 7L' l~
The support reactions under the loads PJ.
and P a were given ·the symbols qe and qe, respec- Finally,
tlvely, Positive up. These forces did not enter
into the strain energy expression so that, a i- rG l r:. -, ~ -, ~ ;:'l r;; -,
~
L:1m ~1~ - ~:CJ ~sr~ ~s~
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A8.21
1.286 •.l25
- .':::;CJ.., -.423L
1 - .'+<:.c .857
=L
.143L -.256L
-1.256 -.428 Loading for column {gi2} of the matrix [gim]
1.714 -.428 Fil{. AB.29
'R2
8 panels at 20" :: 160"
~
Fig. A8.28
3 5
Solution: 20 1 0 0
20 2 .0 0
The internal generalized forces (qi,G ) 20 3 1.0 0
j 20 4 .5 .5
employed the axial loads in the various
~ere 20 0 0 1.0
~eabers. These ~ere numbered from one to 20 6 0 .5
thirty-one as ShC~~l on the fl~1re. The member 20 7 0 0
fle:Krbility 80efficients in this case 'Nere of 40 a 0 0
the form c , = L/AE (Rei'. Fig. A7.35a). The 40 9 -.25 0
• 40 10 -.75 0
coeffic1ents arf written as a column matrix be- 40 11 o~
-. , 0 -.25
low. (They 'NE::ce employed as the diagonal e le- 40 12 -.25 -.75
ments of a square ~~rix in the matrix multi- 40 13 0 -.75
plications, but ars written here as a column to
~1~
1 40 14 0 -.25
conSer.e space.) a1 =~
40 10 0 0
Membe~ loads qa and qo were selected as 22.4- 16 0 0
redundants. ~ith qa and Qe set ':!qual to zero 22.4 1" _.
0 0
(''But''), un t t loads were applied succ ess ive Iv 22.4- 18 I .56 0
at external loading paints one through four, 22.4 19 -.56 0
t~e four stress distributions-thus fo~nd being 22.4 20 • 561 0
arranged in four columns giving the matrix 22.4. 21 -.56 0
~i~ (below). By way of illustration. the 22.4- 22 -.56 .56
22.4 23 .56 -.56
loading fi~ure used to obtain the second column 22.4 24- -.56 .56
22.4 25 .56 -.,:)6
of ~i;] 1s shown in Fig. A8.29. 22.4-
m"
25
27
v
0
- -.56
.00
•
AS. 22 STATICALLY INDETERMll'{ATE STRUCTURES
I~
05 9 .111 .0065 .0035']
1 2 3 4
[ .0035 .0065 .111 .059J
1 1.0
2 .50 -.50
3 The calculation was completed as per eq.
4
(23) to give [qim] , the values of the member
0
-.50 .50 forces for unit applied external loads.
7 1.0
8 -- .50
9
10
11
12
.70
- .25
.25
.25 Ix 1 1.0
1 2
0
3
0 I 0
4
,
I
2 .529 -.445 .003 .001
13 .25 - .25
3 .059 .111
14 .25 - .75
4 .031 .059
.006
.059
.003
15
16 l.12
- .50
5 .003 .006 .111
.031
.059
17 '6 .001 .003 -.445 •.529
-1.12
18
19
- .56
.06
-.56
.56
7
8
0
- .50
0
0
0
0
1.0
0
20 - .56 .56 9 - .764 .222 -.001 0
-- .017
2 .56 -.56 I
10 - .294- .167 -.003 .002
22 11 - .045 -.085 -.032
23 12 - .017 -.032 -.oas - .04.5
24 I 13 - .002 -.003 .167 - "'G'
.':::',,7
25 I
14 0 -.001 .222 - . , ::" ...."
.....
1rom Z F on Joint C.
~
..... B
"' " /
C
::I....... '" -1.0236 P,
~ K
,
__
-,
-, ~~,
I"~
"I ......" P
3
From Z F on Joint B.
II- J--:::--
--
- -'.---
G
H
F --- .081759 Cl.1. - •18334 Cl.. - q. -.07617 Cl.1.. '" .5227 q•
Fig. A8.31
q•• - .91593 Cl.u , 0
{
-.1410 q1.'" '" _.8523 Cl. ..
~ER N"lJ11BER L2NGTH AREA L/ A And so forth, for the other Joints.
Note that in each case the equations were
cli 1 35.47 .565 52.8 arranged with the applied loads (P ) and the
RC 2 25.43 .49~ 51.0 n
BD 3 9.20 .165 .-
00.""~
redundant q's (q33, q33, q.~) grouped on the
CD 4 25.43 .499 51.0 right hand side at the equal Sign. This ar-
DE 5 35.47 .565 62.8 rangement was observed for all 21 equations,
?G 6 35.47 .565 62.3 atter which the equations were placed in matrix
GH 7 25.34 .555 44.85 form as
HI 5 5.00 .165 30.3
IJ 9 25.34 .565 44.85
GJ 10 9.20 .165 50.8
J"
AG
'1
12
35.47
40.4
.565
.630
62.8
64.1
BG 13 15.0 .165 90.9
i,j = 1,2, •.••••. 24
3H 14- 27.57 .500 55.14-
He 5 11. '70 .395 29.62 n = 1,2,3
Ie i5 11.70 .395 29;52
s = 22,23,24
Dr 17 27.57 .500 55. "4-
JJ ,Q 15.0 .160 90.9
SJ ..,
;
40.4- .630 64.1.
(Note that there were 24 equations ~ere,
the adcitiona1 truee equations being the ident-
20 37.36 .565
.' J
FJ
"~.
-."
22
37.36
26.23
.565
.500
66.1
66.1
52."~6
::.ties
sa 23 17.60 'Q.
.... ..,0 1.06.6
".
.U 2.:1 ,-.60 ; .-
• ....0 0 lOE.S
coefficients of the q_. The inverse was rocnc (see appenc Lx)
o
The ::latrix equation NaB solved ~or t~e q.
by finding the " _. [5555
- .2580 l
'2~7~
Thus,
lnver~e of [C
1 j]
(see appendix).
trs-J 10 -.2550
.2775
3.C~12
2.3015
-~ '2:" -
<::'..... ,-,·1.:)
3.023
~
where .18 2259
IG "I =_Ia
LSllJ - LTS -'J LTllJ
ICi "I = 10 -~
-be."=,
","
-238 '"'n .".
-.<:v1.v
- 31A 121.~
1;\ 1 a
2~727 5.;'7~
3
I Z2
,
Zl
I ~
o
1
2 -
.8412
.00017
1~...,.=::
2.203
.~IV 1.296
.~38
1 .5415 -2.:!C59 0
-
, · " ..
.. · ..
.:).5<> 0 3 .01,523 .31.53 .1190 -
zr-
.' 0
·.
.,. ~:: . o
,- .0
4 .00017 -2.206 .E;39 -
_2.72:7 ,:;,,60
.. .. - -- o
:~
0
4.0.59 _7.~ e •..;..,5 5 .6529 -3.008 i -1. 557
, s s ,
-c . ."'~
.."' 6 -1.4-04 1.5247 i .332
.,
.. · .. ,
,
, 7 - .8443 .7352 I .761 -
. ·
,.
• .06<:
· .. as ·-3.n4
.1 ..l" - ...17
.0 0 8 .1094 - .095<::'. I .096::.
· - .0= -
2.," .. ,
9 .84.50 -1.260 -1. -t-50
· ..
, . -i · •.:.009
" - .aiee
.4415
.5 ,5 ,
.54.
."'"
0 ..
· .01<+5
00
.eess
-r.v-a
, ,
-1. ..35
c
0
·
r..4J.S
.3523
0
,
10
~1 -
.0973
.0494
.0205
-1.607 I -l.S'~2
.3338 -
,.·
_2.340
.zao
-.... 5
-
...
.&C
.22
0
0 12 .6638 - .3896
.3143 .733
. . .W
·1.W9.
·
-
,-
'."
.:...,27 · .o se 13 - .2350 - .2641
~
.27E"
,
z . , -L'" .0
,·
.2445
."S06
1..;,01
-'-.302
, _,.~l
1.102
14
15 -
.9185
.5116
- .7998
. , -
445-
.6223
.461
oe ·
0
o
•• 0
0
.~O'"
1.0
.
0 16
17
- .5120
.9193
- .4457 I
.B002 I -
.461
.829
18 - .1061 - .3638 .3205
The member flexibility coefficients L/ A 19
20
.1434
.2023
.2411
1.6471
- .6956
.8512
were arranged as the diagonal elements of the
matrix [u 1j ] Then, multiplying out according
21
22
- .2660
.0002
- .3'796
.543 i
.556e
2.269
,-
23 - .1364- .238 - .2010
to eqa , (17) and (18), 24 - .03 4 .176 .122.3
[177,9
-1137.0
-6430'~ ~ar.mle Problem 15
The doubly symmetric four flange idealized
~~
195.0 - 151. 7 -2263.0
box beam cf Fig. A8.32 is to oe ~n~lyzed for
-154.4 161.6 2273.3 stresses due to load applicat:on ~t the six
paints i~QicateQ. Fl~~ge areas· taper linearly
from root to ~1p while sheet thicknesses are
-114~
[ 2970 1150 constant in ~ac~ panel. The bea~ :s ~o~ted
rl~1dly at t~e ~oot, provid1ns ~u:l restra~nt
~r;] 1150 1221 -1035
-1148 -1035 1224-
• These are the "effective areas", being the flange area plus
• see references in appendix. adjacent effective cover sheet area plus one- sixth of the
web area. (The factor of one-eixm provides the same
moment oC inertia as the distributed web area).
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A8.25
against war?ing of the root CT2SS section ~ue The ~e~ber :lexlbillty coefficients were
to any torsion loadings. collected :n ~trix :crm as shown below. Note
t~at entries for u," U e e • Q v v and u~o,~o were
collected irom ~NO stringers each (as well as
being doubled as discussed above). Coefficients
for these tapered stringers were computed from
the formulas of Ar~. A7.l0.
1)< , ! a IaI , ,
, s I • . i a I 10 U !Z I IJ I
" ,! ie ,
, iaa
,
,
' I
, ,
,
,
" ace
.. ..
··
,-.o.~
,
. · ..
,
c.
,
.
.
i
.
15.2.il ::7 .5
" , .~ , .,
8 .00 ,200
'\.'1:" ,Qle 9 5.341 2.571
10 5.341 2.571
~~ \~
.08_87
12
13 ,0444S .08887 ,1333
14 6.231 4.006 2.003
IS 5, 1 4.006 2.003
,'.~~,
ants q s, q.., and q u. The calculations are
illustrated by the exploded view of the end bay
in Fig. A8.34 showing the calculation in that
part of the structure for q~ = 1. Note that
q:z = 1 was applied as a self-equilibrating pair
of shear flows acting one on each side of the
q, "cut;". The r-ibs were considered rigid in their
own planes.
Fig. A8.33
?ut q~ =1
I From equilibrium of end r t o : (1: M, Z F)
A8.26 ,
STATICALLY INDETERMINATE STRUCTURES
I~ .;'" ~q
.~E;--'-<-S\l25
Per eo.. (18) :
•
q
.1769 .0520j
[774
\q _ lOll
~r~
,40",-25.04 .75x~=.5625 25.04 .1789 .2578 .07322
..
IS"
- E
.0520 .07322 .1254-
~\~
2.5 j 15
Per eq. (17) :
-IX~~
_.008216 ,008216 -.005150 .005150 _.002278 .002278
Fig. AS. 34
e-: = E
10 IS
-2.562
- .1137
6.137
-2.321
-2.521
9.499
Then
q, = -1.0
q, = 1.0 ~sJ =- ~rs -j ~r;;]
From eq'.l111brlum of 200 rib:
[0699 -.0699 .00598 -.00598 .00016 -.00016
q. = -t .5625 - .25) = -.3125
= .03593 -.03593 .03409 -.03409 .00403 -.0040
q. = .3125
.01568 -.01568 .01872 -.01872 .01578 -.0157
q. = 0 , by hypothesiS
Finally, the true stresses were (per eq.
So on, into the next bay. In the idealization
used here l the ribs have zero stitfness normal
to their own plane so that the axial flange-
loads are transmitted directly to the flange
~i~ = ~i~ - ~1J ~~
ends ot the adjacent bay (see q~, q" q .. q"
of Fig. A8.33). I~ 1 2 3 4 5 5
-1~J
P a = -1. Thus, one :inds that under this con-
44 -15.0 o
dition there 1s a root flange load of =; -15.0 38.0 o
·00 38.0
= 3.47 = 0.71 [
ql. - 2.76 Ibs. o 0 -15.0 144J
for a torque of 15 inch Ibs.
472 .888 .0525 .028
AS.11 Redundant Problem Deflection Calculations by •388 1. 665 .0975 . 0525
Matrix Methods.
Datlections (and in particular, the matrix
of influence coefficients) are readily computed
from the results of Art. A8.10.
Assume that the redundant forces { have qs}
[
~0525
.028
.0975 1.665
.0525 .888
.888
.472
J
Finally, the sum of the above eNO matrices gave
been det.er-atned from eq . (22). The total de-
~'#
~lection of external loading points is then -15.9 - .052
eaSily computed as the sum of, 01~, the de-
flection due to external loads acting on the
"cut;" structure, [amnJ {P n } (ref. eq , 16) and,
TNO, the deflections due to the redundant
~,,;;] ,. ~
- 15.9
-
-
.052
.028
-
-
36.3
.098
.052
-
-15.9
.098
36.3 -"J
-
-15.9
144
.052
Solution:
= ( ~mnJ - ~:nsJ ~r(] ~r;] ) {pn} This product was formed readily, inasmuch as
[G ] was evai tame tram example problem 15.
im
The matrix expression set off in paren-
theses above, giving as it does the deflections The result was
for unit values of the applied loadS, is the
matrix of influence coefficients. Let
----
4363
3500
3500
4363
1522
l243
1243
1522
29.7
-29.7 29.7
-29.71
1522 1243 897.8 594.2 33.7 -33.7
1243 1522 594.2 897.8 -33.7 33.7
- • 33.7 - 33.7 .8 -36.8
9.7 - 33.7 33.7 -36.8 36.8 1
so that
AS.12 Precision and Accuracy in Redundant Stress
- - - - - - - - - - - - (26) Calculations.
Matters of prectsion are dependent upon the
number of Significant figures obtained and re-
~ample Problem 16 tained in dealing with the geometry of the
Deter.nlne the matrix of influence coeffic- structure and in the care with which arithmetic
ients for the redundant truss of example oDerations are perfo~ed. In the discussion to
problem 14. follow it 1s assumed that all due caution 1s
exercised with regard to the preCision of the
Solution: work.
~~t~ers of accuracy'r4ve to 10 with the num-
From the previous work the following ~trix ber of Significant figures :lnally obtained in
~roducts wsr9 fo~ed the answer as influenced by the ~er of formu
A8 28 STATICALLY INDETERMINATE STRUCTURES
lation of the ?roblem. ~~e accuracy ~f the ~e inverse ~ar~ot be ro~c (r sr , appendix).
sult may be affected by a n~~ber 0: factors, two condition is terrib12. :::xampl-a:
of the most important of wh i ch are discussed
here.
Two factors infl~encin~ accuracy often are
considered together under the heading ~choice of
redundants n • They are:
• - ACCURACY OF INVERSION OF ri
Lr~
:l ; (bl
~
(dl
THE CONDITION OF THE nATRIX.
Fig. AB.35
The characteristic of the mat-~x
.. ~ [0 rs ]
which determines the accuracy with which its in- Fig. AB.35(b) shows the c~oice o~ general-
verse can be computed is its condition. The ized torces. Only two (Ql and qa) a~Q required
condition of the matrix is an indication of the to describe the strain energy, but the central
magnitude of elements otf the main diagonal support react1or~ were also given s~bols as it
(upper left to lower r1&~t) relative to those was desired to consider them in the dis~ussion.
on. The smaller are the relative sizes of Then
elements off the ~in diagonal, the better is
the condition of the matrix. A weJl conditioned c-
2/3 1/6 0 0
-
matrix is more accurately inverted than a poorly
conditioned one. 'rNo extreme cases are now
given tor illustration: 1/6 2/3 0 0
a) the diagonal matrix. Its ott-diagcnal C:I1J =..1.EI 0 0 0 0
elements are zero, so that it is ideally con- 0
ditioned. Thus. the inverse of
,-0 0 2..
FIRST, suppose the beam was Iade deter.ninate by
selecting the support reactions q~ and q~ as
redun1ants. The ~cut structure~ in this case
may be visualized as the Jeam of Fig. A8.35(c)
whose central supports have been removed.
Application of unit redundant ~orces q3 = I
Is eaSily and accurately obtained as and q .. = 1 gave
~ 3 4
2 1
1 - 3 L - :3 L
tIultiply1rlg out,
:0:
J -W-'
dx Im:EIdx .wh'... ch rorm
' thie on-a..
.,. agcna.J..
,
r,s=l,2
m,la
The condition of this redundant ~~trix is
obviously better than that obtained with the m
z[ ~
first choice of redundants. There is less
cross-coupling between the redundant forces. (b)
Fig. AB.36
Thus the analyst, by choice of redundants, determines
the condition of the matrix. The choice may be critical in the
case of a highly redundant structure, for it may prove im-
possible to invert a large, ill-conditioned matrix with the St~dy of t~e ~it-load j1agr~~ fo~ ~he
l1.mited number of significant figures available from the initiaJ. redundant choice ql' q~ (Fig. A5.35b) reveals
data. The following statements and rules-of-thumb may be that the cross-coupling sh~uld 88 small her~
useful in the treatment of highly redundant problems.
since an 1rrteetr-a lJ ':J".......
v • .9 r' orm J:nlffi'EIl ex can
(1) It is always possible to rind a set of ~ave a contribution from the center s~an O~ly.
r-
redundants for which tIle cross-co'J.pl1m; is~t a Hence, the J ::I1.:n'l, ct.:.:: is obvt cus Ly c ons tder-ab Ly
~inimum-zero, ~n tact. Theoretically t~en, by 0'
~roper choice of redundants, tte ~a~r!x [Q ]
rs sma11 er ·h
cnan 31ex or J r ::I~
~I ex Th
.. us, a
~y ~e reduced to a diagonal ~trix (idea~ly Visual ~nspect1on of Fig. A.36 reveals that q1.,
conditioned) . qa is a better choice of redundants ~han is
(la) The choice of red~~cants which gives q, c...
zero cr oss-c oupt inz ("orthog:Jnal func-O:ions") 1s
no'\: reaai i .' / found in :::ene p :;' . I!1 some special
structures, such as r tngs anc rr-ames, orthogonal .. see eq. (8) Art, A8.7.
AB.30 S!ATICALLY INDETERMINATE STRUCTURES
0
0
1
- 2 L
l
property may be formed trom the m~, m~diagramB
by proper combination. Thus, if one halt the
1
1
C: eJ
1
m" diagram is subtracted rr-cm the --m:
d'Iagrarnto - 2
fOrm one newstress distrIbUtion and one halt 1
the ~ diagram ~ subtracted from the m~ dia- - 2 1
gram to form the other new stress distribution,
the results are as shown in Fig. A.37b. There
ta. oovtous jy less "overlap" of the diagrams for Now form the ~trix of redundant coefficients
these new combir.ations. for the ~ew unknowns (subscripts p, 0; [giP] =
[gj,,] )
L'
24"::' .
Fig. AB.37
The condition of this ~trix is ~eatly
New unit-redundant-force stress distributions (b) obtained by improved over that obtained :or q~, q~ alone
(previously computed), ViZ••
combining previous d1st.r1butions (a).
~ -:/J
, the problem ~y be completed in the "0, 0
2 L
:3 !3 L • system". The appropriate equations ~re abtai~
ad from eqs . (14-), (21), (23) and (25) Simply
, ! L 2
L
1/2 by replac1ng all "r, 5," by "~, 0". Thus
3 - :3
1 0
(29 )
0 1
= { 9.49}
-101 •
o
and compares the result element-by-dement with the 1
matrix previously computed,
-.8330
3774 .1789
'0520~
I~ 1 2 3
10' .1789 .2678 .07322
1 - 1.00 E: [ .07322
.0520 .1254
2 1.00
3 1.00
4 -25.04- It remained to select a determinate stress
5 25.04 distribution Which would reduce the ~gnitude of
6 1.023 - 1.00 the redundants. For this purpose, the engin-
7 - .7103 1.00 eering theory of bending was employed to comoute
stress distributions satisfying equilibrium for
9 - 1. 023 1.00
-29.38 each application of a unit external load. The
"
10 29.38 result was (refer to Example Problem 30, Art.
A7.11).
11 1.0285 - 1.00
12 . - . 833D 1.00
13
14 - .0lD
- 1.0265
- .026
1.00
-31.71 ~ 1 2 3 4 5 5
Fig. AB.39a
95 -3495 -1841 1841 0 0 ~
[gip]
G
New unit redundant stress distributions,
~p~ =~ 1167 -1167 1554 -1554 -1450 1450
1082 -1082 1445 -1445 1802 -1802
10'
E
·t
obtained by :n.ultiplying out per eq , (Soa ) ,
2584
-.011'32
.0000175
-.01132
.2566
-.03124
.000017J
-.03124
.1254
the "p,o system" using the matrices [giP} Capo}
[giro] and [Opn] in eqs • (31) and (32).
,-.Ae·-f
-<rtf
.:{-.J::-i!:
'~-:;:::;:-::;,?
" ~':.)"4
A8.34 STATICALLY INDETERMINATE STRUCTURES
then the virtual work done by a single general- The general expreSSion for the virtual 'Nork
done by Virtual axial loads u in a bar under
ized torce 1s qi ~iT • The quantity ~iT will be varying temperature T is
called the member thermal distortion. The total
virtual work throughout the structure is ob-
tained by summing, giving the deflection at the
w = u aT dx J
r~ cut as the matrix product Where a is the ~terial thermal coefficient of
expansion.
(34)
Where ~i Is the value at the qi due to a unit *n will be convenient later to designate by qST the solution
to eq, (36) when the mechanical loads P n are zero, the
(vtr-tua r ) load at cut r , Note that the term 6 stresses in such a case being purely "thermal".
iT
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES
AB.35
qj~llql
is (see Art. A7.8, Ex. Prob. 24).
u = qJ +
ql - qJ x
L IYrdx
Ti llll1lUJ.llill T i -h-
Tl - Tj
T - Tj • L
x Fig. AB.40
where
6.
jT
e a
qj~T"::1 x
h = constant
Fig. AB.42
Note that variation in the cross sectional
area of the bar does not affect the distortions or
'IT' 'jT' w=
The alternate choice of generalized forces
for the bar under varying axial load is shown where
in ,ig. A8.4l. By a
derivation similar to
that above one finds
e ' ; (31Yr 1 6+ IYr ~
aL (1Yr1 +6 20T~
AjT ;;; h
Fig. AB.41
wner-e Special forms of the thermal distortion
expressions for beams of varying depth may be
derived readily as reqUired.
Example Problem 19
The upper surface of the beam of Fig. A8.43
is subjected to a temperature 6T above that of
the lower surface, varYing linearly ~s shown (i.e.,
The simpler cases of uniform load (qi =U
r~=-x
constant) and unifo~ T (T = Tj ) follow im-
i
~ediately by speCialization of the above forms.
For example, for a bar under constant load d (
qi = qj' and constant temperature, T i = Tj = T, ~.~on~
L -1 L
I f',...
./'
"
~
I
AS. 36 STATICALLY INDETERMINATE STRUCTURES
Solution:
solving
{qs,,}=
In the illustrative example of Art. AS.12
this structure was analyzed by emploYing as EI a OT,{.267}
generalized internal forces the bending moments h .933
q, and q .. over the central supports (see Fig.
A8.35). The two central reactions were denoted The final stress distribution was
by q~ and q~. The matrix of redundant coeffic-
ients, considering q~ and qa as redundants (the
better choice, it will be recalled) was (ret.
Art. A8.12)
=0 ... EI a 6T, 1 0
L
h
6EI 0 1 .257
I~ 1 2
1 1 0 .267 }
EI a. OT 0 933
2 0 1 h :399/L
{
-1.60/L
2 1
3 - C C Thus the reactions were
1 2
4
C - C EI a. 6T o
Lh
Member thermal distortions were computed. (Note
EI a 6To (negative indi-
that OT was negative according to the convention c, = -1.60
Lh cates DO'>,/'N)
adopted earlier).
Example Problem 20
aL
2 X 3! AT ,
v The symmetric sheet-stringer panel of Fig.
~lT 11(-) A8.44(a) is to be ar~lyzed for thermal stresses
6
developed by heating the two outSide stringers
to a uniform temperature T above the center
stringer. Assume G • O.385E.
Solution:
""h.
; - (-) (2 x ~ OT6 + ~ 0
4
T0"" (""
+ -
h
(-l The panel was divided for convenience into
three bays. The numbering and placing of gener-
alized forces is shown in Fig. A8.44(b).
'" - 3:2 l""i 0 T~ Transverse members (ribs, not shown on Figure)
were considered rigid in their own planes - a
satisfactory assumption for symmetric panels.
Then from eq. (36) Because of symmetry only one half the panel was
handled. All member fleXibility coefficients and
~ ~ {q'J ~ ~ ~!
aL6T, thermal distortions were doubled where appropri-
3h ate.
..1.
6EI 14 q, =(-) 0 - 1
01 2 2aL -T The matrix of fleXibility coefficients was
3h 0 c set up as (VOIDS DENOTE ZEROES)
C- L-
o
o
P'
2. 44.4 0 - .0 0
31 200
31,200 0 0
31 2 0
- .5
1.0 0 0
was
The matrix of ~ember thermal distortions
~1~ =
0 1.0 0
0 0 1.0
40 - .025 0 0
40
.025 - .025 0
20
o 0 .025 - .025
o
o MultiplYing out per eq. 18,
o
o
o
1~'8J
13.8
[2,2
Here, for example AlT = 2 x 2 x 10 = 40 (doubled ~r~ = 1
E
13.8 172.2
once because ql acts on two stringer ends and 0 13.8 86.1
doubled again to account for the other halt or
the panel).
and
~ "1I
-.025
", ""'"
1
-.005 o
.025 "
~ -."'-' ao
",,02
PCyp 1=== 60"
T
F=
I- =
=
(bl 'i.
Then the redundant equations, per eq. (36),
FIg. AB.44 were
The structure was three times redundant.
Stringer loads q~J qo and qo were selected as 13.8
redun<1ants. Fig. A8.45 shows the unit redundant
load Sketches for q. = 1, qo = land qo = 1. 172.2
13.8
Solving,
n3lf '\-
{ } = a ~ S
{
. 1082 }
.?9944
."C02
~ 0~ - 50.1 -
:::: x 10 -. 108.2
99.44
is 16}1 0&<I.1n~ fi. 100.2
- 2.70
Fig. AB.45 0.219
- 0.019
A8.38 STATICALLY lNDETERMrnATE STRUCTURES
1
temperature T, unt rcr-a spanwt se , above the other
three flanges.
,. Liz ILiZ I,
L/2
"', ' 1
~~ ~ I
1\ ,\~,. "\~'''' Liz Liz L/2 L/2
). ~':\~,
~'~ •
~,~~- q\~\~
\
Mult 1plying out per sq. (16) ,
~) ~\
:\~, ~
~' .500.. B:33
, \ i:"iC7 . .625 .375 .1:5
~:,.., \1
[iki''' ... .375 .l25
\~'~
.'25
l.·k~(b'::)
~'!J . '", t
.s-e .37' ~ ...::.::.3 .:25
--L
~.--I
(aJ
Fig. AS. 46
~ ~ "\~ .125 . 125 .125 ~ . .JS33
f}
statics that the gener-al.iaed forces for adjacent webs and
cover sheets are equal, as are the loads in front and rear a T Sa 5
spar caps at any given station. Hence an economy of
numbering in the generalized force scheme is possible.
Much labor is saved in the handling of data when the same
[]rLl{"IT}=~ i
symbol can be employed on several members whose loads
are known to be equa1. Then the solu-::ion to 3q. (36) w~s ~Titten as
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES AS. 39
4 ' 079 }
Gt a L T 1.938
{qST} = - 16 Ib + c)
{
.8562 at the external points (compare with eqs. (35)
.2413 . and (37) ). Simultaneously the redundant cuts
will experience relative displacements.
Finally the .conptece set at thermal stresses SECOND, the redundant cuts are restored to
were (still for the case L3k 3 = 1) zero displacement by the application at redund-
ant forces (thiS problem was solved in Art.
A8.l3) .. The qs are given by eq. (36); they pro-
4.079
2.039 L duce additional deflections at points m
1.938
3.008 L
.8562
3.437 L
.2413
The total deflection of point m is then
3.558 L
Example Problem 22
Co~pute the rotation occurring at the right
_ a L 0 To
- h 1.2 0 !.J
hand end at the beam ot Fig. A8.43.
Solution:
aL:ST o
- .289
To compute the rotation a unit couple was h
applied (positive counterclockwise) at the right
hand end. An additional generalized force, It is apparent that the values of the re-
called qa, also was added at that point. Then, dundant ~oments ql, qa could have been chosen
arJ1trar1ly in {g1~}' above, with8ut af~ecting
,
"5T--(-)
_ ct, (3g 0To + 20 T0)_- -4 a L 15 To
the result. Here, clearly, {gi;n} ror any "cut "
h 6 9h
structure visualized w1l1 lead to the same re-
and, uSing some results tram Example Prob. 19, sult, qa being equal to ~ity in all cases.
Example Problem 23
compute the axial ~ovement of the tree end
L 15 To of the central stringer of the panel of Fig.
9n A8.44.
2, 5
Solution:
Note that q~ and q., the intermediate support
reactions, were omitted from consideration. An additional generalized ~orce, qlOJ was
They do not enter into any expression for the added aXially to the free end of the central
internal virtual work or the structure; or, stringer (ret. Fig. A8.44b). T~en
, !31.20:> !
i
i=1,2,5 i 2.2.21
,I 31,200
44.':'
40
From Example Problem 19, the true thermal 40
stress distribution was 20
o
{ q! } =
EI a 0
h
To {.267}
.933
o
a
o ! 1, 2, 5 o
o
(qa was zero by inspection) o
o
To determine [gi~J' a unit couple qa =1
Using the qjT as obtained in Example ?roblem 20
was applied. Taking as the Wcut Wstructure one
Where q. = qa = 0 one has Simply (With qlo = 0), the follOWing product was
formed:
-11.81
-13.45
- 6.655
11.81
Then SUbstit~tlng into eq. (38), 13.45
x aT
{Omr }• ts 0!J(~.
9h
{'} ....La 1n, "0 0T,{.257})
6
40
6EI 14 -
012
h .933
0
6.656
-84.24
6.833
•.5 928
2.402
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A8.41
I'hen , subs t t tut t ng into eq , (38) J a unit load qlo =1 gives (writing the transpose
of {glm})
5", = aT L&nJ 28.19
26.55
13.34 L&nl~ = L· 50 .50 .50 0 0 0 .025 0 0 1~
11.31
13.46
6.656
Then ~ultiplying out for the thermal deflection:
-84.24 O:.r = 34.3 a T (same answer).
6.833 It is apparent ~rom the above result that
.5928 the Simplest deter~inate ("cut") structure should
2.402 be used to compute [glmJ; it is completely
adequate.
It remained to find the determinate stress dis-
tribution for Qlo = 1. CLC6URE
As the determinate distribution the {gim} The general the~al stress problem is
stresses due to a unit load Q~o = 1 were com- complicated by the fact that the material prop-
puted in the "cut" structure obtained when erties E, G-and a vary with temperature. The
q, = qll = q:s = O. (Note that this is a differ- problem created ther-eby is primarily one of book-
ent choice of redundant :uts from that employed keeping - co~puting the member fleXibility coeffi-
in computing the thermal stresses in Exa~ple cients (U and the member thermal distortions
i j)
Problem 20.) rnus (d for a structure whose properties vary from
iT)
point to pOint with ths temperature. The vari-
o ations of E, G and a with T Will, of course,
o have to be known from test data.
o Two additional complications, not considered
1 here, are the lowering of the yield pOint with
1 heating (and the attendant increased likelihood
1 at developing inelastic strains) and the phenom-
o enon of "creep" (the time-dependent development
o of inelastic strains under steady loading).
o Should it prove necess~ry to analyze for
1 the~l stresses under ~ore than one temperature
distribution, the me~ber thermal distortion
and finally,
matrix { d may be gener'a Ltzed eaa i Ly into a
1T}
o",T = aT LO 0 0 1 1 1000 1J 28.19 rectangular form such as
26.55
13.34
11.81
13.46
6.656
-84.24 where d is the member thermal distortion as-
iR
6.833 sociated with rorce qi from thermal loading con-
.5928
2.402 dition R. The matrix [C f j ] of ~ ther.na.l
34.3 a T (AJlSWER) coefficients consists of the constant coef~iclents
in the d expressions preViously presented. wr-ile
As a matter of interest, a different unit iT
stress distribution was employed With a differ- T would be the temperature associated with qj
jR
ent choice of "cuts". If the forces q., q , and for condition R. (compare with eq. (26b),
qll are set equal to zero ("cut"), app Ltca't t on of Art. A7.111.
A8.42 STATICALLY lNDETER1vfiNATE STRUCTURES
100"
20"
"I
[-.10'j
4000*
!
A
jA
f- 100" B C,
Fig. a Fig. b
I- 60
+ 60 -j
(1) Determine the load in all the members Fig. g
at the loaded truss shown in Fig. (a). Values
in ( ) on members represent the cross-sectional
area in sq. in. tor that member. All members 500.
at same material. (6) In Fig. f, AS is a steel wire both 0.50
(2) For ~he structure in Fig. (b), deter- sq. in. area. The steel angle frame C3D has ~
mlne the load in each member tor a 700# load at 4 sq. in. cross section. Determine the load 1n
joint B. Areas at members are given by the member AB. E = 30,000,000 psi.
values in ( ) on each member. All members made (7) In Fig. g find the loads in the two tie
at same material. rods BD and CEo lac = 72 in." ; Abd = 0.05 sq.
in. Ace = 0.15 sq. in. E 1s same for all
60"
members.
1) 30" c
B~.r---aD9ct6~"
Fig. c
IE IF:o-
I
1000* 1000. II 180"
(3) For the loaded truss in Fig. c, deter-
mine the axial load in all members, Values in
parenthesis adjacent to members represent rela-
tive areas. E is constant tor all members.
so.ooos A
'F-100"-I
1-1
Fig. h Fig. i
C
~
(couple and transverse force). Solved in this
I D
C
way the problem is doubly redundant as no ad-
l~iI7T~
vantage 15 made of the symmetry of the st.r-uc'tur-e,
(11) Add ~NO additional members) diagonalS
~_ t FE and ZC (each with areas 1.0 ina) to the truss
Fig.e
I-- 120" + 120" --j
of Fig. A7.85, Chapter A-7. Find the matriX of
Douglas Dc~a airplane. Ph.otograph. showing simulated aerodynamic load being applied to
main entrance door of fuselage test section.
, Cf.
( 'j
A8.44
DOUGLAS DC-8 AIRPLANE. An outboard engine pylon mounted on a section of wing for static and nutter
tests. The steel box represents the weight and moment of inertia of the engine.
CHAPTER A9
BENDING MOMENTS IN FRAMES Al~D RINGS
BY
ELASTIC CENTER METHOD
Assumptions
A9.1 Introduction
In the derivations Which follow the distor-
In observing the inside of an airplane tions due to axial and Shear forces are neglect-
fuselage or seaplane hull one sees a large num- ed. In general these distortions are s~ll
ber of structural rings or closed frames. Some compared to fra~e bending distortions anG thus
appear quite light and are essentially used to the error 1s small.
~lntaln the shape of the bOdy metal shell and
t~ ~rovlde stabilizing supports for the longi- In computing distortions plane sections are
tudinal shell stringers. At points where large assumed to remain plane after bending. This is
load concentrations are transferred between body not strictly true ~ecause the curvature of the
and tail, wing ~ower plant, landln~ gear, etc., trame Chan~es this linear distribution of oend-
relatively heavy fra~9s will be observed. In tng stresses on a frame cross-section. (correc-,
~ull construction, the bottom structural fram- tions for curvature influence are given in
ing transfers the water pressure in landln~ to Chapter A13.
the bottom portion of the hull fraToas which in
turn transfers the load to the hull shell. Furthermore it is assuned that stress is
proportional to strain. Since the airplane
In general the :~ames are of such Shape stress analyst TIust calculate the ultimate
an~ the load distribution of such character strength of a frame, this assumotion obviously
that these fra~es or rings l~~dergo bending does not hold with heavy frames ~here the rup-
forces in tranSferring the applied loads to the turing stresses for the frame are above the pro-
ot~er resisting portions of the airplane bOdy. portional limit of the frame material.
~hese jending fcrces produce frame stresses in
gener~l which are of major importance in the This Chapter will deal only with the the-
strangth ,roportioning ot the frame, and thus oretical analysis tor bending moments in frames
a reasonable close apprOXimation of such bend- ar.d ~ings by the elastic center ~ethcd. FYac-
ing forces is nec~ssary. tical questions of bOdy fra~e design are covered
in a later c~apter.
Such fra~es are statically indeterminate
relative to internal resisting stress and thus The followin~ photographs of a ~ortion ot
consideration ~ust be given to section and the structural framin~ of the hull of a sea-
physical properties to obtain a solution of the plane illustrate both light and heavy fra~es.
distribution of the internal resisting forces.
A9.1
A9.2 BENDING MOMENTS IN FRAMES AND RINGS
Fig. A9.l shows an unsymmetrical curved From Chapter A7, which dealt with deflection
beam fixed at ends (A) and (8) and carrying theory, we have the following equations for the
some external loading p~, P a , etc. This movement of point (A):-
structure is statically indeterminate to the
third degree because the reactions at (A) and -(2)
(B) have three unknown elements, namely, sagnt-.
tUde, direction and line of action, making a
total of six unknowns with only three equations Ilx=Z~~S- -(3)
of static equilibrium available.
Oy := Z i1mds -(4)
£1
p.
P, In equation (2) the term m is the bending
P,
t I
moment on a element ds due to a unit moment
applied at point (A) (See Fig. A9.3). The bend-
\ ing moment is thus
equal one or unity
Fig. A9.1 for all ds elements A~
B
of frame. Cunit moment
A Fig. Ag.3
ds
p P Then substi-
..J I.- B tuting in equation (2) and using value of M from
equation (1) we obtain -
l~~B
is loaded, thus the movement of end (A) under sign as it has been
the loading system of Fig. A9.2 must be zero. assumed acting toward
Therefore, three equations of tact can be
written stating trAt the horizontal, vertical
the right. The ,
distance y to the ds Y
and angular deflection of ,oint (A) must equal Fig. A9.4
element is a plus
zero. distance as it is
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A9.3
measured upward from axis x-x through (A). place of XA, YA and MA respectively.
However the bending ~oment on the ds element
shown is negative (tension in top fibers), thus The axes x and y through the point (0) are
the value of m ; - (1) y = -yo The minus Sign centroidal axes for the values ds/EI of the
is necessary to give the correct bendir.g moment structure. This fact means that the summations-
Sign.
ydS
SUbstituting in Equation (3) and using M Z Er= 0 and
from Equation (1):-
The expressions Z x·ds/EI, Z y·ds/EI and
Z xydslEI also appear in equations 6 and 7.
These terms will be referred to as elastic
moments of inertia and product ot inertia ot
In equation (4) the term m represents the the trame about y and x axes through the elastic
bending moment on a element ds due to a unit center of the ~e, and for Simplicity will be
load at point (A) acting in Y direction as given the following symbols.
=
illustrated in Fig. A9.5. Hence, m l(x) x =
SUbstituting in x·ds yiJds xyds =
equation (4) and Y ds Z """'EI = I y , Z "E'r = Ix <;'
, "" EI Ixy
USing I'! from
equation (1), we x---~ Equations 5, 6 and 7 will now be rewritten
obtain, lj;- USing the redundant forces at point (0).
Fig. A9.5
hence,
- - -(7)
------(ll)
elastic Fig. A9.6
•• y
"'~S -
MX(~\
---r:-:-)
"'~S
Table A9. 1
= \ y - - - - - - -(14) y= x=
IX (I -~) Por- w=ds dist. dist.
TXIY tion 1 ).05. from
x -x y-y
wy2'::2250 wx 2:1440
CD 30=iO 15 12 120 150
If the structure 1s such that either the x 3 Ix: 750 iy=~3=0
12x3
or y axis through the elastic center 48 a axis
of symmetry than the Droduct of inertia Sum 32 0 660 16800 3456
Zxyds/EI = I xy = zero~ Thus making the term
I xy = 0 in equations 11, 14 and 15 we obtain,
=-z ¢s - - - - - - - - - - - - -(16)
tds/EI
_ Z¢sy The terms i x an1 1 are the elastic ~OT.e~t
-r;- - - - - - - - - - - - - - -(17) of inertia of each portIon of the fra~e about
its centroidal x and y axes. Si~ce I is con-
_-Z¢sx stant over each pert ian the centroidal T.oment
Yo - -r;;-
y
- - - - - - - - - - - - - - (18) of inertia of each portion ~s identical to that
of a rectangle about its centroidal axis.
A9.o1 E:m.mple Problem Solutions. Structures with at least
One Axis of Symmetry. To exp'ln In for member AB:- ~ '_ +=t= b
Fig. a
at ;:JOints A and
D, and carrying ly = ~bh3
12
=.l:.- x ~o
12'" x3!'
1 :;:: .09 (negligible)
a single load as 30"
Shown. The Y
The distance from the
8T
problem Is to de- I =3
termine the bend- two reference axes to the
ing moment dia- elastic cer.ter can now be
calculated:-
gram under this
loading.
! I ds=30=b
X •- 'fT11"TT]
A IY- D
- - - - - rrrtrrrr-
=~ : 660 = 20.625 In. ~
-
y I I
T'ne r irst;
step in the
Fig. A9.7 Zw 32
U...l.
-oJ yl ~.!=.!= h
solution is to 0 - I , I 3
x = J::N • 32 = 0
Fig. b
find the location of the elast1c center of the
frame and the elastic moments of inertia Ix
and I y . Having the moment of inertia about axis x-x- we
can now find its value about the centroidal
Due to symmetry of the structure about the axis xx ot the frame, by use of the parallel
Y axis the centroidal Y axis is located midway axis theorem.
between the sides of the frame, and thus the
elastic center (0) lies on this axis. Ix ~ Ix - ~w(y~) = 16800 - 32 x 20.625~= 3188
Table A9.l shows some of the necessary I y : I y - "w(x') = 3456 - 32(0) : 3456
calculations to determine the location ot the
elastic center and the elastic moments of The problem ~ow consists in solving eq~
inertia. The reference axes used are x·-x~ tions (16), (1:) and (18) for the redundants at
and s-s, the elastic center, namely
- ..
--~
-~-
. ..
_ ~ _ Area of static Mil diagram along the neutral axis of the frame members.
Mo - ZdS!I - Total elastic weight ot structure
In Fig. A9.10 the area at the Ms/I diagram
Moment of' static Mil diagram about equals 0 s = 22.5 x 24/2 = 270. The centroid by
X - ZTsY - x ax1s siillPle calculations of this triangle would fall
'
o - x - Elastic moment of inertia about x 10 inches from B. Fig. A9.ll now shows the
ax1s irame With 1ts MslI or lts 03 load. Os is
0 s = 270 -l' .. positive since I1s is
Moment of static Mil diagram about B-...;;:,·:-2 C pos rt ive . The next
y axis
Yo • -Z¢sx = ==""'-==::'-::T;~="'--;=::'-:; 1·375" step is to solve the
fl
Iy Elastic moment ot inertia about y x---- ..:- x equations for the
axis 0: r-edundant at the
: 20.625" elastic center. The
Thus to solve these three equations we ; I signs of the distances
~ust assume a static frame condition consistent A 'Y D i x and y from the axes
with the given irame and loading. In general I x and yare conven-
there are a nlli~ber of static conditions that can Fig. A9.1l t i ona L,
be chosen. For exanple in this problem we
~ight select one ot the statically determinate Thus,
conditions 11l~strated in Fig. A9.8 cases 1 to
,. - (270)
= 32 (from Table A9.1 = -8.437 in. lb.
, 10
,~I Xo
= Z¢sY = 270(9.375)
Ix 3188 =0.7939 lb.
I I
-:-:T1
--ZTsx _ -(270)(-2)
Yo - y - 3456 = 0.1562 lb.
Case 1
t -F'9
Solution No.1 8.437
A
B[rig.lo. ,e
DolA9.9~
Bn Ms is posrnve ,
on inside of frame.
tension
Fig. A9.12
,
7.5#
,
,""'- A
'2.j~ Fig. A9.10
MA =- 8.~7 - .1562 x 12 + .7939 x 20.625 :;:
6.06 in.lb.
The ~rame static bending noment Q~a~ram axes through the fr~e elastic cer.ter are then
~an now te calc~lated and dra'NTI as shawn i~ calculated as -12 ~d -0.625 inches ~espect
Pig. A9.17. ively.
-30 ; - E0s _ - (-390) ;
I
(5) -30 Eds/I - 32 12.19 in. lb.
T
9.375 \ .
1
x = 12 ...J
(6) Xo ; Z0sY
Ix
= -548
3188
= -0 • 203 Ib
c ,
I" C
I ,,0s'-150- "'l50 = -Z~;X ; - t.;~0) = 0.1562
t\I I
1[1:1
I
.- -
,y=0.625'
-
~:i;c:n~er
i
I
Yo
20.625
I 20 - - -(19)
1 Pin
Fig. AS. 17
D
Pin
SUbstituting -
Mg =-30+12.19-(-0.203)9.375+0.1562(-12)
= -17.76 in.lb.(checks previouS solu-
tions)
1 2 3 4 5 6 7 8 tions)
Mom. A' x = dist. Y = dist. The final complete bending moment diagram
I Os = from Y from X
Dia- Area would of course 08 the s~me as dravm in Fig.
gram of
Por- Mom.
for
Beam A
Section I
axis to axis to
C.g. of C.g. of
° 5, A,y A9.13 for the results of Solution No.1.
tion Dia. A, Os
Example ?~oblem 2.
1 ~450 3 -150 _12.00 _0.625 1800 93
Fig. Ag.1a shows arectan~~lar closed
I 2 . 60 2 . 30 _10.67 9.375 320 -281 fra~e supported at pOints A and B and carrying
the external loads as shown. The reaction at B
3 15 2 7.5 - 6.67 9.375 - 50 70
due to rollers is vertical. The ~rame at po~nt
4 45 2 22.5 . 4.00 9.375 . 90 211 A is continuous through the joint but the
react~on is applied through a ?in at the center
5 -180 2 - 90 8.00 9.375 . 720 -844 of 'the j oint . The pr obLem is to d et.ermtne the
6 _450 3 ·150 i2.00 -0.625 -1800
bending moment; c tagr-an.
93
I
'
Sum ~390 - 540 -658 w = 10*.1 in. .2401 (2) ~240
r
I
I .I ! ,L,L I ii'~""""-----7
IB • j
L :: 30 c: \1 B C 3
In order to take mo~ents of the ~5 values
I' 1
y4
L'24I~2f)·\ y
ii,
I
in cojumn (4) of the te.oIe , the centroid of
~~;4
I n
,P1"2M I 1=3
-~E. z 1
each ~ortion at the diagran must be ce~e~ined.
' T - - - ~I .-C:- x~
18
For example, the centroid of the ~wo triangular
bending morr.ent corticns marked 1 and 6 1s
12' y=9f6;'L=30 9.67 cut
x
,
.667 x 30 from the lower end or 20 inches as
I 1,A t
r I:2 D .1. 1 (5' 240
sho'Nn in ?lg. A9.l7. Thus the 1istanceS x "'" '""*' A\oo--15~ 240D
enc y from this 05 location to the y and x I Fig. A9.18 Fig. A9.18
Ms diagram for load Pi
I-I c
"~~""':'O:i'";;"
-: '~~:.-
A9 8 BENDING . :>10YlENTS, .
IN FRAMES ,
AND RINGS
c cnc t t ion en- ce t erntne stc '::1'.:: \"5 j~:-;;i '.:.; .' 'If
·~1
1 r ,
[E7. (7)
S1:::c:li:ity ::12 ,"._" me-tent c ur-ve :',2.5
' : ; - ' 0 " · ...
~ ~
. ':::=:::5:::::'-2,-
h ... ~ . .
"",.,..~
,,-~
~o"',.,
out I
--:- cut 9 l!1g only one of tns ... ext.er-nn l - ' t_",,,, i. -:;"'3.::'.5
.~,,-
,.,~o""
-~ 0
0"
, ,_.,
A
~-4500
D1
I,
iA
.D 3b ...
360
on the s t.r-uc tur e . ?L;s. A.~.::'2, ..
these resulti::~ bend tng ::J,c;;:snt
• C
, 20 S:~:'N
C::'::;V,:;S. ::- 9
po!'tions :J1' tnes a benc t ng ::lO::;:.en: ,j 1 '-':.:;-:'·:o..'~.s
Fig. A9, 19 Fig. A9.20 numuer-ec '- to 10 2.:-8 5 ['1 own tnc par-errtnes ~2 ,~
....
:''..18 nex"': sts'J '" ;:,'ae s-:lutioD ::':'.3:'3t3 '::l~
SoL;t~:n: r Inc tng the area of the M~/I 2i~;r~~s '"-7'.': 'tr;e
first ::lO!1!en'C Gf t ae se j .: 3. ~ :2..i7_S 2.::'0'..1.': tne x 3.:'.:::
T~,e first step is to find the j ocat i cn of y axes t nr-ougn t::e 8l3.S~~:: c ent cr-. T;;'858 s i:::~l
the elastic center of the rr-ame . Due to calC'-llations can jest '.::e cone in tab l e "er-n 3.3
sy.mnetry o~ rr-ane about the y ex.;s , 'the elastic i Ll.uat.ra ted in ~2.ble .'.9.3.
center 1.'111 be on a y axis 'thr-ough tt).~ middle
Table A9.3
o~ the rraae . The ver-t.t ce.I d~s"tance y
measured from a ax i s through AD equals, Portion Area Os ,
, dist. y , dist.
of M, of Ms to Y to X
Axis °sx I. OsY
3°x24~ Diagram Portion A Axis
-y = Z(dIS) Y = 4 ::>
3 x...1'",::;X<::, 372 ( ) , A I T ,
os = 33.5 = 9.67"
Z
T
30+( 24)2.30
4 3, 2 1 - 1440 3 ~ 480 -15 10. 33 7200 - 4958
= 3600
r-edundant for-cas. 1';:',8 calculations 'Nith r-er ar-,
= 5289 ence to fig. A9.21 are -
'y
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A9.9
4124
,'-------'r0 1 Ix = I y = nr~ = n x 18~ = 18300
Fig. A9.21
The next step in the solution 1s to assume
a static ring condition and determine the static
MA = 2018 14-1. '28 x 15 1.22 x 9.67 = -113 (Ms) diagram. In general it is good practice
to try and assume a static condition such that
M8 = 201S 141.28 X 15 + 1.22 X 14.33 = -84 the Mg diagram 1s symmetrical about one or 1t
~ = 2018 + 141.28 X 15 + 1.22 X 14.33 = 4153 posSible about both x and y axes through the
,- = 4124 elastic center, thus making one or both of the
MD = 2018 + 141. 28 x +0 1.22 X 9.67
r-edundarrt a Xo and Yo zero and thus r-ecuc Ing
considerably the amount of numerical calcula-
Combining the bending ~oment diagrams of tion for the solution of the prOblem.
Figs. A9, 18, 19, 20 with Fig. A9.22 would
give the true or final bending moment diagram. In order to obtain symmetry of the Ms
diagram and also the Mall diagram since I is
Example Problem 3. Circular Ring. constant, the static condition as shown in
Fig. A9.24 is assumed, namely, a pin at (e) and
Fig. A9.23 shows a circular ring of rOllers at (f). The static bending moment at
constant cross-section sUbjected to a sym- points (a ) , (b), (c) and (d) are the same
metrical loading as shown. The problem is to magnt tude ann. equal,
determine the bending moment diagram.
50 50. Ms =50(18 - 18 cos 45 0) =265 in.lb.
I~
i0~rack:~r~ ~\\', ~
SUbstituting and multfplying by
are four portions labeled (1).,
4 since there
M=88"#
-,....-,_.-,c-,-;"o:
"
\~
The area of portion labeled (2) equals,
Prae(l - cos a).
Moments due to redundant
forces. Fig. A9.26 Since there are two areas (2) we obtain,
( 1·1
A9.10 BENDING MOMENTS IN FRAMES AND RINGS
Length cis
Fig. A9.2S shows the values at the acti~g Member cis I w' "f y 'NY IX'X' : tx .. wy2
elastic center and the bending moment diagram
prOduced by these forces. Adding the bending BDB' 94.25 1.5 62.8 t!: .. 30 3080 54.00.. 151200: 156600
moment diagram at Fig. A9.25 which is a constant n
value over entire trame of -177 to the static : 49. i
moment diagram at Fig. A9.24 gives the tinal
bending moment diagram as shown in Fig. A9.26. AB 60.0 1.0 60.0 0 0 18000 ... 0 , 18000
Example Problem 4. Hull Frame A'B' 60.0 I. 60.0 0 0 18000 ... 0 , 18000
odl){)"
lJ
In the last col~~ of Table A9.4 the te~
l x is the moment of inertia of a particular
member about its own centTQidal x axis. T~~s
for member 80B;
D
~ ~:_jY
tx - 12 b L h"
h'"''
Fig. A9.28
-L ~~~~. = 12x2".....
1 1""3' '''04'-920
'-'. "±A~ - -
"wy
y=-zw= 1468
221.2 ;z 6.64 in.
Ix = Ix! - 6.64"(Zw)
262140 - 6.54"(221.2; = 252400
78400
y ,
r{l ... cos a - - - - )
sin au
2. ,
30(1 • 0.867 - ¥) tic center. It makeS no dltterence where the
frame t s cut to form our residual cantilever,
a - sin u 0.524 - 0.5 it one or the cnt races 1s attached to elastic
center and the other Is considered fixed. With
= 10 in. the elastic properties and moment wel~~ts known
the redundants can be solved for:
For portion (2) of the Ms diagram the area of
the ~/I diagram Which equals ~s 1s _(,.=.10",Cf7=0:::00~+ .=1,;;72;:;8,.,0;;0...:+...:1:::5",6",80",0'-".) ,
- 221.2
= 1,007,000
X
o = ZX'sy
I xx
::
Example Problem 5
Distance to centroid of ~/I curve along line
;"C iroJ1 A. ?ig. A9.31 shows the general details of one
A9.12 BENDING MOMENTS IN FRA:YIES AND RINGS
Table A9. 5
Calculation of frame Elastic Properties Calculation of Moment Weights and Solution of Redundants
CoL 1 2 3 4 5 6 7 8 9 10 11 12 13
0 Elastic Arm w y' w y'2 Arm Static Moment es y ~10 -Xoy -Total
- " =.fiE: :
Weight to '0 Moment Weight Moment
- I=~
0
Ref. Axis Os :: M
~,j = ds
~
w Axis XX MS
~
" ~.= 1 X'X' ~ Y 1 0
ds I :: y'
1 7.6 .065 117.00 26. 1 3050 79700 17. 2 0 0 0 -9020 16650 7630
0 0 0 6000
2 7.6
7.6
.065 117.00 24.4 2850
2497
69500 15. 5
0 0 0
"
.. 15020
12100 3080
3
7.6
.065 117.00 21. 4 53300
35020
12. 5
8.4 0 0 0 . 8130 --6020
890
4 .065 117.00 17.3 2025
3. 1 0 0 0 ..
6
5 7.6
6.4
.065
.25
117.00
25.60
12.0
5.9
1403
151
16820
892 - 3.0 1220 31. 2 - 637 93.7 .. -
3000
2910 ·10710
- 8. 9 - .. - 8610
7 6.4 . 25 25.60 0
- 6. 3
0 I 0 2880 71. 6
- 6160 .. -14840
8 6.4 .25 25.60 -161 1017 -15. 2 6230 159.3
286.5
2420
- 13220 .. -14720
-20830
-17510
-18590
9
10
6.4
6.4
. 25 25.60 -12.6 -323 4060 -21. 5 11260
18470 473.0 - 21250 . -27100 -17650
6.4
.25
.30
25.60 -19. 1 -488
-543
9320
13850
-28.0
-34.4 29020 618.0 - 33100 .. -33300 -13300
11 21. 30 -25. 5
..
.- 52400 -
12 6.4 .34 -32.0 -602 19250 -40.9 43180 814.0 -39700 5540
6,4
18. 82
27600 58980 1110.0 .. -45600 4360
13 .34 18. 82 -38.3 -720
_840
-47.2
--111300 ..
14
15
6.4
6.4
.34
.3'
18. 82
18.82
-44.6
-50.8 -959
57500
48750
-53.5
-59.9
77330
99030
1455.0
1865.0
77800
.. -51700
-57900
16610
32110
111700 -61100 41600
.....
~.hine -63.0
16 7.0 12.0 O. 58 -55.4 - 32 1780 -64. 3 149700 86.7 - 5570 -62200 78500
f-: 7, , 7.0
7.0
12.0 O. 58
O. 39
-58.4 - 3'24
- 131
1980
1470
-67. 3 221400 128. 3 -
-
8650 -65100
.68100
147300
194100
18 18.0 -61. 4
-
-70. 3 271200 105. 8
-
7440
.
19
20
7,0
7.0
34. a
50.0
o. 21
O. 14
-63. 2
-64.8 - 9
840
588
-72. 1 I 3086()0
-73.7 327900
64.9
45.9 -
4680
3385 .. -70000
-71400
229600
247500
Keel -75.0 332900 I -9020 - 72600 251280
Totals 811.48 7228 423237 7315.0 -348106
2401b
Columns 11 and 12 record the values of l.-. 6' 6" ------ooi
Mo and -Xoy for each station poIrrt . For
examp.Ie , trie va Iue of -Xoy for station (1) c , -r-r-
equals - (-368 x 17.2) : 16650 and for stat~on
(20) = -[-968(-73.7)J = -71400,
T 6"
,
1
Y::5.242
I
'1
~cment carve as the result of the values in 120~ :E.C.
c ojunn 13. ~;1
I
I x- --.. h
A9.5 Unsymmetrical Structures. Example prootem Solutions. 9" I 5. 032
.L
Fig. A9.34
Fig. A9.34 shows ~n l~~s~~etrical frame
carrying ~he leads as shewn. Dete~ine tte
jenQ~ng no~ents at p01~ts A, B, C and D.
Soluti.::m: -
The distance x f r'cm tne j me AB to the Fig. A9.36 shows the centraid loc~tlQns uf
elastic center is, the 05 values along the center line of the
frame. The ~cment of these 0s values about tte
x 6 10 x 12
x = (15)0 + 6
31
+
= 5.032 in.
x and y axes will now be calc~lated.
1_f:Jx
5.242
t U· l
6"
11 • Ms + ~o - XoY + Yox, for exampl e ,
t'" 2)
D 9.758
'1-
258
OS2'::-
. C.
600 6 750
Consider point A.
1 "I ~1080
M s diagram 10S3"'-~60
A .o~
MA = -2520 + 923 - 68.46 (-9.758) +
(-132.36)(-5.032) = -263 in. lb.
Fig. A9.35 FIg. A9.36 Point 8. x = -5.032, y = 5.242, ~s = -1440
The 05 values which equal the area at
the Ms diagram divided by the I values of the Ma = -1440 + 923 - 58.40 x 5.242 + (-132.36)
part1culqr portion will be calCUlated. (-5.032) =- 209 in. lb.
4500.7
P2 = lOOt _ 1
I Member BO, I y - 'j] x 31 367 x 30' = 791.7
Member CD, I y -T
- 20 X IS' + {z x 20(11" : 4501.7
Member AD, I y -_ 12
1
x:r1 x 30'
= 1125.0
Totally = 10919
Calculation of Ixy -
Due to Pl~l,
"
Calculation at Ix -
dem
Y bar·AB , -x
T-lx-l..-(,e3'''+'1'''-7'')=1'7.
3 1.5 ~ . "'tV ..l. .<)~. . . .-;;:~1
0 . .:::. . . . .
=D
=1967.48
(10" I
The ~ext step is to compute the '~lue of
Member AD, Ix = (3 (l1.657::1) 1
+12x 30 x
0s for each ?ortion of the ~oment di~gran. 0 5
20) is the area of the !'1s / I diagram. For reference
(0.5)" =2038.·30
the ~ortions of the Ms diagram have been labeled
1 to 4.
.t»
A9.16 BENDING MOMENTS IN FRAMES AND RINGS
B
" ,
Portion \0). e;s ~
31.67
-2- x ---t-
T-O
:: 39:~8
I~
-165
18.343! ~l~C-. ~
Portion (4). - el s
z0's = 13125
= -5CO(120) _500(1 ° )
2
= -,3000 !
L.-i
:
~ :4.674
~
. 8.343
"31. 1)71-;
.
11.657 \
- 5000 :: 8125
-Ai
;""-15
:
11--"":""'-15''------1
iD.L- 167,§,
167'4;j[~307
fi'"1307
Point c. !'Is =0
. .
i
~ 0 s .:5000
---
Me =0 - 124.8 - 31.07 x 8.343 + 4.674 x 15
= -314 1::1.1'J.
r-- 15'-L.........j
Fig. A9.39 Point D. Ms:: 0
Z0sx = 6000(-15) + 3167(-5) =-105835 MD = 0 - 124.3 4.574 x 15 + 31.07 x 11.657
MO =~
Zds/I
~ -(8125)
65 •.58
= -124 • 8 1
n,
1b
•
Fig. A9.42 shows a rectangular frame and
loading. This frame is identical to example
problem 1 of Art. A9.3, except it is pl~ed at
I points A and D instead of fixed.
= Z0s y - Z0s x ~)
x., Ix (1 _ ;xy') The first step will be to dete~ine the
x"y elastic weight of the fr~ne, the elastic center
location and the elastic moments of inertia
=208659 - (-105835)(-~) = about axes through the elastic center.
6451 [1 -
10919
(-1753.9')
x 6451
J 31.07 lb.
The term dslEr of a 'Jeam element of length
ds represents the angle change be~Neen its ~#o
end :aces when the element 1s acted upon by a
=- [l0sX - l0sY (-¥;~ unit moment. In this chapter this tern has
Yo I J been called the elastic weight of the element.
I y (1 - IXly ) PhYSically, the elastic weight is the ability
of the element to cause rotation when acted
upon by a ~it moment. When a unit ~oment is
= -[-105835 208659C1753.s11
6451 ~_
10919[1
-
(-1753.9')
10919 x 6451
4.674 lb.
J- applied to a rigid support, the support suffers
no rotation since the support is ~igid, there-
fore a rigid support has zero elastic weight
and therefore does not figure in the fr~e
elastic properties.
A9.17
"F' 1I"
L
230
i !
,I
Fn
-ld(~L
Fig. A9.45
Fig. A9.45 shows the bending moment dia-
gram due to the redundant XO ' Adding this
diagram to the original static diagram gives
Fig. A9.46
A B '" yJE.C.
Y
the final bending moment curve in Fig. A9.40.
Fig. A9.42 Fig. A9.44 Fig. A9. 43 Ag.7 Analysis of Frame with One Pinned and One Fixed
Support.
Due to symmetry of structure about the
centerline y axis the elastic center will lie Fig. A9.47 shows the same frame and loading
on this axis. Since the two hinges at A ~d as in the previous example but point D is fixed
B have infinite elastic weight, the centroid instead of hinged.
n4i
or elastic center of the frame will obviously
lie midway becNeen A and B. Fig. A9.43 shows The support D
the elastic center 0 connected to the point A has zero elastic
by a rigid bracket. weight and the pin
at A has infinite
tor frame is infinite because or
~ds/EI
elastic weight,
the hinges at A and B. I Fig. A'. therefore the elastic
center of the frame
The elastic moment of inertia about a y
axis through elastic center is infinite since
x-Ai - - x D lies at point A.
Tne total elastic
y wei~~t of frame 1s
the hinge supports have infinite elastic weight.
infinite because of
Ix is calculated as follows: - pin at A.
Fig. A9.44 shows the static frame con- The static frame condition will be assumed
dition assumed to obtain the Ms values. the same as in the previous example problem,
hence 0s = 270 and acts lOW from B (Fig. A9.48).
r,I
The value of 0s for member Be equals the
area ot the Ms curve divided by I tor 8e, / ' " 5270
hence ¢s = 4S x 24 x 1 = 270. The centroid s - l1S ' n-11.92-~ A-11.92
2' 2' r~lO...i "' W1l
of this ¢s value is 10 inches from pOint B. rw ~
t
'I I 28.45 'a
'~e redundant forces at the elastic center can 3~' r ~
now be solved for ' !MO
tl;!- Xo """ 5038#
~
'i 19
~1 •
~6.1'
, ~.
-270
ir.f:ni ty 'Yo i. 2582* Fig. A9. 50
y
a
= -Z¢5
1y
X = -(270)10 = 0
infinity
Since the frame is ~~symmetrical, the x
and y axes through the elastic center at A are
not prinCipal axes, hence
_ ~0sY _ 270 x 30 =0.482 lb.
Xo - iX - 16500
CHAPTER AlO
STATICALLY INDETERMINATE STRUCTURES
SPECIAL METHOD . THE COLUMN ANALOGY METHOD
AID.l General. The Column Ana Logy- method is a Now assume we have a frame whose centerl~ne
method tr~t 1S widely uSed by engineers in ce- length and Shape is identical to that of the
termlnlng the bending moments in a bent or column section in Fig. AIO.I. The width of each
~lng type structure. The method considers portion of this frame will be ?rOportional to
only distortions due to bending of the lIE! of the member. Fig. AlO.2 shows this
structure. assumed ~e. Furthermore, assume tr~t end B
of the frame 1s fixed and that a rigid bracket
The numerical work in using the column is attached to the end A and terminating at
analogy method 1s practically identical to point (0) the elastic center of the frame. The
that carried out in applying the elastic center frame is SUbjected to an external loading,
method of Chapter A9. W"l.1 w{I, etc.
Y
AID.2 General Explanation of Column Analogy Method. Ell I
Fig. AlO.l shows a short column loaded in
w,
compression by a load P located at distances
(a) and (b) from the principal axes x and y ot
the column cross-section.
x-
.i,
EX
:f),Mo
'-Xo
jyo --r
-ly
A
f-x :11-
Fixed
Fig. AlO.2
Fig. AlO.!
This cantilever structure wIll sutfer bend-
ing distortion under the exterr~l load system
W)., Wv e'tc , , and point (O) will be displaced.
Point (0) can be brought back to its original
undeflected position by applying a couple and
two forces at (0), namely, Mo, Xo and Yo as
shown in Fig. AlO.2. Since point (0) is attached
To tind the bearing stress between the to frame end A by the rigid bracket these three
supporting base and the lower end of the forces at the elastic center (0) will cause
column, it 1s convenient to transfer the load point A to remain stationary or in other words
p to the column centroid plus moments about
to be fixed. Therefore, for the frame in Fig.
AlO.2 fixed at A and 8, the moment and two forces
the principal axes. ~hen it we let cr equal the
bearing stress intensity at same point a acting at the elastic center cause the static-
distance x and y rr-ce the yy and xx axes, we ally indeter.ninate moments M1 when reSisting a
can wr t te given external loading causing static ~oments
MS. The fir~l true bending moment M at a paint
on the frame than equalS
M=!'1s-+!"!i'
wnere A is t~e area of the calumn crass-
section and pa and Pb the moment of the load From ?ig. AIO.2 we can 'Nrite for a point
P about the xx and yy axes respectively. If on the frame such as 8 t~at the lndeterminate
we let Pa = Mx and Pb = My, the above equation bending moment Mi equals,
can be written,
M. i =:1
0 + Xoy -+ Yox - - - - - - - - - - (2)
- - - (1) In Crapter A9, Art. A9.3, the equations
~or Mo, Xo and Yo were derived. They are,
-Method of Analysis due to Prof. Hardy Cross. See "The
Column Analogy" Univ.lll. Eng. Expt. SU. Bull. 215.
AlD. I
AlO.2 THE COLUMN ANALOGY METHOD
(4-) M at any
~ - Z0sx
Mo = Zds/I ' YO = Iy
~
10
j7. 5
they are similar. In other wordS, the inde- 2.5
Fig. AlD.3
ter.ninate bending ~oments Mi in a trame are Fig. AlO.4
analogous to the column bearing pressures 0,
hence the name nColumn ~~alogyn for the ~ethod SOLUTICN NO. i
using equaticn (1). ~ith this ~eneral ex-
planation, the method can now be clearly ex- ~e first c8r.sider the ~r&~e centerline
plained by g~ving several example prablem shape as shewn in Fig. AIO.3 as the crcss-
solutions. section of a shart column (see rig. AlO.5). The
width of each portion of the column section is
AlO.3 Fra.mes with One Axis of Symmetry. equal to llEI of the ~ember cross-section.
Since E is constar.t, it ~i1l be ~de w~i~y a~d
From the previOus discussion we can write, the widths will then eq~al 1/1.
=-76.0
-30 .90 Me = Ms-Ml =-90- (-76.0) =-14.0 In s Ib ,
Fig. AlD.9
4' Fraille Point D. x = 12, Y = -20.525
•
::~ + 2531(-20.525) (-9160 )12
Ml 32 3186 3456
(4) ;' -99.82
(1)
-10 I -30
MD =Ms -M1 =- 90 - (-99.82) = 9.82 1::1. lb.
= -36.06
s
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES AlD. 5
2401b
-1~40
L 6" -.;... 6" .., Fig. AlO.14
at , C
I
JI
, I
120'--r:
1=2
I' I
L'T I
10"
1-1
·1080
_5"'00"'3"'2 , Y
9" I
1= 1 1
!
P 1. =- 1440 x 3/2 = - 2160
1 I -'- P II =- 1440 x 15/1 = -21600
-l IA
'::T:7
p, =- 1080 x 4.5/1 =- 4860
Fig. AlD. 12
Fig. AlD. 13 zp =- 28620
AIO.6 THE COLUMN ANALOGY METHOD
These loads act on the centerl~ne of the Fr~e Po1rot C. x = 6.968, Y = 5.242
frame members and t:rough the centro1~ of the
geometrical moment d1agram Shapes. These M1 =-923+58.37x5.242+132.36x6.S68
centroid locations are indicated ~y the heavy
dots 1n Fig. AI0.l4 and their locations are = 357.7
given With respect to the centroidal x and y
axes. The loads Pl, Pa and 23 are now placed ~ =Ms - Mi = 0 - (357.7) =- 357.7 tn .lb.
en the column in AIO.13, acting upward because
they are negative. FY"ame Point D. x = 6.968, Y =-4.758
i1:x =- 2160 x 5.242 + 21600 x 2.258 + 4860 ?'In = lis -I'1i = 0 - (-326) = 326 1::.10.
x 5.758 = 70290 The above results check the solution of
this same problem by the elas~1c centeY" method
= 2160 x 3.032'" 21600 x 5.032 + 4860 in Art. AS.S of Chapter A9. The st~Gent should
x 5.032 = 139700 solve this problem by chOOSing other stat~c
frame conditions.
To solve equation (6) we must have the
terms ~' MY and k ,
(*)
AlO.6 Problems
~ = l1x = 70290 x 217.74/606.51 (1) Deter.n1ne the bending moment dlag~am ~or
the loaded structures of Figs. AIO.IS to
= 25234 AIO.20.
400 400
11'Y =11Y (~)
I -
y = 139700 x 217.74/942.96 ~ lSi 15 +151 w=lO Ib/in.
= 32258
L_45"
I 3 = f H~'; I
IL=30"
ix(
L=30"
11 L = 24 L = 24
= (1 = (1
-
_ 217.74' ) I =2 =2
k -
x y
) 606.51 x 942.96 II =6 I I =6
29-
I
:L== 20
I
1L== 20
10 L; 24 i 10 I II 5 I 5
whence, 1=4 i-.i. ~ .....
1-8 8+i""; i---- 24" -------!
!'1 i =- 923 + 68.37y + 132.36x -----(7) 100 100
Fig. Ala. 17 Fig. Ala. 18
For Frame Point A. x =-5.032, Y =-9.758
f.., 1=~
= - 923 - 667.15 - 666.0 = - 2256 1/,-;--"
4~",~~~
I
. 1-~-2 /,10" \ I
1 1
-264 i'l.lb. 2
.
,~4'~12'~''''
I
1.5 200 200
ME = 11 8 -~ =-1440- (-1230.6) •
(2) Solve ?rob:ems (2) and (3) at ~he end of
-209.4 l.n.l'tl. C~~pter A9, Art. A9.9.
CHAPTER All
CONTINUOUS STRUCTURES· MOMENT
DISTRIBUTION METHOD
All.l Introduction. The moment distribution set up in the two beams at B. The question is
method was originated by Professor Hardy Cross.* how ~uch of this moment is developed by each
The method 1s Simple. rapid and particularly beam. The physical condition which establishes
adapted to the solution of continuous structures the ratio of this distribution to the two beacs
ot a high degree of redundancy, where i t avoids at B 15 the fact that the B end of both beams
the usual tedious algebraic manipulations of must rotate through the same angle and thus the
numerous equations. Furthermore, it possesses unbalanced moment at 270 will be distributed be-
the merit of giving one a better conception at tween the two beams in proportion to their abili-
the true physical action ot the structure in ty of resisting the rotation at their Bend thru
carrying its loads, a tact which Is usually a common angle. This physical characteristic ot
quite obscure in some methods of solution. a beam is referred to as its stiffness. Thus let
The method ot procedure in the Cross method it be considered that the stiffness factors of
is in general the reverse of that useo -tn the the beam BA and Be are such that 162 is distribut-
usual methods where the continuous structure Is ed to BC and - 108 to BA as shown in Fig. AI1.4.
first made statically determinate by removing (The question of stiffness factors is discussed
the continuous feature and the value of the re- in a following article).
dundant then solved for which will provide the Referring to Fig. AII.4 again it 15 evident
original continuity. In the Cross method each that when the elastic curve rotates over joint B
member of the structure is assumed in a definite that it tends to rotate the far ends at the
restrained state. continuity of the structure beams at A and C, but since these jOints are
is thus maintained but the statics pt the struc- fixed, this rotation at A and C is prevented or
ture are unbalanced. The structure is then moments at A and C are produced. These moments
gradually released from its arbitrary assumed produced at A and C due to rotation at B are re-
restrained state according to definite laws of ferred to as carry-over moments. As shown by
continuity and statics ~til every part of the the obvious curvature at the elastic curves
struc~~e rests in its true state of equilibrium. (Fig. All.4), the carry-over moment is of oppo-
The general principles of the Cross method site sign to the distributed moment at the ro-
can best be explained by reference to a speCific tating end. The ratio of the carry-over moment
structure. to the distributed moment, referred to as the
Fig. All.l shows a continuous 2 span beam. carrJ-over factor, depends on the physical prop-
Let it be reqUited to dete~ne the bending mom- erties of the beam and the degree of restraint
ent diagram. T~e first arbitrarily assume that of its far end. 1Carry-OVer factors are discus-
each span is completely restrained against rota- sed in a following article. For a beam at con-
tion at its ends. In the example selected ends stant section and fixed at the far end, the car-
A and C are already fixed so no restraint must ry-over factor is -1/2). In figure AI1.4 a fac-
be added to these points. Joint B is not fixed tor at -1/2 has been assumed which gives carry-
so this jOint is 1magained as locked so it can- over moments of 54 and -81 to A and C respective*
not rotate. The bending moments which exist at ly. To bbtain the final end moments we add the
the ends of eacri member under the assumed condi- original fixed end moments, (.the distributed bal-
tion are then determined. Fig. All.2 shows the ancing moments and the ca~J-over~aments as
moment curves for this condition. (For calcula- shown in Fig. AII.4. With the indeterminate
tion and formulas for fixed end moments see fol- moments thus determined, the question of shear,
lowing articles). Fig. All.3 shows the general reactions and span moments follow as a matter
shape of the elastic curve under this assumed of statics.
condition. It is noticed that continuity of the
structure at 8 is maintained, hOwever from the All.~ Def1n1t10na and Der1Tat10ns 01 Term8
~oment curves or rig. A1I.2 it 1s found that the
1. Flxed~end ~oments:
internal bending moments in the beams over sup- By ~:ixed end moment 1s meant the moment
ft
port 8 are not statically balanced, or speci:i- which would ex1st at the ends of a member if
cally there is an unbalance of 270. The next these ends were fixed against rota~ion.
step is to statically balance this joint, so
it is unlocked tram its imaginary locked state 2. stiffness Factor:
and ObViously joint B will rotate (See Fig. The stiffness :actor of a ~ember is a value
All.4) until equilibrium is established, that Is, pnopor-t tonaj, to the tnagnf tude of a couple that
until resisting moments equal to 270 have been must be applied at one end of a member to cause
unit rotation of that end, both ends of the
( Paper - A.S.C.E. Proceedings, May 1930) member being assumed to have no movement at
All.l
'."".':_'~':.""'~~-:;;-
'{~1~:
.111.2 THE MOMENT DISTRIBUTION METHOD
r-
200
15''-01
VI: 10.,/in. L 4. Distribution Factor:
Fixed~ . ! I" . 1= Fixed It a ~oment is applied at a joi~t Nr.ere :NO
, .I. i" L·1.3~ ----1 or more members are rigidly connected the distrl-
A B Fig. All. 1 C
but10n factor for each member 1s the prcport~onal
part of the applied moment that ~s resisted by
that member. The distribution factor for any
I ~ember which will be given the symbol D equals
KlZK, where K equals stif~ness ~actor of a par-
ticular member and ZK equals the sum of K values
for all the Joint members. The sum of the D
,-480 values for any joint must equal unity.
5. Sign Convention:
Due to the fact that in ~ny problems where
members come into a joint n-om all dn-ect icns as
is commonly found in alr~lane structure, the
Elastic Curve for Fixed End Spans customary sign convention for moments ~y produce
contusion in applying the ~oment distriJutlon
method. The follaNing sign convention 1s used in
Fig. All.·3 this book: a clockwise moment acti~g on the end
of a member is positive, a counterclockwise one
is negative. It tollows that a moment tending to
Elastic Curve due to Rotation of Joint B rotate a jo Int clockwise is negative. It should
be understood that when indete~lnate continuity
Balancing Joint B moments are determined by the ~oment distribution
i
-1081162 ! method using the above adopted sign convention,
,./ --=:::::::::... ! Carry Over Moments that they should be transferred i~to the conven-
54 -811 from B to C & A tial signs before proceeding with the design of
1 ~ ::..::::::..:c.:::...:....:c.::.._ _
I
I
i
831
Conventional + + tension in bottom fib-
.26~~~====~~~~~~~~~~~~~~~~/~1,
moment Signs ers 1s POSitive bend-
Final Moment ing moment.
Ar- C Curve
E:xam:ple 3 Pab
M t~ L~
• 1.L... _ S'3M a
Fig. All. 9
External applied ~oment
at joint in structure Fig. AI1.10
Since the fixed end moments are statically Since the change in slope of the elastic
indete~inateladditional facts must be obtained curve bet'.. . een ends (1) and (2) is zero, theorem
from the laws or continuity i~ order to solve (2) as applying to fixed end beams can be re-
for them. In this book the theorem of area stated as follows.
moments will be used to illustrate the calcula- The sum of the areas of the moment diagram
tion of the fixed end moments as well as the must equal zero. And fram theorem (1) the stat-
other te~ which are used in the moment dis- ic moment of the areas of the moment diagram
tribution method. -(Ret. Chapter A7) about any point must equal zero or in equation
The following well known principles or form:
theorem of area moments will be used:- ZM = 0 - - - - - - - - - - ----(1)
(1) The deflection at any point WA" on the ==0 --------- ----(2)
elastic curve of a beam away from a tangent to For a beam with variable moment of inertia the
the elastic curve at another point WB" is equal conditions for fixity are:-
to the moment a r the area of the M: diagram be-
jl1dx/El = 0
El j!".xdx/EI = 0
tween the points A and 8 about point WAft.
(2) The change in slape between two points Figs. All.ll and AIl.12 show the static and con-
~A~ a~d ~Bft on the elastic curve at a beam 1s tinuity moment areas, the total area of each por-
equal to the area of the M diagram between the tion and its e.g. location.
El Substituting in Equation (1)
two paints ~A" and "9". M = Pab M1L MaL 0
7+7+2= ------(3)
The "area moment" theorems will be illus-
trated by the applications to the solution at a
simple problem. Fig. AlI.5 shows a simply sup- and rr on equation (2)
ported beam of constant ~oment of inertia and Mx about lett end = Pab L + a ~"hL L
~odulus of elasticity carrJlng a single concen- 2'-3-+2·3+
trated load. Figs. All.6 and All.? show the
static moment curve acd the shape of the beam - -(4)
elastic curve. Now assume that the ends are fix
ad as shown in Fig. All.S and let the value at The values of M1 and t1a tor any value at a or b
the fixed end moments be required. Fig. All.9 can now be found by solving equations (3) and (4)
shows the shape of the final moment curves wade Table AlI.l gives a summary at bema fixed
~p of the static moment curve and the '~~~O~trl
end moments for most at the loadings encountered
tr-api zo i ca.; moment curve for:ned by the ur...iknown in routine deSign and analYsis.
end moments. Fig. AIl.IO shows the shape of the
elastic curve, the slope at the two supports be- w TABLE All. I
ing made zero by fixity at these points.
P
'--
I
a -t--- b -----:1
1 EI is constant ,
Fig. All. 5 r;~( ,
1 L _ _--I ;.-' 12L" 4aL-3a)
@~ Fig. All.6 w
WL~_1< + !:Jr::WL
-8- 8
~ - - - - - - -..- FIg. Al1. 7 . 2 ' 2 .
Fa iP
I
L
b=1 Fig. AI1.S
,
;0 (5u ... 2v)
::::;.-:;::::rITn
v ::::' , 1 I ; I L L'
L ----{
6O(5u ... 3v)
All.4 THE MOMENT DISTRIBUTION METHOD
':'hus
(ML x 1(3) _ ML
6EI (Positive Shear
2EI
~5~" {lOL <I_ 10aL+ 3a 'i ~~
60L {5L~ 3a)
<I
Neg. Slope;
~ x 2(3 ~
"-
~Jj
= 2'E~I ,~
3EI
wa• a a.
30L<I(lOL -15aL+6a)
p/ wa'
20L a ( 5L - 4a ) Let 98 unity
Then 1 ML or Mg = :lEI stiff~ess Factor of'
3EI r::-
Seam SA of Fig. AII.13. A ~c~ent applied at 3
produces no ~oment at A since end A is freely
supported. Thus the carry-over factor for a
beam freely supported at its ~ar end is obvious-
ly zero. Consider the beam of Fig. All.14. uue
to complete fixity at end A, the slope of the
-tj~X(L-XI' ~
~o
t{WX'(L-X)dX
dx
elastic curve at A is zero.
1/3 + ~ 2(3 =0
o
(Ref.N.A.C.A. °A !1ll!: 2EI
2EI
General case - Any loading T. N. #534)
or
MA = -~
2
All.4 Stiffness 7actor; Carryover ?actar: -
Derivation af: Thus the carryover factor for a beam fixed at
its tar end Is 1/2. Using the conventional ~om
(For definitions of these terms see page ent signs, the carry over moment is of the oppo-
A11.2) site sign as shown by the above equation. How-
Consider the beam of Figure All.13. By ever, for our adopted sign convention inspection
~Ohr's theorem (see Art. A7.l2), the slope ot a of the shape of the elastic curve as shown in
Fig. All.14 tells us that the sign of the carrJ-
EI is constant over mament is of the same sign as the rotating
moment at the ~ear end. That ls, the moment act-
i~g on each end of the ~ember 1s ir. the same di-
L
rection, and therefore of the same sign.
Fig. All. 13
Fig. All. 14
Te.ke 9B
L
;
L
unity and let EI = K for simplicity.
L
Problem #1 ~A
~1.5 ~
11 11
gil
]} tt tt tl
! all
W!!'J 51.51
Figs. All.lS and AII.IS show these results. Carry 0.-.,.. 1/1 57.5 _51.5
Fig. All. 17 shows the general case, the r ixa- 2IId Bll!aJlCUlIj: c c «c
tion factor at A being F. The difference be-
." '"
tween F'fga , AII.IS and AII.I? is that the slope
at end A has changed but eB the slope at end B
FI.llaJ. Moll:lefttll
• I•
·o.rerlla:lg MOIll.nt : .5';51.5)[15.75 • . 15x$1.5':17.11: alI3
• •
remains the same.
tlA=O Example Problem 1 shows a two-span cont1nu-
F;i~B ous be~ with over-hanging ends. '..le first as-
A· "". sume all the joints B. C and D to be locked
MA=~2K MB=4K
against rotation or the beams Be and CD are
Fig. All. 15 arbitrarily assumed in a fixed-end condt t t on ,
The !irst line in the solution gives the
stiffness factor of each beam. Fr-om Art. AIl.S
the general expression for stiffness factor is
F=O
~
A
MAsO
=.5 8B'"1,
MB-3K
- EI(3 + F)/L. where F ; fixation factor of far end,
wni en equals , since all joints have been con-
"
steer-ed rtxec. ThUS, stiffness factor K ; EI
Fig. All. is (3+ l)/L; 4. EI/L. Since EI/L is the same for
each span Be and CD, the Stiffness factor has
The change in :noment at end A when changing been wri tten as one since it 15 only the rela-
beam AIl.lS to that of All. 17 = 2K- 2KF ; tiva values that are necessary. The cantilever
2K(1-F). Since 9B is xept the same value, one- SA has zero s t t r rneas ,
narr of the moment change at end A appears at The second row gives the member distri but t on
end 8 but of cppoat te sign, or factor D at each jOint or 0 ; KlZK. ?or eXalIlple
Mg=4K-l/2 2K(l-F) ;3K-KF=K (3 + F) ;
ZI (3 +F) 0) = o and to BC
distribution factor
.
at jc ~::.t 3. the distribution factor to 3A = 0/(1
lI(c • 0) ; ;- . At joint r !:-o. e
to CB ; 1/(1 + 1) ; .5 and
..
L
eA' F(. 5) Lfkewt se to v-c• •
The thll'd row giyes the carry-over factors.
F"F~ .
From Art. All.5, the carry-over factor C ; 2;;'/(3 +
MA'= -2KF MB-K(3+ F) F'I = 2 x 1/(3 + 1) .5 assummg the far end rrxe d
or F = 1 for each member.
Fig. All. 17 The next step as shown by the 4th horizont-
Thus the ge!leral expression tor the stiffness '11 row is the calculation of the r ixed end non-
factor or a beam of constant section equals EI ants. The signs or tl:e :n.oments are according to
C- our adopted sign convention, t~t is an inter-
(3 + F') • ':'he carry-over factor from 9 to A ; nal r-es i st.mg moment wni cn tends to rotate the
All 6 THE MOMENT DISTRIBUTION METHOD
end at a member clock'.... i ee 15 positive. 4. Calculate the carry-over factor C for each
We now begin the solution proper by first end of a:t.l :nembers C = 2F/(3+;-). Thus ror
unlocking joint 8 from its as suzned fixed state. beams f~xed at far end F = 1 and thus C = 112.
We rind a moment of -883 on one side and 765 on For pinned at far end F = 0, hence C = O.
the other side of JOint or a sta t1 c unba.Iance 5. Calculate the fixed end :noments (t) for giv-
of -115. Joint 9 therefore will rotate until a en transverse member loadi:Igs or support oer rec-
resisting moment of 115 is set up iY1 the member-s tions, using equations in summary ~able All.l.
EA and Be. ~~d ~oments which 'tend to rotate end of :Jearn.
The resistance of these members to a rota- cIockwf se are ~ositive moments. (See Art. Al1.2)
t10n ot Joint B 15 proportional to their stiff- 6. Considering one joint at a time, unlock it
ness. The distribution factor based on the from its assumed fixed state, all other joints
stiffness factors 15 o for BA and 1 for SC. remaining locked. It an unbal~~ced moment ex-
Thus 1 x 115 = 115 15 distributed to SC at Band ists balance It statically by distr~butlr:.g a
Ox 115 = 0 to EA. Joint 8 15 now imagined as counter acting moment ot oppOSite sign among 'che
again locked against rotation and we proceed to connecting members according to their D or dis-
Joint C, which 15 now released from its assumed tribution factors.
locked state. Since the jOint 15 already stati- 7. These distributed balancing moments produce,
cally balanced, no rotation takes place and the carryover moments at far-end of member-s equal to
distributing balancing moment to each span 15 the distributed moment t taes the carry-over rae-
zero. Next proceed to joint D, and release it. tor C and of the same Sign as the dlst:oibuted
The unbalanced mament IS, 115 so the JOint 15 moment. Record these carry-over moments far ,~
_0
balanced by distributing -115 between DE and DC ends for all distributed moments.
as explained above for joint 8. 8. Repeat the pr-oceedur'e of unlocking each
As painted out in Art. All.5, when we ro- joint, distributing, and carrying over moments
tate one end of a beam It tends to rotate the until the desired pr-ecf s t on Is obtained, stopping
far fixed end of the beam by exerting a moment the solution e rt.er- a distr~bution. The final
equal to some proportion of the moment causing moment at the end of any member equals the alge-
rotation al: the near end. For beams of constant braic sum at the original r fxed end Jl.Oments and
section and rtxed at their far endS, the carry all distributed and carryover moments.
over factor 15 112 as explained before. Thus
the distributing balancing moments 1n line 4 ~ample Problem #2
produce the carry-over moments as shown In line
5 ot the table. This completes one cycle at the
moment distribut10n method, which is repeated
unti~ there Is nothing to balance or carry-over,
*
.SJ/lIl.. f lJ/lll..
}i.S+---~____t__72
" I' -L
EI l.o Can.w>t
F! q
ge------J..s l . ,~
.
. 12 :
or in other worca until all artiticial restraints I ,
To continue with the second cycle, go back D~~:'IOI>~ ,I. .~:l8!. S72 .s.s .sn!.Hs 110 I
to joint B and release it again tram its assumed
locked state. There is no unbalance since the
Carrr·Oo.r C
factor e .s sl.s sl.s
I
... 'I'
,
carry over moment tram point C was zero, thus fl.om E%ld _a, l&S • ..l...L1 •• ,lllliu:lOl."Lz-'_dzi43:z--.l. .. L1'_43 z
lZ ,11 _ 12
l,e8 "-!.."Ll
11
• _ 7M1883 i
-~"
I
Carrr-Qru
n Il5
72
X 1«lm
57. slo
::::><: 01
0
961_9&
X~
~><~
c _S7.S _121:
1t balanced Wlder the carry-over moments at 57.5
and -57.5. Thus the distributing balancing mom- :lod8ll\alw:lIl.t: u _72 X2~_5!'32_9 ~X2.9!24.6~ 7210
car,.,-Oo... ·(2.3 _3slo X;6.5116.5 0136 12.31
ents are zero. Joint D Is l1kewise in balance
since the carry-over moment tram C 15 zero. All 3"" 8IIl:uI<:..... 0,IZ.3~lS.ti20.6~~X°:.1.1S.?::<11.JIO
joints can now be released Without any rotation car,.,.OvJr 7.1 6.110 10.l!.10.3 *6,1 _:71
since all joints are in equ1ll brium. To obtain ~tlI 8IIlaAc ..... 01_1.7~_1.6 ::':'X~X.SZ.6 ~r,:io
the tinal moments we add the original raxec end e...., ..,·oou I-l.] .~ o _1.611.8 013.9 I.J!
moments plus all distributed balancing and car- Sib SOWle .... olt.3 I. 7 1. ~
," _1.1 _I. 1 _l.J10
ry over momen ts . ll.oal..,..._ .. .~318U -erc 41610,4 _3+413+4 _610. tlalO. 4 _8331883
and unlock, all ot~er joints remaining fixed necessarJ. These modifications usually involve
against rotation. The unbalanced moment 1s rather long expreSSions :or expressing the stiff
(-768 + 432) = - 336. Balance by distributing ness and carryover factors of a me~ber in terms
.428 x 336 = 144 to CB and .572 x 336 = 192 to at the fixation given by adjacent members. It 1s
CD. Proceed to joint D, and release. The un- felt that it is best to keep the method in its
balanced moment is zero which means that joint simplest fOrn which means that very little is to
D is in equilibrium. thus no distribution is be remembered and then the method can be used in
necessary. Proceed to joint E and F and balance frequently Without refreshing ones mind as to
in a similar ~er. The distributed moments many required formulas or equa't Iona ,
will be the same as the values for joints Band There are however several quite Simple ~odi
C due to symmetry of structure and loading, f1cations which are easily understood and re-
however, the signs will be oppOSite under our membered and which reduce the amount ot arithmet
adopted sign conventions. The next step is the ic required considerably.
carry-over moments which are equal to 1/2 the For example in Problem 12, joints Band F
distributed balancing moments. This operation are 1n reality treely supported, thus it is need
is shown clearly in the table. Values of all less arithmetic to continue locking and unlocking
moments are given only to first decimal place. a jOint which is definitely tree to rotate.
The first cycle has now been completed. Cycle Likewise due to symmetry of structure and load-
~NO is started by again releasing JOint B. We ing it is only necessary to solve one half of the
find the joint has been unbalanced by the carry- structure. Due to symmetry jOint D does not ro-
over moment of 72. Balance the joint by distri- tate and thus can be considered fixed, which
buting - 72 x 1 = - 72 to·BC and zero to EA. eliminates the repeated locking and unlocking of
Proceed to joint C. The unbalanced moment is this Joint.
57.5. Balance by distributing - 57.5 x .428 = A second solution of Problem #2 is given in
- 24.6 to CB and the remainder - 32.9 to CD. gxaap te Problem #3. As before we assume each
Proceed to jOint D. There is no unbalance at jOint locked and calculate the fixed end moments
this joint since the carry-over moments are in Now release or unlock jOint B and balance as ex
balance, thus no distribution is necessary. platned in previous example #2. Before proceed
Proceed to joints E and F in a similar manner. ing to jOint C, carry over to C tram B the car-
The carry over moments equal to 1/2 at the ry over moment equal to 115 x 1/2 = 57.5. Joint
2nd set of balancing distributed moments are now B is now lett free to rotate or in its natural
carried over as shown in the table. The second COndition. Proceed to joint C and unlock. The
cycle has now been completed. This operation unbalanced moment = (-768 + 432 + 57.5) =- 278.5 or
has been repeated five times in the solution 278.5 is necessary for equilibrium. This moment
shown, or until the values at the balancing and is distributed between eNO beams, CD which is
carryover moments are quite small or negligible fixed at its tar end D and CB ,Yhich is treely
The final ~oments equal the algebraic summation supported at the far end B. The stiffness fac-
or the original fixed end moments plus all dis- tor is equal to (3 + F) EI/L (See Art. All.5).
tributed and carry over ~oments. One require- Hence for CD stiffness tactor = (3 + 1)
ment of the final end moments at any joint is EI/L = 4 EI/L. For CB stiffness factor = (3 + 0)
that the algebraic sum must equa L zero. The EI/L = 3 EI/L or in other words the stiffness ot
other ~equirement consistant with the cammon a beam freely supported at its far end is 314
slope to all members at any joint is given by as great as when fixed at its far end. Thus the
equation (5) of Art. All.S. The results at stiffness factor at CB at C is .75 x .0104 =
jOint C will be checked using this equation. .0078. The carry-over factor C to B is zero
since B is left free to rotate. (See Art. All.S)
.5 t. l",jc = ~
<ll'Igd -
t.l1cb - .5 Hlbc Kcb Example Problem #3. Simplified Solution Of Prob-
lem /12
SUbt. values
(610.4-432) -.5 (344- (-432)J _178.6-44 =
..., r,:-H.
.~. l'IUI. FlU<!,
[-610.4 -(-768)J - .5 (883 -768) 107.6-57.3
1.343 s
e
. c
12~
B:l!anc. 10"" C
sents the "Cross" method in its flli~damental and Ca,..,.""" to 9. 0
most elementary detailed form. Many modifica- B.t.J..:u>c. ,<lUll 0
tions of the general method have been presented,
..
Car..,_onr to C
.a831883
in the most part for the purpose of eliminating Final mOm.nUl
Con.enuon:U &lll'l_
part of the arithmetic or the n~ber of cycles
I
c, ..
~
All.8 THE MOMENT DISTRIBUTION METHOD
Joint D. Since the mOment of inertia is con- Since the fixed end ~Oments are due to bot~
stant and the spans are cons tant , the relative lateral loads and support def:ect~on, the values
stiffness factor of the beam is 1. In the so- as listed in the solution table Nill be explain-
lution shown since beam ~s freely supported at ed in detail.
3 t.nt s joint is lett tree to rotate after re-
leasing and thus the stiffness factor of beam Fixed End Moment For Lateral Beam Loading
CB Is 3/4 x 1 = 3/4, when compared to one haVing The distributed airload is trapezoidal in
full fixity at B. shape. The fixed-end ~oments for a trapezolcal
Since the first step in the solution proper loading from Table All.4 are:
1s to assume the joints fixed against rotation,
it Is evident that deflecting one support rela- Ml_2=L.a (5u+2V) (See Fig. (a))
tive to an adjacent support will produce mom- 60
ents at the ends which are assumed fixed against M2_1==L.a (5u+3v) (See Fig. (a ) I
rotation. 65
All.a Fixed End Moments ..L_ -- -- -=
Due to Support
Deflections
ll.6 ~~
Ef ,~
i~ ~ -~~
4EI distance d relative to
the support paint C. It
the me:nber ls or constant
f
CD
L
"
@
6
1 LJ ~~O~~~~:~~~~~ -~~~lP~~~i
at the beam mid~oint and
For Span Be:
40'
Fig. All. 18 ME - Me. !1bc (5x3+1) = 426 ~n. lb.
60
By moment area prln- 40'
cipal deflection !'leb (5x3+1.5) 440 in. lb.
a !1L L -J1L 5 L
60
4El (3 4El (3 For Span CD:
or d == ML a 40'
5El I'lcct 60 (5 x 3.5 + 1 I = 494 In. lb.
hence 40'
I'l 6EIa 'tha.magnt tude for the fixed end Mct.c = 6'Q" (5x3.5+1.5) = 507 in. lb.
-V
mO:nent due to a transverse support settlement ?lxed End Moments Due to Sunnort Movement
of d.
From Art. All.S l1 = ~Eld/L.a
Examole Problem #5. Continuous beam with de-
flected supports. For span Be:
Deflection of B relative to C = 5/16 inch. Hence,
General
data; 3 f/in. 4 '''/In. 3'/- It-·049
A1.
! ; ; t
U1;J (Alloy tube !'!be = Me'::l =6 x 10,000,000 x .03339 x .3125/40.a == 390
o ": • , I j , tt.:,1:.03339
in. lb.
I!
-jlOj-- 40 ---r--
I
40 - - 40 40 ~ --110~- F'Qr Span CD:
ABC o Defle<:,tion of Supp?rt!l
t i 3/ 16 "
DI --- ?'lcd =t""'dc :: 6 x 10, 000, oeo x .03339 X .1875/40
a
;:: 234
- -f'11e<I due 10
1::1. 10.
Solution:
A iB
&) i
C
I For si;~s of tte ~oments due to tr.ese deflections
see Art All.2. Havir~ dete~ined the fixed end
Stiffness factor K 3/4x1~. 75 1 ~ooer.ts t~e ~eneral distributing and carrYing
Distl'1.bution Factor ~ (t,i ZK 011 . i21l .571 n I over process follows as indicated in the solution
C.u-ry.over Factor 1/' o 1/2 112 table. Thus at joi~t B, the unbalanced moment ::
Ftxed
End
Due to
Lateral Load
50 426 507 (50-426 -+- 390) :: 14. Balance by df s tr-t out t ng - 14
Momentll Due to support
deflection
o 390 390 234 ,,. xl:: - 14 to Be and Zero to EA. carryover.5 x
-14 :: - 7 to C. Considering joint C, the unbal-
Balance joint B
anced moment == (440 + 390 -+- 234 - 494 - 7) :: 563. Bal-
Carry_over 10 C
Balance joint C
ance bYdistri but tng - 563 x .571:: - 322 to GD and
Carry·over ~o B. D ·161 -563 x .429 :: - 241 to CB. carryover.5 x -322 ::
Balance joint 0 o ·580 - 161 to D. A~ joint D the unbalanced ~oment =
Final Moments 50 _50 580 ·580 (507 + 234 - 161) = 580. This is balanced by dis-
Conventional Signs tributing zero to DC and -580 to the fixed support.
All.I0 THE MOMENT DISTRIBUTION METHOD
factor D for each ~ember at each ~oint which Member - K. Value c. O. Factor
equal K/~ is recorded in c=lon each member' ~(hF)
2F
around each jOint. Thus any balancing ~oment 3- F
is distributed bet~een the joint members as per 2x1 =1
OA:-l..(3+1J "'.4,
these d1stribution factors. The carryover 20 ~ 2
factors for all members is 1/2. The fixed end 2xO
moments due to external loading are computed OB.i·~{3+0) :.3, 3 +0 • 0
tor the three members. For member AO, the tix- 1 2x.25:.
ed end moments equal PL/8 = 120 x 20/8 = 300 in. OC=25{3~.25)=~ 3 •. 25 154
lb. The other two members having no transverse ZK=.83
loading, the fixed end moments are zero.
In this solution the or1er of joint con-
sideration r~s been AOEe and repeat. starting
With joint A the joint is released but since
the member AO is actually held by a fixed sup-
port, no rotation takes place and the balancing
moment of 300 is provided entirely by the sup-
port and zero by the member AO. The carryover
moment C, to 0 is zero. Releasing joint 0, the
unbalanced moment of 300 is balanced by distri-
buting - 300 between the three members accord- SOL tJrION .2
ing to their D values, thus - 300 x .416 =
- 125 to OA; - 300 x .168 = - 50 to OC and Fig. All. 22
- 125 to CB. To prevent contusion it is recom- by distri but.fng .75 X 25 =18. '75 to CO and the r-e-.
mended that a line be drawn under all distri- mafnder- of 6.25 to the support, s tnce the fix!.. ty -r;
buted balancing moments, thus any values above the support at C has been assumed as 25 per-cent _ A
these lines need not be given tur~her consid- line is cra~m ~lde~ t~e 18.70 and the ca~ry over
eration and only values below the lines need moment of 9.37 is taken over to 0.- One cycle has
be considered in later balanCing ot the jOi~tS.
now been completed. Retur~ing to ~olnt A, ~e
Immediately a:ter dlstribu~ing the moments at find -62.5 below the line. This is balanced by
joint 0 the proper carry over moments should be distributing zero to OA and 62.5 to the t!..xed
taken over to the far end ot each member, thus support. A line 1s drawn ~~der :he zero distri-
- 62.5 to A, - 62.5 to Band - 25 to C. Joint buted ~oment to AO and the carryover woment of
8 is next considered. The unbalanced ~oment is
zero is placed at O. Considering joint 0 f~r
-62.5 and it is balanced by distributing 62.5
to 80 slnce the pin support has zero sti:tness, the second time the unbalanced mo~ent is 9.37 +
31.25 + 0 = 40.62 or the S~~ ot all vaiues below
or no resistance to rotation. A line is erawn the col~ horizontal lines. The joi~t is bal-
under the 62.5 and the carryover moment ot
anced by 1istributing - 17 to OA ~~d 08 ~~d
31.25 is placed at O. Joint C is considered -6.62 to OC. Lines are dr~Nn under t~ese bal-
next. The unbal~lced ~oment o~ -25 is balanced ancing momen~s as shown in Fig. All.21 and t~e
SOL111'ION'IIl
carryover mcments are taken over to the far
eTIQs Jefore proceeding to joint 3.
This general process is repeated until ~oint
A has been balanced 5 times and the ot~er jOints
4 ti~es each, as indicated ~n the figure the dis-
tributl~g values have become quite ~ll and it
SWmellB Factor K
[ a
OA=I:- 10 : .10
is evident tha~ a ~igh degrs8 of accuracy has
5;,lO
been obtained. The ~inal end ~oments at each
08= lis joint equal the algebraic sum of the values in .
DC = 225 =~ each COIWL~. A double line ~s placed above the
ZK·JointO=,24 r inaa noaerrts as a df s t tnguf sht ng symbol . In
C.O,F",ctor_1/2
far All Memo.
the fi~~e the letters band c ~efer to balancing
and carr; over ~oments, the sUbscri~ts ~etarring
to the member of the balancing or carry over
o=eration. Any order of joint consideration can
be ~sed in ~eaching the same result.
Sol~tion #2 of ~oblem 7
Fig. All.22 gives a second solution. ~ith
the end conditions ~own at A, Band C, the ~odi
~ied stitfness factors of the members can be
found together with the ~odit1ed carryover fac-
tors, thus making it ~ecessary to balance ~oint
Fig. All. 21 o only once and car~y over this final far end
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES All.13
~oments of each member. The fib~e gives the assumed to give 50% fiXity to these joints. In
calculation of the modified or actual stiff~ess Table A11.2 a modified stiffness factor is cal-
and carryover factors. ~ith these !u~c,vn the culated tor ~embers GI, Fl, and FH using a 50
solution is started as before by computing the percent fiXity at their tar ends. The last
fixed end ooments due to transverse loading on column of Table All.2 gives the s~tion of the
~e~ber AO. Joints Band C are released and ~ember stiffneSS factors for members intersect-
si~ce no fixed end moments exist, no balancing ing at each jOint.
1s required and the joints are left in their 120*
true state 0: restraint instead of locking and 3200"-' J. l-
,- .12'", G 1
unloCki~ as in solution #1. Releasing joint 0
~:/7\ ~~~:'~d
tram its imaginary fixed state the ~~balanced
moment is 300 which is balanced by distributing /\ /
- 300 betNeen the 3 connecting members accord-
ing to the new distribution factors at joint o.
Thus - 300 x .482 = - 145 to OA; - 300 x .361 =
- 105.2 to OB and - 300 x .157 = - 47 to ce.
~~e carry over ~oment to A = - 145 x .5 =- 72.5
l~7;''t ~~-~
Engine
--t1d'r-
MOWlt 2400\
,~~
I~
A
12400".
F
r
H
'\ /
t.o B = - 108.2 x 0 = 0 and -47 x .154 = - 7.2 to
Exa~ple Problem #8
"'\ -, f -,
/
Landing
Gear
Fig. All. 23
34.'
, 1.0240
.0240
0.582
61
.S18*
3.06
A
IS
C
D
S 80
4 81
6 95
6.14
all the ~a~bers due to the_eccentric joint ~om -- 2 -- .049 30.0 .0588 1 6 ,sa
,.. _3 - .,
en~s ana two lateral loads. The effect of joint
..
D' 1. 96 7 4.42
.0449 6 G 3.21
tr?-nslation and secondary ooments due to deflec- • - •• -- •• 34. • •
..
:I 2•
o s•
3 .033 D.
t10TIS and axial loads is to be neglected in this
ex=..m.ple. • 1_1_1 2 - .049139.5 0178
Be
- - • ". (lSBB (II: 1. eoutaJIt
ell 8 - o (I:l~141 0 •
,• fol' Lll _ -
So1.:J.tion: 1_ - >en)
~able All.2 gives the calculation or the
( .t1ffn••• fLctor _ 7/8 ~ beca~ of ~OS f1xity)
stiffness factor for each truss member. The L
~uselage truss aft of joints I and H tave been
Fig. All. 24
..
_1.98 (
.1.82 b
-,
(n5
-141
b .178 _187
. ,a ee .", "'-, --. ·,
"
• 18 b b -:....11. .286
-"
·.,84
1)---i
0-'.
c-t
-'-'
b
'-."
...:l.o
'-.-.i..
c ·n ·"',....
«n .195 0-'-
.l15
•
"'."' ......,..,....
:::~ sr- -s
" .
..
__ .834
:...JJMb
na --o...ill
~b aa .,29 b
_ _ b· 140
n•
-
70 • 140 -10 ->;.286
',0 b ~
" ,
. ~b
-,
,
~ . --1)·15
':""""55'
m" b-rot ; ~
~"
.\11.14 THE MOMENT DISTRIBUTION METHOD
Fig. All.24 gives :he solution of the 8Y distributing 183 to AC and 15~ to AS with
problem. The procedure in this solution was as carry over mo~ents of r~l: these values to ends
follows: C and B respectively.
The stiffness factor K for each member as The student should now be able to c~eck the
computed in Table All.2 15 recorded in the cir- rest of the solution as given on Fig. All.24.
cles adjacent to each truss member. The carry The solution could be ~de with any order of
over factors for all ~embers 15 1/2 except fcr joint consideration. If any particular jotnt ap-
modified members GI, FI and FH for which the pears to be nearly balanced, it 1s best to skip
carryover factor to the 50% fixed ends is .286. it tor the time being and consider those joints
The distribution factor for each member at each which are considerably unbal~~ced.
jOint is recorded at the end ot each member, and 7he ~inal moments at the end 0: each me~ber
equals K/LX. are given below the double :i~es.
The next step in the solution is to cocpute
Example Problem #9
the flXed end moments due to the transverse
loads on members. Fig. All.25 represents a cross section of a
For member BE welded tubUlar steel tuselage. The top and bot-
MBE=Paba/L li =120x29.25x12.11/41.2S ll =-2J8"# tom members which are web members in the top ~~d
bottom fUselage trusses are subjected to the
l1E:B =120 x 29 .25' x 12/41.25' =725"# equipment installation transverse loads as shown.
l1cD = 100 x 20' x 10/30' = - 445 Let it be required to deter.nine the end bending
l10c = 100 x 10' x 20/30' = 222 moments in the rectangular trams due to these
transverse loads assuming full cont:nuity thru
These moments are placed at the ends of the joints.
members on Fig. All.23 together with the eccen-
tric joint moments. The process of ur~ocklng Solution:
the joints, distributing and carrying over mom- Fig. All.26 shows the solution. The dlstr1-
ents can now be started. In the solution as bution factors based on the member sti!fness fac-
given the order of joint consideration is tors are shown inD at ends of each member. The
ABCDEFG and repeat, and each joint has been bal- first step 1s to compute the fixed end moments
anced three t1.mes. due to transverse loads, on members AS and CD us-
Consider joint A:- ing equations from 7able All.l. The ~gnltudes
Unbalanced moment = 2400. Balance by dis- are 1890"# for AS and 2025~# for CD.
tributing - 2400 as tollows:- Joint B is now released f~om its assumed
To AC = - 2400 x .527 =-1258. Carry over fixed state. The unbalanced moment of 1890 is
to C =-634 balanced by distributing - 1890 x .247 = - 467 ~o
To AB 2400 x .473 = -1132. carry over SA and the remainder of -1423 to BD. T~e carry
to B =-566 over moment to A = - 46S-x .5 = - 233. Due to
Proceed to Joint B:- symmetry ot struc~~re and loading only one t~lt
Unbalanced moment = (-566 + 3200 - 298) = of trame need be considered and hence these car-
2336. Joint 1s balanced by distributing - 2336 ry over moments to A are not recorded. However,
to connecting members as tollows:- • in balanCing joint A 1t will throw over to 8 the
To BA =- 2336 x .569 =- 1330. carry over to same aagrrrtuee of carry over moments as thrown
A = - 665
over to A from B but of opposite sign since the
To Be = - 2336 x .310 = - 724. Carry over to original fixed end moment at B is minus. Thus
C - 362
233 comes to 8 tram first balance of A as shown
To BE =- 2336 x .121 = - 282. Carry over to in the figure. The distributing moment to a of
-1423 produces a carry Qver moment of -1423 x .5=
E = - 141
The convenient device of drawing a line
-712 at D. ~27"
300# 300#
I
under all balancing moments is used to prevent 10··--t- 7" -r-10"-j
contus1on in later balances of the joint.
1"-.035 Tube
T
Proceed to Joint c:- K=· 0~~37::. 000458
Unbalanced moment = (-634 - 362 - 44S) =-1441
The joint 1s balanced by distributing 1441 as ~
.
. ,. 0
CB = 1441 x .214 = 309. carry over to B = 155
CE = 1441 x .064 = 92. Carryover to E = 46 '~I~
<;gOM :<;g
0
11
CD =1441 x .282 =406. carryover to D = 203
This process is continued tor the ~emainder or
the trJSS jo rrrts, ar tar all joints has been
I~ ~ 600# III~_~.J
I. :
balanced once, on ret~1r~ to joint A we rind -Llr13.S'~13.5"-W
below trie lines an unbalanced moment; of (318-
665) =-347. The jotnt is balanced a second time 1~118 - .035 Tube
~o 11~
Returning to joint B, the unbalanced as
recorded below the single lines is (233 - 446) L _ a - -(5) (Area of ll/I diagrams
= - 213. ~o balance 160 is distributed to SD
equal zero)
and 53 to SA. Carry-over 80 to D and bring
over from A to S .5 x-53 =- 27. Continue this
i
L~ -O
~rocess until joints A and D have been balanced
4 times or 4 cycles have been completed. The - (6) (Moment of the WI dia-
tinal moments are shown below the double lines. I - gram as a load about
?ig.All.25a shows the resulting moment diagr~ o either end equals zero)
on frame.
(Note: Since E is usually considered constant
it has been omitted trom denominator or the above
171
r :rwo®~1710
Final
§
z
equations. )
TABLE All. 3
_...
~ J-
Moment
~ Diagram
_.......- Static all C1U"'N Trial -
t
<:Ur"N Trial ~C1U"'N
r
,... ""
("
4050 i I
,
Fig. All. 26a
a
a
'"
,...
".
'" ,...
2.67
6.:n
U.
:l1ll0 U3 I 6.07
". r.. ...
"•
•• r 2. 67 1
,::~ ::.9 ,~~~~ U.
..
10 11l
All. 10 Continuous Structures with Members of
'"
,
•
..,.,...."
1120 14.10 111600 1130 13.i" 1810. 70.0 1O
Variable Moment of Inertia
I • 18.07 26000 1110
22 06 ,38900
17.96 li761 90.0 18.06 U30 I
...".'" , i " .,
1760 21 92 1970 110 0 :z:.:I cs
In Arts.Al1.3, 4 and 5 consideration 'HaS
• I 615 211.88 i41600 67.3 2ll.1l8 117<lO!~~~.1l 26.04
I
2lJ. 77 aeoec 40.7 29.76 1 121.5!117.0 211.94 I
given to the deri~~tion of expressions for
fixed end ~oments, stiffness and carry-over
ractors for beams of ~iform crass-section.
I •
"r
,,. .... 179Ctl1l
:i·g
33.llJ i4180 17.11 33.8.
M.5.7
, ~~I 33.1• "00
111122:2 1
743 8
•
" 1812:1:
)
factor for a beam with var~able ~omen~ of in- _·''IT·
ertia. ~he effect ofax:al load on these ~ac-
All. IS THE MOMENT DISTRIBUTION METHOD
00
r curve equate to zer8.
Hence
845.7 M ....~,....
(a) 743.8x 15.5.;.l'CO ~.Ax .:::V.~ = 0
whence
MA = - 53.2"#
(b)
~heretore carry over :~ctor 3 to
A =- 53.2 = - .532
Fig. All. 30 100
1'+' 19 = loa
ac xx40l ' whence c = .7 0 7 bent car:ying a distributed side load on one
leg as shown. The K or IlL values for each
10 was taken as the value at the A end or 1. member are g~ven on the figure.
(See Fig. All.27). Under the given loading it is obvious that
Therefore KAB = .707 IA/L the frame will sHay to the right or in other
Similarly ~or end 8:- words, jOints 8 and C will undergo considerable
100 x 40 horizontal movement. The moment distribution
;:gB = 4(743.8-.532x845.7) = '!exl method assumes that only joint rotation takes
whence place. To make this assumption true for thiS
structure we will add an imaginary support at
c = .85 and KBA = .85 IA/L joint C which will prevent sidesway of the frame
as illustrated in Fig.All.34. The end moments
All. 11 Frames with Unknown Joint Deflections Due in the frame will then be found by the moment
to Sidesway
distribution process. Fig. All.34 shows the
In the example problems so far treated the results of this process. To explain, the Solu-
~oi~ts of the particalar structure were assumed tion begins with computing the fixed end ~oment
to r8tate without tra~slation or with a definite on member AB !,.[L"'/12 =
(300x25 "')/12 =
15.63 =
amount of translatory ~ovement. Translation of thousands at foot Ibs. This value with the
the joints may however, be produced by shorten- proper sign is written at the head of a column
ing and lengthening of the members due to axial of figures on member AS as shown in Flg.All.34.
loads and by lateral sway due to lack of diag- Now considering Joint B, the unbalanced
onal shear ~e~bers. The problem relative to joint moment of 15.63 is distributed as follows:-
the effect of jOint translation due to axial To Member SA =- 15.63 (40/190) =- 3.27. Carry
stresses is treated in a later article. In over to Joint A = - 3.27/2 =- 1.63
this article only the effect of sldesway of
rectangular frames on the frame bending moments To Member 8C = - 15.63 (150/190) =- 12.36. Car-
ry over to Joint C = - 12.36/2 = - 6.18
';;111 be considered.
:1g. All.31 illustrates cases of tr~es
where only rotat1on of joints takes place (neg-
lecting axial deformation) whereas Flg.All.32
ll1~strates conditions in which sidesway takes
4 6' I 10'
1
6
'r-
place and the joints suffer translation as well 1= 10, 500 cl
T
B
as rotation. I
K=L'= 150
0 °
~
~ 0
~ g " 25'
-~~!J
~
°° "
~ -1-'
'"
A D
rnr
Fig. All. 33
1 Symmetry of Structure
& Loading
Frame prevented
from sidesway by
external restraint
Fig. A11.31
~Ll'>~II~I'~I~I~
- . ,-
(O.OO)f (0.00)
Symmetry
of Loading
<:"lOOIO'I_ -12.36 b 1 . -
+ 2.44 _ _ bl.
-6.18
4.88
... , ,,' II'"
but Structure '" " '"
.... .c.c..::::t.o .1.93b", -0.96
Symmetry of Str-uctur-e, Unsymmetrical + 0.38 . . - b 2+O.76
Loading Unsymetrical
Fig. A11 ....3.2
I;[ i ~
• O. 30 b~__
.O.04b ... -
-11.75
-0.1:>
_b,.,+0.12
_0.02
b..... 0.02
There are several ~ethods of dete~~ning -1. 53
the bending mcments Gue to sidesway. Only one
~ethod will be presented here and it can Jest
be explained by the solution of example prob- Fig. All. 34
lems.
~ample P:oblem 1.
9230
As explained in example problem 1, the
moment distribution 's carried out in two steps,
one for jOint rotation only and the ~ther for
effect of sidesway or horizontal translation of
- 11.75 - 1. 53 joints B and C. Fig. Al1.41a shows the moment
,
~
9.81
~
+ 20.03
distribution for no sidesway by placing an
imaginary reaction Rc at joint C. The process
is started at joint A and the order of jOint
IB C
balancing 1s ABCBC. As soon as a joint 1s
Fig. All. 39 I
~~,
~~ ~ :lJ~
_-10
balanced the carry over moments are 1mmedlately
carried over before proceeding to the next
~
~,
+1, jOint. When a joint 15 balanced a horizontal
line 15 drawn.
~r
~ ~~
.. ~~~
F.E.M. ·200 200 F.E.M.
"
~ ~
d..;j . Bl. ·143 - - - 71.5
- 4 2 . 8 - - 85.7 a ,
A '7'~
mil\'( ~ " II ...
" 1I II
D B. 24.4 - -u:2
- 4.1-- -8.1 B~
9.81
~I e
~
_"='~"b7T
M M • B,-W--U
...:..., -363.2 ~ B lI
J <I~~
.... ~~ ::i
1 .
47.3
~. N g:
.-
l>.o~E-
e .~
~
.IN
r;::1~
el,-.
,
<D
.
t: "
"'''''''
.. " ..
"'''''''
4 1.07 22.73
0
• _- •0
~<r;:: ~I< ;:
0 OO~I~I='I~
o.a 0 . " .
<'") _ _ co ...., <'") C01_1"'IIM
Fig. All. 40
,_ 'cD
M '"1"'1 0 1.;
...
I
I
I
"
,
...
I
,':);EifF"'c
,-'-..' -•.-':).:::';>it..
All. 20 THE MOMENT DISTRIBUTION METHOD
61..,. _
3.15 C
-----Rc' 61.85 ::i
144.6 ---l......Q. 53
162.6 -" .
'r
B
"'l .... I'l
"T"
363. 2 ~ H 65 HC=3.15 c-; ~ ;:Q C:U::l
~~I.''''I= cl
B" oiIMI~"'; IB
<:0 i
t-
;~<:Di~ F~ I I
10JI
35 A
!
D1_ 3 . 15 'i
0
~ ;1 0°110 QA
r l' 01
•
;ic5'
oJ
Fig. All. -lIb
. ~
Fig. All. 41c ~
;l;'"
oJ
To find Ha take moments about A, .; 6.0L _ p=6. 02" 10.85
B 10.85 =16.87 lb.
100Xl2+362.3-24Ha = 0, Ha "" 65 lb.
144.6 '0.85_")'62.6
..r:-:~ 02 C I,
To find He take moments about D, ~.
,
,
now must liquidate this reaction RC and per,nit To liquidate the reaction Re = 61.85 it
:rame to sldesway. will therefore require 61.B5/16.37 : 3.67 tioes
the moments in Fig. All.4lc. Therefore the
We will assume that the frame 15 :ixed !lnal bending moments equal those in ?~g.
ended and that an unknown horizontal torce ? All.41a plus 3.67 times those in Fig. All.4lc.
at C will deflect the trame Sideways. The Fig. All.4le shows the results. T~e bendl~g
fixed end moments for equal horizontal de- moment and shear diagram and the ~xial loads
flection of a and C will be proportional to now follow as a s iapje aat t ar of jsta t i cs .
EI/L a ot the vertical member. -363.2 47.3
~ ..ll2..Jl.
F ormemer I . 24 _
bAB 'P-24: 0417 """'lS6:8 643. J
a - '
B Ci: ~1:;
F or memb er DC , r::aI -- 1510 " -- • 0445
ti gil; li~ ~I~
Mlt:Il_
",
For convenience we will assume 417 in.lb. as
the fixed end ~oments on AB and 445 on DC. I DOl O
The assumed fixed ends will now be eliminated
by the moment distribution process as shown I
in Fig. Al1.41c. The order of joint balance
was ABDCBCBC.
IA
\1 00110 Fig. All. He
Fig. All.4ld shows the shear reaction on SXample P:oblem 3. 3enc~~g ~Oilients ~n ~~uss
the vertical members at Band C. These forces Involving One Panel ~ithou~ 01a~Gnal Stear
reversed on the top ~ember show an unbalanced Member.
force of 6.02 + 10.85 16.87 lb. Therefore=
a force P = 16.87 was necessary to ~roduce the Frequently, in aircraft struct~r~s ~ :rU2S
bending moments that resulted on the frame is used i~ which a diagonal member ~us~ os l~f~
due to sidesway. out at one or ~are truss bays. T~e sy.:er~al
shear load on such bays ~ust je carried by :he
truss chord members in bending. Fig. All.42
shews a 3 jay truss witt no 11ag~na: ~e~ber in
the center bay. The jending ~cments on tte
truss members will be dete~1ned rJ~ the truss
______________________•••••••••••..,RIlIR.'.·t!liiilll
.,
10 10 10 To Member cb =4/7 - .57
a 'c ld
~d
Gi ® .:zJ -r I
To Member cr = 1/7 = .14
~~
-7.1
de
3,0
-3.4 0.5
g!1 -0.8 -0. 5
-3.6 -0.4 """0:0
-6.7 -0 1 ~
ha -0.4 -11. 8 3.0
~ -0.9 <s!
1.2 o 1.5
0.3 -0.1 Q..2.
'.3 -TI.":'T i -0.3
Q.Jl
I
0.0
" 16 I '~. !' 13 I ~
I 2.0
~
-12.2 -24.4
10.3 5. 1
-2.6 -U
1.1 0.6
~ ---l!...i
-22.3 Fig. All. 43
All. 22 THE MOMENT DISTRIBUTION METHOD
Therefore the fixed end mcments on ~ember to axial loads ~rcduces :r~~sl~tiQn of the
gt will be 75 percent of those on member bc. joints of the structure. Since the ar~les be-
tween the members at a joint re~aln the same
Assume 100 as the fixed end moments on due to continuity, this translation of joinGs
be. Then a 75 x 10 : 75 1s the accompanying bends the members between joints. (2) ~~e
fixed end moment on gr. bending ~omen:s In ~he ~e~o~rs d~e to exte~nal
joint or lateral loads or those ~ue to ~oin~
w~ now remove the imaginary fixed supports rotation produce lateral deflection of the
on the center truss and let them rotate to members Jetween joints. The member axial loads
equilibrium by the moment distribution process times these lateral deflections produce
as shown in detail in Fig. All.43. seconda~J ~oments. These secondary ~ament3 can
be r~ndled by the general method of mcment
The first step is to record the assumed distribution however the stiffness and :arry
fixed end moments of 100 at each end ot ~ember over factors and the fixed end ~o~ents are ~ct
be and 75 at each end at member gt with due constant but become nL~ctions of the axial
regard for Sign. The moment distribution pro- loads.
cess will be started at jOint C. The unbal-
anced moment is 100 or -100 is necessary :or Fig. All.45 shows a prismatic beam Sl~ply
static balance. Using the distribution fac- supported at A and fixed at B, Nith a moment
tors on joint c, we find -57 goes to Cb, -29 MA applied at A and carrying an axial compres-
to cd and -14 to cf. Short lines are drawn sive load? Sub. Figure a, band c show the
below each at these numbers. Fifty percent of 3 parts which make up the ~cment jiagr~ on the
these balancing moments are carried over to beam. Without the axial load P the pertien (c)
the tar end of each of these members. This would be omitted.
process is repeated at each jOint ~ntil the
remaining balancing moments are negli~ible.
In Fig. All.43 the order of jOint balance was
ctdebghaCfdebghacfdebg. If the student will
tollow this order he should be able to check
the figures in Fig. All.43.
,
i i i
,
,
,
Section ! ! , i ,I :
, ,
r,
• , I i,
,
! i ! !
!
In deriving expressions for fixed end mom-
ents, stl::ness and carry over iactors the beam i
,
! I :r-, e_ - , !
coltwh~ to~ulas ot chapter AIO, must be used
, I I I I 1
i I i !
, , ,
1
I , V
0 ,
sions tor these factors and has provided graphs i ,
, ,pb- I
,
A<" I
I
of these factors for use in routine design. , ,
~~ I
Figures All.~6 to All.56 inclusive are taken •, Ii
,
i /, , I'"
:rom this thesis. The use of these Figures can / I I
, , , , i
best be illustrated by the solution of several
pro b.Lems •
.. .:.---- ,
/1
I
I I
!
,
i
i
,
,
,
,
,
i
I ,
!
j
, ,i
Example Problem #12
., I ! I i~ 1
,
! I
F~g. AII.57 shows the same continuous beam ,, ,
,,
,
tor for stiffness factor = .775 when tar end is I I I r_I..o.'. tu' _ 'tl_., '/ I
fixed and .36 when far end is pinned. Hence,
r. •
,
, ,, ,/ I
,
I , , , ,
y ,, , !
, ,
,
NY j I
,
",-<:: 'A''''
, , ,
,
i i ,
Span CD .. ,
"C ! \
I 'i \i·-e_.... l_. fU . . . IU"
I ",
/
I~',
'r
All. 24 THE MOMENT DISTRIBUTION METHOD
, ,
,·fJ
, ,
i ,
,
",
_,1"II./I.a. ,
!",-I 'Io~
,
w
! iA i ,
!
,
! ,, 1 , V , ,
,
R
,
, I,
I ... • - .. ·,t I ,
,
,
,
u
i ,;I ,
I ,
!
i
i ! ! i r : rtr
" I, y-t··- r " ,
I
,
1Ira
:
, ,
, '\!
I 1 Y ,
, , ! 1 ! I ! '\
" ,
I A I
I
,
! 1
! ! : 1\ ,
,
1
r1
1
,
, V , 1
! 1 I I, ,
r ,
\i ,
,
u
, , , !
'" ! , , ! 1
~I! \
, , 1
u
I ! ,
,
,
, , . ,
i ,,' f . . . ..1... , ! , I
! '\J I I
! ! ,I 1 ! ;-," I
-'I-'-1--'HH-i---'+--+--+_--+--'
"
I
,
I : i\
i : ,
,
! \
" 1 I ,
"
\,
,I '\ I , I I 1 I
• ,
!\
, I
! I ! '\
I I
\i
.u ,
,
! i I
,
,
\ 1
\ I
,
,
, '"
Fixed-end moment coefficient.
", Fixed-end ~oment cceffic1ent.
uni rora Load. UnifOrmly varying load.
Axial .compression.
Fig. All. 48 Fig. All. 49
.. ~ .... _~ i
,
"
'. (" I ~Ii I ~/2
\") .., ,
I
~
~
, . ,
! ' i
, :
1
: ,
, : " • -"to • ~
T_ _ /
m
, ,I
I
"
I , ,
! i ! ,
,
! I ;~ I , , :
'Y
! " i
" ,I 1 ! , U , ,
,/
. ,
I i I I I, /1 ,
1
', , ,
..
,
I ,I i 1 V ,
,
, I ! ! ,
J ,
V, , I ! /1 I
I I 1 , 1
,
•
i I
, ! 0.
,
I : 1
,
, >,, ! I ! , ,
,
'.• I ! 1/1 ! i I : Y, ,
! ,
,
,
1
~ll
, , ,
'. ,I / i
•
,
: !
!
~
I v, 1 , ,
1
! i :~I I
,
I
,
1
I 1
,
j--{' ! ~.~--
,
»
I I ! i i ,
r
, I 1 : i
I ,
~;
I I ~ !
,
I ,
I ,
,, \, ,
"
, Y
,
,
! !
·
,
I ' , ! I 1 i , 1 ,
i
• ,
, ....--1
, ,
I
,
I ,
I 1 ! , ,
\ I
I ..:--
, , , ,
i i , ,.llJ I : '{
·
, , : !
• i I
..J...--+- 1
! I
I
i i !\
,
1
,
,
.
,
I
,! I , I I ! \
,
·
, ! , , ,
! , ,
w , 1 1
• • ,! • , ,
'" coefficient.
c
• a
'I' coefficient. •
Fixed-end moment ?lxed-end ~oment
Uniformly vary'ing load. Concentrated load at ~id-span.
Axial tension.
Fig, All. 50 Fig. All. Sl
. --.-.-.------- IIIIIII!!l!!!!~
ze=
f-I.l.l~ a 1"0 ~~ I -,
r.e I- \~ 1-+-l-+7'f'=--f-+--l--t---+-t---+-t---j-+--1
!.fA, .. WI;
'b 2
D..L) Col
"
v I \
I-
4.0
I
J.5
i
.,
2.0~
I 1/ j .. 1 ~ a compreu1on--..
. JiO
4.0 !
!
.1
6.0 .......
I
.J
t- I
.. .5
alt or bit
'.0 I
., •• .9
I I
I I
I I I l.~(~~ I
I I I 1:.-- i
! ' 'I 'N.
I !
! ;
1/'f'_ _
V 1/1// i
., ., . . . -. . . '''01_' !If/"
~
(/'1
1/1
.
'. ": ":
, ": 0
iii Y YI!
. I' ! I iI Ii! i I
i I I' I : I
II !II
,
•
I
, ,
,
I Ii ,
Iii Ii :
,
;1 ,
~~Irl-'--;",,,1_,,';'-'f;p,.-,,,,--:.C;,,Tr.. ~
I
I
,,
.1 ,, I
,
I : I ii I, ,i !
I' /' F l.2L._
r -r-
= -,." " , : : , , "
.1,
,
:I I
- - -"' II
. I ..... 1
,~
I , I •,
,.,:
,
Ii i II II II '! I
I : i
i~·L-.~
, i
, I
-e
'.• ,
, i
I, I I
,
i \ \ i , II ij I iI
~ ~. ,I ,, ! ,, I i Ii i
! i I
\l-~i, .,
_.-
,
. ,
\ I
; ... ."
L~ ... ! ;1
:
,
'; , ,i I ',.1 'II., , I I I ill,,
, Fig. All. 53
' '."" \ ,
\ , , ,
\ v \i '\ 'I~ I I / II
, : ,
•
,
.• " ••
"'1 , II
,
I
All 26 THE MOMENT DISTRIBUTION METHOD
IXrnr&d~IFB
1
1.a
ri-f ~
,
i
' I:
I
,
,
,
!
I
I
I i ! III, I <;.:......eO" B' lO~" A' 84" A lOOt~60" C
j ~~. I w.3~1 w:32 ~16
L
I,
I
I I I i
i::l::O 1 ttl tIl t 1 CIT' 1 r~~
2
J
, I
.. ,
I I
! !
,i
I
,
~ 7~ 1..@j.@;2JJll.. .:L..iQ...15.J.Q:6Jg:2 ~ T FB
i I
I
I
I
1 1
,
i
! ! 1
,
C'
Fig. AI1.58
B' '~....
.. ,
............ B ,.. '
A' '....cabana /A
',strut~ "
,,,'B
v....
C
IT' ,:.,.,
a, .. " c:,....
I I ! I I ,.q_lIl_ Spruce . . ,__ uael. /:.~~\.
! i
,
II I
i
I
I
I
T
, I ,
I
r,
, i
i 6" 6"
Boam Loads
", 1 I
1 [;U75BJ -~ ~
1
• r.
i ,% I
I I 1
Y II 1=17.21 1=22 A' A B
./'! I' I I I Span·AB, A'B' Span4AA'
,. ./ ~ i : ] i I
Fig. All. 59
r-c, I i I I I
Fig. All.59 shows the total beam axial loads in
NI I I the various portions of the beam by adding the
, I
values shown in Fig. All.S8. The outer span AB
r-, I due to the axial loads 1nduced by drag wires ot
I ! ~
drag truss 1s subjected to a 'mrJing load. As is
customary in design of such wing beams the ax1a1
1
&
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES All 27
I"", "
f "J ~::..'
All. 28 THE MOMENT DISTRIBUTION METHOD
in Chapter A9 provides a rapid and Simple diagram. Thus to find the distance fram E x
method for deternining these values for curved to this centroid we take moments of the MlEI
members. The use or the elastic center method diagram about B and diVide by ~ the total area
in deter.nining the value of stiffness and at the MlEI utagran, Thus
carryover factors will now be explained.
X : -2( .667L) + l( .333L) :..:...1..= L
All. 17 Structures with Curved Members. - 2 + 1 - 1
Before considering a curved member a Thus the centroid of the total ~oment weight ~
which equals one lies at paint A or a distance
strai&~t member or constant EI will be con-
Sidered. Fig. All.66 shows a beam treely L to left of B.
,i ",, ,
v'./
All. 30 THE MOMENT DISTRIBUTION METHOD
Fig. AII.70 The total area of the M/sr curve ~or the
curve in Fig. All.72 it calc~lated ~rould equal
The bending moment at end A equals one or unity as explained in detail for a
straight ~ember. The centroid of this M/EI
diagram would if calculated fall at 901nt A.
Thus in Fig. All.73 we 2.?ply a unit ~s load at
A and find the redundant force at O. Due to
By definition the stiffness factor is the symmetry of structure about a vertical or y
moment at A which is yeQUlred to turn end A axis the elastic center lies on ~h~5 3ymnet~ical
through an angle of I raddan , Thus 42I/L axis. The vertical distance ITom :~se ~:ne AB
Is the stiffness factor and thIs result to elastic center equals Y= .6366~. (See
checks the value as previously derived In page A3.4 ot Chapter A3).
Art. Al1.4.
The elastic moments ot inert~a Ix and 1y
The bending moment at B in Fig. All.70 can be calculated or taken ~r0m reference
equals, sources such as the table on ~age A3.~.
Fig. All.71 shows a curved member, namely, Solving the equations for the redundanta
a halt circular arc of constant EI cross- at (O)J remembering ~s = 1 and located at pOint
section. The end B Is fixed and the end A is A.
freely supported. A moment MA Is applied at
A of such magnitude as to cause a rotation at = .=.....ill.::: - E: I
A at 1 radian as illUBtrated in the Fig. Me - - nr nr
Fig. All.72 shows the general shape of the ZI
bending moment curve which IS statically in-
determinate. In FIg. All.73 the support at A V
_
- Z¢sx _ - (l)(-r) _ 2EI
·0 -
Iy - nr.3 - nr"
ZSI
= Z¢sy = (l)(-.6366r)
Xo Ix O.2g7e~ 3 = -
SI
Fig. All.74 shows these forces acting at
the elastic center. T~e Jendi~g ~oment at A
equals the stiffness factor ~or the curved Jearn
Fig. All. 71 Fig. All. 72 :1xed at rar end 3 and :reely supported at near
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES All. 31
end A. The ratio of- t~e :ending ~o~er.t at The method of hew to handle this re~aining
point B to that redundant force can best be explained by pre-
at point A 'Nil:' senting some example problem solutions.
cive the carry-
Jver fac:or. All. 19 Example Problems. Continuous Structures
Involving Curved Members.
Bend.tng
:!loment at A, Ex~~ple Problem 1
Fig. All. 74 B
Fig. All.76 shows I =1
MA = nr - 2.14 (0.6366r) -
-SI;'I
;'3 (281)r
nr " a frame consisting of
both straight and
= -0.318EI 1.:;81 _ 0.536 SrI: 2.314- :::1 curved members. Al-
r r though simplified L::r60"
relative to Shape, I •2
Send~ng moment at 3, this frame is somewhat
representative of a
:13 ::: _-,,0-,-.3,,',,""8",E",I, 1.36EI + O.636EI fuselage frame with twa L::80" L=80"
1" 042 E1
r r • r cross members, one be- 1=1 I =1
tween A and C to support
T~er9fore the stiffness fac:or for a half- installations above
circ~lar arch of constant EI is 2.314 EliI'. cabin ceiling and the
other between F and 0 w: SOl/in.
The carry-over factor equals the ratio of to support the cabin I I
Mg to ~A or (-1.042 El/r)/(-2.314- EliI') = floor loads. The frame
L=60"
C.~2. It sr.ould be noticed that the carry- supporting forces are 1" 6
over mo~ent has the same sign as the applied assumed provided by
~cment at A as compared to the opposite Sign the fuselage skin as I :: 2
~or straight members. In other wor1s, there shown by the a~ows
are t~o points of inflection in the elastic on the side members. E
curve for the curved arch as compared to one Eccentricity of these
for the straight member. skin supporting forces Fig. All. 76
relative to neutral
axis of frame member is neglected in this stmpl!-
tied example problem, since the main ~urpose of
The fixed end moments on a curved member this example problem is to illustrate the appli-
for any external loading can be dete~ined cation of the moment distribution ~ethod to
q~ite rapidly jy the elastic center ~ethod as solVing continuous structures involving curved
11:~strateG ir. Cha~ter A9 and thus the ex- members.
?lanaticn will not be repeated here.
SOLUTION
The student should realize or understand
that ~hen the end moments on a straight member Due to symmetry of structure and loading,
in a ~cntinuous structure are found tr~m the no translation of the frame jOints takes place
~c~ent cistrijution ~rocess. the remaining due to frame sidesway.
end f~r:es are statically dete~inate, whereas
for a c~~~ed ~emjer· in a continuous structure, The ~rame cross members AC and FD prevent
Lr.cwing the end ~c~ents does not ~ke the horizontal movement of joints A, C, F, and 0
curvet ~e~ber statically ~eter.n~r.ate, since due to bending of the two arches. .~y horizontal
',':8 have six unknowns at the two supports as movement of these joints due to axial deforma-
il:~strated in Fig. All.75 and only 3 equations tion is usually of minor importance relative to
~~ static equili- caUSing bending of frame members. Therefore it
brl·~. ~ven when can be assumed that the frame joints suffer
tne end mcment s Fig. All. 7S rotation only and therefore the ~oment distri-
are de t erninec bution method 1s directly appllC~ble.
t: om the ncment
Calculation of stiffness (K) values for
pr-oc ess there each member of ~~a~e: -
still r-enatna
one unknown, Upper curved memJer ABC: -
n~ely the hori-
zcntaa reaction K,il,Be::: K CEA::: 2.314 E_I. mh'..e va 1ue was
,_~,
at one of the
beam encs , der:ved in the ~revlous Art. AII.IS.
All. 32 THE MOMENT DISTRIBUTION METHOD
2 "2~' l
K, ..."d.... X - = .0770 (E is con-
,;: K, ;:
ABC 30
:BA
s~ant and therefore omitted since only ~elatlve 'The curved :r.enbe:-s :::r' the .:'ra:J.e have ::c
values are needed for the K vaLues v ) applied external lcadin~s ~ence the .:'~xed end
caments on t~e 2urved :r.emjers is zero.
~he relative moment o! i~ertia of the
cross-section ot eac~ fr~e ~ember is given For ~embe~ AC, :lxeC end ~cment equals
on Fig. All. 76. WL 2/12 = 10 x 60 2/12 = 3000 in. lb. and tor
member F'D = 50 x 60 11/12 ;: 1.5000 in .lb •
K,.., =2.314x2=.1540
=X,.,
,'~D 30wEii' Since the supporting skl~ .:'arces on the
side members have jean ass~~ed ac~1::g along
For all straight members, the sti~fness factor centerline of .:'~2.Ee membe:-s, the ~1xed end
equals 4EI/L. Hence, moments on ~e~bers AF and CD are zero.
K
AC = KCA = 4x 2/60 0.1333
Moment DistriJution Process: -
, '- A .5121 I C
D = .0770/0.2503 = 0.296
illC
1
.
2 ( 09750/Y
DF
-15COO 15000 !
D
the 0 at each joint. The process is started inertia about x and y axes through the elastic
by placing the fixed end moments wi~h due re- center.
gard to s~gn at the ends of members AC and FD,
n~nely, -3000 at AC, 3000 at CA, -15000 at FD Bly
and 15000 at OF. ~e now unlock Joint A and I.. 1
- 1 - ---l
i
find an unbalanced moment of -3000 which means d:; 4.6r:-T
a plus 3COO is needed for static balance. - ---.-~ I
Joint A 1s therefore ba:anced by distributing , I 19. 1"
.512 x 3000 = 1536 to AC, .296 x 3000 =
880 to
k
ABC, and .192 x 3000 = 576 to AF. Short hori- Fig. All. 78
zontal lines are then drawn under each of these
distributed values to indicate that these are
balancing moments. Carry-over moments are
immediately taken care of by carrying over to 19.1 x n (30/1) :
joint F', .5 x 576 = 288. From A to C the y = 14.4-9"
'IT x 30 60
carry-over moment would be .5 x 1536 =768 and 1
+-
2
therefore the carry-over :rom c to A would be
.5(-1536) = -768 which is recorded at A as (NOTE:- 19.1 equals distance from line AC to
shown. For the arch member ABC, the carry-over centroid of arch member ABC.)
moment from A to C would be -0.452 x 880 =
-401 (not shown) and therefore fram C to A = Calculation of moment Of inertia I : -
-0.452 x (-880) = 401 as shown at joint A in x
the figure for arch member CBA. Joint C in r.:ember ABC
the figure has been balanced once for the
purpose of helping the student understand the Ix = -1- + -rtrd
.3r~
1-
a
~Sl = 1094 x n x 30/1 = 103000 46 lb. The member AC also suf:e~s an ~x~al l~~~
due to the shear ~eactions at the tJ~ o~ =e~:~"2
0., = 60 x 2353/2 = 70590 FA and DC. ?ig. All.Bl shsws ~re8 te118S :: ~he
s1de member ?A and DC wi:h the en~ =~,e~t~ as
~S~ = (-.667x60x4500)/2· -90000 found in ~ig.All.77.
The shear reaction
l:0 s = 63590 at A and C can be
found by taking
Fig. All.30 shows the ~s values concen- ~oments abo~t lower
trated at their centroid loca~ions and referred ends. Thus for FD, 80"
to the x and y axes :hrough the elastic center. RA = (1259 + 1791)/30
• 38.1 lb. Likewise iF
~-.:...
~
,
,-'
jo
He = 38.1 lb. These 1791 1791
y ~eactions react on
B Fig. All. 81
cross member AC in t~e
l • 0$ = 103000 oppOSite directions
4.61"......"
! .,.II 0
'x--~--~,---x
0 I thus Giving a compression load of 35.1 lb. i~
Gl, .70590. 14.49 member AC, which ~ust be added to the tension
A' It c-L- load of 46 l~. :rom the arch reaction to cb~ain
~:: ~90000 Fig. All. 80 the fir~l load in the cross-member.
= - ~ 05 -03590
Z d8/1 • 124.24
=- 673 Lns Ib , 8753
'I· ·····\.\/592
Fy D F~ID/
Xo = Z/8 Y = [102000 (4.61) + 70590 t "~-
3
T \ _ •IE.c.
...i....X_\_ __ 3.34" 19.1
x..:.:.L:.. ';1 1 . / ,...., :::J
x (-14.49) _ 90000 (-14.49)J / -"';: I -::/ f I 22500 .
16410 =46 lb. E
Iy Fig.AIl.SI E
3 -
Fig. AIl.82
Yo = -z1¢sx ~ 0 because x ~ zero.
y = 19.1 x n(30/2) z~ = 57.1
Y n x 30 + 60 I
The bending moment at any point on ABC 26
or AC equals that due to Mo and Xc plus the
moments in Fig. All.79. Ix = .3X + 15nx3.34'1 = 4578
For example,
Z30.3
(60/6)(15.75)' = 2478
Ix = 7056
At point A on member ABC,
2rom P:g. All.82,
MA = 1094 - 670 + 46 x 14.49 = 1091 (should
-8753 x 60/6 = - 37530
be 1094 since moment as found in Fig. All.77
is correct one. Small error due to slide r~le .667 x 22500 x 60/6 = 1500GO
accuracy.
05 3 = - 5962n x (30/2) = -326000
At paint B on ~ember ABC: -
z 0s = -265500
MB = 1094-670-45xl5.51 :::II -289 in. lb.
So Iv tng for Redundant; "or-ces i"LJ anc Xo (See
The horizontal reaction at A will not :;'::'g. ).11.33)
~roduce any bending on nember AC, thus t~e
values in Fig. All.79 are the true moments. M - - Z 2iS = - (-265500) = ..',.650
'0 - Z dS/I 57.1
The axial load in ~ember AC by statics
from F:g. AIl.BO equals Xo or a ~er.sio~ load of
22£ su E
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES All. 35
x .Z0s Y: - 8753OX15 . 75+15CCOOx15 . 76- 328000 (- 3 . 34 ) Example Problem 2. External Loads on Curved
. o -x 7056 :-lembers.
X o = 294. lb.
Fig. AIl.85 shows a frame which has ex-
y ternal loads applied to the eurved member as
F I D well as ~o the straight members. The fra~e
c~ ),Mo'll4650./
1 - ' -
l~. 76
supporting torees r~ve been assumed as acting
uniformly along the side members AD and CEo
x _ - - ;Xo=294J-X--+- The problem will be to determine the bending
i 14.24 moments at frame points ABCDE.
E
Fig. All. 83
1000 B 1000
Bending moment at F on member TED, \ .,.. \.15 I
Curved member ABC. total elastic weight of ~emjer and its elastic
mcments o! iner~~a abo~t ~es t~rJugh :~e
The ~lxed end moments on this curvec e Ias t Lc cent.er •
~ember due to the ex:ernal loads will be de-
termined by the elastic canter method. ~he TOTAL ~~STIC ~~IGHT
nr
': Z ds/I =-- = n x ~O
l.v
assumed static frame condition will be an arch
pinned at A and su?ported on rollers at 8. = 62.83
(See Fig. All.B6)
Distance y from line AC to elastic center of
Fig. AIl.S7 arch ABC equals 0.6366r = .6366 x 30 ': 19.1 i~.
shews the general 1000 =P B 1000 : P
Ix =0.2978r 3 / I =
2978 x 30 3/1.5 = 5350
!~\I
Shape of the static
moment.cu~re. :or
the fr~~e portion I' /
between the re- I 00:" 30~_ .- .- C _---,.<~ Os~ • 167300
AA, -r-~--
actions and the load
paints, the bending Fig. All. B6 ~,0Mo
-/-.---'~ - "0-
moment equation is
M=P(r-r cos a).
(2) I i
0'""'l -Os'1 =288001 19 • 1
For the beam (1' I \ I
portion between the (1) -- - - 'lIT....-.%-.
C
two 1000 lb. loadS, Fig. All. 87
the bending moment Static Moment Fig. All. 88
1s constant and Curve.
equals, Calculation of redunda.~t forces at elastic cer--
M=P(r - r cos 30 0 ) tar.
Calculation of 0s Values. Z 09 = -(28800 +' 67300) =
110 - - Z osvt 62.83 in.lb.
The ¢g values equal the area of t~e X/I
C~les. The moment curve in Fig. AII.S? has _ Z 0sY _ 167300x5.7+28800(-9.l) =
been broken down into three Darts labeled (1) Xo -~- ;:)350
(1') and (2). - 129.3 lb.
y = r(l- cos a -~) = 30(1- .867 - 0.5/2) = MA = Me = Mo + Xoy = - 3122 + 129.3 (- :9.:;
c c s in c .524-.50
= - 652 in. lb.
ro in.
Xament ~istribut1on Process
no:
):/sa
=- Prae
I
(1 _
cos a ) . e = 120 0, a• 30 0•
Having determined the fixed end ~cmen"'Cs,
distribution and carry-over tacters, ~~e =oment
¢., = 1000~30'X2.1 (1- .867) = 167300 distribution can now be carried o~t. Fig.
.0 AIl.89 shows the solution. The first cycle wi::
be explained. Starting at joint (A) the un-
Vertical distance y from line AC to centroid of balanced moment is -3000 - 652 = - 3652. The
.elSa equals, joint is balanced by dis~ributlng .~2 x 3652
= 1543 to ACj .364 x 1543 ': 1333 to AbC and
2r sin 8/2 = 2x30xO.S67 .212 x 3652 = 776 to AD. The carry-over moment
r » e 2.1 = 24.8 in. from C to A = .5(-1543) : -772; from C on
member eBA to A = - .452(-l332) ': 601; and trom
Fig. All.SS shows the 0s va1ues and their lo- A to D = .5 x 776 = 388. Now ~roceedl~g to
cation with respect to x and y ~es ~hrough joint 0 the unbalanced TIoment is -7500 + 388 =
the arch elastic center. -7112. ~~e jOint is balanced by distributIng
.75 x 7112 = 5340 to DE and the re~ainder 25
The elastic 8entar ~ethod reqUires the per :ent = 1772 to DA.
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES All. 37
B
__ over from C
_.-lS2. (_6S2.\ = car·,
Static Mom.
(/r : ~
u '" ...,'" ""ao ....'''' «>''''' -,<:>,.,.....
• 366
(1)
Curve
"I'
~ :;ii/ \~
( 1)
1::1"'''''$'''''-''''''''''''''<'«'''-
<"i'~: ";'1'7'71' 'I' ",
A .422.1-",;;=------:"::0'"01 C
,.212. -3000 i.t.M. I 2513
I I Static Mom •
• 302./
151 Curve
-l:~/
.~~/
Fig. A1l.90 -r/a wLi!. = 4500
A '<, C
?ig. All.90 shows the bencing moment tis = -90000 '" 75390 = - 14610
c~ves for ~2mbers ABC and AC as ~de up of Fig. All. 91
5 parts labeled: to 5.
Solving for redundant forces at elastic center.
Calculation of 0s val~es Nhic~ equal area
of each ~oment curve ji,1ded by r of member. 0s = -218290
Mo = z l: ds/I 92.83
=-2356
2~Sl = 25800
X =..usL= 167300 x 11.83+ 36800 x 6.13 +
0·'. = 167300 o Ix 12713
28800 (-2.971 - 14C:O (-12.97)= 181
0s , = 58~ x rt x 30/1.5 = 36800 12713 lb.
,,
20" 400~ I =6 1~400
]:::4 1=41 10
Thus at ~oint A on member ABC, -~
MABC = 584-2356+181x12.97 = 554 in. lb. Fig. 7 Fig. 8
which is the same as found by the moment distri-
bution process.
100 400 600 400
~_100
At point B on arch: - :-12 ~ r-I O Pin ': 4"
ME tram Fig. AIL90 = 584 + 4000 = 4584 1- 5" ..1 1·4 I I '" 6 1= 3 AI=l :"'-8"-
Mg from Fig. All.9l -2356 -181 x (30 - 12.37) I--- 16" ----'
I
t:.t= '
(4) Fig. 12 illustrates a continuous 3 span
1100 1100 w= 101b/in.
Wing bea~, carrying a uniform air load of 20
1f-8"-+-16"-----li "I
~
I • lb./ln. Deter.nlne the beam bending moments at
24"----1 a strut points A and 8. Take
Fig. 1 Fig. 2
IAB=171n., IAA I = 20 in., and E = 1.3x:l.O e ln~
'fo w=201b/. 200y" . . .
I::2 1=1 fll'=P ----{~
h+i·to16
In. Neglect e~fect of support deflection due to
strut axial defor.rnat:on.
I-- 16'.--1---- 12"-~ Fig. 4 ~50"~ 110,-"---100'+ 110'~50'1
Fig. 3 ;
100 f , I , • •• • .<W=.20.rb~in,'
....
,
12~ , 1~ 12 ......
t
.A '¥!
"",
hI 1_2
f- 14" 36"--~ 16" -j
Fig.S Fig. 12
I : I *i ;
(5) Figs. 13 to 15 are leaded structures :hat
suffer jOint translation. Jeter.nlne be~ding
Ii
j
CHAPTER Al2
The problem 15 to dete~lne the relation- distance !J. which causes end moments M~ and M":., .
ships between the displacements Of the end In deriving the slope deflection equations each
supports or a beam and the resulting end of the three beam deflections will be considered
moments on the beam. separately and the reSUlts are then added to
give the final equations. Fig. A12.2 shows Fig.
Fig. A12.1a shows a beam restrained at c repeated.
ends 1 and 2. It is assw~ed unloaded and at
constant cross-section or moment of inertia.
This bea~ is now displaced as shown in Fig. b, Fig. A12. 2 L
namely, that the ends are rotated through the
angles eJ, and e a plus a vertical displacement
dJ, and d lil of its ends rr-cm its original pos i ,
tion, which ?rOduCes a relative deflection 6
of its two ends. The angle 0 representir~ the
swing of the member equals 6/L.
A12.1
A12.2 SP ECIAL M E l' HOD S - DEFLECTION METHOD
S L 0 P'-"E--="'-'-"c:..::...:.:.:::...:.:.....:::.-"-''-''-''-''--- _
assumed also ~osltlve, as the algebraic solu- I Fig. A12~6 shows ~he ~oment diagr~ ~0r ~~
tion will deternlne the true sign of r.~. applied bending noment ~~ which rotates en~ (2)
thro~gh an angle 6 2 Nhen other ~nd (1) is ~!xed
~NO ~oment area theore~s will be used 1~ (See ?1g. A12.5). In a similar ~nner as
deriving the slope-deflection equations, desc~ibed befo~e,
namely (I) that the change in slope of the
beam elastic curve between two ~oints on the Sa = (-M'lL/2) + (-~~L/2) = :-~(~l1~"~.;+~,,;.~~~)"_L _ (5)
beam is equal in magnitude to the area of the E1 ~I 2E1
M/EI diagram between the two points, and (II)
the deflection of a paint (A) on the elastic ~ak1ng moments of M/EI diagram about (2),
curve away from a tangent to the elastic curve
at (B) is equal in magnitude to the first (l1~L)!: + (l1~L) 2L = 0
~oment about A of the MjEI diagram between (A) \2E1 3 2E1::3
and (B) acti~g as a load.
or, - - - - - - - - - - (6)
In Fig. A12.2 since Sa 0, then 6~ will =
equal the area of the M curves divided by Sl. whence,
) = (11'~+i1~)L
e =
1.
(11\L/2) +
EI
(11~L/2
E1 2E1
(1)
e. = M~L
2£1 an
d Mil - 2ElS"
~ - L
(7\.
The deflection at end (1) away from a Then from equation (6)
tangent at (2) equals zero, thus we take
moments or the moment diagram about (1) and
equate to zero. .
M" --
4E16 a
L
- - (S)
(lille)
2E1 3
h (!ill,,) 2L
+
2E1 3
Fig. A12.8 shows the third ~art of the
beam deflection, namely, support (2) is jef:ec~sd
downward a dista~ce 6 assuming both ends fixed.
whence -tl'llI =!1'J2 -- (2)
or I1' 4E1el.
,- L - - - - - - - (3)
,
Fig. A12.9
M'''.,h
, 2 M'a·~t
Then !rom equation (2)
11'. =
-
2EI6j
L - - - - - - - - - - - - - (4)
M'{' c:==:====-==
.............-- . , '"
M
Fig. A12. 10
The change in slope of the elastic curve
between the enes equal zero, thus Jy the moment
L
Fig.AI2.5 area theorem the area of M/EI diagram equals
zero.
"r (, ~_"'-=- __
+ M~ )L
I---.i; L ----r-- Hence,
2El
o
Whence M:~ =-:1':- - - - - - - - (9)
Fig. AI2. 6
Frem the deflection theorem,
M'~ (10)
Mi,': ""T e:t====="
Subt. (9) in (10)
Fig. A12. 7
= - 6EL\. (11 )
~-
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A12.3
+ 6£1:.6- (lea)
= --'i-'T'-
, 41 (12)
(lSb)
The ccmo tned effect of deflection Or , o,
an~ ~ can now be obtained by adding the separ- A12. 3 Hinged End. Slope Deflection Equation.
3.te results.
Consider that end (2) of beam 1-2 is free-
ly supported or in other wards hinged. This
means that Ma_l. is zero. Thus equation (18)
can be equated to zero.
Let K ~ ~C and 0 ~ ~/L
(13)
whence
ana ;1a_ .. = -2K (28,,+9 .. -30) (l4)
2K~ ~ -K81. + 3K0' - .SMFa
These end ~oments due to the distortion
of the supports must be added to the moments Substituting this value in equation (17)
due to ~ny applied loads on the beam when
considered ~s :lxed ended. Let these fixed
ecd moments be l~beled MF l and MF". Then A12.4 Example Problems.
equations (13) and (14) can be written includ-
ing these mcments. Problem 1
(15) Fig. Al2.11 shows a ~NO span continuous
(16) beam fixed at ends (1) and (3) and carrying
lateral loads as shown. The bending ~oments at
points lJ 2 and 3 will be dete~ined. Relative
MCCI?I~D SIGN CO~~ION
values of I are shown for each member.
Squatlons (15) and (16) have been 100 lb.
jeveloped uSing the conventional signs for ~ 12" ~ w = 24 lb/In,
~ 1 -.:21Ji!i
bendi~g ~cments. In general there are ad-
V2~ntages of uSing a statical Sign convention
2.3 '.'las ueed in the moment d t s t r tbut ton method
1 § L ; 24"
I~24
i L • 24"
1=48
in :hapter All. Therefore, the following
sign :onvention will be used in this chapter: - Fig. A12.11
Ml._,. = Z8,. - 300 - - - - - - - - - - - (a) Fig. A12.12 shows a loaded 3 span beam
fixed at points (1) and (4). ~hls ~~xity at
M.- = 2K (28 + 8:1.) +i'IF '1
1 a (1) and (4) causes 81 and 8~ to be zero.
M'1 _ .. = 2xl (26. + 0) + 300 Fixed end moments:
l'l~, = 48'1 + 300 ---- - ---- -- It) MFl,,_g =- PL/8 = 100 x 24/3 = - 300,
Beam 2-3. 6. = 0, because ot fixity at (3) . !iF -'1-1 = 300
11 11-= 2K (28 a + 6,,) +i'IF •
" MF-'1_::I = (100x8xlS Il / 24 ) + (lOOx16xa 2 / 2 4- 2 )
=-534
M._ a = 2x2 (26. + 0) -1152
M._ a = 86 - - - - - - - - - - - (e)
a-1152
Ml;"
'''-a
= 534
l'l_. = 2.'{ (28" + 8 +MF" a )
l'lF._. = 400
11 3 = 2x2 (0+9,.) + 1152
_ .
At joint (2) l:l'l = 0 for static equt r rbr-tum, !11._11 = 2 X l( 0 + 8:a) - 300 = 28 2 - 300 (a)
- 534 - - - - - - - - - - - - - - (e)
SUbstituting trom (b) and (e)
4Ba: + 300 + S9 a -1152 = 0
+ 534 - - - - - - - - I co )
whence 8 11 71
::
!13_. =2 x 1(28" + 0) -400 :: 4e 3 - 400- (e)
With 8 11 known the final end ~oments can be
round by substituting in equations (a), (b), M~_" = 2x1(0+8,,) +400 = 28" +400
(0) and (d) as follows: _
Joint-Moment eQul1ibril~ equations: -
!11_. =2x71-300 =-158 in.lb.
JOINT (2)
M.._.. :: 4 x 71 + 300 = 584
whence
!1._3 :: a x 71 - 1152 = - 584
48 g + 3 0 0 + 8 8 a+48 3 - 5 3 4 = 0 or
M"_II = 4 x 71 + 1152 = 1436 - - - (6)
Changing the resulting ~ament signs to the
conventional sign practice, gives JOINT (3)
80:]1'"48 a+534+48.,-400 =0 or
Solving equations (g) and (h) for Sa and 8~ or free to rotate we use equation (19) in writ-
gives, ing equations for Ma _ 1. and M3 _ 4 •
!1Fa_ = 250
1. A12.5 Loaded Continuous Beam with
Yielding Supports.
i1...
- ,~ =!'!F ~_a = 0 (No load on span 2-3)
The
~ovable contr81 surfac~s of an air~lar.e,
the elevator, rUdder ~G aileron aTe
11~
. 3-4 = FL/8 = 200 x 20/8 = - 500, na~ely
attached at several ro ints to the s tact i icer ,
!14_~ = 500 fin and wine r-esr-ec t tve Ly. T:"_ese suppor t t ng
surfaces are usua:ly ca~tl1ever strUctU:-8S and
Slope-def1ection equa~ions: - ~hus t~e suppcrting ~olnts ~o" ~~= 7.cvab~e ccn-
~rJl sur~aces suffer a ~lsp1ace~ent :hus ~"o
Since suppor:s at (1) ane {~; ~re ~i~~ed
~o~:~~ a cJ~tin'..:.ous ~e~~ cr. Jlelc~~s cr de~:ec:ed
-"."',
A12.6 SPECIAL METHODS - SLOPE DEFLECTION METHOD
14!1IilllilTIVIJll!,j!!
2 t3 , f \
'r Calculation of ~ val~es.
~10'~40"- ~10't-
*--1.
" I
40" -
I
I
40"--+-- 40"
'
Fig. A12. 14
\'
tube
wi~h
Span 2-3. The settlement of support (2)
respect to SUDDor"': (3) :: (0.5 _ .':'875) =
.3125 inches = ~
~/16'" ---------. ¢ =- 6/L =- ,3125/40 =- 0,007812 cad,
deflected sUPPJrts.
Fig. A12.15 (Since (2) noves co~terclockN1se with ~espect
to (3) the Sign of 0" is nega t tve ,
Fig. A12.14 is representative of an ele-
vator beam attached to detlec~ing stabilizer Span 3-4. ~ = .1875 inches
str~cture at the five reaction points as
indicated in the figure. The elevator beam ¢ =-/l/L =- .1875/40 =- .004688 cad,
15 a round aluminum tube 1-1/4 - .049 in size.
The moment or inertia (I) of this tube equals Since the be~, external loadly.g and supper:
0.03339 and the modulus of elasticity (S) or settlement 1s sJillmetrical about support (4), the
the material equals 10 million psi. The air slope of the beam elastic cur-ve at (.;) is nor t ,
load on the elevator beam is variable as zontal or zero and theref0re 9.. = O. T~us cnly
indicated in Fig. Al2.l4. one-half of the structure need be considered in
the solution. Due to the fact :hat (2) is a
Fig. A12.15 shows the Shape of the de- Simple support with a cantilever overhang, the
flecting supporting structure, which means moment M~_l is statically dete~l~ate and equa~s
that supports (2) and (3) move downward 'thr-ough 5 x 3 x 3.33 :: 50 m.rc, Then for stat ;c equili-
the distances indicated on the figure. The brium of joint (2), I1 11- 3 must equal - 50.
problem will be to determine the bending
moment at the supports under the combined Substitution in slope deflection equations
action or transverse loading and settling of (17) and (18): -
supports.
80LlITION: -
M.a_3 =2K(28 11 + 8 3 - 30") + I1F 11-3
- 50 = 2X8347 [Z9.a +93 -3(,007812)J -426
Calculation of fixed end moments: -
or = 333886 11 +1669483 +15 =0 - - - - - - - (1)
For a trapeqoidal loading as shown in
Fig. a, the fixed end ~omen~s are, ~_a = 2K(29 3 + 911 - 30') + :11'\_11 = 2 x 3347
[293 +9'i1 -3(-.007812)J +440
_ L"
M:i._ - 60 (5u + zv)
Whence ~_II = 333888 3 + 166948::1 ... 831 - - - - (2)
_ L" (ao - '"\'
Ma-:i. - 60 (5u + 3v) ,'"'... - Uj
SUbstituting in these
equations tor the loading Fig. a Whence !13_.... = 333888" - 259 -----(3)
values as shown in Fig.
A12.4, M..._" = 2K(28... - 8" - 30')
40"
M'F.a_3 :: W (5x3+ 1) =-426 in.1J. ~_3 =2 x 8347 [0 + 9" - 3(-.004688)J + 507
!1 ..._" = 166940 + 742
MF3 _ 11
=:gil (5x3+1.5) = 440
whence 3 - - _ (4)
converting these signs to the convent1onal I11o_.lII =2K (28:1, + e.) + MF:1,-.
signs would give Mil =-50, M" =--582 and
11 4 :::: -580. M1o_.lII =2 xI (26 1 - 91 ) - 480
L: 24 L: 24 which gives,
1 0 0 - I = 40 I =40 lot-- 100
K = 1. 67 K = 1. 67 281. - 480 + 6.676 1 + 3.338" + 300 =0
whence 8.678:\, + 3.338" -180 = 0 - - (e)
JOINT (3 i.
which gf.vee ,
Fig. A12.16
6.676;, + 3.3361. - 300 + 26" + 334 = a
?lg. A:Z.16 shows a closed rectangular
fr~e subjected to loadings on its four s:des
whence 8.678" +3.336:\, +14 = 0 - - - - - - - (f)
wh:c~ hold the frame in equil~brium. The
Solving equations (e) and (f) for 8" and
bending ~oments at the frame joints will be 61. gives,
deuerrmned ,
SOLUTION: -
81. = Z6.3, 9;, =-13.94
A12.8 SPECIAL METHODS - SLOPE DEFLECTION METHOD
Subt. in equation (a) Due to the flXity a~ j~i~ts (1), (3) ~nd
(5), 81.-a = 9:5-" = 9&_., = O. "het-e r or-e , the
Ml.-e =2x25.3-?80 =-428 in.D. unknowns are e a -~, e.. _~, g e _ e :::. nd 6.
233 M e-!l
Fig.A12.17 f~
T 4 T i 6
I
Li,J I
i 15
194
.I'_a,
'-'
....1,. ,J.,--H~
Me_t'l
M~_...
A12.7 Frames with Joint Displacements.
Fig. A12.19
In the previous example problems. the
conditions were such that only joint ~otations Treating each ~ember sepa~tely. we ta~e
took place, or it any transverse joint dis- moments about the upper end ~.d equate to zero
~lacement took place as in the ex~ple problem and then solve for the horizontal reaction at
of Art. Al2.5, these displacements were known, the bottom end. The results are,
or in other words the value of ¢ in the slope-
deflection was known.
Exa~ple Problem 1
SOLUTEN: -
Fig. A12.18 The relat:ve ~8rnent o~ :~ertia of each
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A12.9
The angles .0
M,,_ .. = Z8, - 0.60 - - - - - - - --- (e)
Fig. A12.20
are proportional to 11_, = Zxl(Z8, +0-3xO.10) +0
the ~/L for each
member where ~ is 11,~ = 48, - 0.60 - ---------- (r)
same for member-s
(1-2) and (3-4) and Equilibrium equations: -
zero for member (2-4).
(See Fig. A12.10). JOIN]' (Z).
Hence, Substituting: -
" .. = 10
)(.1,,_ 1 = 0.100 whence, 9.33361i1-0.2667~+3.3338,-40.32
=0 - - - - - - - - - - - - - - (g)
o JOIN]' (4). I1.._ a + I1,_;s =0
.0:a_-4.= 12 = 0
FL"'CZD 2ND :-!OMEN'I'S: - 6.6676, + 3.3338 11 + 144 + 48", - 0.60 =0
tlember 1-2. whence, 10.6676, + 3.3338.. - 0.60 + 144 =0 - (h)
I1F
.-. =-(48xo"x9)/lS" =-69.12 tnv Lb • Writing the bent equation (See Eq. 20).
- - (1)
MF..-a = 144
I1e_ +!"',",-_., + lOR., =0 Ing the symme'tr t ca.L cf s tr-Lbut.ed load as shcwn .
w: 12 lb/in.
..
Substituting values of
equations, gives,
M~_~ and M~_~ in above I 3
10"
t
H, =- o.se. + 0.120 - - - - - - - - - (k)
3 4
SUbstituting values of H 1. and H~ in equation
(1), gives,
M..~ = ,; (-6.22) - 0.6 x 196.9 = -142.9 MFa_ = 400, MF .._., = 400, MF;,_.. = - 4-00
l ' /t/i E
3"1 3
\C/.
0"-0- =-
= 2K(2S",
28. - 0.150 - - - - - -
2'
3'--_ = 2xO.5(O-8. -3x .050') +0
i M",_ =- e:a - 0.150 - - - - (h)
I
I JOINT (2). Equl1i~rl~~ equation.
,2
_
3
'
10"
.--L-
240 20,,-----f
=2K(28
MJ _
M~_a
a
= 2x
J + S. -30) +MF
.892(0+8. -3xO.10+400
,-. 1 1
,,-,- H 1
Fig. A12.26
M \-.11
:1._ J =2x .892(-28. +O+3xO.:0) +400 ane in Fig. A12.26, taking ~oments about (3),
I1 .._ J
=-3.5888" +0.53520+400 - - - (f) !'! a-" + 11 3-a + 240 x 20 - 20 x 12 x 10 -lOH .\1-3 =0
"+M,, 2+24:00
wr.ence, :f a-~ = 10
A12.12 SPECIAL METHODS - SLOPE DEFLECTION METHOD
Shear equa t 1or- , H .\1_1. = H 2_" ?:~. Al2.27 ShC~2 t~2 ~~2~1~:r:~ :e~~:~;
cement c iacr-em, ~::'rs: ::'.::"'2.','.7', tn oar-as i.'cr 2S.,:::j
whence, member of t~e fr~~e ant ther: addec :0 fCrT, :~e
cO~Dosi:e d:.a~~~. ~lJt:ec. :ir~c:ly :r: :~e
f::"'2.~e.stears an::'. ~~~2.l l::acs ~cllcw as
:~e 2.
~atter of statics. ~lth t~e 2cment ~i2.~r~
Y~ovm the fr~s det1sc~sd ~~~pe ~5 ~ead~~y
Substituting values for the er:d Iome~ts: - c3.1c:11a:ed.
Sa + 0.150 + Z8!i + 0.150 :::; 3.56882 - 0.53520 A12.9 Comments on Stope-Deflection Method.
20 10
The ex~~pls problems solut~ons i4c::a~e
- 400 + 1.7848 2 - 0.53520 + 4,00 + 2~OO that the met.hoc c: s Iope-def Lect t cn nae t.vo ad-
10 vantnges , nameLy, (1) i t !'SCc:.C8S t::s number- 0:'
equatior.s to )8 solved s:~~1:2~e:uslY a~G (2)
Whence, it ,resents equations trzt are ea5:11 a~c
rap:cly ~or.nulatec..
0.38528a - 0.1220 + 240 =0 - - - - (j)
Thus for structures ~lth a hi~~ ~e~ree o~
Solving equations i and j for S.\I and [2,?;i vee red~ncancYJ the slo~e-deflectio~ !1ethod should
be considered as 80s51j1y the jest !1e~hod o~
[2 =2805, ea = 265.8 sclution. The solvin~ of the equations in :~is
method are readily programmed ~or solution jy
Substit~ting these values in equations (a) to high speed computing ~achlnery.
(h), the end moments are obtained as follows: -
A12. 10 Problems.
M1.-a = 6.. +0.150 = 265.8+0.l5x:2505
= 666.5 Ins Ib , (1) Determ,ne the bend~"g Go~ents at suoport
points A, a, C, 0, for t~e con~lnuous te~~
MlII-1. = Z9 a + 0.150 = 531.6 + 420.7 = 952 shown in Fi~. A12.28.
M lII-=3.5689.. -0.5352.0-400
3 100 100 150 lb.
= 3.568 x26S.S - 0.5352 x2S05 -400
=- 952
~7'!..l.6"...L7" +-12"
+ + I l I
w,., 101b/in,
•
M::I_... = -1.7846.11 + 0.53520 -400:::; - 628 (2) Same as problem (1) but consi::'.er sup~crt A
as ~reelY supported instead of fixed.
M._. = 952
(3 ) Determine the bendinz ~oment d~agram ~or
M ..._ 6 = - 952 the variOUS loaded structures :n r:g.Al2.2S.
w,., 12 lb/in.
M:~_ ... =-686.5 200 lb. I : .: .. I ;
C~D
~8"'" I I
A
L:Il 16
I =32
~~ 12"
300 lb.
100~=30
I
I
.60
= 36
L=30
1·60
, '
100 lb.
oJ
1
Pin,e,t-
15 i
, I
...1.,TTm
A
M IS
I
m'\'l..L.-
B
1000 1000 lb.
,
w,., 20 Ib/in.
952
I ~5"~5"
C 1 _ 2 D
"
L,., 16
686.S • 1
,., 20 20" 1=3
, A
-L ~
Fig. A12.27
A Fig. A12.29 ,--20"---......J
CHAPTER A13
BENDING STRESSES
~
y
however in this expression. the term Oc 1s not
z-I-t.:z zero, hence the term fc
yda must equal zero and
x LJ
Beam this can only be true i ! the neutral axis coin-
Section cides with the centroidal axis of the beam
Fig. d cress-section.
A13.1
A13.2 BEAM BENDING STRESSES
The neutral axis does not pass through the Thus, Y~ • (y - x ta.~ 0) cos 0
beam. secticn centroid when the beam is nonhomo- • yeas 0 - x sin 0 - - - - - - (2 )
geneous that is, the modulus of elasticity is not
constant over the beam. section ~~d also when Then Sq , (1) bec ames
Hook's Law does not apply or where the stress-
strain relationship 1s non-linear. These beam o :::; k (y cos 0 - X s tn 0) -
conditions are described later in this Cr~pter.
This equation contains three unknowns, v,
A13.2 Equations for Bending stress, Homogeneous Beams, k, and 0. For solution, two additional equa-
Stresses Below Proportiooal Limit stress. tions are furnished by conditions of equili-
In the following derivations, it will be brium namely, that the sum ot the ~oments of the
assumed that the plane of the external loads external forces that lie on one side of the
contain the flexural axis of the beam and hence, section ABCD about each of the rectangular axes
the beam is not subjected to torsional forces X-X and y-y ~ust be equal and opposite, respect-
whiCh, it present, would produce bending stress- ively, tc the sum of the moments of the internal
es it tree warping of the beam sect~ons was re- stresses on the section about the same axes.
strained, as occurs at points of support. The Let 11 represent the bending ~oment in the
questions of flexural axes and torsional etfects plane of the loads; then the moment about axis
are taken up in later chapters. X-X and Y-Y is Mx : :; M cos g and My • M sin Q.
m The moment of the stresses on the beam section
about axis X-X is / a da y. Hence, taking
P moments about axis x-x, we obtain for equil-
Plane ibrium,
of loads c
J1 cos Q = / c da y
- / k (cos 0 y"cta - sin 0 xyda)
x-----'b-tt-.::::.:y---'- X
= k cos 0 / y'da - k sin 0 / xyda (4)
m
In similar manner, taking moments about
the Y-1 axis
Fig. A13.1 y Fig. A13. 2
m 11 sin Q = / 0 de x
whence
Fig. A13.l represents a cross-section of a
straight cantilever beam with a constant cross- M sin Q = - k sin ¢/ x~da + k cos ¢ / xyda(4a)
section, subjected to external loads which lie
in a plane making an angle Q with axis 1_1 Al3.3 Method 1. Stresses for Moments About the
through the centroid O. To simplify the figure, Principal Axes.
the flexural axis has been assumed to coincide In equation (4), the term. / y~da is the
with the centroidal axiS, which in general is moment of inertia of the cross-sectional area
not true. about axis X-X, which we will denote by Ix, and
Let NN represent the neutral axis under the the term / xyda represents the product at in-
given loading, and let 0 be the angle between ertia about axes X-X and Y-Y. ~e know, however,
the neutral axis and the axis X-X. The problem that the product or inertia with respect to the
is to tind the direction of the neutral axis and principal axes 1s zero. Therefore, if we se-
the bending stress 0 at any point on the section. lect XX and IT in such a way as to make them
In the fundamental beam theory, it is as- coincide with the principal axes, we can write
sumed that the unit stress varies directly as equation (4):
the distance tram the neutral axiS, within the
proportional limit of the ~terlal. Thus, Fig. M cos Q = k cos 0 Ix - - - - (5)
Al3.2 illustrates how the stress varies along a p
line such as ~ perpendicular to the neutral
axis }I-N. In like manner, from equation (4a)
Let c represent unit bending stress at any
point a distance Yn from the neutral axis. Then I1 sin g k sm 0 Iy (6 )
the stress 0 on da is P
(7 )
A13.5 Method 3. Stresses from Moments, Section Prop-
erties and Distances Referred to any Pair of
The minus Signs have been placed before each Rectangular Axes through the Centroid of the
term in order to give a negative value for crb section.
when we have a positive bending moment, or Mxp The fiber stresseS can be found without
is the moment of a couple acting about Ap-X p' resort to pr tnc i pa.L axes or to the neutral
positive when it produces compression in the axis.
upper right hand quadrant. Myp is the moment
Equation (4) can be written:
about the Yp-Y p axiS, and is also positive when
it produces compression in the upper right hand MX = k cos ¢ Ix - k sin ¢ I XY ----(11)
quadrant.
where Ix :; / ya da and Ixy :; / xyda , and Mx =
BENDING STRESS EQUATION FOR SYMMETRICAL M cos 9.
BEAM SECTIONS In like manner,
Since symmetrical axes are principal axes 11y : - k sin ~ ly + k cos ~ I XY - - - - - (12 I
(tei~ / xyda ~ 0), the bending stress equation
tor bending about the symmetrical XX and YY axes SolVing equations (11) and (12) for sin ¢ and
1s obviously, cos 0 and substituting their values in equation
(3), we obtain the following expreSSion for the
cr = _ MxY _ Myx - - - - - - - - - - - - - (7a) fiber stress crb: -
b Ix Iy
(11yl x - 11Xl xy) x _ (11xl y - 11yl xy l Y _ (13)
A13.4 Method 2. Stresses by use of Neutral Axis for
Given Plane of Loading.
IXl y - li:r Ixl y - I~y
The direction of the neutral axis NN, mea- For Simplification, let
sured from the XX p principal axis is given by
dividing equation (61 by (51. K. : lxy/(Ixly - I\y)
K, ~
Ix/(IXl y - I\y)
It frequently happens that t~e plane of t~e Al3.8 Illustrative Problems. Example Problem l.
bending moment coincides with e~ther the X-X or Fig. A13.3 shows
the Y-Y axiS, thus making either Mx or ~y equal a unsvnme
J~~~
t r i ca I one f, P y " 6000#
to zero. In this case, equation (15) can be cell box beam wi th i ;1 1 .
simplified. For example, it My = 0 four corner r lange / ,/ IPx:1600~i
member's a, o , c and d. I - ,1/
~' I I
I /
I,. I
1/1
J'
'I I, I - - - - - - - o - t i !
axial stress in the N ,1/ 1/
and It l1x = 0 / II II
._-
Ix ____ - - - - - - - - - - - - (17)
four corner members
due to the loads ?x
and Py acting 50"
1/ II
/.10",'
L // "
,;/ ,
il/'
tan 0
I xy from the section
abed.
a" II T .5°" ," ! I
,yf'l
A13.6 Advantages and Disadvantages of the Three In this exa.. n ple b I: JIT
I I
Methods. solution the s nee t 12" I' at!
Method 2 (bending about the neutral axis
tor a given plane of loading) no doubt gives a
ccnnec t.i ng the corner
members w::.ll be con -
xtL d ( -LX'
80' ,#1/.'-------,-16"''''- -"-:I:, .40"
better picture of the true action of the beam atder-ed inarrect ive in' ' 1
relative to its bending as a Whole. The point bending. The stresses Fig. Al3. 3
at maximum fiber stress is easily determined by will be determined by
placing a scale perpendicular to the neutral each ot the three me-
axis and moving it along the neutral axis to thods as presented in this chapter.
tind the point on the beam section farthese away
tram the neutral axis. In airplane design, SOLUTION
there are many design conditions, which cr~ge
the direction of the plane of loading, thus, The first step common to all three meth~ds
several neutral axes must be computed for each is the calculation of the moments of inertia
beam eectlon, which is a disadvantage as com- about the centroidal X and Y axes. Table A13.1
pared to the other two methods. gives the detailed calculations. The proper-
In determining the shears and moments on ties are first calculated about the reference
airplane structures, it is common practice to axes x'x r and ylyt and then transferred to the
resolve air and landing forces parallel to the parallel centroidal axes.
airplane XYZ axes and these results can be used
directly in method 3, whereas method 1 rsquires TABLE A13.l
a further resolution with respect to the prin- Area
cipal axes. Methods 1 and 3 are more Widely Mem. nA" y' x' Ay' Ax' Ay,2 Ax,2 I Ax'y'
used than method 2. 256.0 -192
a 1.0 12 -16 12 -16 144.0
Since bending moments about one prinCipal
axis produces no bending about the other prin-
b 0.5 8 0 4 00 32.0 0 I 0
c 0.8 0 -16 0 -12.8 0 204.8 0
cipal axiS, the principal axes are convenient d 0.4 0 0 0 0 0 0 0
axes to use when calculating internal shear flow Totallll I 2.7 16 .28.8 176.0 ! 460.8 -192
distribution.
A13.7 De!lec:tions.
Calculation of Centroid of Section: -
The deflection can be :ound by using the ZAx'
beam section properties about the neutral axis x =
lA = -2:7
28.8 ~
- 10.567"
tor the given plane of loading and the bending
moment resolved in a plane no~l to the neutral lAy' 12..- = 5.926'"
axis. The deflection can also be :ound by re- Y= lA = 2,7
solving the bending ~oment into the two prin-
Cipal planes and then using the properties about Ix ::: 176 2.7 x 5.926~ = 81.18
the principal axes. The resultant deflection is
the vector sum at the deflections in the direc- I y = 460.8 - 2.7 x 10.667~ = 153.58
tion ot the principal planes.
I XY = - 192 - (2.7 x - 10.667 x 5.926) = - 21.33
- \,,:~'.-->,~. _;_-, -/::.--~_~)_:34· ~~~::"-~-""-"_!'><:';~~::i.~_':',,~" ~"~::",";:-"1~',;-,; :'.,~,: ,~.: c~. :.•._I;,~.,,;·=.--c:· ~"'":: :)r"'~~ .:':~.~-J'~.~T;.-:"/:~~::~'1:.'~'i:~i,.'!._
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A13.5
Solution by Method 1 (bending about principal axes) (Positive moments produce compression in upper
The angle 0 between the x axis and the right band quadrant I.
principal axes is given by the equation,
Calculation at Stresses
2 I xy 2 (-21.331 - 42.66
tan 2 0 = ,,------'''i_ = 153.58 - 81.18· 72.40 = _ Mxp Yp Myp xp
I y - Ix
"D - - - . - - -
"lCP ryp
= - .589
Stringer a
2 .0 = 30° - 30' , 0_15°=15'
Xp =- 6.74, Yp = 4.45
I x p = Ix cos~ 0 + Iy 5in~ .0 - 2 I xy sin ¢ cos 0 hence
268700 x 4.4.5 - 156200 (- 6.74)
= 81.18 x .9648 a + 153.58 x .2636 a - (- 21.33 x °b = - 75.38 159.34
! f-6.7·l_~ a I 9.75,,--..j
Xp 'II -
- 268700 x - 7.12
3.58, Yp 'II _ 7.12
- 156200 (- 3.58)
I /
Xp -..i.,. __
"10'
L'D.667" b
O'b 'II 75.38 159.34
~~ = 6000 x 50 : 300,000w~
My = - 1600 x 50 = - 80,OOOw~
Let a = angle be~Neen neutral axis NN and
the Xp axlS~
These ~oments will be resolved into bending
moments about the x p and yp principal axes. IXp tan Q - 75.38 (- 0.5816)
tan a • - ~'--,~-
159.34 - .275
Mx P • 300,000 x cos 15° - IS' - 80000 x
hence a = 15° - 24' (see Fig. A13.3b).
sin ISO - IS r = 289700 - 21000 = 268700 w• Since the angle be~Neen the X axis and the
neutral axis is only 9' , we can say
i1y =- 300000 x sin 15° - 15' 80000 x IN = Ix = 81.18.
p ResolVing the external bending moments
normal to neutral axis, we obtain
cos ISO 15' = - 79000 - 77200 = _ 156200 wi
A13.6 BEAM BENDffiG STRESSES
Stringer a Stringer a
x=-5.333", y = 6.074"
Yn = 6.074 + 5.33 x .0025 = 6.087"
300 200 x 6.087 _ 22500i/ln. 5I 0b - - [.00674 (- 80000) + .00177 x 300000J x -
81.15
Stringer d
y x e 10.667, Y =- 5.926
Q' : 14°581
~::z 15015' 0b = 8 x 10.667 - (3697 X - 5.926) =
Q:30011'
plane of .I:0ading
~ 85 + 21900 = 21985
X 9' 52' N
-15°15' NOTE: The stresses Ob by the three methods
were calculated by 10" slide rUle. ~ence the
"p small discrepancy betNeen the results for tte
three methods.
Yp Fig. A13.3b
Y
Error in Stresses Due to Assumption that Section
Bends About X and Y Centroidal Axes Due to Mx
Solution by Method 3 (Method Using Properties and My.
About X and Y Axes)
(Equation 7a)
Mx =300000, My = - 80000
- 21.33 Stringer b
12016 = - .00177
y z 2.07, x = 10.567
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A13.7
I
14260*
181.17 437
I I I I: II I'"\ ,I
11 I 1 I I I I
'I
11 II 11 7150 - 5700 12850~/sq. in.
I II 11761l1]1 I 1 ' I 11 "
I I I: II , I I I II ,1 11 Stress on Stringer 9:
"1
I! i II I I 1 I I II [I II
I! I' I I I I I I II I' II 50"
I; I: II I I I I I II II II Yp = 9.04", xp • 14.24"
',J..I II II! I I i 11 II II
III r- 4-, I1 I I I ~ I 1I II II
! III
1
, I-
II
-+-r -J.1r..-l-
1- Ir+- --+ i-I -r-t r- T1
II I I : I l i t I ,I II "b -- (700000 x ,).04) - 140000 x 14.24
111 181.1/ 437
III: II I I a
I'I[ II 3 4
;11 2 4560 ::. - 30340t/sq. in.
= - 34,900 +
I
X-.-_I Stress on Stringer 12:
B
tion, and the location of these resultant loads Let Q equal the angle between the plane of
is 50 from section ABC (Fig. A13.4). the resultant moment and the Yp axis.
;he bending stress intensity at the cen-
troids or stringers number (1), (9) end (12) Then a = 9' + 0 = 3° 3' + 8° - 16' 11<;1 - 19'
will be calculated uSing all t~llee methods.
From equation (8), the angle be~Neen the Xp axis
Solution by Method 1 (Bending about Principal Axes) and the neutral axis ~ Q, and
The bending moment at section ABC about the
X and Y axes is, I xp tan 9 181.17 x (- .200)
-tan a
I yp 437 =
Al3.S BEAM BENDING STRESSES
Y YO - 36.41
I;;JXY) = 186.46 x 431.7 - 36.41;;J
Plane of
Resultant
~',.
gl-!
f
Moment- I "
= - .00046
\
\
\ I '
36.41 ; J
= 431.7 = .C0545
793000
= 186.46 .0'02355
793000
Stress on Stringer 1:
Fig. Al3.5
Y1 = 4.39", Xl = - 1""'.41"
~ .083. Hence, a :;; 4 G - 45'
MN = 714060 x sin 83 G - 26' = 709350 in. lb. - [.00545 x 713000 - (- .00046 x - 3EOOO)J 4.39
Stringer 1: ab
-TAr
.-
I
Yl = 4 + .39 = 4.39 r-TB 3"
2.64
-----, -L
I
Xl = - 33.15 + 15.74 - - 17.41
2" --t
...<--cB·2;64 1 3"
a b = - MxY CA' l
Ix
=- 713000 x 4.39 1- 38000 x - 17.41) = Fig. A13.8 Fig. A13.9
ab 186.46 431.7 stress intensity at bottom edge of portion A is
- 18330#:!i n :l .
ah 12 . 25 / 3 ) = 0.75 Ob.
Y. a 6.89, Xg = 15.39
= - 713000 x 6.89 (- 38000 x 15.39)
ah 186.46 431. 7 The distance from the neutral axis to the
centroid of the trapizoidal stress loading on
- 25000#! l n :l . portion A is 2.64 inches. In like manner,
Example Problem 3.
TB = CB' = (0 +2 Ob) 3 x 0.25 = 0.375Ob
The previous example problems were solved
by substituting in the bending stress ~quatlons.
The student should solve bending strs's prob- The arm to TB is 0.667 x 3 = 2 inches.
lems by equating the internal r-es ts t Lig moment
at a beam section to the external bel.ding mom- For equilibrium of the beam free body, take
ent at the same section. To 111ustrlte Fig. moments about point (0) and equate to zero.
A13.6 shows a simply supported 10adfd beam. The
shear and bending moment diagram for the given ZMo = - 500 x 48 + 2.54 x TA + 2.64 CA' + 2 TB +
Fig. A13.7
°b = 11-I-e where 11 = 3"
- 24000
e = Moment
I = Neutral ofaxis.
inertia about
Calculat10n of r ,
For Portions A and A'
-
12*/in.
Portion Area Average Total y : arm Resisting
' 1·1' I I II A Stress Load to N. A. moment
i~l~
I
5#/in. See in
, (2) (4) x (5)
Fig. terms x (3)
---I
n
I--- 100"---1. 30"-;:-'00" A13.13 of c b
-11-r-
I
Fig. A13. 10
J
RA
1
He i
-4-
-th-2"
1/2 _-<=
__
Jl-li rI I
2
1.045
1. 00
-.337crb
-.757
-.353crb
-.757
1. 39
2.35
- .490 O'b
~1. 780
I 'I I I 3 - 3' 1.00 -.837 ·.837 2.62 -2.190
(~t~I 2.9;..
6 1. 50 . 821crb 1. 2370b -2.57 -3.180 Cb
11 44
t=-3 Totals 6.19 O. OOOOb -8.82 O"b
~
1
Fi•. AI3.11 -.--
This method of solution involves ~ore cal- For equil18rium, the total co~pressive
culations than that required in SUbstituting in stresses on the cross-section of the beam ~ust
the bending stress fo~ula, however, the student be equal to the total t ens i Le stresses, or l.H
should obtain a better understanding of the in- must equal zero. Col~ 4 of Table A gives the
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A13.11
R
total load on each portion of the cross-section 1 1
T~
and the total of this column is zero. -- 21"- --..j 2"1-
T
The uending stress on the lower fiber of
,:. r
4.1
the cross-section 1s directly proportional to
3
the distance from the neutral axis or, ab1 ower = I Maple 3.33"
LI1j I
11.,l-NI·A.
hence
SOLUTION USING BENDING STRESS FOR:-mA.
- 36670 x 3.09
I N•A• = 27.2 in." 1Fig. Al3.14
rr I
1"
! ]" I
I
~teel
Fig. A13. 15
2.73"
the stress in the spruce sect:on. Since the ~e alloy. ~he original area of 0.1 sq. in. 9ach
intorcing strips are maple, the stress at the have been ~ultiplied by these st1f~ness ratio
top edge of these maple strips would be 1.23 values.
times (- 3900) = - 4880 pSi.
The bending stress at the lower edge of the
transformed beam section of Fig. A13.l6 would be: T
4.65
_
60000 x (- 2.73) -_ 3200
+-
i
°bt = 51.30 ' ps.
r
steel 10 • .10 • •
Ix • 0.6075 x 4.65 5:35 = 28.27 in. •
ra "",':rues. ,1
+ 0.531 x
1 MagneSium
c .1 Alloy .10
1 4
" ---j
Alum.Alloy
Area of Each
stringer :II) sq. in.
I xy = 0.446 (- 2.365) (4.65)
0.1615 x 3.635 x 4.65
0.10 (- 2.365)(- 5.35)
0.431 x 1.635 (- 5.35)
- - 4.90
= 2.73
= 1.27
- - 3.77
TOTAL - - 4.67 in. <6 =
Fig. A13.17
sowrrON; The bending stresses will be calculated
by USing method 3 at Art. A13.S.
Since the 4 stringers are made of different
materials we will tranSform all the materials From Equation (14) Art. A13.5 --
into an equivalent beam sect ron with all 4
stringers being magnesium alloy.
Salum.allOY = 10,500,000 Iy
~1\..ill1. = 10.5 = 1.615 Ixl y -
E",ag. 6.5
Ix = 28.27 = 0 1797
Using the ratio of stiffness values as indicated Ixl y - 157.4 •
above gives the transformed beam section of F1g.
A13.18 where all ~ter1al 1s now magnesi'~ 10,000 in. lb. My • 5000 in. lb.
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A13.l3
that the ~imu~ failing ccmpressive stress A13.19 shows that :or a given strai~ ~~ tr.e ~n
occurs at a s~raln of 0.01 in. per inch. The elastic range the resulting tensile stress 1s
problem 1s to deterwlne the ultimate reslstln~ ~1gher than the ~esultlng compressive stress,
moment developed by this round bar and then and :urthe~ore from internal eqUilibrium the
compare the result with that obtained by uSin~ total tensile stress on the cross-section must
the beam bending stress for~ula based en linear equal the total compressive stress.
variation o~ stress to strain. The problem as stated assumed that a
,-- compressive unit strain of 0.01 caused failure.
Fig. b thus shows the strain picture on the
beam just before failure since plane sections
g i '!.l .r J remain plane after bending in the ~nelastic
~.
Col. 4 e = strain at midpoint", (y - .0375)/103.75 !
Thus using the ~ame failing stress at :he given in most books on advanced engineering
far extreme fiber the beam for~ula based :n mechanics and will not be repeated ~ere.
linear stress-strain relct~onship gives an ul- It is convenient however to express the
timate bending strength of 38000 as compared to influence of the beam curvature in the for.n of
a true strength of 56735 or only 67 percent as a correction factor K by which the stresses ob-
::J.llch. tained by the beam formula for straight beams
Fig. c of Pig. Al3.l0 shows the true stress can be multiplied to obtain the tr~e stresses
:istribution on the cross-section, which ex- for the curved beaIT~. Thus for a curved beam
plains why the reSisting ~oment 1s higher than the maximum stress can be calculated from the
Nhen a triangular distribution is used. equation
The problem ot the ultimata bending
strength of struct~al shapes is discussed in cr = K M c - - - - - (19)
I
ietail in Volk~e II.
A13.11 Curved Beams. Stresses Within the Elastic Table A13.3 gives the value of K for
Range. various beam section shapes and be&~ radius of
The equations derived in the preVious art- curvatures. The table shows that far only
icles of this cr~pter Wer9 fer beams that were rather sharp curvatures is the correction ap-
straight. Thus in ?ig. ~13.2l, the element or preciable. In general for airplane fuselage
length (L) used in the derivation was constant rings on frames the curvature influence can be
over the depth of the beam. The strain (6L/L) neglected. However there are often fittings
was therefore directly ~roportior~l to 6L which and mechanical structural units in airplane
had a linear variation. construction whose parts involve enough curva-
In a curved beam, the assumption that plane ture to make the influence on the stress of
sections remain plane after bending still primary importance. ~he concentration of
applies, however the beam se~ent of a curved stress on t~0 inst1e Bd~e of a curved unit in
beam cannot have squal width over the depth of bending may In r Luence the rat tgue strength of
the beam because of the curvature as illustrated unit considerably, thus a consideration of the
in Fig. A13.22, or in other words the length of possible influence of curvature should be a
the segment is greater on the outside edge (L ) regular ;art of design procedure.
In the inelastic or plastiC stress range,
than on the inside edge (Ll). Thus in calcu-
the tn rjuence of beam curvature should be
I- consider~bly less since the stiffness ot a
4L--! f-=-i material in the inelastic range is much less
~-I--"ivl than in the elastic stress range and changes
rather slowly as the stress increases.
~ I I,
Fig.A13. 21
A13. i2 Problems.
Straight Beam
(1) Fig. A13.23 shows the cross-section of a
single cell beam with 12 stringers.
Assume the walls and webs are ineffective
in bend1ng. Calculate load in each
stringer by use of beam formula. Also
calculate stringer loads by equating in-
ternal resisting ~ament to external bend-
Stress ing ~oment. Bach stringer area is same
FIg. Al3. 22 Distribution and equals 0.1 sq. in. applied bending
Curved Beam moment Mx =-
100,000 in. lb.
, I I
lating the strain distri~ution over the beam L5"-r5"--.l....5"-S"_·_' 5.. -1..5.. -'
I, I
" _'
I _I _ I'
depth the change in length ~L at a point must be d
divided by the segment Width at that point. bee f
Thus even though plane sections remain plane
the strain distr1bution over the section will
not be linear. '!'he width of the segment at any
point is directly proportional to the radius of
curvature of the segment and thus the strain at
a point on the segment is inversely to the
radius ot curvature. This gives a hyperbolic (2) Same as problem (1) but change external
type of strain distribution as illustrated in bending moment to My = 200,000 in.lb.
:ig. A13.22, and if the strains are within the
elastic limit the stress distribution will be (3) Fig. A13.24 shows a beam section with 4
similar. The development of a beam :ormula stringers. Assume web and walls in-
based on a hyperbolic stress distribution is ef!ectlve in bending. Stringer areas
A13.15 BEAM BENDlliG STRESSES
SECTION
R
C
FACTOR K
IN"",E
FmER
vUTSID,t,
FmER
.'
R SECTION R
c
FACTOR K
lliSIDE
FIBER
OUTSIDE
FffiER
'-R 1.2 3.41 0.54 0.224
I 1.2 3.55 0.67 0.409
Pi
1.4 2.40 0.50 0.151 1.4 2.48 0.72 0.292
1.6 1. 96 0.65 0.108 1.6 2.07 0.76 0.224
I.a 1. 75 0.68 0.084 1.8 1. 83 0.78 0.178
2.0 1. 62 0.71 o, 069 2.0 1. 69 0.80 0.144
2.0 1. 33 0.79 0.030 3.0 1. 38 0.86 0.867
~I
4.0 I. 22 0.84 0.016 4.0 1.26 0.89 0.038
6.0 1.14 0.89 0.0070 6.0 1. 15 0.92 0.018
6.0 1; 10 0.91 0.0039 8.0 1. 10 0.94
10. a 1. 08 0.93 0.0025 10.0 1. 08 0.95 ~·~~~lO
1.2 2.89 0.57 u•• o, 1.2 2.52 0.67 0.408
dJl
1.4 2.13 0.63 0.204 1.4 1. 90 0.71 0.285
~tI--41 l' f- I,
I
1.6 1. 79 0.67 0.149 1.6 1. 63 0.75 0.208
1.8 1. 63 0.70 0.112 I.a 1. 50 0.77 0.160
I
3T~
I 2.0 1. 52 0.73 0.090 2.0 1.41 0.79 0.127
3.0 1. 30 0.81 0.041
'L_: I 3.0 1. 23 0.86 0.058
I-R-j 4.0
6.0
1. 20
1.12
0.85
0.90
0.021
0.0093
.L
r...:-
,
'
R ----oj
4.0
6.0
1. 16 0.89 0.030
1.10 0.92 0.013
0.92 0.0052 I
6.0 1.09 8.0 1. 07 0.94
10.0 1.07 0.94 0.0033 10.0 1. 05 0.95 g.g~:~
1.2 3.01 0.04 2.37 o. 73
ib--.j I 1.4
1.6
2.18
1.87
0.60
0.65
~:~~
0.168
1.2
1.4
1.6
1. 79
1. 56
0.77
0.453
0.319
TtIJT
0.79 0.236
1.8 1. 69 0.58 0.128 1.8 1.44 0.81 O. 183
1 1i
2.0 1.58 0.71 0.102 2.0 1. 36 0.83 0.147
3.0 1.33 0.80 0.046 3.0 1.19 0.88 0.067
II 4.0 1. 23 0.84 0.024 4.0 1.13 0.91 0.036
6.0 1.13 0.88 O.all 8.0 1.08 0.94 0.016
~R--I 8.0
10.0
1.10
1. 08
0.91
0.91
0.0060
0.0039
8.0
10.0
1. 06
1. 05
0.95
0.96 g. ~~~~
1.2 3.09 0.58 0.336 1.2 3.28 0.58 0.269
1.4 2.2:5 0.62 0.229 1.4 2.31 0.64 0.182
-oj I
t=:JTIi
1-3b 1.6 1. 91 0.66 0.168 1.6 1.89 0.68 0.134
1.8 1. 73 0.70 0.128 1.8 1. 70 0.71 0.104
2.0 1.61 0.73 0.102 2.0 1. 57 0.73 0.083
b 1 I 3.0
4.0
1. 37
1.26
0.81
0.86
0.046
0.024
3.0
4.0
1. 31
1.21
0.81
0.85
0.038
0.020
--401---1 6.0 1.17 0.91 0.001 8.0 1. 13 0.90 0.0087
r-
R 8.0 1.13 0.94 0.0060 8.0 1. 10 0.92
g. g~~;
r 5b --I
10. a
1.2
1.4
1.11
3.14
2.29
0.95
0.5.
0.54
0.0039
O.,5.
0.243 r-
Trent I
41
-! I
10.0
1.2
1.4
1. 07
3.55
2.48
0.93
0.67
0.72
0.409
0.292
~e::Jl
1.8 1.93 0.62 0.1'19 1.6 2.07 0.76 0.224
I. a 1.'14 0.65 0.138 2
, 1. 8, 1. 83 0.78 0.178
, 1- ! I
2.0
3.0
1. 51
1. 34
0.68
0.76
0.110
0.050 4~
I I
t,
t/2
t
2.0
3.0
1. 69
1. 38
0.80
0.86
0.144
0.067
4.0 1. 24 0.82 0.028 4.0 1. 26 0.89 0.038
1.1 I'
t;-I 6.0
8.0
10.0
1.15
1.12
1.12
0.87
0.91
0.93
0.012
0.0050
0.0039
f j... C
I- R-1
~ I 6.0
8.0
10.0
1. 15
1. 10
1. 08
0.92
0.94
0.95
0.018
0.010
0.0065
1.2 3.63 0.58 0.41.
i<it.<!1- I 1.4
1.6
2.54
2.14
0.63
0.67
0.299
0.229
• e equals distance from centroidal axis to neutral axis.
T
~'=ill ., 1 I,
......JCf-
, 1.8
2.0
3.0
4.0
1.89
1. 73
1.41
1. 29
0.70
0.72
0.79
0.83
0.183
0.149
0.069
0.040
References: Wilson and Quereau. "A Simple Method ot
Determining stresses in Curved Flexural
Members"
"Advanced Strength of Materials", by Seely.
f.- R~ 6.0 1. 18 0.88 0.018
8.0 1.13 0.91 0.010
10.0 1.10 0.92 0.0055
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A13. 17
s~own in () on fi~Jre. ExterTIal applied (7) Fig. A13.27 shows 3 dif~erent Je~ sections.
Je~ding ~cments are; They are wade af aluminun alloy whose
stress-strain G~agram is the sa~e as t~at
Mx= -500,000 in. lb. a~d My = 200,000 in. lb. plotted in Fig. AI3.19. Determine the
Find stress on all four stringers by all ultimate internal resisting moment it the
three ~ethods which were explained in this naximum compressive strain is limited to
chapter. 0.01 in./in. Consider that upper portion
is in compression. Compare the results
1.2)
a obtained with formula M = 0bIlc, where crb =
compreSSive stress when unit strain is
(0.5)
"i -,--
5"
0.01.
1 1/4"
in.lb. and My 2000 in. lb. Find bending
stresses at pOints a,b,c,d.
I
-n'-
I-- 1"--1
Fig. Al3. 27 8 fu4
i
'~
314 .L
Lx
"L 1-;;'
L 15 c
... d
-Jl/4j-
--..L
Fig. A13. 25 500*
c
50011
2" - - - 1
~
MX= 80,000 In. lb. T
Et"'".ag.= 6.5x10 s 1/1/8 .4;~~O.I"
Esteel = 29 x lOS
Steel
-=-
r "T" 2'•..., 1/16 -L~ '0.1" L
0.6"
(6 ) Fig. Ai3.26 shows (8) Fig. A13.28 shows a curved beam, carrying
a cross-section of a two 500 lb. loads. Find bending stresses
wood beam composed at points C and C", when beam cross-section
at 3 kinds of wood is made 3 different ways is indicated by
labeled A, 3 and CJ sections 1, 2 and 3. Use Table A13.3.
glued to~ther to
for.n a composite
beam. If the beam
is subjected to a
bending moment
Mx = 75000 in.lb.,
find intensity of
bending stress at Fig.AI3.26
tap edge of beam.
Also find total end load on ?ortions 3
and C.
A = Spruce. E = 1,300,000 ?si
3 = Maple. E = 1,500,000 psi
C ::: Fir. E = 1,600,000 psi
A13.18 BEAM BENDING STRESSES
DOUGLAS DC-8 AIRPLANE. Over-all view of the test wing section representing center wing section of DC-8.
c -_.I ___
a".
A ctoae-cp view of wing test section showing details of wing ribs, stringers, etc.
CHAPTER A 14
BENDING SHEAR STRESSES - SOLID AND OPEN SECTIONS
SHEAR CENTER
Al4. 1 Introduction.
In Chapter A6, the shear stresses in a Load -idxi-
Fig. A14.2
member sUbjected to pure torsional forces were D' F'
, ,
¥
considered in detail. In Chapter A13, the sub- _ _ _ ryo __
N. ""----' ~l_'_,o -X
ject of bending stresses in a beam subjected to
pure bending was considered in considerable de- I "Co0l;
dA • ---"-'- Ct " , ,
tail. In practical structures however, i t , ,
seldom r~ppens that pure ~endlng forces (coup- Beam seC~ion t!===±=:!==,:===j
o F
les) are the loading forces on the bea~. The Fig. Al4.!
usual case 1s that bending ~oments on a beam cr c -jdxt"- cr '
are due to a transfer of external shear forces. ~
N'A'~'
JA~-:r::r
Thus bending of a beam usually involves both
bending (longitudinal tension and compression
stresses) and shear stresses.
'~lL/ 0 ~J,j~
C1tf
The same assumptions that were made in
Chapter A13 in deriving the bending stress
equations are l1kp.wlse used in deriving the
equations for flexural shear stresses. With
AU,
crt at
F\.
crt crt
c. .
flexural shear stresses existing, the assumption Fig. Al4.3 Fig. A14.4
that plane sections remain plane atter bending
1s not completely true. since the shearing is as indicated. The stress crt is greater than
strains cause the beam sections to slightly cr~because the bending moment due to the given
warp out at their plane when the beam bends. beam loading is greater at beam section DO' than
This warping action is usually referred to by at FF'. Now consider that this beam portion dx
the term nshear lag~. However, except in cases is further cut as indicated by the notch DeEF in
of beams with wide thin flanges, the error in- Fig. Al4.l, and this segment is shown in Fig.
troduced by neglecting shearing strains is quite Al4.4 as a free body with the torees as indicated.
s~all and therefore neglected in deriVing the Let crt = ~lmum tensile stress at a distance
basic flexural shear stress formula. The prob- c from the neutral axrs ,
lem of shear lag influence is considered in Then the stress at a distance y tram neutral
other chapters. axis is cry = at y/c.
The total load on an element of area dA of
AB.2 Shear Center. the beam cross-section (see Fig. Al4.2) thus
T~en a beam bends without tWisting, due to
equals C1ct / y dA •
some external load system, shearing stresses are
set up on the cross sections of the beam. The Now, referrin~ to Fig. Al4.4, the total
centroid of this internal shear force system is tensile load on each face of this segment will be
often referred to as the shear center for the calculated.
particular section. The reSUltant external
shear load at this section must pass through
the shear center of the section if twist of the Total load on face CD = °ctJC Y dA - - - - - (1)
section is to be prevented. ThUS, if the shear Yo
center is known, it 15 POSSible to represent the
external lead influences by eNO systems, one
that causes flexure and the other which causes
only t'Nist.
Total load en face FE = crt
c
.r Yo
c
y dA (2 )
AB.3 Derivation of Formula for Flexural Shear Stress. From Chapter A13, the equation for flexural
2ig. A14.1 shows a loaded Simply supported stresS cr was derived, namely crt = Me/I. Let M
beam. :dhen the beam bends downward due to the equal the bending moment at beam section DO' and
given loading. the beam portion above the neu- Ml tr~t at beam section FF' and let I and I' the
tral axis is placed in compreSSion and that be- moment of the inertia of the cross-sectional area
low the neutral axis in tension. Consider a about the neutral axis at these same beam sections
short portion dX of the beam at ~oints OF on the respectively. Then substituting value of crt in
beam and treat it as a free body as shown in equations (1) and (2) we can 'NTite,
Total load on face CD = ~JCy dA
Fig. Al4.3. The variation of tensile and com-
pressive stress on each face of the beam portion - - - - - (3)
Yo
Al4.l
A14.2 BENDING SHEAR STRESSES SOUND AND OPEN SECTIONS SHEAR CENTER
Resultant horizontal load· M-I M' fC ydA. neutral axis are given on the tigure.
Yo SOLUTION: -
For equilibrium of segment in x-x direction, Shear stress at neutral axis x-x
Z Fx =- -M-
1 -M'
- JC ydA + ~ b dx = 0
y, "t"x_x • "0'
V J3.09
oY d.A
M M' JC dA
hence ~ • I ; dx - - - - - - - - - - (5)
y, Table A shows the calculation of' the term
However M ---ax
M' dll
= a:x=V=the external shear on
the beam section ..
3 . 09
[ , Y dA.
v
hence "t" ::: IE JC ydA (6 ) TABLE A. REFER TO FIG. a.
y,
PORTION Area d A Y ydA
It Is Important to nots that equation (6) 1 2.09 % 0.5 , 1. 045 1. 045 1. 096
applies only to beams ot unitorm section (con- 2 3 x 0.5 , 1. 50 2.34 3.510
stant moment at inertia). In airplane wing 3 0.5 x 0.5 , 0.25 2.84 0.710
structures the common case is f'or beams to vary 3' 0.5 x 0.5 :; 0.25 2.84 0.710
in cross-section or moment at inertia, and it
this variation is considerable, equation (6) SUM 6.026
should not be used and resort should be made to
~
equations (3) and (4). This tact Is Illustrated
in example problem 2. This matter at variable 2 3' ~}"h~
,
cross-sections is discussed later in this ,
850 x 1.42
"MAX. =~ (I1 Ie) =,- (79200 x 4)
38.67·
(The width b = 0.5 + 0.5 for the two portions For a paint 1 inch from either edge of the
3,3' i. beam
The Shearing stresses as calculated act in the
plane of the beam cross-section in the Y di- a = +- (79200 x 3) = ± 6135
38.67
psi.
rection and also with the same intensity paralle
to the Z axis which is normal to the beam TAJ:ILE AB.1
section. z
Example Problem 2. VARIABLE 110l1Th'l' OF INERTIA.
Fig. A14.6 shows a cantilever beam loaded On Top tnt lI\t I
With a single load of 600 lb. at the end and BenQ1ng or 1" From 'I 2" From 1
Moment Bottom Top or Top or i p~,rt~~n ~,rt:~D i ~rt~~11
acting through the centroid of the beam cross- M fiber Bottom Bottom' A B r 'C
section. The beam section is constant between y-4",y-3" 1,"2,,1
stations a and 132, then it tapers uniformly to ,
the sections shown for stations 175 and 218. 132 79,200"f =.8180 :t6 U 5 +4090 7157i1 i H12' I 1023* I
;~:~ : ~~~: i
I 175 1I 105, OOO"fl .!oU80 +8135 +4090 1023 ,
The shear stress distribution on the beam cross-
section at stations 175 and 218 will be deter- i I
Z1. 130,800'" =.8180 +6135 t4090 1023 i
:nined.
From the results in Table Al4.l, it should be
'-43"~43"'- +--- 1 3 2 " - - - -
Fixed
!!-__;-__'-_-,
1- I -
~~i-----;;c:-~O;::+-;-;;;;-;;;:r-;o.,-------;;2
f"0. noticed that the change in moment of inertia be-
tween the three stations is directly porportlonal
to the change in bending moment, hence the same
Sta. '218 Sta:175 sci.. 132 Stl.O value for the bending stresses for all three sta-
tions. Columns 6, 7 and 8 give the total bending
~.::1
stress load on portions A, B and C of the three
cross-sections (see Fig. A14.6). These values
equal the average stress on the portions times
B the area of the portion; for example, for station
132, the load on portion A:
load = 6180 +
2
6135
x 1 x 1 7157 Jil, and for
portion C:
sts, 218 sta. 175 Sta. 132 load = 40902 + a x 2 x .25 - 1023*
I : 64.0 in. 4 I ::; 51. 33 in. 4 I: 38.67 in. 4
Fig. A14.7 shows the tension and compressive
stresses acting on a portion of the beam between
._lS4
.
A14 4 BENDING SHEAR STRESSES SOUND AND OPEN SECTIONS SHEAR CENTER
stations 175 and 218. Fig. A14.8 shows the re- shear stresses between stations l32 and 175
sUIting horizontal shear stress pattern result- would be the same, since the change in bending
tng trom the loads in Fig. A14.7. For example, moment and moment of inertia have been ~de the
it we take a section along the beam 1" from the same as between stations 175 and 218.
tap or bottom edge ot the beam and treat this Figs. A14.9 and A14.10 show the shear
portion as a free body as shown in Fig. A14.12
applying ZH = 0, stress patterns i f the formula -r = V ~~ dA be
used for each station. The discrepancy is con-
ZH =- 7157+7157+'t'x43xl.O= 0, hence 't'" =0 slderable as the equation does nat apply to
beams of varying section.
~
613 -~.~
7157. X - 7157 * To illustrate the calculation by the shear
stress formula. the shear stress will be calcu-
09 ~'---B IB - -- _10224*
lated at the neutral axis for the beam sect~on
r
1023*
1- 1023• at station 175.
--- --- --- -
~a
.!.Qill. I--- 1023* V
't'=Tb oYdA
18180
Iv ._-B
··AI
-
I IB ---10224*
Ik:!--7157f
where f> cIA =1 x 1 x 3.5 -+ 2 x 1 x 2.5 + 2 x
~
;;
.;
Fig. Al4.7
-
~
~
~
0.25 x 1 • 9.0
~
409
II 20'
TABLE A14.2
,--J ('4'2
~
49:6 MAXIMUM SHEAR STRESS FOR SIMPLE SECTIONS
477*/0" 1r431 !.--420
LOCatiOD of
to.~·
1
C ross section Max. Shear Stress
Max. Shear Stress
~ ~
to 59.5
I I
True
Shear
Pattern
Sta 218 Sta 175
Shear Stresses
By Equation (6)
N-B-N I 3V
l' = 2A e• 0 I
43"
I
I I II
0 -il-1/4"
7157
-- --
15336
F sx43x.25
7157
10224 N-ON 4V
't' = 3A
I
II
I
e• 0
Fig. A14.13 ,
D7~
43"
~7 N-~N II -r = 4V(
I
3A 1'"
Dd)
na .. d. e• 0 ,
"'1"1- ====~--
F sx43xl
I
i
Fig. A14.12
eL -'
!
""'~=01
'.t@;r
~
00(>:"1
I,
d--
I
qx •
--e
q
Z
qx
"
~
the cross-section, namely: -
If the calculated shear flow is directed toward the
boundary line between the two intersecting planes of the par-
ticular free body, then the shear flow on the other plane is
also directed toward the common boundary line, and con-
versely directed away if the calculated shear flow is directed
away.
Fig.A14.22
r Fig.Al4.19 T' Fig. A14.14 shows the sense of the shear
Fig.A14.20 flow pattern on the beam section as determined
for the given external loading.
s.u
j."::'--';::;;;'_"""i.;H~_
""= .""
= 18.75
= - 37.5 + 375 (- 0.5)(1 x 0.1) = - 56.3 lb./in.
~b = 1 x 37 .5/2
. lb.
A~I"l
a = - - - - - - - - - - - - - - - (10) The section has been
divided into 4 portions
labeled 1, 2, 3 and 4.
,Method 3. The Method USing section properties V = 10000 lb. . 4'J: ..L-
about centroidal Z and X axes. For k :1"
brevity this method will be called
the k method. Fig.AB.26
hence 2 ¢ = 2S 32.2'
0
- or $ = 12° _ 46.1' Now SUbstituting in eqUation (15)
sin ¢ = 0.2210 and cos ¢ = 0.97527 .
The moments of inertia about the principal axes qb = - 0.~16 (1 x 0.1 x 1.5136)
can now be calculated.
Ix = Ix cos· ¢+ I z sin· ~ - 2 I xz sin ¢ cos ~ 6792
- 0.02782 (1 x 0.1 ) ( - 0.1184) = - 1756 + 2890
p
,
= 0.6035 x .97527 + .0574 x .2210 - 2 (-0.1305)
--'
= 1134 Ib/in.
x .97527 x .2210 = .63316 in'. Calculation of shear flow at point (c).
I z := Ix sin' ¢ + I z cos' ~ + 2 I xz sin ~ cos~
p
,
= 0.6035 x 0.2210 + .0574 x ~ + 2 (-0.1305)
. For portion (2) area A = 1.4 x 0.1
zp =
.70 x .97527 0.6825 in.
xp = .70 x .2210 = 0.1547
=
0.14
x .97527 x 0.2210 = 0.02782 in~. The shear flow at paint (c) equals the
The equation for shear flow q is, shear flow at point (b) plus the effect of the
portion (2) between points (b) and (c)., hence
- - (15) 7348
~ = 1134 - 0.63316 (0.14 x 0.6825) -
Vz • 10000 x cos 30° • 8667 lb. The shear stresses at these two points (b)
and (c) would equal q/t = 1134/0.1 and - 5260/0.1
Vx = 10000 x sin 30 0
• 5000 lb. or 11340 psi and - 52600 psi respectively,
Resolving these z and x components further SOLUTION BY NEUTRAL AXIS METHOD. (Method 2)
into components along the principal axes we In thiS solution n~
obtain,
it is necessary to find \ i
= 8667 x .97527 - 50000 x .2210 =7348 lb. the neutral axis for
the given external load- : .O~~6" \1
iag. In Fig. A14.28,
=8667 x .2210 + 5000 x .97527 = 6792 lb. the angle Q is the angle zn: .186
("""~
( '.) r
A14.l0 BENDING SHEAR STRESSES. SOUND AND OPEN SECTIONS. SHEAR CENTER.
The component or the given external shear load [3.257 x 8667 - (- 7.406 x 5000)J L z A, whence
normal to the neutral axis n-n equals
Vn • 10000 x sin 45° - 29' • 7130 lb. q - - 235438 Z x A - 65258 Z z A
~ITe
beam composed ot heavy
tlange members and a curved
thin web. For bending
=- •01762
0.1305
- - 7.406 about the X-X axiS, the
Iz 0.0574
web on the compreSSive,
side ot the beam absorbs
x--:-~
~ i I
.
1<0 = Ix Iz - Irl = .01762 =3.257 very little compressive
stress, since buckling ot q ' I ,,~
over a particular web. In other words in the EXA!1PLE PROBLEM - RESULTANT OF A CONSTANr FLOW
FORCE SYSTEM.
shear flow equation q : Vz I z d A, if the area
Ix Fig. A14.30 shows a constant flow force
of the web is neglected then q is constant be- system thru points ABC o E with q • 10 lb.
tween flange members. per inch. The resultant or this force system
is required.
RESULTANr OF CONSTANT SHEAR FLOW FORCE SYSTEm
Fig. A14.Z9 shows a beam assumed to be
car-rying a downward shear load (not shown) and
to cause, bending about axis x-x without twist.
Assuming the two flanges develop the entire
bending reSistance, the shear flow q is con-
stant on the web and acts upward along the web
to balance the assumed external downward load.
The resultant of this reSisting shear flow
force system will give the lateral position of
the shear center for this beam sectlon. The
problem then is to find the reSUltant ot the SOLUTION: -
shear flow system.
Let q = load per inch along web (constant). Draw clOSing line between the beginning
and end points ot ror-ce system. (line AE). The
Let R = resultant of the q torce system. length h ot thiS closing line is ZO incheS.
From elementary mechancis, From eq, (17) R = q h = 10 x 20 = 200 lb.
The direction of the resultant is parallel
R =.jz qx -+- Z qz ,where qx and qyare the x to line AE or horizontal In this problem. To
tlnd the location of R take moments about any
and y components of the q forces along the web. point such as (0). D~ lines trom point (0) to
SiLce q is constant, Z qx is zero, hence, points A and E. The enclosed area (A) equals
(17) s x ic- s x io- .5nxS· -+-lOx 50 = 189.3 sq. in.
sheet to form the webs and walls. The flange in order to indicate at what point the shear
members are numbered a to h and the areas of flow ~ is being calculated.
each are given on the figure. It will be as-
sumed that the webs and walls develop no bending Qab = - 2.5 Za z A = - 2.5 x 5 x 0.1 =
resistance and thus the shear flow between ad-
jacent flange members will be constant. The = 1.25 Ib./in.
problem is to determine the shear center for the
beam section. The first letter of the subscript refers
to the flange member where the shear flow q is
SOLUTION: - being calculated and the second letter indicates
8n which adjacent side of the particular flange
Since the beam section is symmetrical member. Hence qab means the shear flow at
about the X axis, the centroidal X and Z axes flange (a) but on the side toward (b).
are also principal axes, since the product of
inertia I xz is zero. Qab = qba = - 1.25 (Since no additional flange
The vertical pOSition of the beam section area is added, and thus
centroid due to symmetry is midway between the shear flow is constant on
upper and lower flanges. sheet abo
To find the horizontal pOSition of the
centroid; take moments of the flange areas 'lbc = qba - 2.5 "b Z A
about the left end or line be: - - 1.25 - 2.~ x 5 x 0.4 - 6.25 lb./ln.
x= lo A x = 0.4 x 15 + 0.2 x 10 ? 0.2 x 5 = - 6.25
ZA 1.6 Sco - 2.5 !c z A
= 5.625 In. = - 6.25 - 2.5 x (-5xO.4)= - 1.25
I z = 0.8 x 5.625
2
? 0.2 x 0.625
•
? 0.2 x
qer = 0 - 2.5 "e z A =o - 2.5 (- 5 x 0.1) = 1.25
• qre = qer .. 1.25
4.375 ? 0.4 x 9.375- = 64.4 In~.
qrg = 1.2: - 2.5 "r z A = 1.25 - 2.5
HORIZONTAL PCSITION OF SHEAR CENTER: - t: 5 x 0.2) = 3.75
The horizontal pOSition of the shear center
will coincide with the centroid ot the shear
%r = qrg = 3.75
flow system due to bending about axis xx with- qgh = 3.75 - 2.5 "g z A = 3.75-2.5x5 x 0.2 =
out ~Nist. For simplicity, to elimir~te large 1.25
decimal values tor shear flow values an ex-
ternal shear load Vz =
100 lb. will be assumed
qhg = qgh = 1.25
and the internal reSisting shear flow system qha = 1.25 - 2.5 loh z A = 1.25 2.5 x 5 x 0.1 =
will be calculated for this external loading. 0 (checks free edge at h).
From equation (8)
The Sign or sense of each shear flow is for
the shear flow in the y direction as explained
SUbstituting values at Vz and in the derivations of the shear flow equations.
Ix The procedure now is to determine the sense of
the shear flow in the plane of the cross-section
100 " z A 2.5 " z A or in the xz plane. It is only necessary to
40
determine this sense at the beginning point, that
We could start the solution at either at is in sheet panel abo The surest way to deter-
two points (a) or h since these points are free mine this sense 1s to draw a simple tree body
edges and thus qy is zero. In this solution, sketch at flange member (a) as illustrated in
Fig. Al4.3l. The shear flow on the cut face is
we will start at the free edge at paint (a) and qy(ab) = - 1.25 and
go counterclockwise around the beam section.
The area or each flange member has been concen- this value 1s shown
trated at a point coinciding with the centroid ~n the free body. By
of each flange area. In solving for the q val- s1mnle rule given at
ues the subscript y will be omitted, and sub- the end of Art. Al4.6,
scripts USing ~~e flange letters will be used the shear flow in the
plane of the cross- Fig. A14.31
----
"'" : ' .~. <: ~ -v- ;,. ::.. '~ .~_:~'-_ ~ ~·f,;~~-;'.. ":<h:~2.:-._;~ '".:: ~ :. ,>:. ~:, :·.L" ~~: ~ '..<~~' ~ :', ;-: ," .~,,;)~":':' ~.~ ;: ~.,~ ~ ~ '~ :,:::,. ,~.;:.;. '~·~.r . ~~? ·5J-:~~. . t';-· ~
ANALYSIS A D DESIGN OF FLIGHT VEHICLE STRUCTURES A14.13
section is also directed toward the cammon We will again start at the free edge ad-
boundary line and thus qx(ab) has a sense as jacent to flange (a) where qy - O.
shown in Fig. Al4.3l. The sense of the shear
flow on the cross-section will now continue in
~b =- 1.55 Za x A =- 1.55 (- 0.625 x 0.1) =
this direction until the sign changes in the 0.0971 Ib./in.
origianal calculation, which means therefore the
shear flow sense will reverse. Fig. A14.32 qba = 'cab = 0.0971
shows a plot of the shear flow pattern with the qbc = 0.0971 - 1.55 Zb x A = 0.0971 - 1.55
sense indicated by the arrow heads.
(- 5.625 x 0.4) = 3.592
'leb = qbc = 3.592
'led = 3.592 - 1.55 Zc x A = 3.592 - 1.55 x
5.625 x 0.4) = 7.087
'ldc = 7.087
'lde =7.087 1.55 (- 0.625 x 0.1) = 7.184
qed = 7.184
qet = 7.184 - 1.55 x 4.375 x 0.1 = 6.504
Fig. A14.32
qte = 6.504
The results will be checked to see it
qrg = 6.504 - 1.55 x 9.375 x 0.2 =3.589
static equilibrium exists relative to Z Fx and qgt = 3.589
Z Fz O. 'lgh ·3.589 - 1.55 x 9.375 x 0.2 - 0.674
Z F z = 100 (ext. load) - x 6.25 - 10 x 3.75 +
~~
Fig. Al4.33 shows the plotted shear flow
1.25 x 0.5 x 4 - 1.25 x 0.5 x 4 = 0 (check). results. The signs of the calculated shear
flows are for shear flows in the Y direction.
Z Fx = - 5 x 1.25 + 5 x 1.25 - 5 x 1.25 + 5 x Simple Consideration at a free body of flange
1.25 =0 (check). member (a) will ~ive the sign or sense of the
shear flow in the plane of the beam section.
The shear flow force system in Fig. A14.32 Thus in Fig. Al4.34 qx must act as shown when
causes the section to bend about axis xx with- qy is positive.
out twist. The resultant of this system is 100 .674
.0971
lb. acting down in the Z direction. The posi-
----
tion at this resultant will thus locate the
lateral pOSition of the shear center.
Equating the moments of the shear flow
system. about some point such as (c) to the mo-
i b a
3.589 11
~~4
ment of the resultant about the same point we
obtain: -
100 e = 10 x 3.75 x 15 - 1.25 x 0.5 x 2 x 5 -
1.25 x 0.5 x 2 x 10 + 1.25 x 0.5 x 2 x 15 e=6.43"
~
CENTER.
-----For convenience as betore, we will assume I
• 1".,...l I
a shear load Vx = 100 lb. and compute the re- Ox _5.625 -I
sisting shear flow system to reSist this load I
in bending about axis ZZ Without twist. The a t
resultant at this shear flow system will give
the vertical location ot the shear center. The
shear flow equation 1s,
Fig. A14.34 1
S. C.
6.43
•
q = _ Vx Z x A ::I' _ ~Z X A - 1.55 Z x A Fig. A14.35
y Iz 64.4
A14.14 BENDING SHEAR STRESSES. SOUND AND OPEN SECTIONS. SHEAR CENTER.
Re=ZIl - 21.333
15019 .00142
100 e • 5 (.0971 + .674) 11 + 10 x 3.589 x 15 +
.5 x .0971 x 5 - (.5 x .674 x 10) + .5 x .674 x Iz 170.667
= 0.01136
15 + .5 x 6.504 x 15 + .5 x .674 x 10 - (.5 x Ix I z - 15019
.0971 x 5) Ix 90.667
0.006037
100 e = 643 Iz - I xz
' 15019
• = 643/100 • 6.43 inches Resolving the given shear load of 141.14 into z
and x components, we obtain,
Fig. A14.35 shows the resulting shear center
location for the given beam section. Vz • 141.14 x sin 45° 100 lb.
EXAMPLE PROBLEM 2. unsvnmetrr ica). Beam Section. Vx • 141.14 x cos 45° ~ 100 lb.
Fig. Al4.3S shows a tour 1'lange beam seC-
t ron, The areas at each flange are shown ad- From equation (14) -
jacent; to tlange. The external shear load V
equals 141.14 lb. and acts in a direction as qy = - (k 3 Vx - k 1 Vz ) Z x A - (k. Vz - k 1 Vx )
shown. The problem is to tind the line ot Z z A
action ot V so that section will bend Without SUbstituting
twisting.
used.
To solve this problem, method (3) will be ~b =- 0.7457 X 1 (- 5.333) - 1.278 x 1 x 6.667
To locate centroida1 x and z axes: - - - 4.544
= ~b = - 4.544
=
1 x 12 + 0.5 x 8
3 =5.333 in. = - 4.544 - 0.7457 x 1 (- 5.333) - 1.27S x
1 (- 5.333) = 6.249 1b./in.
= (0.5 + 0.5) 16 = 5.333 in.
3 = %c = 6.249
A L 5 A E51 o v A14.15
qed = 6.249 - 0.7457 x 0.5 x 10.667 - 1.278 x A14. 11 Shear Center Location By Using Neutral Axis
Method.
0.5 (- 5.333) = 5.680
In a beam subjected to bending there is a
~c = qed = 5.680 definite neutral axis position tor each differ-
~a = 5.680 - 0.7457 x .5 x 10.667 - 1.278 x 0.5 ent external plane of loading on the beam. The
shear flow equation with respect to the neutral
x 2.667 = 5.680 - 3.977 - 1.704 = axis is,
o (checks free edge at d where qy
must be zero.)
qy =_ Vn ~ Zy,A (20)
In
Fig. A14.37 shows the resulting shear flaw re- where, Vn = Shear resolved normal to neutral axis
sisting pattern. The sense of the shear flow in In Moment ot inertia about neutral axis =
0"",
Zn = Distance to neutral axis
I
a
liR'
V= 141.14
141.14
d
U~
Ox
a
Ftg.A14.37a
In finding the shear center location ot an
unsymmetrical section, it is convenient to as-
sume that the Z and X axes are neutral axis and
q =5.680 find the shear flow system for bending about
each axis by equation (20). The resultant ot
b qe 6.249 c each of these shear flow force systems will pass
through the shear center, thus the intersection
Fig. A14.37
ot these two reSUltant forces will locate the
shear center.
the plane or the cross-section 1s determined in
web at flange member (a) by the simple tree Example Problem
body diagram or stringer (a) in Fig. A14.37a
Check Z Fx and Z Fz to see if each equals The same beam section as used in the
100. previous article (see Fig. A14.36) will be used
to lllustrate the neutral axis method.
6.249 x 16 = 99.99 (CheckS Vx = 100)
Fig. A14.38 shows the section With the
" Fz =- 12 x 4.544 - 8 x 5.68 =- 99.94 (checks cerrtr-cfda.L axis drawn in, The X axis will now
Vz=lOO).
The resultant of the internal resisting
~!
~
a I d
Y 100i + lOO~
~~.47
shear flow system equals • 141.14 --l. _ _ _ -x
lb.
To locate this resultant we use the prin-
5.333
I
1*1ll66
b I C b c
Ciple ot moments. Taking pOint (b) as a moment
center,
Fig. A14.38 Fig. A14.39
141.14 e = 8 x 5.68 x 16 - 4.544 x 30 n be assumed as the neutral axis for an external
212 plane of loading as yet unknown. T~e 1'1111 further
hence e = 141.14 = 1.50 inch. assume that when this unknown external loading
is resolved normal to the X neutral aXis, that
Therefore external load must act at a distance it 1'1111 give a value of 100 lb., or Vz = 100.
e = 1.50" tram (b) as shown in Fig. A14.37. The From the previous article Ix = 90.667.
load so located will pass thrti shear center of
section. To obtain the shear center location, Since the X axis has been assumed as the
another loading on the beam can be assumed, and neutral axis, equat10n (20) can be written
where the line ot action of the resultant of the
resisting shear flow system intersects the re- qy _;!:z. Z ZA, hence,
lx
=
sultant as fo~~d above would locate the shear
center as a Single point. It the shear center
location is desired it 1s convenient to assume Qab 9~~~67 x 6.667 x 1.0 -7.35 lb./in.
=- =
a unit Vz and Vx acting separately and find the
horizontal and vertical locations of the shear 'Ibc 7.35 - 9~~~67 x (-5.333)1 = -1.47
=-
center from the 2 separate shear flow force
systems. qed = -1.47 - 9~~~67 (-5.333) 0.5 1.47 =
Fig. A14.39 shows the resulting shear flow
values.
A14.16 BENDmG SHEAR STRESSES. SOUND AND OPEN SECTIONS. SHEAR CENTER.
ZFx =1.47 x 16 =23.52 lb. Take moments about (b) and let (e) equal
distance to "esultant R.
ZFz = 7.35x12+8xL47 =- 99.96 lb. (check)
100.8e = 3.125x1'tx5 a+3.13xSx16
R =..; 100· + 23.52 = 103 li
lb.
e = 753/100.8 = 7.47 In.
tan Q = 23.52/100 = ,2352
Fig. AI4.40 shows the position of this re-
hence Q -= 13° - 16' l0o-tS R = 103 sultant torce. T~ere it intersects the ?revlous
resultant force gives the shear center location.
=
Let e distance fram 23.52
A14.12 Problems
resultant R to point b.
(1) Fig, Al4.42 ~1+1+1"1
shows the cross-
It
Equating moments ot b b -.-
sectlo~ ot a wood
resultant about (b) to that of shear flow
system about (b), beam glued to-
gether on lines i a a
lO3e = -7.35 X 6ft a + 1.47 x 8 x 16 a-a and a-b. The
beam Is subjected 4"
e =- 644
103 = -6.25 In.
to a vertical
shear V : 2400 I
Fig. A14.40 shows the location of the
resultant. We know the shear center 11es on
lb. Determine
shearing stress
on sections a-a
r-f""'4c----,
i
-L
r i---.L
the line ot action of this resultant. Thus we and a-b. Find
maximum shearing
Fig. A14.42
stress on beam
section.
"~~
~.
is. c.
~1-1"-1
as shown. Find -l2'I'
must obtain another resultant force which the shear stress L-J'l / I
passes through the shear center before we can at sections a-a 1000 lh. ' I¥-,-
det1nitely locate the shear center. Theretore
we will now assume that the Z centro1dal axis
and b-b by three .
different methods.
is a neutral axis and that a resolution at the
external load system gives a shear Vx = 100 lb.
qy =- ~Z XA, Iz = 170.667
Qab * - 17~~~67x (-5.333)1 =3.).25 lb./ln.
I"
_ 100
qbc - 3.).25 -170.66 (-5.333)1 = 6.25 lb./1
tan Q = 112
0Z
= .125 12.5~ r---1O" -----j Fig. A14.46
~
,...
10"
( 10"
I
1-+
I 10"
_+_
l~o,~/1
I
r- "-i -.r 20
Ll50"~
Fig.A14.51
FIg.A14.52
r--l2-J-" 8 t10""
1 sq. in.
c 1.5°" 1. 0 0.8 1. T
R1I7.5 .5 sq. in.
1~' 2a'
I
I liS
1.2 o. T>" 0.8 1.0 1
.8 sq. in.
20'
Fig. AB. 53
.1 !-7+-15"-+8-1
Fig. Ai4. 54
1----19" I .5 sq. in. {g} In Fig. AI4.53, the tour stringers a, b, c
Fig.Al4.50 and d have the same area. Assume the webs in-
effective in resisting bending stresses. De-
termine the distance (e) to product bending
about the horizontal axis without twist.
(6) Determine the shear center tor the un-
symmetrical beam section of Fig. A14.50. As- (10) For the wing cell beam section In Fig.
sume sheet connecting the four stringers as A14.54, determine the location ot the shear
ineffective. Areas of stringers shown on Fig. center. Assume webs and 'Malls ineffective in
bending.
A14.18 BENDING SHEAR STRESSES. SOUND AND OPEN SECTIONS. SHEAR CENTER.
Al5.! Introduction. The wing, fuselage and em- open sections we could start the summation at a
pennage structure ot modern aircraft Is essen- tree surface Where q would be zero, thus the
tially a single or multiple cellular beam With summation to any other point would give the true
thin webs and walls. The design ot such shear flow qy. In a closed cell there Is no
structures involves the consideration ot the tree end, therefore the value of qy Is unknown
distribution ot the internal resisting shear for any point.
stresses. This chapter introduces the student
to the general problems of shear flow distri- gquatn on (1) g1ves the shear d1str1but10n
bution. Chapter A14 should be covered betore tor bending about the X axis lit thout twist.
taking up this chapter. The general procedure Is to a~ a value or
the shear flow qy at same point ~d then find
AlS.2 Single Cell Beam. Symmetrical About One Axis. the shear tlow pattern tor bending without
All Material Effective in Resisting Bending twist under the given external load. The cen-
Stresses.
troid of this internal shear flow system will
Fig. Al5.l shows a single cell rectangular be the location where the external shear load.
beam carrying the· load of 100 lb. as shown. should act for bending without twist. Since
The problem Is to find the internal resisting the given external shear would have a moment
shear tlow pattern at section abed. about this centroid, this unbalanced moment
must be made zero by adding a constant shear
10"" flaw system to the cell.
To illustrate we will assume qy to be zero
at point 0 on the web ad.
I
I
I qo = O. The term ~= 100/62.5 = 1.6
I • x
L _
'lao = -1.6 Z~zA = -1.6 x2.5 x 5 x 0 .1 = -2 1b/1n.
/ .05
/ " qba = -2-1.6 Z~zA = _2_1.6x5x20x .05 = -10
.05
1---'20" --~
-~-
.05
x-f
z
qo'b
Cleo'
= -10-1.6 L:~I zA=-10-1.6x2.5x5x .05
.05 =r -10
Fig. A15-1 'led 3 -10 -1.6 z~ zA = -10 -1.6(-5)20 x .05
= -2
Solution 1
qod = -z-1.6 z~ zA =-2-1.6(-2.5)5xO.1=0
Due to symmetry ot material the X cen-
troldal axis lIes at the mid-height of the
beam. The shear flow equation requires the Fig. A15.2 shows a plot at the shear flow
value of I X I the moment ot inertia of the reSUlts. on the vertical web the increase in
section about the X axis. shear is parabolic since the area varies
directly with distance z.
Ix = {zX.15XIO;'+Z[30X.05X5 l1
] =62.5 in ....
The intensity at qx and qz In the plane at
From Chapter A14, the equation tor shear the cross-section Is equal to the values of q
flow Is 1 found above which are In the y direction. The
sense of qx and qz 15 determined as explained
qy • -~
x
Z z.A - - - - - (1) In detail in Art. A14.6 or Chapter A14.
This equation gives the change in shear
flow between the limits of the sunnaatn on, In
Al5.!
16~'
./.
A15.2 SHEAR FLOW IN CLOSED THIN4WALLED SECTIONS, SHEAR CENTER.
2 >/In. '--!44~-
10 H/in.
1. 66 Jii/in.
Flg. A15-2 -6.34 ft/in.
o = _ 16.21 x 10 6.21 x 24.28 + 1Qi of the cell, the web and wall thickness does
.04 .025 .04 influence the amount of twist for a given
torsional load. In the shear center solution,
+ 24.28q = a it 1s known what portion of the shear flOw 1s
.025 due to torque or pure tw1st, and also that due
to bending Without twist, which fact 1s some-
hence q : 8.26 lb. in. times of importance.
Adding this constant shear flow to that TORSIONAL DEFLECTION OF CELL
or Fig. Al5.8, we obtain the shear flow at
Fig. Al5.10. The angular twist as gl ven by the f1nal
shear flow pattern of Fig. Al5.8 equals
The lateral position of the shear center 2QAG =- IqL/t, whence
is given by the location at the resultant at
the shear flow system at Fig. A15.10. 2QAG = -6.21x24.28_16.21Xl0= -10082 __ (3)
.025 .04
The resultant R ::. ./ IF + IF i
' x ,I Z After finding the shear center location,
ZFx = a we found that the external load had a moment
of 1330 in. lb. about shear center, which was
ZFz = 10 x 7.95 + 2.05 + 10 = 100 lb. resisted by a constant shear flow at - 8.26
lb/In. The angular twist under this pure
Theretore R = 100 lb. torque shear tlow should therefore give the
same result as equation (3) above.
Equate moments or R about point A to
moment at shear tlow system. about; A. 2QAG = - 8.26 x 24.28 _ 8.26 x 10. _ I0082
.025 .04
Re = l:MA which checks the result or equation (3).
100e =2.05 (2 x 80.52) A1S.5 Slngle Cell-Three Flange Beam. Constant Shear
Flow Webs.
e = 3.30 in.
It should be noticed that the web or skin For such a structure, there are six un-
thickness dOes not influence the magnitude or knowns, namely, the axial load in each stringer
the shear flow system in a Single cell beam. and the shear flow q in each of the three sheet
A change in thickness, however. effects the panels that make up the cell. For a space
unit shearing stress and therefore the shear- structure, we have six static equations ot
ing strain and thus tn computing angular twist equilibrium, thus a three f Lange single cell
Al5.6 SHEAR FLOW IN CLOSED THIN-WALLED SECTIONS, SHEAR CENT ER.
Fig. AlS-14
I
( I "be 'lab,
-' B
+
Fig. A15-15 Fig. A15-16
statics.
To find ~a take moments about point B The three unknown resisting shear flows
and equate to zero. will be assumed with the sense as i~dicated by
the arrow heads.
ms 100 x 5 - 25 x 10 + C1ca (128.54 x 2) o
To find Qac take ~aments about 9
~
hence 'lea - - 257.08 =- 0.972 1b!ln.
2MB =100 x 5 - ~c (128.54 x 2) = 0
To tlnd qab take ZPz = 0
Qac = 1.945 lo/in.
ZFz = 100 - 10 x 0.972 - 10~b = 0
ZPx =- 15 x 1.945 + 15Qbc = 0
~b = 9.13 1b!ln.
qbc = 1.945
To tlnd qbc take EFx = 0
ZFz =100 - 10 x 1.945 - lO~b =0
ZPx - - 15 x 0.972 - 25 - 15qbc =0 ~b = 8.055 1b!ln.
hence qbc =- 2.639 1b!ln.
The algebraiC signs ot the unknown Q value
The signs of Qca and qbc came out negative, all come out POSitive. thus the assumed direction
hence the sense of the shear flow on these cell of shear flows in Fig. A1S.l4 is correct.
wall portions is opposite to that assumed in
Fig. A15.l2. The resulting shear flow pattern To make the cell VNlst zero. we must add a
is plotted in Fig. Al5.13. constant shear flew q to the cell (see Fig.
A15.15). The relative twist under the shear
The student should realize the thicknesS flow of Figs. 14 and 15 will be equated to zero.
or the wall elements does not influence the
shear flow distribution if we assume the three Z qLt =_ 1.945 x 20.71 1.945x15 + 8.055x10
flanges develop the entire resistance to the .03 .025 .04
bending moment. + 20.71q + lOq + l5q
.03 .04 .025
=a
AlS.6 Shear Center 01 Single Cell-Three Flange Beam.
Constant Shear Flow Webs. T~ence, q = 0.322 Ib/in. with sense as
assumed in Fig. A1S.lS. Adding this constant
Let it be required to determine the shear shear flow to that of Fig. A15.l4 we obtain the
center location for the beam in Fig. AlS.ll. shear flow system ot Fig. A15.16. The resultant
The shear center is a point on the beam cross- R of this shear flow system is obviously - 100
section through which the resultant external lb., since the external load was 100 lb. The
shear must act it the cell is to bend without location of this resultant R will therefore
twist. locate the horizontal pOSition of the shear
center. Equate moment of resultant R about
The shear center location ~ill be deter- point 8 to the moment of the shear flow system
mined in two steps, first its horizontal about B, Whence.
location and then its verti~al location.
100e =1.623 (128.54 x 2)
C~lculation of horizontal location:-
or e = 417/100 = 4.17 in. from line AB.
We will assume any vertical shear load, (Fig. A15.16)
as the example, the same vertical shear as used
in the prOblem of Art. Al5.4, namely, a 100 Calculation of Vertical Position of Shear Center
lb. load acting five inches from A, as
illustrated in the following Fig. A15.l4. A convenient ~orizontal shear load will be
.
~ '~ ~~~ ;~~~'~ ",~' ;: :',~,,/ :/~ _~'< :", \ ",:~:,;)',:,:;:c~:1 :~~: <-~~-~;. "'., ._:~~ ':'s- ~>:~: ,:L<~;.; <~_~ "'-:~," ~- I:,~~ :.,~.<'~ ~ ~-;':~-/,,",:r;~;'6:"~~--'~"~:
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES
Al5.7
assumed acting on the cell. Since we used a A1S.7 Single Cell-Multiple Flange-One Axis of
25 lb. load in the example problem of Art. 15.5, Symmetry.
we will assume the same load in this solution.
Fig. AlS.l7 shows the loading and the assumed Fig. A15.20 shows a Single cell beam With
directions at the three unknown shear ~lows. 8 flange ~embersJ carrying a 100 lb. shear load.
The resisting shear flow system will be calcu-
lated.
A 2S
...
.1!l4\~
C
~-t:J
'lea
qc
Sab
B
+ q
q
q _
R:l2.1. T
1.01
e
I
~
B
~
S"
.40" .10" .10" .20"
z
10"
Solving for the three unknown shear flows
in Fig. A15.17. __L = 'l-d =i:. !_
OMB = - 25 x 10 + qca (12S.54 x 2) • 0
To check ZMy =
zero take moments about
same point such as C.
IJ:.' I .'.0 _I
.--1
.80-1 ! 10.667
d
.1 .•
"1
_--=:b+-_ 2.074
0
928
1.82 5.333 I
Fig. AlS-23a
6000f =V z
~~8'j
cut
I b 1600=
Iq=449 .78 I - Vx
L.J,~o~-J q=75.8
Fig. Al5-26
q =J1.
2A = 261160
x 160 = 169 . 88 Ib .in
•
qy = -71.21 Z ZpA + 19.59 Z xpA - - - -(6)
(A = area of cell = 160)
! t f-
Al5.10 SHEAR FLOW IN CLOSED THIN-WALLED SECTIONS. SHEAR CENTER.
These shear flows are practically the qy = (k~VX-k.V2)ZXA - (k.Vz-k 1VX ) ZzA
same as obtained in solution no. I as recorded Substituting in above equation,
in Fig. A15.24. Discrepancies are due to
slide rule accuracy.
qy ~ -( .002355 x 400 - .0004598 x 1000)l:xA
For equilibrium of moments, take moments -( .005452 x 1000 - .0004598 x 400)ZzA
about (b).
qy • -0.4822 ;:xA -5.268 ZzA - - - - - (7)
Z!1b '" 6000 x 8 - 1600 x 8 - 448.92 x 12
Since the value of the shear flow is un-
x 16 = -50993 In.lb. known at any poi~t on the cell walls, it will
A constant shear flow q around cell must be assumed that the cell wall is cut between
be added to produce 50993 in. lb. tor equil- flange members 1 and 10, thus making q zero on
ibrium. This balancL~g shear flow is, the sheet panel numbered (1-10). Then using
equation (7) the shear flaw is calculated in
going clockwise around the panel. Columns 1 to
10 of Table AlS.l show the calculations in
solVing equation (7). FOF explanation on how
Which Is the same as in solution no. 1. to determine sense of shear flows qy in Columns
9, 10, and 11, review Art. A14.6 of Chapter
Example Problem 2 A14.
Table A15 1
1 2 3 4 5 6 7 8 9 10 11 12 13 14
0 0 0 0 0 0 -20.6
1 O. 14 4.396 0.615 -17.41 _2.437
2 0.14 .446 0.902 -13.54 _1.896
0.615 -2.437 1. 175 - 3.240 2.065 55.2 114 -18. 53
3 0.38 7.396 2.810 -- 1. 517 -4.333
9.11 ~3. 462 4.327 -7.795
2.089 --22.794
7.991 5.902 44.2 261 ~ 14.70
constant negative shear flow around cell where AlS.9 Two Cell-Multiple Flange Beam. Symmetrical
magnitude equals About One Axis.
Fig. A1S-28
34.
-.
1199 19.
~
5 - 6 7 8
_~It-,~
AIS.12 SHEAR FLOW IN CLOSED THIN4WALLED SECTIONS. SHEAR CENTER.
shear flow at ~y pOint in each cell is ~~known. zerQ. Thus 'the shear flow in the ver-t ica t web
Therefore, to make the flexural shear flow cc ' Is dete~lned by the stringer c alone,
statically determinate, ~ value for the shear na:nely
flow q in each cell will be assumed at some
paint, and the flexural shear flow for each c'
'lcc ' =-4Z c zA=-4xlX5 =-201b.lin.
cell will then be calculated, consistent with
the assumed conditions. These resulti~g static We can now continue around cell (2) starting
shear flow systems Will, in general, produce a with stringer en where we were ,revlously
different total shearing st~ln around the stopped. _
perimeter of each cell, or in other words,
01'
produce a different cell twist. Since full ~Idr = qb' c '+ 'lee' -4Z,zA
continuity exists between cells, this condition c
= 0 2 0 - 4 x (-5) 1 = 0
cannot eXist, and therefore an unknown constant
shear flow of q:l. in cell (1) and c, in cell (2)
must be added to make the ~Nist of both cellS 4 x (-5) 0.5 = 10 lb./in.
identical. This fact gives us the basis for
one equation and the other equation necessary qe ' e = 10 4 x (-5) 1 =30 lb./in.
tor the Solution of the two unknowns q:l. and qa
is given by the requirement of eqUilibrium,
namely, that the moment of the external and
the internal shear forces about ~~y point in =
Qdc 10 - 4- x 5 x 0.5 ~ 0, which checks
the plane of the cross section must equal zero. the assumed value of q = 0 in panel cd.
In Fig. Al5.31 the flexural shear flow has
been assumed as zero just to the left of The shear flows in cell (2) could of
stringer c in cell (1) and just right of course been found by starting in ~anel cd
stringer c in cell (2). The balance of the where the shear has been ~ssl1med zero and
flexural shear system consistent With this proceeding clockwise around cell as for exa~ple
assumption is calculated as follows:
Clde =a - 4- x 5 x .5 = -10 lb./in.
lOf/in.
Qee' = -10 -4xSxl::a -30 Ib./in.
c
qe'd' = -30 4x (-·5) 1 =-10 lb./tn.
50>/iD. --l'0:" -30./1n. '1i' c' .. -10 - 4x (-5) 0.5 :: 0 lb./ln.
The magnitude of the results are the same
as previously calculated but the Signs are
opposite. As emphasized previously the shear
tlow calculated together with its Sign Is in
Fig. A15-31
the y direction or qy. The direction of the
The general shear flow equation is, shear flow along the cell walls 1n the xz
plane can be determined by drawing Simple free
1000 ZzA = -4ZzA body diagrams as illustrated in Cr~pter A14 but
qy
V
=---". ZzA = -
it Is simpler to use the automatic rule of
Ix 250
Art. A14.11. To illustrate, the solution
Cell (1). Starting in panel cb where the started in panel cb and proceeded counter-
shear has been assumed zero and proceding clockwise around cell (1). By the rule this
counter-clockwise around cell means that the shear flows in ~he xz plane will
have the same sign as qy. Since the signs of
qCb =zero (assumed) qy were negative, the d1r9ction of ~he shear
tlows in the xz plane .vjLl, be counter-cjoctcvtsa
%a = -4ZzA = 0-4x5xO.S = -10 lb./in. around cell (1).
Saa' = -10-4xSx2 =-so lb./in. To obtain the shear flow in the vertical
web eel it was necessary to start at c and go
~rb' = -50-4x (-5) x2;:: -10 lb./in. toward c!. T~iS direction Is counter-cloc~~lse
with respect to cell (2) or clockN!Se with res-
qb'c' • -lO-4x (-5) xO.5 = 0 lb./in. pect to cell (1). qy for this panel from the
equation was found to be -20. If we consider
We cannot proceed beyond stringer o' be- web cc n as part of cell (1) then the direction
cause there are two connecting webs with un- in solVing the s~atlon ZZA was :loskwise wi:h
known shear flows. ~e can get around this respect to cell (1) and by our ~Jle the shear
difficulty by going back to stringer c, where flow in the Z direction will have the 0P9bs1te
the shear flow on each side of c was ass~~ed sign to qy or plus 20, which ~eRns clockwise
" -~,.~. ,s.'~:c .:':_: __ .,~.~:_" .. ~ . . . . . '~:.,,; :<,:,',/: ,~,:~~._. -~~:. '_. ':~, '.,=;,_-. -v-: ''': __-'.'''\.' ':;"-•.:.;-,,,.•..-',- -",".;:.:-~~i;:;
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A15.13
Solving equations (5) and (6) for q s and qa' the reSisting internal torque must therefore
we obtain qa = - 2.0 lb./in. q~ : 5.00 lb./in. equal -3600. Therefore,
Therefore, taking these values of ql and qa
in Fig. A15.32 and adding the results to that qt(l) = .00254(-36001 = -9.17 lb./:n.
of Fig. Al5.31, we obtain the shear flow
pattern of Fig. Al5.34 which is the shear flow SolVing for qt(2)
system for bending without twist about X axis.
The centroid of this shear system locates the
shear cent er ,
_ lr-ac"A, +
qt (2) - z[
31190000
auA] T
In Fig. Al5.34,
+ 10 + -lQ) 100 + .!9. x 20;1
ZV = 0 = -10 x 45 - 10 x 28 - 10 x 27 =- 1000
lb., which checks the external shear of 1000
= 21 Q19..
(.03 .03 .05)
31190000
,03 ~Im= 00245T
.oj' .
lb. ZH = 0 by observation of Fig. A15.34.
hence: qt(2) = .00245x··3600· -8.85 lb/in.
v- 1000,*
8 Therefore, it we add to the shear flow sys-
-5
tem ot Fig. A15.34, a constant shear flow of
-9.17 lb./in. to cell (1) and -8.85 lb./in. to
cell (2), we will obtain the true internal re-
28 Sisting shear flow of Fig. A15.35, which checks
-4
solution 1, any discrepancy being due to slide
rule accuracy.
-5 5 (Values in lb./in.)
Fig. A15-34
-- .-
Cell (1) Final stresses - Fig. A15.35 cell the same, since it any ~Nisting takes
200A _ZgL_(14.17x5)2. (4.17X5)2.54.17X10 place, all cells must sutfer the same amount.
• t - ( .03 ) ( .03) .05 Furthermore, for eqUilibrium, the moment of
the internal shear flow system plus the moment
26.68x10 = 8000 of the external shear load must equal zero.
.03
For bending about axis XX, the flexural
Cell (2) Final stresses shear flow will be assumed as zero at a point
200A _ZgL_(1O.85x5)2. (0.85x5)2.26.68 x 10
just to the left of stringer a in cell (3) and
• t ( .03 ) ( .03) .03 just to the left and right of-stringer c in
cells (1) and (2) respectively. one might con-
-19.15 x 10
• 04
= 8000 sider the cells as cut at these three points •
Fig. A15.37 shows the flexural shear flow under
these assumptions. Since the leading edge cell
A15.10 Three Cell - Multiple Flange Beam. Symmetrical
About One Axis.
(3) has no stringers and the covering Is con-
sidered ineffective in bending, the shear flow
Fig. A15.36 shows a 3-cell box beam sub- will be zero on the leading edge portion since
jected to an external shear load of 1000 Ibs. the shear flow was assumed zero just to the left
as shown. The section is symmetrical about of stringer a. The resulting flexural shear
axis XX. The area of each stringer is shown flow tor the 3 cell section will therefore be
in parenthesiS at each stringer point. The identical to Fig. Al5.3l and the calculations
internal shear flow system which resists the for the flexural shear flow will be identical to
external load of 1000 lbs. will be calculated those in Art. Al5.7.
assuming that the webs and walls take no
bending loads, or, the stringers are the only
effective material in bending. The moment of
inertia about the XX axis of effec.lve ~terial
equals 250 in ~ (Note: this beam section is l~
-lOi/in.
I'
I
10 t/in.
---I
j~
identical to the two cell"beam ot Fig. A15.30
plus the leading edge cell (3). I I 30./in.j
I ~ 50.JI/in. I 20t/in.
I I I~
-
-,
I l:\
laflin. 10tt/in.
L 1(2.01
3'
1(·5) I
5"---1-5"----1- 5" -
b' .03 c'
1(1.0)
Fig. A15-36
d'
1(.5)
5"(-1-5"--1
.03
I(La)
e' sense of each has been assumed pOSitive in each
cell.
r - ---r
If <1 I
"I <3If I
Solution No.1 (Without use of shear center) I
(21
The system is statically indeter.ninate, to I
the third degree, since the value of the shear
flow q at any paint in each cell is unknown. I
Cell (3)
-12.12/F/in. -2.12*/in. -7 2.91 ill/in.
2QA G = ZqL
e t
I'
ZQ x39.4G = 10 x 50 ... 15.71 q ~ ... 10~ _ 1031. f
.0S ~ -;-os- .00 22.91 I/in. ~';
or Q1 =10.5q.-2.55q.+127 - - - - - - - - (1) ( 1)
-37.620/ln.
(2)
-24.97 if/in.
,
Cell (1)
2QA G = ZqL
• t I in.
-12.12*/in. -2.12li"/ in. _7.0911/in.
2Qx10OG = -50x10_2(10X5) +10x20+3(10Q.)
.05 .03 .03 .03 Fig. A15-39
... 10q1. _ 10q~_10qll hence Q:} = 6q1.-q3-1. 67 Q a A15.11 Shear Flow in Beam with Multiple Cells. Method of
-:05 --:ob:-cr;-
Successive Approximations.
- 33.34 - - - - - - - - - - (2)
The general trend in airplane structural
Cell (2) design appears to be to the use of a relatively
large number of cells. There are various
2QA G =ZqL reasons for this trend some of which are:
• t (1) USing multiple interior webS, the detri-
2Q x 100 x G =_ 20.03x 10 + 2 (10.03x 5) + 10.040
x 30 + mental effect at shear defo~tion on bending
stress distribution is decreased; (2) the fail
sate characteristic at the wing is increased
3x 10 Xqa ... 10q,_lOq; hence QJ = 6.25Q. _ because the wing is-made statically indeter-
.63 .04.03 minate to a high degree and thus failure of
1.67q1. ... 20.83 - - - - (3) individual units due to fatigue or shell fire
can take place without greatly decr-eas tng the
Taking moments at the internal shear flow over-all ultlmate strength of the wing;
systems at Fig. Al5.37 and A15.38 and the ex- (3) the ultimate compressive strength of wing
ternal load of 1000 lbs. about stringer a and tlange units is usually increased because
equating to zero:- - column action 1s prevented by the multiple webs
which attach to flange members.
l:Ma = 10 x20 x 10 + lOx 30x 20 - 5 x 1000
In Chapter AS, Art. A6.l3, the method at
+ 78.6Qe'" 200q1.'" 200Q. = 0 successive approximatton was presented by de-
= 3000+7S.5Q.+200Q. +20Oq,=0 - - (4,) termining the t: eetsutng shear flow system when
a multiple cell beam was subjected to a pure
Solving equations (1) (2) (3) and (4,) tor the torsional moment. This method at approach has
unknown q1.' qll' qe and 00 we obtain: now been extended to determine the resisting
shear r Low when the beam is s ub jec t to flexural
q. = - 2.12 lb./in. bending without ~Nist·. USing these ~NO methods
the shear flow in a beam with a relatively large
Q. = - 7.09 1b./in. number of cells can be determined rather rapidly
Q. = -14,.5
1b./in. as compared to the usual method of solving a
number of equations.
Q1 = - 19.9
PHYSICAL EXPLANATION OF THE =HOD
Adding these constant shear tlows to the flex-
ural shear tlow of Fig. A15.37, we obtain the Fig. AlS.40 shows a 3-cell beam carrylng
true internal resisting shear flow as shown in
and external shear load V acting through the
Fig. A15.39.
shear center of the beam section but as yet un-
known in location. In other wordS, the beam
bends about the symmetrical axis X-X without
twist. The problem is to determine the internal
reSisting shear flow system for bending without
twist. In this example, it is assumed that the Now consider each cell as a separate cell.
bending moment is resisted entirely by the The static shear flow qs acting on each cell
flange members as represented by the small will cause each cell to ~Nist. Since zero
circles on the figure, which means that the ~Nist is necessary a constant shear :lo~ q~ to
shear flow will be constant between the flange cell (I), q~ to cell (2). and q~ to cell (3)
members. must be added as shown in Fig. A15.42. and the
magnitudes of such value as to make the ~Nist
The first step in the solution is to make of each cell zero. However. the cells are
the structure statically deter.ninate relative actually not separate but have a common web be-
to shear flow stresses for bending Without tween adjacent cellS, thus the shear flow q~
twist. In Fig. A15.41 imagine each cell cut acts on web 2-1 which is part of cell (1). and
at points a, band c as shown. For the given thus causes cell (1) to twist. Likewise cell
shear load V, the static shear flow qs can be (3) is twisted by q~ and cell (2) by both q~
calculated, assuming the modified section bends and q~. Therefore to cancel this additional
about axis X with nO twist. Fig. A15.41 shows cell twist. we must add additional constant
the general shape of this static shear flow shear flows q:, q; and q~ as shown in Fig.
pattern. A15.43, and considering each cell separate
again. However, since the cells are not separ-
ate these additional shear flows effect the
twist of adjacent cellS through the common web.
-x
~ll Cell Cell As before this disturbance in cell VHist is
(1)- - (2) - -(3) - -X Fig.A15-40 again cancelled or made zero by adding further
closing shear flows q~. q:, q:
as shown in
Fig. AI5.44. This procedure is repeated until
the clOSing shear flows become negligible. In
general the converging of this system is qUite
rapid and only a few cycles are necessary to
% -q;- give the desired accuracy of results.
a b c Fig. A15-41
The total closing shear flows q:.. q" and
e-,
--- 'Is q, are then equal to -
= q~ + q~ + q~ +
-
".~--t ---
1- -I
'Is
q.
q. = q~ + q: + q: +
-
:<I'
I hi q. = q~ + q; + q: +
\
1"2
~
(21 I
___t The t 1nal shear flow on any pane I then
'>':! equals, (See Fig. A15.45)
'Is
". q = Qs + q:. + q" + qs - - - - - - - - - (1)
=(~L
cellular beam as illustrated, and the numbers
in the Table should be lined up with respect =.§£9. = .2336
to the cells as indicated. C a-l. L: L 856
.1:
The solution as presented in Table A15.2
Is carried out in 17 simple steps. The first We are now ready to start the solution
step as given in row 1 of the Table is to proper by successive apprOXimations. In row 5
compute for each cell the value for L: ca J. of the Table, the first apprOXimation is to
t assume a value q' added to each cell which will
where qs is the static shear flow on each cancel the twist due to the static shear flow
sheet panel of a cell; L the length of the when the cell webs are not cut, but each cell
panel and t its thickness. Values for qs are is considered separate or independent of the
taken from Fig. A15.48. other. This constant closing shear flow ql
equals,
For example, tor CellI L
zqgt"
ql = Z.b. •
The minus Sign is necessary
Z qg ~ = 2(36.36 x 10) .~ = 18180 t
The sign is positive because Qs Is posi- because the twist under the static shear flow
tive. (Clockwise shear flow on a cell is must be canceled. The values for q' are re-
positive.) Row 1 in the Table shows the values corded in row 5 of the Table.
as calculated tor the 5 cells.
For example for cell (1),
The second step as indicated in row 2 of ,- 18180 - _ 21.238
the Table 1s to calculate the value of the q--856-
expression L: Lit for each cell.
For cell (2),
For example, for CellI,
qI 3 _---gsa-
27275 -_ _ 28 . 71
z...h.=...lQ.... (10)2. 10 = 856
1. t .064 .04 .05
Steps 6 to 13 as recorded 1n rows 6 to 13
For cell 2, at the Table are identical in operation, namely,
the carryover influence tram one cell to the
adjacent cell because of the cammon web between
z...h.= 10.2(10) • .12.. = 950 the cells. As a closing shear flow 1s added to
• t -:05 ~ .04
each cell to make the cell twist zero when they
The third step as indicated in row 3 is to are considered separate, this result is contin-
calculate the value for the Lit of the common ually disturbed because at the crnnmon web.
web be~Neen two adjacent cells. Gradually these corrective shear flows become
smaller and smaller until the cells reach their
For example, for web bb' between cells true state and possess zero twist. In the
(1) and (2),
Table, arrows have been used for two cycles
to help clarity the operations.
f-\tL) lo_
=..l£..
.05
=200 For example in row 6, the carryover shear
flow from cell (1) to cell (2) is,
The fourth step is to determine the
carryover factor from one cell to the adjacent - 6.700 x .2105 =- 1.414
~ell. The results are recorded in row 4 of
From cell (2) to cell (1), the carryover value
the T2.ble.
1s
Referring to equation (3) for general
expla~ation. the carryover factor from cell (-4.480 - 8.330) 0.2336 =- 3.000
(1) to cell (2) is,
From cell (2) to cell (3),
200
(-4.480 - 8.330) 0.250 =- 3.215
Cl._<;J .. 950 3 .2105
, b' c d' 5 ,
Fig. A15-47
.04 .04 .04 .03 " .03
Flange and Cell
Web Data 10" X .064 (I) .05 (2) .04 (3) .04 (4) .03 (5) .03 x
Fig. A15~48
Assumed Static
Condition ~or
~'
t,
36.36
Cell
(I)
r ut
54.55
(2)
c'
ut
63.64
(3)
d'
ut
72.73
r ut
81. 82
(5)
f '
roo]
,
,
,
1
Shear Flow qs 0 0 0 ( 4)
To
a 36.36 54.55 c 63.64 d 72 73 e 81.82
Table AIS. 2
Row OPERATION
I. Z q" L!t for each cell 18180 I 27275 I 31820 I 48487 I 87880
2. Z L!t for each cell 856 950 1000 I 1250 I 1333
3. L! t of cell web 200 250 250 333
4. Carn OVer Factor (C) .2105 I .2336 .250 I . 2633 . 200 I .250 .2501.266
5. 1st Approx, q' == -EQs Lit!!. Lit ~21. 238...... _-29.7" ~31. 82 .... -38.79_ ~65. 910
Fig. A15-49
Closing Shear
1---1 r~ --t~----tj----tfl---
Flows to Make I II II II II
Twist of Each
Cell Equal Zero
f q'33.51 It q=52.45 II q=63.43 II q=73.91 It q.64.39
I If
----~~-- - tl--- tj-- ---
l 2.65
! 2.10 0.27 1. 18 2.57
Fig. A15-50
f1
Final Shear Flows.
(Fig. A15.48 plus Fig. A15.49)
2.10
, 0.27 t
10 46
.
1. 18
10.48
2.57
15.61
In row 13 of the Table, the carryover condition where the static shear flows in the
values are 50 small that the process Is termin- webs should be much closer to the final values
ated. The final constant shear flow that must and thus hasten the convergency in the succes-
be added to each cell to cancel the twist due sive approximation procedure.
to the static shear flow equals the algebraic
sum of the values from the beginning row 5 to Thus in Fig. A15.51, we have assumed the
row 14. The results are shown in row 14 of the top panel in each cell as cut to give the static
Table. condition. The static shear flow is now con-
fined to the vertical webs and zero values for
The results in row 14 are obtained after a top and bottom sheet. Table A15.3 shows the
·considerable number of multiplications and calculations tor carrying out the successive
additions of numbers, thus it is easy to make a approximations and needs no further explanation.
numerical mistake. To check whether any It should be noticed that atter the first ap-
appreciable mistakes have been made, we take proximation was made in row 5, only three carry
the values in row 14 and consider these values over cycles were needed in rows 6, 7 and 8 to
of constant shear flow in each cell as that obtain the same degree at accuracy as required
causing zero twist if cells ~~e separate. Then in 8 cycles in Table A15.2 for solution 1. Fig.
bringing the cells together, through the common A15.52 shows the final shear tlows which equal
webs causes a disturbance in tHist and this is the constant shear flows in each cell fram row
made zero by the carryover values. This step 9 of Table added to the static shear flows in
in the Table is referred to as a reiteration Fig. A15.51. These values check the results of
and is indicated in row 15. Then adding the solution 1 as given in Fig. A15.S0, within slide
values in row 15 to the initial approximation rule accuracy. In Table A15.3 no reiteration
ql in row 5, which value is repeated in row 16, steps were given. The student should make it a
we obtain the final value of q in row 17. The practice to use such checks.
values in row 17 are practically the same
magnitude as in row 14, thus no appreCiable A15.13 Example Problem 2.
mistakes have been made. If the difference All Material EUective in Bending Resistance.
was appreCiable, then a second, and if needed, The general trend in supersonic wing struc-
even a greater number of reiterations should tural design is toward. a large number ot cells
be carried out. In the Table a second reiter- and relatively thick skins, thus in general, all
ation is shown in rows 18, 19, 20 and the cross-sectional material or the wing is effective
results in row 20 are practically the same as in resisting bending stresses and thUS the shear
in row 17. flow va!ies in intenSity along the walls and
webs at the beam cells. Fig. Al5.53 shows a ten
It will be assumed that the solution was cell beam with web and wall thicknesses as
stopped after first iteration, and thus the shown. It will be assumed that all beam mater-
valueS in row 17 are the constant shear flows ial is effective in bending. The shear flow
that must be added to the static shear flows resisting system for bending about the horizontal
to produce bending without twist. Fig. AlS.49 axis without ~Hist will be determined. The cen-
shows these final closing constant shear flows. troid of this system will then locate the shear
Adding these values to those in Fig. A15.48 we
obtain the final shear tlows in Fig. AI5.49. center.
Fig. A15.54 shows the static condition that
The lateral location of the shear center has been assumed. namely, that the upper sheet
for this given 5 cell beam coincides with the panel in each cell has been imagined cut at its
centroid of the shear flow force system in midpoint, thus making the static shear flow zero
Fig. Al5.50. The calculations for locating at these points. The static shear flow values
the shear center are given below Fig. Al5.50. Qs are shown on Fig. A15.54. To explain how
they were calculated, a sample calculation will
Solution 2 of Problem 1 be given.
In solution 1, the assumed static condi- The moment of inertia at the entire cross-
tion involved cutting all vertical webs except section about the horizontal centroidal axis lS,
the right end web. Thus the static beam
section became an open channel section and the Ix for top and bottom skins,
resulting static shear flows must obviously be = (50 x 0.125 x 2.5')2 .. 78.0
far different t~ the final true shear flow Ix of all webs = (0.912x5~)/12 = 9.5
values, since the webs always carry the greater Total Ix ::; 87.5 in"
shear flows in bending without twist. This
fact is indicated by the relatively large num- For convenience an external shear load
ber of steps required in Table A15.2 to reach Vz ::; 8750 lb. will be assumed ac~ing on this
a state where successive corrections were small beam section.
enough to give a desired accuracy of final re-
sult. Thus it is logical to assume a static Hence, q =-~ZzA
Ix =_875
87.~
9 ZzA =-100 ~zA
I ~
. .)
,. .......1
AIS.22 SHEAR FLOW IN CLOSED THIN_WALLED SECTIONS. SHEAR CENTER.
Now consider Fig. A15.55 which shows a Starting at point (al in cell (1) where the
sketch or cell (1) plUS halt or cell (2). As shear rlow is zero and going counter-clockwise
previously explained the upper cell panels were around the cell. the static shear flows are
assumed cut at their midpoints (a) and (ml. as follows: -
Solution II out
0 0 I 0 '~ 0 0 0
I
out out out out
Fig. Al5-S1
Assumed Static
Condition foE" 36.36 (1) 18.18 (2) 9 09 (3) 9.09 (4) 10.09
01 (5) 18.18
Shear Flow eu and
ReIlU1t1ng lis Values.
0 0
1 I
Table A1S. 3
Row OPERATION
Fig. AlS-52
F1na.l Shear Flows
2.87 2.081 0.211
I 1.206
2.5 S
Fig. Al5-53
.
g (1)
.
~
~
(2)
..
~
~
(3)
.=
~ (4)
.
g (')
..
i!: (') -
=
~
(7) ;;;
= (S) "~
~ (9) "0
~ (10) 8T
5"
-
Bottom Skin - • 125 Inches Thick I--- 5"--1
1
-- -100 - - -1" - -
78 7So 78 7S o 78 78 078 7S o -
7878 0 78 78 0 7878 0 7S 7S o
-1" - -15"
j
78 a 7878 a 78
- ;7~5~f
_7 8 -rse -m '-.156f- .,..156
Fig. A15-54 _9
" "•.. 185. 185.4 IS5.4 18S.4f 181.3 181.3 178.5
_7
-0-
_78
-0- -0- -7878
0_
_n78 -~m
_0 _ _ 0_
-7878 .78m
_ 0 _
-7878
_0 _
.78m
_0 _0
Table AIS. 4
I,
..0.89
-10.42
-3.21..{).36 -1.01-0.32 -Q.38..o.81 ..0.3 -2.56..0.8 -1.34 -2.74 1.53 -1.241.S3 -1.6 7.15
0
_3.57
0 0 0 -6.92 0 _5.63 0
1.70
20.91
-11.31 -1. 33 -1.17 _2.8S -9.07 -4.27 -5.04 5.55 22.61
·89 31 ee.es .JH.57 74.H .79.33 76.67 ·7".17
- ". .
-
15.12_87.07 ~8.93 ·~2.Z7 13.73_~3.04 7~9
- .,- -
83.55 _55.39 [00,61
109.3 t 17'1'.66 183.16 185.24 187.11 191.59 176.SU 182.07 167.9\ 181.H 120.51
Fig. Al5-S7.
- ."
Flnal Shear Flow Values.
--
ee.es ·81.' 14.4' .79.33
-
76.87.79.17
- -
78.33 .so.se - - --
rs.ra .a7 ,~7 5R.93 ·aZ.27 .3.13 _d3.04 ;2.96.72.45 - - -
U.S5 _55.39 100.61
(Note: Shear Flow at Ends of Webs Equal Sum of Shear Flows in Adjacent Sk1n Panels. )
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A1S.23
Procaading to (g)
cut
TXT
b a h
qg = qhg+100Z~ = 156+100 (1.25x2.5x
h
.094) = 156 +29.4 = 185.4 1b./in.
5" c - - (1) - - - - g J - - ( 2 ) - - rZA
qrg = 185.4+10OZ = 185.4+100 (-1.25)
11 Qa =0
d
Fig.Al5-55
g
(2.5 x .094) = 156 1b./in.
Due to constant clOSing shear flows as Fig. A1S.58 shows a Single cell distributed
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES
A15.25
flange beam. Consider the beam acts as a canti- flange members, see Fig. A15.60. Again ZFy =0,
lever be~~ with the bending moment existing at
section (A) being greater than that existing at whence ~P3 + ~P~ + qyd =°
secti~n (B) and that the bending moment produces
compression on the upper surface. By the use of _ (6P,. 6P.)
orqy-- d
the flexural stress equations, the bending
stress on each stringer can be found, which if
multiplied by the stringer area gives the Therefore starting at any place where the
stringer axial load. Thus at beam section (B), value of qy is known, the change in the average
let p~, P~, P3 , etc. represent the axial loads shear flow to some other section equals
due to a bending moment M. The external bending
- - (1)
t!A
~A
'i
B
~ 20" -+- B
30" ----I
10--ro" ~ bl"J"
Ole Fig. AlS-83 Fig. A15-84
~-l c' I b' ...1.. Solution No.2. Considering AP Loads in Flange
. k" +."~ 11-5"+5"..,1 Stringers. (£guation 1)
SECTION A-A SECTIO;," B-B TO "'-C Bending moment at section AA = 1000 x 50 ::: 50000
I x=SxS 2x2=250in4 Ix=3x5 x2=150in
in.lb.
Fig. Al5-62 Bending moment at section BB = 1000 x 30 ::: 30000
obtained by increasing only corner stringers b in.lb.
and c. The shear flow on section A-A will be
computed using equation (3) which applies only Considering Section 8-B:
to beams with constant section and also by Bending stress intensity at midpOint at
equation (1) which applies to beams with vary- stringers by the tlexural formula:
ing moment ot inertia.
- M.:, _ 30000 x 5 - 1000 pst.
"b-I 150 -
Solution 1. USing Shear Flow Equation for x-
Beams at Constant Cross-8ection. Axial load in each at the stringers a, b, and
(Equation 3) c = 1000 x 1 = 1000 lb.
d'~h,4
/ J'" I";""'~ /Iw'
'lac
'leA = -10 e • b
1000
'tcc ' • -10-25QX5x2=- 50 Ib/in. l ~V '~~ .. -
Proceeding around the cell the balance ot the
shear flow ccufd be ca.Lcu.Ia'ced , but due to sym-
/ .!'I;;~I ;§Or ":lIb'
metry enough values have been found for the ~ Fig. Al5-8!i ~ Fig. Al5-66
shear flow to draw the complete shear flow piC-
ture tor bending about the X axis when it is Having tound the AP flange loads over a
assume~ that one-halt ot the area of stringer length (d) of 20 w , the shear flow can be com-
(a) acts With each adjacent web. Fig. A15.63 puted by equation (1).
shows the resulting shear flow diagram. The
resultant at this shear tlow pattern is a It will be assumed that one-halt at the 4P
1000 lb. torce in the Z direction and its load in stringer (a) will tlow to each adjacent
location through the midpoint at the box since wet. However, there 1s no AP load in stringer
the tlow 1s symmetrical. The external load at
1000 lb. also acts through the midpoint or the
(a) hence Qab Qac = =
O. Then from equation (1),
cell hence the external load 1s in eqUilibrium qbb' = O-Z~= O-~= - 50 lb./in.
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES AlS.27
t.Pb' -1000
qb I a I == -50 + Z -d- :: -50 - 20 == 0
should be noticed that the true shear flow is Thus having the shear flow on the stringer
greater in the top and bottom skin than that edge of the sheet panel, the shear flow on the
given by equation (1) which applies only for large end of the tapered panel can be found by
beams of constant cross-section. equation (5).
A15.19 Example Problem of Shear Flow in Tapered
Multiple Flange Single Cell Beam..
8.5"
L , a'
Ix=271
b' c
Sec. SU. 90 qX(ab) = 24.87x5.5/6 =- 23.2 1b./in.
3.25 2.5 75
qY(bc) = 24.87-471/30 = - 40.57
1000f
qX(bC) =- 40.57x5.5/6 = - 37.2
40.57-196.6/30 = - 47.12
1
10"
23.2 37.2
40.0~
t
j
Fig. A15-74
J 88.9
133.3
'23.2 b' 37.2
8" 1
lOa
8"
-
;
j -
66.7
44.4
Fig. A15-73
/ I
746 471
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A15.29
The forces in Fig. AIS.74 will be checked Taking moments about point a of the forces
for equilibrium. in Fig. Al5.74 but replaCing the shear flows in
the top and bottom panels by the values found
ZFz = 1000 - 266.6 -200 -133.4 -10 x 40 = 0 above, we would obtain an unbalanced moment at
7970 in. lb. instead of 8670 preViously found.
ZFx = 0 by observation. The correcting shear flow would then be g =
-7970/240 =
-33.2 instead of -36.1 as previously
Take moments about stringer (a), rcuno , The tinal shear flow pattern would be
as shown in Fig. AIS.76, which values should be
l:Ma = -3 x 1000 + 200 x 6 + 133.4x 12 + 10 x 40 compared to those in Fig. A15.75.
x 12+(+88.9 - 44.4)10 + 23.2 x 6 x 10
+37.2x6xlO = 8670 in.lb. i'.4 2.1
'TD
Let VZ(net) be shear load to be taken by flow diagram..
cell walls, 400 lb.
This corresoondS to value ot 23.2 in (4) Flgs. Al5.79 and AlS.eO show two loaded
previous solution (see Fig. A15.74). single cell - 2 flange beams. Assume the flanges
develop all the bending resistance. Determine
qY(bC) = 17.8+~gX5X3 = 31.1 lb./in. the shear flow reSisting pattern by two solu-
tions, namely, without and With use or shear
As canpared to 37.2 in Fig. AlS.74. center.
qy(CC') = 31.1+~gx5x2 = 40.0 lb.ln. (5) Fig. AIB.Sl shows a Single cell - 3 flange
beam SUbjected to loads as shown. Assume the 3
which is the same as in Fig. Al5.74.
-j r
)
TC
A15.30 SHEAR FLOW IN CLOSED THIN-WALLED SECTIONS. SHEAR CENTER.
flanges develop the entire bending resistance. (11) Fig. A15.84 shows a ~ultiple flange-ctrcu-
Determine the internal shear flow reststing lar beam section. Find the reststing shear
system. flow pattern when carrying the external shear
300 lb. 60"" load of 5000 lb. as located in figure. Assume
100 lb. 1,1
r~.02 f
45°
10"
Q
A
.0' '.02
cell skin ineffective in reSisting beam bending
stresses •
t1000 lb.
l-_~
.8....., I"
I
bl--
'\_1"
B
l -F-
.03
( . ' \ 500 lb • 0.5
tV= 3000 lb.
.1
10'Y
r 81< "'"1
rU
.04 I
Fig. A15~81 Fig. Al5-82
.03
.3 .1 .1 .1
(6) Find shear center location for beam in
Fig. AlS.81 if the 3 flanges provide the entire [-'0'1'0''"1 I 4 panels a 5" = 20"
bending resistance. Fig. A15-85 Fig. A15-86
(7) Find the internal resisting shear flow (12) Determine the shear flow resisting system
pattern tor the 3 flange-single cell beam of for the beam section of Fig. A15.85. The 6
Fig. AlS.82. Assume webs or walls ineffective flanges r~ve areas at 0.2 sq. in. each. Skin
in bending. is .032 thickness. Assume skin ineffective in
bending.
(8) Determine shear center location for beam
at Fig. A15.82. Webs and walls are ineffective
in bending.
.3
20
1 00* .4
(13) Find the shear flow resisting system for
the unsymmetrical beam section in Fig. A15.8B.
Flange areae and skin thicknesses are given on
figure. Assume skin ineffective in bending .
Top Skin .03 (14) neterrune shear center location for beam
.04[ section in Fig. A15.SB.
Fig. A16-a3 .035 10"
Bottom Skin .03 T f 500 lb.
.3
f--- 8 panels at 5"
.05
.03
1--12"--+-8" -I
[:303
.03
.04 $02 I
i 8" 1 1000
lb.
flow equaticn for
lat! .SCI" .5C1" ;-12.5''1 beams of const~~t
I 04 Cell Cell
. Cell Cell 03
section. Campare
the results. All six
12" . (I) .05 (2) .04 (3) .04 (4) . flange members have
.L .03 .03
Fig. A1S-92 0.2 sq. in. area
each at section A-A
and taken uniformly
1 - 1 2 " + 8"--j to 0.1 sq. in. at
end c-c,
FIg. A15-89
1000f
leaving a 2 cell beam section. Find the 'S
resisting shear tlow system under the et r5",5"1
,r 1,
gl ven loading.
j ~iH
-r-r- ,
(b) Now add the lett curved sheet to form -=---120'~'- - - - '
cell (1) thus giving a 3 cell beam. Find
shear flow system.
(22) Add two interior webs to the beam of Fig.
(0) Now add the right curved sheet to A15.9Z, connecting flange members a-a~ and b_b n ,
form cell (4), thus giving a 4 cell beam thus making it a 3 cell beam. Find the shear
section. Find the shear flow system. flow resisting pattern at section A-A by the
liP method.
-x- f·1 ~
S;,-.I!R1S!3}0G 10 .1 -x-
ing stresses.
-~c:-=-::-;-;::;""---I
\ 0.1
7 cells a 6" : 42"
.08
Fig. A15-91
-'
i:t:L
'I
AlS.32 SHEAR FLOW IN CLOSED THIN-WALLED SECTIONS. SHEAR CENTER
.:1,f!;1
~£!W.~~~::'~'~ ~
;~'i
DC-a Fuselage Main Cabin Framing Structure
;;r:;:~s.!:.~, "':~~' -.~. ,~:.,,·~;~-,;i Also Window Belt Treatment
~·.;..t. " ~ :.,. :~r~;::·;-l~J;i;;;.iMf·:;:;·S!:~
•
&
CHAPTER A·16
MEMBRANE STRESSES IN PRESSURE VESSELS
ALFRED F. SCHMITT
Al6.!
i ../r-,
i -.;- '/
Al6.2 MEMBRANE STRESS:c.S IN PRESSURE VESSELS
absent due to the symmetry of the problem. The its support, Is neglected.) In the lower hem-
included angles between the pairs of nerldlo~~l
and tangential rcr-ces are :is!ll/Hm and dSt/:Rt,
respectively.
Summing forces normal to the dlf:erential
element, one has
• •
or,
Fig. Ale.4a Fig. A16.4b
p - - - - - - - - - - - - - - - -
Here p is the Internal pressure, positive out- ispherical portion meridional strssses are re-
ward. Note that the shell wall thickness does quired as shown in Fig. A16-4b. Hence, in this
not enter eq , (1). The pressure p may vary in class of problems it is best to derive the nec-
the meridional direction but is constant in the essary Second equilibrium equation (correspond-
ing to eq. 2) by considering the individual
tangential direction by hypothesis (rotational
symmetry assumed). characteristics of the struct~e.
Eq. (1) is one equation containing two un- Al6.3 Applications to Simple Pressure Vessels.
knowns. Another equation may be obtained by the
condition of equilibrium ot a portion of the Example Problem 1
shell above or below a parallel Circle. ThUS in Determine the membrane stresses 1~ a cylin-
Fig. A16-3, the pressures acting dmvnward on the drical pressure vessel of circular cross section
lower portion at the shell are equilibrated by (radiUS Ro), haVing hemispherical ends, i! the
the upward vertical components at the meridional internal gas pressure is p. Also find the
stresses, Nm. greatest combined normal strese.
Fig. Al6.5
Nm = P Rt - - - - - - - - - - - (2)
2
Nt = P RQ
R, 2.7 Ho P
Stress =Nt =p Nb = 2
t t
Therefore
The meridional stress in the cylindrical portion
Is found from sq. (2); Nb cos ~ = 2.7:, p (.92.)
p R, = 1.25 Ro P
2
Finally, the compressive ring load, F, is (ret.
Stress =P R o Fig. A16-6c)
2t
Since shear stresses are absent everywhere 2F = r:1.25 R, p sin 9 R, d 9 = 2.5 R~ P
in the meridional and tangential directions)
these are the directions of principle stresses.
Hence the greatest combined normal stress Is In computing reinforcing ring stresses
identically equal to the greatest meridional or tram this result it is necessary to include
tangential stress as just computed. It Is seen some effective skin frmn the adjacent shell
to be walls when the ring cross sectional area is
figured.
"max = p R,;t .
Example Problem 3.
~ple Problem 2. Another form of bulkhead used to close a
.~ important problem in pressurized cabin circular pressure cylinder is elliptical in
design concerns the Shape ot the end bulkheads. section as shown in Fig. Al6-7. Such a bulk-
;fflhl1e hemispherical bulkheads (such as used !n head shape provides tangential meridional forces
Example Problem 1) are highly desirable from a at the seam (requiring no reintorcing ring as in
stress standpoint, such fQ~ are uneconomical the last example) and yet is reasonably s!flcient
as regards space utl11zat~on. on the other ex- as regards space utilization. Problem: de-
treme, a flat bulkhead, while ?roviding far more termine the membrane stresses in such a bulk-
useful volume, cannot resist the pressure load- head.
ing by membrane stresses and hence 15 structur-
ally inefficient. A compromise configuration is
that shown in Fig. Al6-6, in which the bulkhead
is a spherical surface of low curvature,
(~dished head~) supporting the pressure loading
by membrane stresses. A reinforcing ring,
placed at the seam, resists the radial component
of these s-tr-eeaes . Problem: find the com-
pressive load acting in the reinforcing ring. Fig. A16.7 1..
J""'~
,~
,1,
A16.4 MEMBRANE STRESSES ill PRESSURE VESSELS
The meridional stress ts found from eq, (2). Summing forces horizontally at the floor joint:
Thus,
Nt = P (RU cos al, + RL cos a.~) •
_ P Rt _ p ("a:..'..;.Y,-'-,--:,--,,--,-,--
Nm--2--~- Assume all stresses are equal and are given by
s.
Substituting the expreSSions for Rm, Rt and Nm
into eq, (1) one r inds Su = s = P Ru/tu
SL = S = P RLitL
Sf = S = f:t (RU cos al, + RL cos a, )
or particular interest are the stresses at Letting the weight denSity of the material
the seam. Here y = a and x = a. One tinds be wJ the weight per unit length (axially) along
the cabin is (w times the developed length of
walls and floor).
Nm=~
w = ',V 12 tuRu (n - a,) + 2tLRL (n - a.)
Nt=pa(l-~) + 2 tt RU sin a:l. } •
This last result is import~~ since it in- Solving for the various thicknesses tram above
dicates that compressive tangential stresses are and substituting, one finds (to obtain a result
possible it a> b VZ. As will be seen below, symmetric in appearance use was made of the
such a situation is undersirable because at the tact that RU sin a:l. = RL sin aa)
large resultant difference in radial expansions
between the cylinder and bulkhead (the bulkhead
actually contracts radially it Nt is negative) W =Z; 1RU [n - (1:l. + ~ sin 2 a:l.) +
producing high seconcary bendIng stresses in the
vessel walls.
R~ (n - a, + ~ sin 2 a,) f
Example Problem 4. Since one may show that the cross sectional ar-ea 01 this
Determine the weight per unit length or the fuselage Is
double-cyllnder tuselage cross section as a
function ot the internal pressure J allowable A z
,r.
RU\'1t - a1 +
1
'2 sin ),1\1t
2 U:l. t RL • 1. )
aa + '2 Inn 2 a ••
stress and the geometric parameters ot Fig.
AlS-8. For structural efficiency it is desired an important consequence 01 this calculation is that the ratio
of shell weight to shell volume 1s
to maintain equal membrane stresses in the skin
and raccr-, W. 2wp
A --.-
and Is therefore independent of the combinations of Re, RL,
U:l. and a. used. The designer 1s thus free to choose these
shape parameters 80 as to satisfy other requirements.
Example Problem 5.
Determine the membrane stresses in a
conical vessel of height h and halt apex angle
Fig. A16.8 a. The cone forms the bottom of a large vessel
f1lled With a liquid of speCif1c weight w and
havtng a head of l1quid H above the cone. The
compt.ete..unt t is supported from above.
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES AlG.S
SOLUTION:
(IJ. Poisson's =
ratio, = .3 for
alum1num)
The meridian cur-ve or the cone has a
radius of curvature Rm » cc ano , at any point a
=~(l-~)=
P R,
distance y down from the top at the cone, the .S5
radius Rt is Et 2 Et
N - ~ - w (H + y) (h " y) tan a
t - p - cos a.
n (h _ y)& tan a a • w (H + y) + P R,
EtSLKllD = 2Et (1 - ,.) = .35
+ w ~ (h - y)' tan' a • {h - y)
Hence, by integration
Solving P R,
= .35 "'"'Et""
tan a
""m ="wG - (h - y) [ H + y + -1 (h - y) ]
cos-a. 3 Thus the cylinder tends to expand more than
the bulkhead - a situation prevented by the seam
bevHeen these elements. It follows then trzt
Al6.4 Displacements, Boundary Conditions and Local the seam experiences a transverse shearing action
Bending in Thin Walled Shells.
It is appraoriate at this point to examine as indicated in Fig. A16-ll. These shear forces
some of the raresaing illustrative cases to de- in turn produce bending moments in the sr.ell
termine Whether or not the membrane stresses wall as shown on the figure.
computed gave ss.t i srac tcr i tv accurate measures
at the shell stresses. Anticipating the answer,
we state trat, while the membrane analysis will 6'olkhd. •
~oTl
give the primary streSS system in a shell-like
pressure vessel, a careful (and often lengthy) T
analYSis of induced bendi~ caused by boundary
effects will reveal localized ,secondary stress
peakS. In static strength analyses of properly
designed- vessels it is the practice to neglect
~ 1'ill-ijjj-,=.....=..;:...::.1
- various codes and standards give proportions ol common Fig. Al6.l!
vessels which will correctly limit secondary stresses.
See for example reference (1).
1("
1_, L
A16.6 MEMBRANE STRESSES ill PRESSURE VESSELS
While it Is not our purpose here to take up shell bending 13% abav o t~e membrane stress (as against only
in detail, some indication of the character and magnitude of 3Sb above for the nemtspber tcz,; bu.lkhead ) (ref-
these bending stresses should be available to place them in
proper perspective. The most striking thing about t.hese wall
erence 2, p , 410). For this type of bulkhead,
moments Is that they are quickly damped out, becoming neg- boiler codes sometimes pemit a ratio of alb as
ligibly sma.!!Jdown to l';b of their maximum value) at a distance high as 2.6, however.
of about 4~ from the seam. Thus. for an instance. In a
circular cylindrical shell of 40" radius and .065" will thick- A16.4 Special Problems in Pressurized Cabin Stress
ness, these moments are so damped at 6.5" from the seam. Analysis.
-------
urized tank. The action is representative of a
fuselage with widely spaced rings inside.
_____ -~,"" pA
1d__-.,::-_-_-_-:.-_-_1r
. _ ~ _ _
I <:"l .I
j .' j
Al6.8 MEMBRANE STRESSES IN PRESSURE VESSELS
In examining the figure to dete~ine wr~t Fig. A15-16 shews the typical cutout
sort of canceling stress system must be supplied, structural arrangement. ~hile analytical
we see that the tangential hoop stresses border- approaches have been tried, it is probably
ing the cutout cannot be canceled by a self- safe to say trAt the ~rue elastic stress dis-
eqUilibrating set since they have a radial com- tribution in such a configuration cannot be
ponent. However, the radial component of these computed. The necessity for avoiding high
stresses will actually be supplied by the door intensity stress concentrations (With their
or window pressing oU~Nard against its fra~e. attencant fatigue likelihood) makes empirical
Hence, it is only the component of the hoop inforTation mast useful in such cases. On
stresses along a chord which need to be canceled the other hand, a simple rational analYSiS,
(Fig. A16-14b)·. based on principles outlined above, will very
The llmmediate problem bec~mes one of de- likely suffice tor a static strength check and
Signing a structure to effectively support a set tar most design purposes. (,:"150 see reference
of uniformly distributed self-equilibrating 8, pp. 16-23).
stresses acting in the plane of the chord con- The above discussion has concentrated
necting the upper and lower edges of the opening attention on the problems of carrying the hoop
(Fig. A16-15a). stresses around a cutout. The longitUdinal
pressure stresses, while being s~Aller them-
selves, are intensified by bending stresses
fram the tail loads. Hence, the longitUdinal
stresses across the cutout ~y make this con-
dition (or the combination) most severe.
Fig. Al6.15
,. ~ '. ~ .: - ~ '"0 .. ; ... ; .-:" "-_·' ... ~.F , ••' • , • , .,'~. ~ .. ' , "' • • -•• -,_ ~~, " ~, • -~., 0 _. \ ••
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A16.9
Douglas OC~8 Under Ccaatructton and Assembly. Fuselage is Pressurized to Permit High Altitude Flight.
~:,
"
At7.1 Introduction.
It was seen in the last. chapter that thin with another SUCh, we write the equations down
curved shells can resist lateral loadings by by a direct appeal to past experience and
neans of tensile-compressive membrane stresses. intuition.
As w1ll ~e seen later, thin flat Sheets, by de- Fig. A17-l shows the differential element
flecting enough to provide both the necessary of a thin, initially flat plate, acted upon by
curvature and stretch, may also develop mem- bending moments (per unit length) Xx and My about
brane stresses to support lateral loads. In the axes parallel to the y and x directions respect-
ar~lysls of these situations no bending strength ively. Sets of tNisting couples Mxy (= - Myx)
1s presumed in the sheet (membrane theory). also act on the element.
In contrast to the membrane, the plate 1s
a bvo-dlmenslonal counter?art of the Jearn, in r
which transverse loads are resisted by flexural
and shear stresses, with no direct stresses in
its middle plane (neutral surface).
The skin may also be classified as either a
plate or a ~e~brane depending upon the magnitude
of transverse deflections under loads. Trans-
verse deflections of plates are small in compar-
ison with the plates' thicknesses - on the order
ot a tenth of the thickness. On the other hand,
the transverse deflections of a membrane will be
. /
yx
r
Fig. A17.l
•
on the order of ten tL~es its thickness.· As in the case of a beam. the cu~rature in
Unfortunately for the engineers' attempt at an orderly the x, z plane, a a.N/aX.:l, is proportional to the
cataloging of croclems, most aircraft skins fall between the moment Mx applied. The constant of proportion-
above two extremes and hence behave as plates having some ality is l/EI, the reCiprocal of the bending
membrane stresses. stiffness. For a unit Width of beam I = t 3 / 12 .
Plate bending investigations ~Ave for a In the case Of a plate, due to the Poisson
longttme been important in aircraft st~uctural erfect, the moment My also produces a (negative)
analyses in their relation to sheet buckling curvature in the x, z plane. ThUS, altogether,
problems. Recently they have assumed new im- With both moments acting, one has
portance With the introduction of thick skinned
construction and still more recently With the
use at very thin low aspect ratio wings and
control surfaces which behave much like large
plates, or even are plates in same cases. Where ~ Is poisson's ratio (about .3 for alumin-
It is the purpose of this crapter to pre- um) • Likewise, the curvature in the y, z plane
sent briefly the classic plate formulas and Is
same applicattons. Appropriate references are
cited in lieu of an exhaustive treatise, which
could hardly be presented in one chapter (or
even one volume) as witness the voluminous
literature on the SUbject. These two equations are usually rearranged to
give the moments in terms of the curvature.
A17 • .2 Plate Bending Equations··. They are written
Technical literature in this field abounds
with many excellent and elegant derivations of
the plate bending equations (references 1 and 2,
aX
i" ay" -
w) (1 )
tor inStance). Rather than labor the subject
(and visa versa)···. It ~s proportional to the In a similar manner, a ~oment summation about
~Nistlng couple Mxv ' A careful analysis (see the x axis yields
references 1 and 2) gives ~he relation as
Q =al1y+~ (6)
l1x y = D (1 - ~) axdy
a' w (3 ) y ay a x
1i /,'..
~ ~/
~4Z
:?-"'" '~~41C~.,
K~.4IC~
»'
..-&P::'I:I":;!';~'
V---..#
~L
::;w
Geomstry
Structural Bending
Characteristic Stiffness
De!lections
a'"'w
Distortions axa ,
0- 12
o-W
~
E:t~
•
aa.,Of
• dxdy
11 - \-'0 ")
I ~
y
ax
EI
,,(
"p
tIC.,
:2
(a)
~ ,)
(b)
' ..... Loa<Ungo
ccceree
oar,
t Latera q
x' ,
M
Fig. An. 2. The differenti3ls are increments which "Hooke's l1oment- !1x = D ('"
ax-").l. ,·W)
ay:r
should be written more precisely as, for instance, La.' Distortion
4Q r • (aQr/eY)4r. Relation a aw oa.,Of) ~ day
My = 0 ( ii? ..... ax" 11"::.I dX'"
The next relations are obtained by summing Mxy =0 (1 - ;..) axdy
,,,
moments in turn about the x and y axes. For ex-
ample, we visualize the two loading sets of Fig.
A17-2 acting s11nUltaneously on the single ele- Equilibrium Moments v = d i'!
ment, and sum moments about the y axis. cx
DiViding by dX dy and discarding the term at Finally, one verJ l.m.portant equation is
higher order gives obtained by eliminating all internal forces
(Xx, My, Mxy, Qx, Qy) between the above six
equations. The result (whfcn the student
shoUld obtain by himself as an exerCise) is a
or, relation between the lateral loading q and the
deflections w·:
rv.: = a l'lx + a l1x y _ (4)
-sx ax a y (7
m, n = 1, 2, 3,
(8)
Mx =- n
• .
(a'""" b'~)' (~'a'-+ "b'r.') s i nm-an-x sin -o-
;lL +
r: ;,: y
My ;:
cmn
- n' (",? +~)' o .n")
m'
7'"" b' Sin~ sin n a y
a --,---
em (1 - :..L) mn mn x n
it y
l1XY =n' ("'+n')'
~ ~
ss cos a cos - b -
q a'
"!lAX = jl to - - - - - - - - - - - - - - - (13) • pp. 55-84, 257-287.
• -p. 128-132.
&
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A17.5
6 = J} ~ (:YJ + dx - a
We note that the limiting case, alb = 0, cor-
responds to the one-dimensional case analyzed
earlier .. Unfor'tunaue Ly, an extraporat i cn of
these ~No-dimensional results to that limit
does not show agreement with the one-1imen-
= ~ n:r dx
sianal result. Presumably the jiScrepancy way
be traced to the excessive influence of inac-
curacies in the assumed deflection Shape Jf the
~embrane as used in the approxDnate two-dL~en
Substituting through the use of eq. (15) and in- sional solutions.
tegrating we find Experimental results reported in reference
4 show good agreement with the theory ~or
gJi a 3 square panels in the elastic range.
O=Z4s't'
TABLE A17.2
Now by elementary considerations I
Membrane Stress and Deflection Coefficients i
o ; s a
E alb 1.0 1.5 2.0 2.5 3.0 I 4.0 I 5.0 i
approx.
----
- -
Now solve eq. (18), the membrane relation, tor
q; call it q",
'0
..
o•
10
a
' ./'
,,.-'
:.--
e---
q = q' + q"
o
o
/
-:.-
100 4 1'0 '00
1 Et 3 1 Et :5
Fig. A17. 7 Large·deflection theories· mid-
q = -a - - b - '''l1AX +,--,-.."..
a40 (_)6 n, a
Ww.x - - - - (20)
panel stresses; simply supported square panel.
a
Because ot the obVious desirability at
Eq. (20), we see,. is based upon summing the in- using the results of the more exact theory, s~e
dividual stlttnesses at t~e two extreme be- or these are presented in Table A17.3. The
havior mechanisms by which a flat sheet can treatment at additional cases (other types ot
support a lateral load. No interaction between edge support) may be found in reference 6, pp.
stress systems is assumed and, since the system 221, 222.
1s nonlinear, the result can be an approXimation
only.
Eq. (20) is best rewritten as TAaLI Al1'.'
,.
(lIILIIGnD P....... l..oui ('ll ~, """'..... ~I
q "
E t 40
= U"'MAX)(~)'
a \~ b
+ 1
~
(WMAX\'
t 7
_- - - - ( 2 1 ) ...
I
- I
or Slr_
c
_It I 'L' •
,430 1'810
"... .. n
I.U
,.
P-..C.,.rtb:1
, ,.n
.. t/D4
no
,.~
,n
,... - ....-
I.U '.n
L. ..• ..• I ...
Fig. A17-6 shows eq. (21) plotted for a i ,,~/1\~ 3.110 $.10 1.1'0 lD.ao ,u 14.:10 ,1$.10 11.011 II. III
square plata USing values at a. and TIl as taken I ~/nal 0.10
,..
•... ........
I 1.10 1.10
, .ee '-" 10. :10
from Tables A17.1 and Al7.2. Also plotted are
the results of an exact analysis (reterence 5). 11 ta.- _It I .IM ·_11.24 1.44 1.1'11 I.M '.M "
As may be seen, sq. (21) is somewhat conserva- I 1 ,U l'.~
--
'ab / gJ i 4." 1.11 jlO.)O 14• • 1'.4
11•• 0
tive inasmuch as it gives a deflection which is
too large for a given pressure.
II
_.'" • 1-
L
.,
y 11E11 1 1. 2I
' ..
'M' ..""'..II
1.40 4.1' '.11
.n.... •0lIl_01. ..........
_ _ 01.
1.'1 1.10
..
II.U 10.10 10.110 L1.»
II
.xac~
&pprox._
---
'00
"0
VI A17.8 Considerations in the Applicationa of Large-De-
neeuce Plate and Membrane Analyses.
Before concluding this chapter it 1s
~ pertinent to note several serious omissions in
the developments outlined above with regard to
»:
100
their application to flat pressure-panel
analyses within a ship hull or fuselage. The
o
o
~-
0.' vlt
-'
1.0
.~~i.~:!..~
l.~
-
2.0 • but 1181ng n2 • . 260 in eq, (19), This value gives the
stresaes at the center of a square panel whereas n2 •
Fig. A17.6 Deflections at the midpotnt 01 a simply .356 in Table A17. 2 is for stresses at the panel edge.
supported square panel by two large.deflectian
theories.
,c<
,O~
A17.8 BENDwa OF PLATES
PART 1
ELASTIC AND INELASTIC INSTABIUTY OF COLUMNS
Ala. 1 Introduction, Thus the total bending moment at section z
will be:-
Part 1 of this chapter will be confined
to the theoretical treatment of the instability M= Pu + Q~Z on upper portion - - - - (d)
of a perfect elastic column and an imperfect
elastic column. The column Is the Simplest of
the various types ot structural elements that 11. Pu + Q(l-a) (l-z) on lower portion (e)
1
are. subject to the phenomenon of lr~tabl11ty.
The theory as developed tor 001umns forms the From mechanfc s of simple bending, we have
basts tor the study of the instability of thin the deflection equation,
plates, which subject Is treated in Part 2.
d"u 11
AlS.2 Combined Bending and compreeeton of Columns. --a = - - - - - - (1)
dz Ely
Consider a column with one end simply Thus the deflection u(Z) ot the column
supported and the other end hinged (Fig. AlS.l) IS,
under the simultaneous action of a compreSSive
load p and a transverse load Q. Without the Qaz
load P the bending moment due to Q would be:- , (0 ~ z ~ 1 - a)
1
p
d"u (1-a) (l-z)
El dZ" = - Pu - Q 1
I
I (1 - a ~ z ~ 1) - - - (2)
a I
If we introduce the notation,
.ll:....-Jf+--
I
Q
I - (3)
I
\ The general solution of eq, (2) is:
\
\ Qaz
U ... C:L cos Kz +. C. sin Kz - PI '
--\u (0 ~ z 1 - a) - - - - - - - - - (4a)
o (I-a) (l-z)
u = C;5 cos Kz + c, sin Kz - 't i '
A
~!¥-
(l-a::z::l) - - - - (4b)
p
Fig. AlB.l Where C~J Ca, C3 J and C40 ~re constants of
integration to be deternined from bo~~dary
On the lower portion or the column conci tions.
_ caz
11, - -1- ----
-- ---- - -- (a)
z
For eqe . (4), s tnce u = 0 tor z = a and
= 1, it follows that:
On the upper portion
C~ = 0 and C3 = - c. tan K1 - - - - - (!)
M~ :: ~ (l-z ) - - (b)
At z = (1 - a) the two portions of the
Due to the deflection u(z), the axial deflection curve given by (4a) and (4b)
load P contributes to the bending moment by respectively must have the s~e deflection and
the amount:- slope. From these two conditions we determine
Cll and C•.
M~ = Pu - - - - - - - - - - - - - - (c)
Al .1
AlS.2 THEORY OF THE INSTABILITY OF COLUMNS AND THIN SHEETS
c,= ~ sin Ka C =_q sin K(I-a) (g) the bOdy would cause only infinitesimal Changes
PK sin Kl' to PK tan K1 in the displacements and the body recovers 1f
the added loads are removed. When the dis-
Substituting (r l and (g) into eq, (2), placements are continuously increased With
we obtain: little or no further increment at loads, the
system is unstable. If the body will remain
Q. sin Ka ca in the displaced position after the removal
u = PK sin K.l sin Kz - PI z ot the disturbance, the body is said to be in
neutral eqUilibrium. Havir~ these definitions,
(05 Z S 1 - a) - - - - - - - - - - -rsa ) we will not investigate the behavior of the
column betore and atter the critical load !s
u =~ sin K(I-a) s1 K(I-z) q(l-a) (l-z)
reached.
PK sin Kl n PI
(l - a S z s 1) - - - - - - -(5b) p p
m=..L.~=~ (8p)
Per 1 1t
The total length ot the column in the
deflected shape 1s given by:- Since p = sin %is a function at a so is the
= _Ll d9 - - - (8h)
elliptic integral K and the ratios P~r and i
-a 12 k ..; cos Q _ COS <l
calculated tram equations (sa) and (80).
or 1 = fa
-a
d9
----(81) Thus P~r is a function or t calculated by
2k/ stn" ~ -sln"9/2 means at tables giVing elliptic integrals.
Thus m can be plotted against 0/1 as shown in
Denoting sin ~ by P and introducing a new FIg. A18.3.
variable 0 :-
011
sin ~= p sin ¢
O~/l
Equation (81) then becomes,
n
l=:f:-iz d9 =2K - - - - (8J)
1<: a / I-p"' sina 0 k
one when 0/1 = 0 and the other (point A) tor (b) At PIPer = 1, or at point (A), a bifurcation
which 6/1 ~ O. The column thus bAS two at equilibr!um occurs and the column starts
possible equilibrium forms, one straight and to acquire two POSSible neighbor pOSitions
one bent. of equilibrium, the straight and the bent.
Let us now assume that at 6 = 0, the (c) Above PIPcr = 1, the column has two pOSSible
column Is displaced by a small disturbance equilibrium positions 0/1 0 and ~ /1 .,. o.
and acquires a deflection O~. For pl C Per'
we see trom Fig. 3 that m s- m1.. Thus p1. Is Thus as far as initiation of instability is
not sutficient to maintain the column in concerned, the Euler load as given by Eq. 7 can
eqUilibrium in the bent form and i t will spring be considered as the critical load. The question
back to its straight rorm . Thus for p.l. ... Per' arises Whether this load has a practical use
the straight form Is stable. for design purposes. A logical design criterion
is obviously the maximum load which a column can
It p1. > Per' then m1. > m. Thus m1. will sustain. We observe from FIg. A18.4 that the
bend the column still further. This means load P increases tor increaSing displacement 0.'
that if p1. - PeT' the straight form of This behavior is due to the development of large
equilibrium Is unstable. The column will deflections due to bending. However J over a
continue to bend until m1. becomes equal to m considerable range of deflections oJ the P -+ 6
(point A in Fig. 3). It the column Is dis- curve is practically horizontal (for instance,
placed further tram A, the deflection becomes between points A and B the ratio 0/1 varies
larger than O. and m _ mJ. at the new position. trom zero to - 0.4). For such large deflections
The column will spring back to pOint A. Point for which the column load does not change
A is therefore stable. practically, i t is obvious that the column
ceases to fUnction properly. Therefore, trom
At P .: Per, the mJ. versus 0/1 Une is this point of view J the Euler load can be con-
tangent to the m curve at the origin. There- sidered that which characterizes the maximum
fore, ror an infinitesimal disturbance, the strength ot the column.
column will remain in equllibrium at the
dIsplaced POSition since for such small The riSing part ot the curve BD holds as
disturbances mJ. remains equal to m. The column long as the material behaves elastically. At
is therefore in neutral equilibrium. some paint OJ however, inside the almost flat
portion ot the c~rvr. the inner fibers of the
AlB. 4 The Failure of Columns by CompreSSion. column acquire maximum stress equal to the
yield stress. If we carry out an elastlc-
In diSCUSSing the ~tability ot a column plastiC analYSiS of the subsequent behavior J
in the previous section, it was shown that we observe that the curve drops almost
below the critical Euler load (Eq. 7), the 1mm.ediately. Again this maximum load PB is
straight form is stable, above the Per the very near the Euler load. For deSign purposes,
bent torm is stable and at Pcr the eqUilibrium therefore, the Euler load J which is a buckling
is neutral. By plotting the curve PIPer versus load, is a very good approximation to the
0/1 as shown in Fig. AIS.4, we observe the Ultimate load which the column can sustain.
following behaVior.
Another argument will contirm the above
(a) Below PIPer = 1, there Is only one conclusion. In discussing the DuCkling of
eqUilibrium position, 0/1 • O. columns in the pr-evf ous paragraphs , we have
assumed that the column 15 initIally straight,
centrally loaded and made of
PIPer homo~eneous material. Actual
columns, however J are imperfect
r
due to initial crookedness (for
P;a./Per instance, due to unaVOidable
tolerances in their manutacture),
due to slight load eccentricities
B
and due to lack of complete
1.0 .~ hamogenity. Therefore, a certain
amount of bending Is always
present even for small loads.
the change of curvature due to subsequent approaches the horizontal line PIPer = 1
bending (after loading) is:- asymptotically. ThiS curve, however, is valid
for small deflections for which the approxl-
mation:-
is valld.
In the differential equation of deflection
one can prove that llR is th8 change of
curvature which for an initial straight column By a treatment similar to that in the
coincides with the curvature itself. Thus in previous paragraph, we will find that for
the present case, where the bending moment is large deflections the load deflection curve
Pu, the equation of deflection becomes:- raises after the paint I (curve FIH). Due to
the onset of plasticity, the actual curve
.- k'u - - - - - - - - (9)
drops at the point I~ (curve FII~Hl). The
failing load at I~ can be either greater or
smaller than Per' but it is usually very near
to it.
Let us express uo in Fournier series:-
~ nuz In the above discussions we have shown
Z 3 sin--
uo .. , - - - -(10) that for all practical purposes the Euler
n 1
buckling load can be considered as the ulti-
Substituting (10) in (9), we find the • mata load which a real or practical column
can sustain. Besides its closeness to the
solution which satisfies the boundary actual ult~te load, the critical load can
conditions (u = 0 for z .. 0, Z = 1) is:- be eaSily calculated from equation (7) with-
~ nz out the necessity of carrying out a lengthy
u ::.1: on sin T - - - - - - - (lla) calculation which will include the initial
n=l imperfections and plastiC effects.
It should be noted, however, that the
where on • ,,-_:;-,,:- - - - - - - - - - (llb) buckling load ~iven by equation (7) is valid
when the uniform stress due to a compreSSive
load (0 = PIA, where A is cross-sectional
area) is below y1eld stress. It a is above
the yield stress, the theory of plasticity
predicts another value for the buckling load.
The deflection of the column at the center Referring now to equations (11) we tind:-
15:-
Pn = n~Pcr (Per tram equation 7)
- -(12)
~n
If we plot the deflection versus the load
we obtain the curve (Fig. AlB.B), which
piPer
Thus as P approaches Per' we see that
c etc.
I'
1.0 A - - ~- -
, G · x -- v
'":Il8. , '·I
_..,;6'i';-,,;_
- PIPer
<,
-;
'H~ In a buckling test we measure a: Omax
- 6, where 6 is the initial deflection at the
middle point. Thus:-
o
Per p - 6 = 6:l,
Fig. AIS.6 - - - - - - - - - - (13)
113
AlB.8 THEORY OF THE INSTABILITY OF COLUMNS AND THIN SHEETS
bending-buckling 18:-
The buckling load for other end conditions
can be derived in Similar manner.
- - - - - - (141
INELASTIC COLUMN STR~IGTH
and the general solution is:-
AlB.6 Inelastic Buckling. Introduction.
U3 c~ sin kz + C. cos kz + C3 z + c. - - (15)
Euler's theory of buckling 1s valid as
-The coefficients C~, C., Ca and C. long as the stress in the column nowhere ex-
depend on the conditions at the end supports. ceeds the elastic limit of the column material.
The various end conditions are:- We have seen that the analysis tor perfect and
imperfect elastic columns leads to the same
Free end: -
d'u
dz lll = ° d'u
dZF = ° result, namely, equation (7).
The case of the inelastic buckling, that
d'u
Pin end: - u =0
dz'" = ° Is, instability under axial load exceeding the
elastic limit stress, presents some difficulties.
As we will see, the perfec~ column analysis
Fixed end:- u ~ ° du
dz • 0
leads to a different expression !ar the critical
load than for the perfect column. This is due
to the tact that 1n the plastic stress range,
Thus we have 4 end conditions. These the material behaves differently under loading
give systems at four linear homogenous and unloading, as illustrated in Fig. AIS.?
equations. A trivial solution of these is Let us now examine the two cases:
the zero solution. For the buckling state,
however, C1 , Ca , C3 , C. are not all zero.
The condition at non-zero solution of the --:::::: Loading
above system is that the determinant at the
coefficients of C1 , C., C3 and C. is equal to I
zero. From this equation, we calcUlate the
buckling load. o
/ /~nJoading
•• /
Fig. AlB.7
c, + C. = 0 C. =0
AIB.7 Perfect Column. Reduced Modulus Theory.
(1 ) That the displacements are small so that The distribution of the compressive (-)
the relation between the radius of stress and the tenSile (+) stress due to
curvature R and the deflection u of the bending is shown in Fig. Ala.sa. The stress
elastic exrs is, becomes zero on line (a a1.), which is at a
distance e from the centroid c. For
eqUilibrium of stresses on the cross-section
- - - - - - - - - - - (17a) we have,
(2) Plane sections remain plane, therefore _ ~~ OS1d.A + ~all OS.dA =0 - - - - (20)
the change of straIn due to bending at a
distance h on the plane of bending IS, and for eqUilibrium tor moments,
h dllu
0< = if = h dz' - - - - - - - - - - (17b)
(3) The stress-strain relation tallows the Due to the linear distribution at stress,
Simple tension curve for the material. we have:-
(4) The plane of bending is a plane ot 00,
symmetry of the cross-section.
os, =-h
a, >
- - - - - - - - - - - (22)
Assume now a column with the cross-section Ocr.
as shown in Fig. Ala.sa be compressed in the os. =-h
a. •
IntrOducIng now (17b), (18), (19) and
(22) in equation (20), we obtain,
where,
Q~ = ~a1 h1. dA , Q,1I = ~a. h.dA (24)
(28)
d
1/2
Since E is a function of Ocr given by eq. (28) b
and the value of Et at the unknown Ocr, the
calculation of the critical stress requires a
trial and error sireultaneous solution of
equations (23), (26) and (28). 1/2
~
cussion with the classical definition of
P = Pt 1 + 1
1+-r) - - - - - - (32)
instability (perfect column, bifUrcation of ~+
equilibrium) strain reversal does take place.
In recent years, hOwever, this tangent modulus
2d 1--r
theory has been proved useful. where
Under the assumption at ho strain reversal
both sides of the cross-section in Fig. AlB.Ba,
will be characterized by the same linear stress It must be emphaSized that the buckled con-
distribution, corresponding to the tangent- figuration Is a stable one similar to that
modulus Et. Thus the buckling equation will be, considered in ~he refined Euler's theo~J.
Shanley has recognized the fact. tnat such a
EtI ~:~ .;. Pu =0 - - - - - - - - - - (29)
stable configuration ~y exist after exceeding
the tangent modulus load.
and the critical stress for sunply supported It R = PIPt , Shanley found that the
end conditions becomes, relation between the variation of stress due
to bending and the compreSSive strain 6t
crt = n'E;;! __ - (30) corresponding to at is:-
AI'
,
~
~
. •
,
\
~;"";----'~j.-+--h,.4-+-+-l----l-i--+
f-+-+-l'::::=f7~~k-+-l-+--+-I-+-
;,
II
I· -I-i--+--+-+-,.4--+~-l----l-i--+-
" \
;. .
• • ~
f'oI._"""". _
• m _
su_ ~
RAno (1..+) I. ,~ * • m ~
".~
HS.T..tOi _ _ .... (0.- i-. oti_l.
, A.n nl .... /(11 L_
• r-, t'l~
FIg. AlB. 11
,
•
...... Modulu ".a •
..,
GfU,T £.. TUTS
. . ,.. t18 1'UI1
FIg. AlB.lO
References:
Ocr .. - (34)
Ret. 1. Engesser F., Schweezeriche Bauer
Zeltung. Vol. 26, p. 24, 1895.
Figs. AlB.ll and 12 show how experimental Ret. 2. Shanley, F.R., Inelastic Column
results check the strength as given by the Theory, Jour. Aeronautical SCiences,
Euler equation USing the tangentwmodulus Et. 1947, p. 261.
..... _.
A18.10 THEORY OF THE INSTABILITY OF COLUMNS AND THIN SHEETS
PART 2
dr
r------ x
atter bending are
,
-d{jX-. ---- f+-l..,.
_ aw aw y
, ix - ax' iy = ay z
y , Fig. 2b
For small de- Let Fig. (2a) represent a thin rectangular
Fig. 1 flections, the Plate loaded by unifOrmly distributed bending
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES AIS.11
B M
x
= -D (a.llw + V a.ll,;) (Sa)
ax.ll oY
My==-D (o"'w
ay.ll + 7J o·w)
ax.:! - - - - - - ( 5b )
EW"3
where D = 12(I-V.) = the tlexural rigidity or
1
the plate. It besides the tlexural moments :M:x ,
My, there are uniformly distributed twisting
z moments Mxy and Myx along the sides of the
FIg. 3 plate of Fig. 2a, these must be equal to the
resultant at distributed shear torces crxy, 0xz
This angle tor small displacements is obviously along the sides at the element at F1g. 2b.
equal to ~;. Thus the horizontal displacement From eq, (3) we obtain:
Ux in the x-direction at a pOint at distance Z
trom. the middle surface is; 2Gz _ a·w
"xy = cryx. Rxy - 2Gz axay ~ - - - - (4<1)
aw
ux ~ - Z ax (The Sign - indicates
negative displacement tor positive z ) ,
MxydX • J h/2 "xyz dXdz , MyzdY =
-h/2
In a similar manner we tind the displacement
in the y-df.r-ec t fon , The complete displacement J h/2 oyxz dydz, Mxy =M -
yx-
system is; -h/2
The corresponding strains are: Equations (5) give the moments per unit length
tor pure bending and twisting at a plate.
a.ll·N Z
Ey =- z ayr == R
y
,
Ala. 12 THEORY OF THE rNSTABrLITY OF COLUMNS AND THIN SHEETS
I
I a
I aw i;!+~dx - - - - -(11)
oy a xay
y
Mxy \ 2 !oJ This expreSSion will be modified later when we
\ Mxy will consider the superposition of compressive
\ loads in the plane of the plate which are
Fig. 5b related to the problem of plate buckling.
w= z
.. .. l: ("
'"1IlIl
sin ~ sin nny
a b
---(14a)
ot w we find the final tonn of the boundary :n=l n=l
conditions along the tree edge:
By substituting in sq. (18) with q given by
eq . (13) we find:
- - - - - - - - (12c)
z z Cmn
m=l n=l
In the following solutions tor various ~
mrtx nny
edge conditions will be developed. I: amn stn-a:- sin-- - - - - - (14b)
b
n=l
1. Simply supported rectangular plates
This relation is identity it:
Lat a plate with sides a and b and axes
x, y, as shown in Fig. 6, be simply supported
around the whole periphery and loaded by a
distributed load q = f (x,y).
l: "mIl ern =
a- ern "F"
nrty
-(13a)
f (x,y) = qo = const.
m=l m=l
H-'- t::
I--
I'
(140 ) :
169 0 0.:> anx nrty
I F
00
s ina- sin --0--
W=n"D Z Z
. m=l,3,S n=l,3,S
mn {m
aa
a
+£)'
b~
y <\
ffiII1Jtq Fig. •
with maximum oernecnton at the center,
- - - - - (15b)
~ ~-l
The boundary conditions are: OZn"D Z Z (-1) 2
m=l,3,S n=l,3,S
W = 0 , Mx ; a at x = a, x =a
W = 0 , My ; 0 at y = 0, y = b, or:
-----(15c)
(1) W =0 (2) ~; = 0 at x = 0, x =a This 1s a rapidly converging series and a
satisfactory approximation is obtained by
(3)11=0 taking only the first term. For a square
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES AlB.15
W=WJ.+W a - - - - - - - - - - - - (17a)
where
(L cos h rnn:y + lL mny sm h mny) sin ~
- - (17b) "'m a '-ma a a
--.
a"W a
ax'
.--, 0
a"w a
ay' (18)
expression into the boundary condItions:
w : 0,
a·w
ay": 0 for g : -
+ b
"2 we tind:
With w a in the form of (16) and satisfying
together with wJ. of eq. (17b) the boundary 4
a'w • b Am cos h am+am Em sin h Cm,+lleme = 0
conditions w = 0, ayr::: 0 at y = - 2' (see - -(21a)
Fig. 7). (An + 2Bm) cos h Om + am 8m sin h am = 0
Substituting (16) into (18) we obtain:
where am = mrtb/2a
2(~ tanhOm + 2)
Am =
nllm" cos ham (V ' = qa::l + (l-V)qa'n' Z
~
ux}mox 8
m=1,3,S
2
8m = - - - - - - - - (21b )
n"m ll cos ham '( 2V
m Am -l-V 8m )
Thus:
- 384""ll - - - (0.68562 - 0.00025· .... ) increment of strain energy due to the incre-
~'D ment ot the deflection by:
40
- qa
= 0.00406 D
Ow
= ... mroc nny
,-,Cmn sin -a- sin "F" - - - - - - (24a)
,rr'ab "
[--4- -mn P sin ~§ . sin n~~ ] Let us consider the case of a uniform load q.
we write the deflection in the form:
6Cmn = a - - - - - - - - ~(26a)
Since 6C~ is arbitrary its coefficient must Where w~ is the deflection of a simply
be zero. Thus: supported strip of length, a, which for the
system of axes of Fig. 8 can be written (see
4P sin a3.. • sin E!!!l
b
Levy's method in previous section):
Cmn = ll
mnx
n"abD (m.a ,. n ), sln- ,
artJT a-
+.-
L _ 4,
,--,..-----r.... x .nm - - 1"(lImli , em=-Dm
By the conditions (25c) we find:
1 .. 0_'_.
1\''''.
!3"1I1(1-1I) co.sIl·%+211coShB.,.-!l(l-!l)~ !ln~a.,..(l-/l·)
:3+11)(1-111 cOSh·s..:" n·l/)·a.,· .. (."1/). ___ _ (25s)
;, 0_'_. (3+111 (l-ll)stnh a., COShs,.·V(l"!I)stn~B.t-II(l-lI)a" CO!h~-(l-II)'3:l
n"",' (3+JlI(l_lIlcos~'3,,+fl_I/)' s,.··(1+11)·
y
Fig. a
Substituting Am, Em, em and Dm in eq, (2ed) we
Assume that the edges x ~ a and x = a are find the deflection. The ~imum deflection
simply supDorted, the edge y = b free and the occ~r~ at the middle of the free edge.
edge y = 0 bUilt-in (Fig. 8). In such a case
the boundary conditions are: AlS. 14 Combined Bending and Tension or Compression of
w -= 0,
a'w =
aX""" 0, for x = 0, x = a (a)
Thin Plates.
dx)(~+a2w rjxl
x +awx
N a aw
- I X Y ax +
(N
ax ax a;{"l ,
y
and neglecting terms of higher order:
Fig. 9
--.
aNx
ax
aNxy
ay --.
aNy
- 0 , ay
aNxy
ax - o - - - - - (28)
Yay' ay ay
-------(b)
In thiS chapter the problem of instability and neglecting terms of higher order:
of pla~es will be examined.
~
2
Thus in eq . (6a) the terms given by (a), (b), 2~1 I (N; +N; -2!INxNy + 2(1+V)Nxy') '- - (a)
(e) and (d) should be added (divided of course
by dxdy}:
During bending due to transverse loads
or/and due to buckling we assume that the edge
loads and consequently Nx , Ny, Nxy, remain
constant. Its variation is thus zero and we
do not consider it in the following. Let us
apply now the transverse load that produces
bending. (~e can also consider bending due to
aut due to the equation of equilibrium (28) the other transverse distUrbance, which is the
two terms inside the parentheses in (e) are case of bUCkling). It u, v, are the displace-
zero. Thus: ments at the middle surface due to the coplanar
loads (which are assumed constant across the
ao.x
dX
+ 0iY
y
+ q aaw
+Nxax a +
thickness) and w the bending deflection of the
plate it can be shown that the strains are:
aaw aaw
Nyayr + 2Nxy axay :: a ------(29) = au + 1.. (a w ) .
ex ax 2 ax
Eq , (29) replaces eq , (ea ) when edge loads are =~+ 1:. ~).
present. Eqs. (6b) and (6c) are, however, still Ey ay 2 ay
valid since they ex~ress moment eqUilibrium _ au av aw oW
of the element dxdy ln wnlen ~ne con~rl0U~~0n
ot Nx, Ny, Nxy is zero. Thus eliminating Qx ,
O'J:Y - ay + ax + ax . ay - - - - - - (b)
fa afa b [Nx
au oV au av]
ax + Ny ay+ NX"J (ay +ax) dxdy =
I.J C
AlB.20 THEORY OF THE INSTABILITY OF COLUMNS AND THIN SHEETS
Obviously the first two integrals represent buckled conf1guratio~ (ceflections different
the work done by the edge loads while the from zero) is possible. It was found in ~he
second integral 1s zero due to the equilibrium case of column that this latter solution
equations (28). Thus the work of the edge (Suler's solution) approaches asynptotlcally
loads is: the first at the l1~it where the deflections
become extreillely large but for even small
IItI -_ II [ Nx ai{+Ny
au ,av au + av] deflections the edge load acquires a value
ay +Nxy (ay ax) dxdy- - (r) very near the Euler's critical value. The
latter teChnique is mathematically more
We assume now that for small deflec~ions convenient and it gives ~or plates also a ~ery
the stretching at the middle surface of the good estimate of their compresSive strength.
plate is negligible. (This 15 the so-called In the following we shall use this latter
1nextensional theory at plates). In this case approach by assuming a plate with edge loads
by zeroing the strains 1n eq, (b) and substi- and no transverse load. Sq. (30) beco~es in
tuting in (f) we find; this case;
"\'IN 1 II (a w) . aw ew awl a"w +2 a"w ." a'w - 1 (N a'w +N a"w +21N ;;.'l.w
= -2 [ Nx ax +N
j, Y
(-)'
ay +2Nxy -
ax , ay
- dxd y ax" ax 3ay2' ay" - D ! X axr
I y if xy axay
u ,
i
(32a) (33)
In the strain energy expression, eq. (a) By solving eq. (33) 'Ne will find trat the
the first two terms cancel each other and the assumed buckling made is possible (w ~ 0) for
strain energy is due only to bending: certain defi~ite values of the edge loads, the
smallest of which deter.nines the critical lead.
li
Vi = ID!I f("lI W+ aayliJ
W' lII _ The energy method can also be used in investi-
2 lax:! gating buckling prOblems. In this ~ethod we
' w a'w
2 (l-lI) [ aax'· a'"
ay" - (axay)' ]1 dxdy - - (32b)
assume that the plate is i~itially under the
plane stress conditions due to the edge loads
and the stress distribution is assumed as known.
In the absence ot transverse loads the work We then consider the buckled state as a possible
of external torces is Simply due to the edge configuration of equilibrium. T~e change of the
loads: work is given by eq. (32a). We inter,ret here
w as a Virtual displacement though we do not use
the variation symbol O. Thus the increment of
- - - - - - - - (320) work ow is ~iven by (32a) and the increment of
ExpreSSions (32b) and (c) will be used in strain energy 6Vl is given by eq. (32b). If
6~Vl for every possible shape of buckling the
solVing the buckling problem by means ot the
principle ot Virtual work. flat equilibrium 1s stable. If 6W~Vi for a
certain shape at buckling then the flat :on-
AlB.16 Buckling of Rectangular Plates with Vartous Edge figuration is unstable and the plate will buckle
Loads and Support Conditions. under any load above the critl:al load. It
OW = OVi' the eqUilibrium is neutral and fro~
General discussion this equation we find the critical load. The
critical load therefore 1s found :rom the
In calculating critical values of edge
loads for which the tlat torm ot eqUilibrium
becomes unstable and the plate begins to
buckle, the same methods and corresponding
equation:
1 II [ N (aw.
- '2 x ax) + Ny (aw •
ay) + 2Nxy
sw s
ax· ay dxdy =
reasonings as tor compressed bars will be D ! a'w +ay')'
zlll(ax"
a'w [a·w
-2(1-11) ax'
a'w a'w Jl
'ay" - (axa/j,dxdY
employed.
•
7
6
5
\\
\
'm'"
"\
\
A:
n=3\
~I:
,\
'
"mO<'
¥
\ x," ~ /
~,,"
".-
,
4 , , ,
3 ,
, ,,, ,;
2 , ,
I ,,
, I, ,
o , Fig. 13
o '16 ~
3/12 4
alb
_ (3'79.)
Fig. 12
znb a (m.+l)b
-+-= ---+ a~d by introducing t~e pa~~et~r:
a JIb a or:
nOD
00
• =-.-
a h - - - - - - - - - - - - (37'))
~=vm(m+l) ---(36d)
b
',.Je obtain:
Thus the transition from one to two half-waves 3
occur for: a ... '18. a
0e (n - n~) - - (37.:::)
(t : thickness) Fig. 14
AIB.18 Buckling 01 Simply Supported Rectangular Plate A18. 19 Buckling of Simply Supported Rectangular Plate
Compressed in Two Perpendicular Directions. Under Combined Bending and Compression.
Lat (Fig. 13) NX 1 Ny the uniformly distri- Let us consider a Simply supported
buted edge compressions. Using the same as rectangular plate (Fig. 13). Along the sides
before expression for the deflections (eq. 35a) x :: 0, X = a there are linearly distributed
and ap~lYlng the energy equation (33) with edge loads ~lven by the equation:
Nx, Ny - constants (whiCh is the sol~tion of
the corresponding plane stress problem) we find:
Nx := No (l - A{) - - - - - - - - - (39a;
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A1S.23
which is a combination of pure bending and rle examine for each value of m the solutions of
~ure compression. Let us take the deflection the system (38f). Starting from m 1 and =
a 6 a i n in the foron: denoting:
oo
w = ""
.z Cmn ~
sm TJtX
-,-. 1 nny
!5 n -b- - - - (39b) Ocr =
(No)cr
- - - - - - - - - " (39g)
m=l h
b
:1:1
m
~
1'[
halt-wave lengths. The regions of the curves
with minimum ordinates define the region or
stabU1 ty as In F1g. (12).
m=l
AlS. 20 Inelastic Buckling of Thin Sheets
"=
£. z Cmn:l- - - Z
Bb' ""
[4 !1=1 1'[:01 n=l The problem of the inelastic buckling of
thin sheets has been extensively studied by
various authors. The main difticulty in such
Equating (39c) and (39d) and solvir.g for No studies is in reference to the stress-strain
we find:
relations at plasticity under complex states
at stress. Many controversial discussions
have appeared in literature without resolving
the theoretical difficulties. For this reason
we will not develop the theory of inelastic
buckling in this chapter. Some ot the better
references on this subject are listed below.
- - - - (39.)
Cnapt er- C4 presents the plasticity
"!he coefficients now Cmu are so adjusted that correction factors to use in calculating the
(Nol er becomes minimum. 8y taking the inelastic buckling strength of thin sheets.
derivative of expression (38e) with respect
to each coef~icient emn and equatir~ these to AlB.21 References.
zero we f:"nd:
(1) Bleich,?: Buckling Strength of Metal
Structures. Book by McGraw-Hill.
(2) Stowell, E.Z., A Unl~ied TheoFJ of Plastic
Buckling of Columns and Plates. NACA
Report S9S, 1948.
(3) Gerard and Becker: Handbook Of Structural
Scabi11ty. nACA T.N. 3781, 1957.
(4) Gerard: Introduction to Structural
Stability Theory. Book by ~cGraw-Rll1
Co., 1962.
'r)
........ '
"·-~~~f.
" ':-
:;:,:;0""
A18.24
This multiplEl exposure photograph of a Boeing supersonic transport model shows the variable-sweep wing
in three configurations: forward for takeoff and landing, swept part way back for transonic flight. and swept
completely back as an arrow wing for 1800-mtie-an whour supersonic cruise.
A19,l TYPical Wing Structural Arrangement relatively thin. In general the wins st~uc~ural
For aerodyr.a~lc reasons, the wins cross- flange arrange~ent can be classified ~nto two
section must have a streamlined shape commonly types; (1) the concentrated :lange type where
refeTT9d ~o as an airfoil section. The aero- flange ~aterial is connected directly to in-
ter~al webs and (2) the distributed flan~e ~J?e
dJ~~~lc ~orces in flight chan~e in ~gn1tude,
~irection and location. Likewise in the various where str1ng~rs are a~tached to s£in between
landing o8erations t~e l~ads ctange in ~agnl internal webs.
tUde, direction and lecation, th~s the required
3tructure must be one that can efficiently A9.3 shows several structural a~an~e
F'~g.
resist loads causing combined tension, com- ~ents for wing cross-sections for superscr.lc
pression, bending and torsion. To provide aircraft. SupersoniC airfoil shapes are
torsional resistance, a portlon of the airfoil relativeLY thin compared to subsonic aircraft.
surface can be covered with a metal skin and
then adding one or ~ore internal metal webs to
produce a single closed cell or a ~ultiple cell
l
wing cross-sect~on. The external skin surface Fig. d
which 15 re Lat tve Ly t ntn for subsonic a i rcrarr
is ef:icient for resisting torsior~l shear
s~resses and tension, but quite inefficient in
------- 1
::-8sisting ccmpressive st~8sses due to 8end1ng
of wing. To provide strength efficiency, span-
wise st~f:ening ·~~its commonly referr~d to as Fig. e
:lange stri~gers are attached to the inside of
the surface skin. ~o hold the skin surface to
airfoil Shape and to provide a ~edium for
transferring surface air pressures to the
cellular Jearn st~cture, chordwlse fo~ers and 0 0 0 0 0 ... ----
ribs are added. To transfer large concentrated
loads into the cellular beam structure, heavy Fig. t /--II
ri~s, cOIT~lonly referred to as bulkheadS, are ~ , i _____ ]
,L
used. " ,
c r--------~
"'----------
L c ,
Fig. b '!'"
,
,
-l F 1::;. h
~
'---.. ,
T
1 T!.,-
)" ----
Fig. c
Fig. A19.1
I F'1,~ •
\::] ::1 L
,
,
I r'
eJ
Concentrated ?ls.ncc8 ":ne of ,';in;:: 8ea:n.
Deshed line represents sec cndar-y s t.ruc tur-e . In Fig. A19.2 C:)rr:Y:lon :'YP8S of '.~·1 ng Seem Flange
aany cases this por t t on ~E :ab::-ic c over-ed, Ar~ange:ner:t.
A19.1
--.
j:.1.7
Al9,2 ANALYSIS OF WING STRUCTURES
~o withstand the h1~h surf~ce preSsures a~d :0 merrt (Fig. ,:"19.2), a cor-t t on c< t:-,e::::::,',',::":::g on
obtain sufficient strength ~uch ~hicker ~i~g the c0mpression si~e is ~a~e effec:172 since ::
skins are usually necessary. Modern ~il:ing 1s attaChed to closely spac8c stringers cr SQrT~
machineS pe~it ta~ering of skin ~h~C:~1es2es. ga t t ons . However , the r Iange s tr tnger s cc t.veen
To obtain ~ore rlanse ~ater~al intesral flange cell nebs are Supported July at rib or j~lkr.ead
~~its are ~achined on the thick s~in as 111~s points ~~d thus suf:er CJl~~ action no~,al to
trated in Fi~. k. the cell coveri::g. ThiS factor reduces the
flan~e allcwable c8mpressive stresses since it
?lg. j is not practical to space wing rl~s less ~han
12 to 18 times the flange stringe~ :e~th. Th~sJ
< ,
][ I[ j[ I[ i~ ~here were no other ~ontr2l11n; =~ctcrs, or.e
could easily ~2ke calc~lations to dste~lr.e
which of the above would prov~ li~htest. In
general, if the tor-s i cna L ror-c es cr; t.ne .v tng
Fig. k are sraaLl , thus recu ir-ma cn::'y a m;n c ove r tng ,
the concentrated flange ty;e of Nln3 st~~cture
should prOve the li~htest.
<c
In general, the flange ~aterial should be
Light Weight Core
placed to sive the largest moment 0: inertia in
Fig. 1 the Z direction, which ~eans ~n general that the
flan~e ~ater1al should se placed bet~een the 15
and 50 per cent 01' ':Iing chord r r'om the leading
edge.
Fig.A19.3 Wing Sections - super-soni c Aircraft The secondary or ~istrij~ting structure aft
In a cantilever wing, the wing bending of the st~~ctural box bea~ should be ~de as
light as pOSSible and thus in ;eneral the farther
moments decrease rapidly sparmise frem the forMara the rear closing web at the box be~, the
maximum values at the fuselage support points.
lighter the wing structure as a whole.
Thus thick skin construction must be rapidly
tapered to thin skin for wei~~t effiCiency, but
thinner skin decreases allowable compreSsive In the Layout of the main spanwise flange
stresses. To promote better effiCiency sand- members bends Or changes in direction should be
wich construction can be used in outer vortion avoided as added weigh: is required in spl~cing
01' wing (Fig. 1). A light weight sandwich core or in transverse stiff~ners whl~h are ~ecessary
~o change the direction of ~he load in ~he r:ange
is glued to thin skin and thus the thin skin is
capable of reSisting high ccmpressive stresses members. If flan~e members must be s01iced, care
since the core prevents Sheet from buckling. should be ~ken not to splice them in the region
of a max imum cross-section. Pur-thermcr'e , in
Al9.2 Some Factors Which Influence Wing Structural general, the smaller ~he ~umber of fit~ings, the
Arrangements lighter ~he structure.
(1) Light Weight: ~ (2) Wing-Fuselage Attachment:
The structural deSigner al'H.aYs strives for If the airplane 1s 0= the low ~nng or the
the minimum weight which is practical 1'ram.a high wing type, the entire ~ing structure can
production and cost standpoint. The higher the co~tinue in the way of the 2i:-plane body. How-
ultimate allowable stresses, tee lighter the ever, in t~e ~id-wing tj~e or se~i-low wing
str~ctures. The concentrated flange ~ype of type, limitations may prevent extending the
wing structures as illustrated Fi~. (a, b and c) entire wing through the :uselage, and some ot
at Fig. A19.l per.nits high allowable compressive the shear webs as well as the wing covering
flange stresses si~ce the fl~ge members are nust be terminated at the side of the ~Qselage.
stabilized by both web and covering sheet, thus If a distributed flange type of cell structure
eliminating col~~ action, which permits design were used, the axial load in the ~lange string-
stresses aporoaching the crippl~ng str~ss of the ers would ~ave to be transferred to the T.embers
flange ~emb~rs. Since the flang~ nembers are extending through the r use Lage . '1'0 cr-evice fer
few in number, the size or thickness required is this transfer of large loads requires struct~l
relatively large, thus giving a high cri~pling weight &~d thus a concentrated fl~nge t:~e of
stress. On the other hand, this type of desi~n tox str~cture X:~ht prove the DRst type of
does not develop the effectiveneSS of the ~etal structur-e.
covering on the compressive si~e, ~hlch WllSt be
balanced against the saving in the weight of thel (3) Cut-ours in Wing Surface:
r range tnember-s , I The ideal arranse~ent where continUity of
In the distributed type of flan~e arrange- structure is aa i nta mec oyer t ne en t i r-e sur-race
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES Al9.3
of the structural box 1s sel~om obtainec in ac- the lower side of the Wi~g. They are usually
t~al air?lane cesigr. due to cut-outs in the fastened to two spanwise stringers With screws
Ning su~~ace ~or such items as retractable and the re~ovable ~anels are ef:ective i~ re-
landing gears, ~il compartments and bomb and s Let Ing bending and shear load. (See Fig.A19.5)
gun bays. r: the distributed ~~ange type of
box beam is used, they are interrupted at each
cut-out, which requires that means nust be
?rovided for drifting th~ flange loads around
Removable panel for
the opening, an arrangement wntcti adds weight Fig. Al9. 5 assembly and inspection
because conservative overlapping assumptions purposes.
are usually made in the stress analysis. The
additional str~cture and riveting to provide
for the transfer of flange load around large
O?enings adds c8nslcerably to the production
cost. Cutouts in the wing structural box destroys
the continuity of the torsional resistance of
?or landing gears as well as many other the cell and thus special consideration must be
installations, the «ing cut-outs are confined to given to carrying torsional forces around the
the Lower- surface, thus a structural arrangement cut-out. This special problem is discussed
as illustrated in Fig. A19.4 is quite common. later.
The upper surface is of the distributed flange
type whereas the lower flange ~terial is con- (4) Folding-Wings:
centrated at the two lower corners of the box.
In the normal flying conditions, the lower sur- For certain airplanes, particularly Carrier
face is in tension and thus cell sheet covering based Naval airplanes 3 it is necessary that pro-
bet~een the cut-outs is equally effective in
viSion be made to fold the outer wing panels up-
ward. T~is dictates definite hinge pOints be-
bending if shear lag infl~ence is discounted.
~Neen the outer and center wing panels. If a
For negative accelerated flying conditions, the
distributed flange ty~e of structure is used,
lower surface is in compression thus sheet cov-
ering between corner flanges would be ineffec- the flange forces must be gathered and trans-
ferred to the fitting points, thus a compromise
tive in bending. However, since the load fac- solution consisting of a ~ll number of span-
tors in" these flight conditions are approximate-
wise members is cammon practice.
ly one half the normal flight load factors, this
tneffectiveness of the lower sheet in bending is (5) Wing Flutter Prevention:
usually not c~itical.
With the high speeds now obtained by modern
Cut-outs 11~ewise destroy the continuity of airplanes, carefUl attention to wing flutter
intermediate interior shear webs of such prevention must be given in the structural lay-
sections as illustrated in Figs. (c and i)3 and out and design of the Wing. In general, the
the shear load in these interrupted webs must critical flutter speed depends to a great ex-
be transferred around the opening by special tent on the torsional rigidity of the wing.
~ulkh3ads on eac~ side of the cut-out, which 'Ahen the mass center of gravity moves aft of the
means extra weight. 25 per cent of chord paint, the critical flutter
speed decreases, thus it is important to keep
weight of the wing fon'ard. At high speeds
where ncompressibility" effects became ~portant,
the torsional forces on :he wing are increased,
Fig. A19.4 which necessitates extra skin thickness or a
larger cell. DeSigning :or flutter prevention
is a highly specialized problem.
~ ------ -,
<_C'--
A19.4 ANALYSIS OF WING STRUCTURES
.- .
.'
, ~ \
::1')1v .,\, / \
~60.500
procedure is to find the stresses or loads 1~
all parts of the cell cross-section at several Vr700 i Mx"'l,lOO,OOO
stations along the spanwise direction and tram ~10"""""
these loads or stresses proportion the required Fig. A19. 8
I Mz",aoooo
areas, thicknesses and shapes.
SOL~ION.
In this example, the internal leads will
be calculated for only one section, n&~ely, ASSUI1?T'IONS: - It will be assumed tnat the
that at Station 240. It will be ~ssumed that 3 flange stringers A, Band C develop the 3r.tire
the design critical loads from the critical resistance to the bending ~oments about ~he Z
flight condition are as follows. and X axes. FJr skin ~~cer compression t~~s
assumption 1s nearly correct since the skin will
Mx = 1,100,000 in. lb. Vz = 11,500 lb. buckle under relatively low stress. Since sheet
can take tensile stresses, this assumption is
l1z = 80,000 in. lb. Vx = 700 lb. conservative. However, since the thin sheet
My = 460,500 in. I'D. cover must resist the shear stresses we will
~ake this conservative a3s~~.~tion. ~he main
Fig. Al9.8 shows these shears and ~oments advantage of this approximate assumption is
referred to the reference axes with origin at that it ~kes the structure statical:y deter-
point (0). Moments are represented by vectors minate.
with double arrow head. The sense at the
moment tallows the right hand rule. Fig. Alg.a shows ~~e wing c~t at Station
240. The wlliL~own forces are the three axial
7.27" loads in the stringers A, Band C and :he three
t.J
3. ii' ,2. 33 shear flows Qab' ~c and qbc on the three sheet
,t
-=::::±':::J)
r-..L~" panels, ~king a total of 6 unknowns and since
I T there are 6 static equilibrium equations avail-
STA.O - 32"-"""t able for a space structure, the struc~ure is
statically determlr~te.
,.."' and 8,
PC!.:J.ts A
I
l:l1z(ab) - - 22 Cy + 60000 = 0, whence
Cy = 3636 lb.
r= was correct.
X ax:s
-
To find By take moments about ~
'thr-u (A).
A 240 6.2 3
whence, Qab = 805 lb./in.
1
1296
"
---22'~'....".0:-'111-
fo!
, In Chapter AZ, the stress analYSiS ot an
2Jil 3181 externally braced fabric covered monopl~~e wing
267,-+ A 'lac 2161 C 6 was considered. To provide sufficient torsion-
45 / al strength and rigidity, two external brace
struts were necessary. However, 1t a wing 15
%c metal covered, a single external brace strut
can be used, since ~he closed cellar cells
for,ned by the ~etal sheet covering and the in-
Fig. A19 9 ternal webs provide the torsional resistance
and the wing can be des fgned as a S1.mply sup-
ported beam with cantilever overhang. An
To find qbc take moments about point (a') excellent example of this type ot wing struc-
ture is the Cessna aircraft Model 180 as shown
:Mar = + 2679 X .5 - 2584 X 11.5 - 316 in the photograph. An excellent airplane
x 0.375 - 45 x 22 + 700 x 12 relative to pertormance, ease of ~nutacture
- 11500 X 39 + 460,500 - coo and maintenance.
(665) =0
To introduce the student to the general
approach Of stress ar41YZing such a wing struc-
A19. B ANALYSIS OF WING STRUCTURES
ture , a Liznl t ed -nscuss i c- -"i':h a few ca LcuLa-. Fig. A19.1D and AIG.Il shows the wl~g
~ions ~111 be ~rese~ted. d1mensions ane genera: str~c:~~3.1 :ayout of a
monoplane wing wl~h o~e external brace s:rlt.
~ The wing panel is attached ':0 ~~sela~e by
:§ Single ;in fl':tlngs at points A and 8 with ~1n
axes parallel to X axis. The ~ating 1~~8 ot
4fii\ the :lttlngs at paint A are ~ace snug :1t but
t' L.
namely, (1) The front spar ACE resists the
bendl~g moments and shea~s due to load w . ,
z
Strut C. y (2) The skin and webs of the two-cell tube
resists all mrr~ents about y axis or broad:y
speaking tors~onal moments, (3) the entire ~anel
cross-sect ton resists the bending moment and
Fig. A19.l0 flexural she£r due to drag load wx, with the
top and bottJm skin acting as webs and the t~o
spars as th~ flanges of this box beam.
General Calculations; -
z The unknown external reactions (see Fig.
AI9.ll) are Ay, Az, Ax, By, Bz and DC., or a
total of 6. Since 5 static equations of
equilibrium are available, the reactions are
statically deter.r.lnate. Reaction DC is also
the load in brace strut CC.
30.27 To :lnd reaction DC take moments about x
"/m·4t~m.
axis through pOints A, 3
~_J ZMx(AB) = (- 170 x 30 x 170/2) + DC(60/99.4)60
-til- =0
Fig. A19.l2 whence, DC = 8979 lb. (T::e Sign cernes out
plus so the sense assumed in Fig. A:9.11 was
correct.) The load in the strut 1s therefore
8979 lb. ~ensicn.
To find Ay take ZFy =0 points A and B. This couple force equals the
moment divided by distance between spars or
ZFy =2141 - 8979 (80/99.4) + Ay =0 15300/27 = 566.7 lb.
whence Ay = 5085 lb. Front spar (ACF) loads due to Wz =30 lb/in:
To tind Bz take ~oments about y axis w'" 30 lb./ln.
through (A). t t t t t lltttt
=+ (30 x 170 x 3) - 27 Bz =0 ===c y-,-"...c:::= __
lliy(A)
7226/ I
whence 8 z = 567 lb. acting as assumed. Cz.:t5329
~~. • 59
To tind Az take ZFz = 0 80 Flg. A19.14
ZFz =170
+ Az
x 30 - 567 - (8979 x 59/99.4)
=0
80"
+- 90" --i
whence Az = 796 lb. actlr~ as assumed. Fig. A19.l4 shows free body of front spar
ACE. To rind strut load DC take moments about
The tittings at A and 8 should be designed (A) •
to take the reactions at these pOints as found
above. The external strut DC and its end lMA =(- 30 x 170 x 170/2) + 60 (DC x 80/99.4)
fittings ~t carry the tension load ot 8979 lb. = 0
increasing from zero at tip to 2141 lb. at would ~3ve t8 be replaced by two fltti~gs, one
fuselage attachment ,oints and varying as y~. on the upper flange ane the other an :he lower
At point C on from spar the axial tension load flange in order to be able to resist a ~x
would be .074 (90') = 600 lb. moment. Fig. A19.l6 shaNs this modification.
The fitting at B could remain as before, a single
The design of the tront spar between pin fitting.
points C and E would be nothir~ more tillL~ a
cantilever beam subjected to a bendir~ torces
plus an axial tensile load plus a torsional
shear flow. The design of the spar between
points C and A is far more complicated since
we have appreciable secondary bending moments
to determine, which must be added to the pri-
~ry bending moments. Fig. Al9.lsa shows a
free body or spar portion AC.
w::; 30
fltltt~
~Mc::;30x90x45
Fig. Al9.16
5085-<:- -. - - - -- '
1 Fig. Al9.15a
j The stress analYSiS of this wing would consist
of the spar AE reslstl~g all the Mx mcments and
The lateral load of 30 lb./in. ~ends the the Vz shears and acting as a cantilever beam.
beam upward, thus the axial loads at A and C The torsional moment about a y axis coinciclng
will have a moment ar.n due to beam deflection With spar AE would be resisted by shear stresses
which moments a&e referred to as secondary in the cellula~ tubes tanned by the skin ane the
moments. To find deflections the beam moment spar webs. The d~g bending and shear fo~ces
of inertia must be known, thus the design of would be resisted by the bean whcse flar~es are
this beam portion would fall in the trial and the tront and rear spars ant the web being the
error procedure. Articles A5.23 to 28 of top and bottcrm skin.
Chapter AS explain and illustrate solution of
problems involving beam-column action and such A19.8 Stress Analysis of Thin Skin - Multiple Stringer
a procedure Would have to be used in actually Cantilever Wing. Introduction and Assumptions.
deSigning this beam portion.
The most common ~ype of wir~ con5t~uctlon
The rear spar SF receives two load systems, is the ~ultiple stringer type as illustrated by
namely a varying axial load of zero at F to 2141 the six illustrative cross-sections in Fig.
lb. at B and the web of this spar receives a A19.2. A structure with ~any stringers and
shear load rrcr.m the torsional moment. The sheet panels is statically indeter.ninate to
rear spar is not SUbjected to bending momen~s. many degrees with respect to inte~al stresses.
Fortunately, structural tests of complete wing
In Fig. A9.l0 the secondary structure structures show that the Simple beam theory
consisting or chordwise ribs and spanwise light gives stresses which check fairly well with
stringers riveted to skin are not shown. This measured stresses it the wing span is several
secondary structure is necessary to hold wing times the wing chord, that sweep back is minor
contour Shape and transfer air pressures to and wing is free of ~jor cutouts and discon-
the box structure. This secondary structure tinUities. Thus it is common procedure to
is discussed in Chapter A2l. The broad SUbject analyze and design a wing overall by the beam
at designing a member- or structure to withstand theorJ and then investigate those portions of
stresses sarely and effiCiently 1s considered the wing where the beam theory ~y be in err~r
in detail 1n later chapters. by USing more rigorous analysiS methods such as
these explained and illustrated in Art. AB.IO
A19.7 Single Spar. Cantilever Wing - of Chapter AB.
Metal Covered
ASSUMPTIO~S - BSAM ~~£ORY
A Single spar cantilever wing with metal
covering is often used particularly in light In this crapter the wing bending and shear
commercial or private pilot aircraft. stresses will ~e calculated USing :he un-
symmetrical beam theory. The two ~in assump-
Suppose in the single spar externally tions in this theory are: -
braced wing of Fig. A19.Il, that the external
brace strut DC was removed. Obviously the w1~g (1) Transverse sections of the 8eam originally
would be unstable as it would rotate about hinge plane before ben=ing remal~ plane a~ter be~di~g
fittings at ~oints A and S. To make the of beam, This assumption means that Longftudf -.
structure stable the Single pin fitting at (A) nal strain varies cirectly as the jlstance from
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A19.11
the neutral axis or strain yariation is linear. corner flange members (a) and (b) are stabilized
in two directions by the skin and webs and
(2) The longitudinal stress distribution is usually fail by local crippling.
directly proportional to strain and therefore
trom assumption (1) is also linear. This Now continuing the loading of the wing
assumption actually ~eans that each longitudinal after the skin has buckled, the stringers and
element acts as it it were separate tram every corner members will continue to take additional
other element ~~d that Hook's law holdS, compreSSive stress. Since the ultimate strength
namely, that the stress-strain curve 15 linear. of the stringers is less than that ot the cor-
ner members, the stringers (c) will start to
Assumption (1) neglects strain due to shear bend elastically or inelastically and will take
stresses in skin, which influence is commonly practically no ~Jrther stress as additional
referred to as shear lag effect. Shear lag strain takes place. The corner members still
effects are usually not ilnportant except near have considerable additional strength and thus
major cut-outs or other major discontinuities additional external loading can be applied until
and also locations where large concentrated finally the ultimate strength ot the corner
external forces are applied. members is reached and then complete failure of
the top portion of the beam section takes place.
Assumption (2) is usually not correct if Therefore, the true stress - strain relationship
elastic and inelastic bUCkling of skin and does not follow Hook's Law when such a structure
stringers occur before failure of win~. In is loaded to failure.
applying the beam theory to practical wings,
the error at this assumption is corrected by In the above discussion the stringers (c)
use ot a so-called effective section which is were considered to hold their ultimate buckling
discussed later. load during considerable additional axial
strain. This can be verified experimentally
A19.9 Physical Action of Wing Section in Resisting External by testing practical columns. Practical col-
Bending Forces from Zero to Failing Load. umns are not perfect relative to straightness,
Fig. Al9.l7 shows a common type of wing uniformity of material, etc. Fig. Al9.l8
shows the load versus lateral deflection ot
cross-section structural arrangement generally column midpoint as a column is loaded to failure
referred to as the distributed flange type.
and fails by elastic bending. Fig. Al9.l9 shows
skin similar results when the failure Is inelastic
bending.
a b
'-~vr--~ z
web c(stringers) web Fig. A19.18 Fig. A19.19
La
Fig. A19.17
The corner members (a) and (b) are conSiderably
larger in area than the stringers (c). The
skin is relatively thin. Now assume the wing 6 '" Central Deflection
B '" Central Deflection
is subjected to gradually increasing bending
forces which place the upper portion of this The test results show that a compreSSion
Wing section in compression and the bottom member which fails in bending, normally con-
portion in tension. Under small loading the tinues to carry approximately the ~aximum load
compressive stresses in the top surface will
be small and the stress will be directly ?ro- under considerable additional axial defo~tion.
Thus in the beam section of Fig. Al9.l? when
portional to s~rain and the beam formula the stringers (c) reach their ulti~ate load,
0c = Mxz/l x will apply and the moment of
inertia Ix will include all of the cross- failure of the beam does not tallow since cor-
section material. As the external load is in- ner ~embers (a) and (b) still have remaining
creased the compressive stresses on the thin strength.
sheets starts to buckle the sheet panels and A19.10 Ultimate Strength Design Requirement
further resistance decreases rapidly as further
strain continues, or in other words, stress is The strength design reqUirements are: -
not directly proportional to strain when sheet
buckles. Buckling of the skin panels however (1 ) Under the applied or limit loads, no
does not cause the beam to fail as the stringers structural member shaIIibe stressed above
and corner members are lowly stressed compared the material yIeld point, or in other
to their failing stress. ~he stringers (c) words, there must be no permanent de-
are only supported transversely at wing rib :crmation or deflection of any ?art of the
points and thus the stringers ac~ as columus and structure.
fail by elastic or inelastic bending. The
A19.12 ANALYSIS OF WING STRUCTURES
~I';' kf-.
the stress at failure ot a member the less -ow,
material required and therefore the less I -
structural weight. The stress engineer 1. it- '\
thus tries to deSign members which fail in
the inelastic zone. Fig. A19.21
Fig. A19.20
:: n
~ "
c:
A 19 14 ANALYSIS OF WING STRUCTURES
spanwt sa ~
"
.
consists of a front and rear beam (spar) with
o't...r .l.ng~~
Tayering in cross-sectiorzl
~
C'':'''''c!
material is obtal~ed
o. . .'
bet .....e'-'•• the ... no hceams , I
3efcre equations 2, 3 and 4 can je solved.
axial load in the stringer.
by decreasing size of ~embers by cuttir~ off the effective cross-sect~cn area ~us~ be known
portions of the spanw1se stringers and corr.er as well as :he ~oments of ine::-tia and ~roduct
flanges and decreasing the skin and web thick- of ~nertla about x and z c9~troidal axes.
ness.
ftJ
--r-
/4 "ii2.
"
/3 /2
" 10 q 8
-
UPPER SURFACE STRINGER AND SKIN ARRANGEMENT
<47·5" '00 70
:ITA FIO'-----"F----'T'---'T'-----''t'---'"IL--q.''!.-2.1
'.0 'PO 120
f-- I
I I I I
::lJ1
II
, -\1 Skin thickness
)-."4 i .03.2,. \ ,0z.5 between beams
1/ -\
rOt: '""-'-I -to- 1. "---i
-' ~
I T Upper Front Flange .050
, 1:
.b4
r--- 2 1/4----..-;
Z' 7
~--L_
•~I----------""='T'-'-,r--,j <f"'----,:;-;;:;:;
- - - - - .. o X Centroidal Axis
-~
~ .c
N ~
z '"••
14 C!
8
.032 lower skin
I
,- Lower Surface Stringer
area of bulb angke e • 11
7/8 area skin '" . 16
I -:27
5"
Ir 3.75
1
area of angles '" . 508
area of skin "'.120
area of web "'. 064
. 692 Lower Front Flange
area of angles '" . 508
r2
area of skin "'. 160
area of web "'. 100
:768
-
,
s
Fig. A19. 25 WING SECTION AT STATION 20. EFFECTIVE AREAS.
Z'
I
. 032 upper skin
.•• ,\\.o-c.
0\ ~e'"
7
•o _r
["".0.. , X' reference axis
~:----=====-~r======-=,,========...::;~===..::"::=======-=..:::j::=i
.- X Centroidai Axis
14
13
12
10_j 9 8
Z' Z "'- .025 lower skin
Fig. A19.26 wrsc SECTION STATION 47. 5
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A19.17
TABLE A19. 2
SEC1'10N PROPERTIES ABOUT CENTROIDo\L X AND Z AlC4S
1 , 3
• 5 II 6 7 8 8 10 I II 12
!
13 !
I, Flange
Mo.
Are"
A Z' AZ'
,
AZ·2 X'
! AX' AX';': AX'Z' Z ~ Z' - z! X ~
.' - x I "b
"
= "'b , I
!
,
I 3z
1 .6<18
.353
5.50
5.90
3.56
2.08
19.61
1':::.30
-0.10
5.65
_0.06
1. 99
0
11.30
_0.36
11.78
6.68
7.08
-17.4
-11.65
_37660
_39940 i _24390
_14070
!;
.300 I 5.85 1, 76 10.28 1l.:W 3.36 37.60 19. 6 7 1 7.03
I -
- 6.10 _39620 ! -11890
i • .300 5.55 I 1.66 9.25 16.85 5.05 85.00 28.10 6.73
,
0.45 -37950 ! _11370 I
o
6
.300 5.05 I 1.52
1, 55
7.65
6.33
22.40
28.00
6.72
9.89
I 150.50
277. 00 34, 00
43.50 I 6.23
5.58
I ::;.10
iO.70
-35115
! -31450
, _10510
-11090
7
.3511 4.40
2. ;.!4 7.95 35.72 23.15 826.00 80.00 4.73 18.42 I _26560 ! _16700
8
.630 3.55
.692 _8.40 -5.81 48.90 ! 35.72 24.70 I 881.00 -208.00 -7.22 I 13.42 .;0750 28230
,
8
10
.2701-8.50
.270 -8.50 i
I _2.30
_2.30
19.55
19.55
27.80
22.60
1.50
6.10
208.00
138. 00
-- 52.00
64. 00
I
I _7.32
-7. J2
!
10. SO
5.30
41280
41250 I
11180
11170
'i.i. .160 -8.30 _1.33 11, 05 16.70 2.67 44.60 - 22.20 i -7.12 !
I - 0.60 40100 I 64)0
12 .27
,
-8.00
I _2.16 17.30 10.90 2.94 32.00 - 23.50
- 11.35
-G.82
I -_11.7
6.40 38400 10390
13
1< ."
.768 I
-7.50
_6.50
_2.02
_5.00
15.20
32.50 -
5.60
(LID
L51
.08 !
8.45
0 0.50
_6.32
-5.32 I _17.40
35570
29950 I!
9600
23020
,· _6.56 • _1.176
Ge ne r a l Nates:"
I I
·
5.584
96~44 ;. 11.3
5" l' ig. A19.25 for Section at Station 20.
Rete e enc e axes X'X' and 2'2' are assumed
Properties are calculated with respect to th"se
shown.
"
238 _ 5.~84 ~ 1.176~ axes and transferred to the c e nt r ci da I X c nd
230.3 in.';
Z axes.
2700 ~ 5.584 ~ 17.3 2 1030 in.';
TABLE A19. 3
SEC1'IOK PBOPERTIES ABOUT CIDfTROlnAL X AND· Z AXES
1"lnn;.;o Area
AZ'~ AX· 2 ,. - . _ 1r ,
)(0. A Z' AZ'
I
X· AX' AX''Z'
" Z' X X· "b • "'"b .J,
,
1 • .;7G 5.(0
I
2.66 14.95 - 0.10 - 0.05 0 - 0.27 6.21 -15.60 _3927:; ~18700
I
!,
I 3
.318
.2136
~.83
~.75 I
1. 86
1, 53
10.80
8.80
4,40
10.00
1. 40
2.56
6.16
26.60
8.15
15.30
6.44
6.36
,, _11, 10
- 5.50
-40790
-4025 ,
_12990
_10710 ,
I, •56 I
.266 1, 44 -38400 I _10200
.266
5.'Z
4.90 1, 30
7.81
6.40
15.60
~1. :::0
4.15
5.64
64.75
1::0.00
22.40
27.60
6.03
5.51 i 0.10
5.70 _3~075
-
9320
1
.318 4.30 r 1, 37 I 5.88 I :':6.70 8.50 226. :;0 35.60 4.91 !, 11.20 -31360 ! _10000
i
i 7 .476 I _7.10
3.60 1.71 6.17
i 31.70 15.10 478.00 54.40 4.21 16.20 -27030 , _12880
!
8 .35'::
i -7.40
-3.93 27.90 31.70 17.51 :iS5.00 -124.50 I -6.';9 16.20
,I 41050 I 22770
I
8
10 !,
.23:;
.':;35 -7.40 I
_1.74
1. 74 I
1:::.90
12.90
:i!5.50
21. 30
I, 6.23
5.00
165.00
107.00
-
-
';6.10
37.18
_6.79
_6.79
I
11.00
5.80 I
43090
43180 I
10140
10160 i
II .1:.:5 I -7.20 ! -0.90 6.48 15.10 1.89 l8.50 - 13.60 -5.59 - 0.40 i .;1900 I 5240 ,
1 c~ · ;:3<: ;_e.. 90 _1.62 11.20 9.60 :':.::6 :0:1.70 - 15.GO -6.29 _55.90 I 40085 I 9410
13 . 23:; ,i _6. 2S _1.4 • 9.20
-
4.50
I -
1. 06 4.7S - 6.61 _5.84 _11, 00 i 36060 I 8470
,
"1 , 50~ -::'.40 I
i
_3.27 17.70 0.10 0.06 0 0.33 _4.79
,,
_15.50 30725 18600
I I - 79.0
..
4.61 _2.80 1~9. 1 71. 3 1804 0 !
-~'. aO/4. 61 . _0.51" I
r . 71.3/4.61 . 15.50"
General Notes:
See Fig. A19.26 for section at Station 4 •. 5
.
Ix IS!!. 1 157.4 i
I,
·. -.9 -1304 4.61 15.5': . 700
°b -14.5 X - 6360 z
~p equals the cbAnge in stringer ~~1al load areas are referred as ~ values (See Fig. A19.27)
over a distance d in the y direction. Column 6 of Table A19.4 records these double
areas which were obtained by use of a planimeter
Since the cell in our problem is closed the Column (7) gives the moment of each shear flow
value qo at any point is unknown. we assume it about the e.g. and the total of this col~
zero on web 1-14 by imagining that the web 1s gives the moment about the e.g. of the complete
cut as shown in Fig. A19.27. Equation (6) thus shear flow system of Fig. Al9.27 or a val~e of
reduces to, 256060 in.lb.
I F
~
Fig. A19.27, remembering that qy qx qz. For point 0 equals q
rules giving direction of qx and qz refer to times twice the area
Chapter Al4, Art. A14.6. (A 1. + A a ) . In most
cases, the area Aa , A. z,
245.9 288.8 ", , 374.6
414.2 can be neglected.
206.'
I - - -;. 4 5, - - -6--- 553.2
l 1
By Simple
-rj geometry, the area
z I A1.= 1/2 rx, Za - X",Z;s)' The moment of the
x -W area qz~ shear flow q on web (2-3) thus equals q (X;, Z.
:
-14070 -12990 39.1
3 -11890 -10110 42.9 245.9 43.8 10780 -328 - E12. I
•s VZ'12r~q:
-11370 -10200 42.5 288.8 42.0 12120 -328 - 39.2
331.3 42.6 14100 -328
!~.~.
-10510 - 9320 43.3 3.3
e ~11090 -10000 39.6 374.6 43.6 16340 -328 46.6
414.2 42.0 17500 ·3213 86.2
7
a
-16700
28230
·12880
22770
139.0
~198. 7
SSJ.2 197.0 108900 -328 225.2 v x-- 27001~ .01--L-
11.8 -
s _ 37.8 354.5 47.4 161300 -328 26.5
My=39~33.3
11180 10140
10 11170 10150 _ 36. 7 315.7 37.2 11800 -328 - 11.3
11 5430 5240 ~ 43.2
280.0 40.8 11420 -328 - 48.0
11 10390 S410 • 35.15 236.8 42.2 10000 -328 - 91. 2
13 9600 8470 - 41. 1 201. 2 38.0 7650 -328 -126.8 Fig. A19.28
L60.1 54.0 8650 -328 -167.9
, 14
1
23020
I
18600 -160.3
0 204.0 0 -328 -328
Therefore moment ot external loads about
Total mq . 256060 c vg, is,
For equilibrium af all the forces ln the ZMc.g. =12000 x 33.3 - 2700 x 11.8 - 390000
plane of the c~oss-section ~y ~ust equal zero. = 41800 In.lb.
ror convenience we will select a morr.ent y axis
through the c.g. of the c~oss-section. The Moments Produced by Inclination 01 Flange Loads With
~oment of the shear flow q on any sheet element Beam Section.
equal q times double the area of the triangle
formed by joining the e.g. with lines going to Since the flange members in general are not
each end of the sheet element. These double normal to the beam secttons, the flange loads
A 19 20 ANALYSIS OF WING STRUCTURES
have components In the Z and X direct1ons. Due to external Loads = 4.1800 in. lb.
Columns (4 ) and (7) of tree Table A19.5 g1ve the
values of these In ;Jlane c:::mponents. The slopes T;;'e~ the total unbalancec :::o:n.e~t = J.C"'~
• ...-•.A) +
dx/dy between stations 20 and 47.5 are f'olU1d by 256060 + 4l8CO =
302795 Lnc Ib ,
scaling from Fig. A19.24. F1g. Al9.29 shows
these 1nduced In plane forces as fOW1d in Table For equt Lfbr tum, this ::!lust be baIancec oy
A19.5. a constant shear- flow q,.
Flange dx
Px = I M e. g• d,
Fz " Me. g, = The shear stresses q, are listed In Column
(8 ) of Table A19.4.
p
No. dY dY
p dx I = Px z dy p.!!!..
dy
·(P z Xl
The final or resultant shear flow qr at an y
I -24390 -.046 1120 7480 .025 ·610 -10600
2 -14070 0 0 0 .022 -310 • 3610
pOint therefore equals
3 ·11890 0 0 0 .022 -262 · 1600
4 -11370 0 0 0 .022 ·250 · 112
qr = q + q,
,
5 -10510
-11090
0
0
0
0
0
0
.022
.025
-231
-278
1180
2970
7 ·16700 ,021 - 350 -1660 .025 -417 7770 The resulting values are given 1n Column 9
,
B 28230
11180
,021
0
594
0
-4300
0
.018
. 017
-508
·190
9350
2000 of Table A19.4. r1g • A19.30 illustrates tr..e
10 11170 0 Q 0 .015 -167 885 :-esults graphioally.
11 6430 0 0 Q .015
"
.-177 · 57
12
12
14
10390
9600
23020
0
0
-.046
0
0
·1060
,
56-10
Q
Q
,017
,018
.018
·173
-416
• 1130
• 2020
• 7250 -_.- _.
1
/ /
2 3 4 -5- ,--, /
~~ II
7160 • 2224
7.~ •
+,.g.
m
NOTES:
--Column (2); P Ircm Table Al9. 2
Column (5) and (8): Values of Z and X are found iII
0' '/
Colwnns 10, 1 I of Table Al9. 2
---
lli
'(L//
13 12
'-_.-
11 10
---8
9
810 417 Fig. A19.30 Final shear flow diagram. For val-
I
t·
3 ues see Column 9 of Table Al9.4.
r"
1 I' 15 16
310 282 250 231 278 Having deter:nined the shear flows, the
shear stress on any panel would be qjt. I:l
+,.g. checking the sheet for strength In shear and
106
I
14 ~3 lil 1,1 .
10 9 8
I
594
combined shear and tension, interaction rela-
tianShips are necessary. The strength deSign
418 1~3 1;7 9
18
d7 1~0 508
of sheet panels under combined stresses Is
covered in considerable detail i::1 Voli.J:l8 II.
A19,14 Bending and Shear Stress Analysis of 2-Cell
Fig. A19.29 In plane forces prOduced by flange Multiple stringer Tapered Cantilever
axial loads. Wing.
The moments of these in plane components about A two-cell beam IS also quite common In
the section e.g. are given in Columns (5 ) and wing structural design. A 'DNa-cell structure
(8) of Table Al9.5. In general, these moments In bending and torsion Is statically indeter-
are not large. minate to the second degree since the shear
flow at any one pOint In each cell Is unknown,
Total I'!oments of All Forces About Section .:.¥; . However, due to continuity between cells the
at atat i cn 20: angular twist of each cell must be the same,
which gives the additional equation necessarJ
Due to flanges = 7160 - 2224 4936 in.lb. = for solving a ~No-cell beam as compared to the
(Ref. Table Al9.5) single cell analysis.
Due to assumed static shea:- :low 256060 = Example Prcb Ieta
1n.lb. (Ref. Table A19.4)
To avoid re'Oet1t1on of similar t e calcu-
£ SKU
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A19.21
lations as was used in the previous single cell The first 7 columns of this table are the same
problem, the bending and shear stresses will be as in Table A19.4, since no stringers have been
dete~lined for the same structure as in the added to cell (1), and the shear q is assuoed
previous example except that the leading edge zero In cell (1).
cell is considered effec:lve, th~s ~king a
2-cell structure. Since there are no spanwise ~o ~e the twist of each cell the same and
st~ingers in the leading edge, very little skin also to make the summation of all torsional
on the compreSSive side will be effective. On forces zero w111 require two unknown constant
the tension side, the leading edge s~in would shear flOWS, q~ in cell (1) and qa in cell (2).
be effective in resisting bending axial loads Thus VNQ equations will be written, namely:
and thus the moment of inertia would be slightly
different from that found in example problem 1. (8 )
Since this pr0blem is only for the pur,ose of
illustrating the use of the equations, the Zl1c .g , = 0 (9 )
leading edge s~in Nill be neglected in computing
the bending flexural stresses. ~lth this The VNist Q per unit length of a cell
assumption, the bending stresses and flange equals
loads at stations 20 and 47.5 are the same as
for the previous prob Lem. (See values in ccj.cnn _ 1
12 and 13 of Tables A19.2 and A19.3.) Q - 2AG 1:qL/t uo:
Shear Flow Calculations: - The modulus of rigidity G w1l1 be assumed con-
stant and thus will be omitted.
To compute the static shear flow, each cell
is assuned cut at one point as shown in Fig. Consider cell (1): (Refer to Table A13.6,
A1S.31, and thus the shear flow is zero at C01UITL~S 10 ~~d 11)
points (a) and (b).
Fig. A19.31
Area of cell (1) =23.5 sq. in.
G L/~
·~r
2 3 4 Q, ; _1 q Lit. Z q, Lit. 1: q.
5 2A,
[
6
~83.5 14 Area
Cell (2)
:z461. 5 sq. in. Q, ;
2 x1 83• 5 ~ + 1278 q~ - 230 qJ, whence
9
1.3 12
'.' 10
Q1 =7.65 q~ - 1.378 q.- - - - - - - - (a)
TABLE A19. 6
SHEAR FLOW 2 CELl. - -
VULTIPLE FUNCJi TAPERED BEAM
(Aver:li:C Sh.. nr Flo.. Between Stations 20 and 47.5)
1 ! , I
3
,
, • I , 6 7
• 9 10 11 12 13 14
Flant:e i
<0.
P :P .l I." ),,~P_ • , LI<
~(~t c'(gl
q (Lit)
Ce~~
'r ,
i at St;... i :J.t_Sta·l_ '; p.
! cO , 4 •• 5 , .1.0> 1-7.;, "' (i;. ) I I ~(~) (2)
'1 '2
Cl-KIl-KI2
,
1 1- 24390 , _18100 206.8
206. sl
48.4 10000 5.15 .04 1<4 29800 0 _317 -1l0.2
2
,
3
I -14070
i -11890
.11310
I
_12990
_10710
_10200
39.1
4:.\.9
245.9 43.S
288.8 4:;:.8
10780
12120
5.50
5.50
.04
.04
137
137
33700
39600
0
0
-317
31 7
I -- 11.1
28.2
42.5 -
_317
331.3 42.6 14100 5 • .50 .04 137 45400 0 H.3
1
,
,
, I
,
S~ I I 256060 1218 2469 I 0 133050
MOTES: ~ length of web she~t betweea flange s.mbers. Cal. ~ .. (Cal. 2 + Col. 3)/27.5
t web thickness.
A19.22 ANALYSIS OF WING STRUCTURES
~1.5 ~733050 q~
A19.15 Bending Strength of Thick Skin - Wing Section
2 X - 230 + 2469
?igs. i and k o! Fig. AIG.3 ill~s~rate
approxi~ate srzpes for air~oils c~ supersonic
G. = 794 - .2495 q, + 2.678 o, - It)
aircraft. Such e u-ro i ;s nave r eiat ive iv low
thlc~~ess ratios and since supersonic ~l:itary
For continuity O~ ~ust equal O~, hence aircraft have comparatively h:gh wl~g loaclngs,
equatlr~ (al and (b):
it is necessary to go to thick skin in order to
resist the wing bending ~cments efficiently.
7.899 q~ - 4.056 qa - 794 = 0 - (c)
The ultinate compressive stress of s~ch struc-
tures can be made rather uni rcrn and o c curr-t ng
For equilibrium the summaticn of all mom- at stresses considerably above t~e yield ~oint
ents in the ,lane of the cross-section abo~t the
of the material. Since struc:ures wust ca~ry
section (e.g.) ~ust be equal to zero, or the design loads without failure, it is neces-
sary to be able to calCUlate the ultimate bend-
Z ric.g. = 0 ing resistance of such 3. wtng section i t t::e
~argin of safety is :0 be given ~or various
The ~~ent of the external loads about the load conditions.
section e.g. is the same as in previOus problem.
= 41800 in.lb. The question of the ultimate bending
!1 external forces resistance of beam sectlor~ that ~al1 at stresses
beyond the elastic stress Yar~e is treated in
The induced moment due to the ~n plane
Article A13.l0 and example problem 7 of Chapter
components of the flange axial loads is likewise A13 and should be studied again betore ~ro
the same as in previous problem (see Table ceeding with the following example ,~oblem.
A19.5).
A19.16 Example Problem
Mdue to flange loads =4936 in.lb.
~a illustrate the ;roced~re of Art. A13.l0,
The torsional moment due to ~he static a portiO~ o~ a thick =kin Ni~g section as
shear flow rr ce Column (7) or Table Al9.6 equals 1l1L.:.strated In F'fg , A19.34 'N111 be cons Lder'ed ,
256060 in. lb. The torque due to the unknown
constant shear flows at q1 and q~ 1s equal to
twice the enclosed area ot each cell times the
6" 6'" 11.-3/ S'"
shear flow in that cell, whence
l~
3" ~l2
3 ~ 1","" T
3 -l13/8"
+
8 2 a
due to q::,. and Gil) = 2x83.5 q~ -+-2 x 461.6
l'l( 1 '"
2-
qll = 167 q:l, -+- 923 qll .t.. 2
---x
ation of crippling and column strength at the The values in colljffiTI (3) of Table A1S.7
sti:fened Skin, a subject treated later. represents the true compressive stress at the
~idpoint ot a strip area when the beam 1s
Curve (A) of Fig. A19.35a is a portion of resisting its maximum or failing bending'nament.
the compressive stress-strain diagram of the The values in column (4) represent the com-
aluminum alloy material from zero to 50,000 PSi. preSSive stress at the midpoint ot the strip
areas it the bending stress is linear and vary-
Due to symmetry about the x-x axiS, we ing trom zero at neutral axis to 50000 psi at
need only to consider one halt or the beam edge of beam section (Curve B of Fig. A19.35a).
section. We divide the upper half of the beam
section into a horizontal strips, each 3/8 inch To illustrate, consider strip area number
thick. Each beam portion along these horizontal (2) in Fig. b ot A19.35. ~oject a horizontal
strips can be placed together to form the areas dashed line from midpOint of this strip until it
labeled (1) to (8) in Fig. b or Fig. A19.35. intersects curves A and B at paints (a) and (b)
Since plane sections remain plane after bending respectively. From these intersection paints
in both slastic and inelastic stress zones, project downward to read values of 48000 and
Fig. c shows the beam section strain picture. 40600 pSi respectively.
Fig. A19.35 In using the linear beam formula, the
stress intensity on strip (2) would be 40600
but actually it is 48000. The ratio between
the two 1s given the symbol K. Thus to modify
the linear stress to make it equal to the
nonlinear stress we increase the true strip
areas by the factor K, giving the results ot
co Lumn (6).
~
' " Approximate 3-3, stringer (6) becomes fully effective and
Load ill '<, ~ thus R = 1.0 for all stringers. The tinal
I'
I
Stringer modified stringer area (A) in column (4) of
(2)
0';;- --'- Table A19.l would then equal the true stringer
area plus its effective skin times the factors
B ----3b ~A KR. The procedure tram this point would be the
same as discussed before. Thus shear lag ap-
Fig. Al9.37 proximations can be handled quite easily by
~odifylng the strir.ger areas. USing these
A19.l9 Application of Shear Lag Approximation to modified stringer areas. the true total loads
Wing with Cut-Out. in the stringers are obtained. The true stresses
equal these loads divided by the true stringer
?ig. A19.38 shows the top ot a multiple area, not the mOdified area.
stringer wing wh1ch includes a cut-out in the
surface. The str1~gers (5), (6) and (7) must A19. 20 Approximate Shear Lag Effect in Beam Regions
be discont1nued through the cut-out region. where Large Concentrated Loads are Applied.
3 2 Wing and fuselage structures are otten re-
I I
D :: :~ I I /Cut-out
II
quired to resist large concentrated torces as
for example power plant reactions, landing gear
reactions, etc. To illustrate, Fig. A19.39
'r--! I
I
r represents a landing condition, with vertical
<, load. The wing is a box beam with 7 stringers
-'H,h ~I
a1.-/
1.-/ <,
'
c ----... .-/ Top Surface A
,--' "I <,
b
l ~ <, Effective-
-, 1-3b_ f---3b -.I ness Tri-
-, an 1e
I
:-- I , b -;1
-, '-'
1 I
3 2
Fig. Al9.38
II I j I ,
~ Dead wt.
inertia forces
It 1s assumed that the ef:ectiveness of
these 3 inter~~pted stringers 1s given by the
triangles in the figure. At beam section 1-1 ~
~ Bending Mom.
these strir~ers have zero end load. The
stringer load is then assumed to increase IV~
linearly to full effectiveness when it inter-
sects the 51des at this triangle whose height Fig. A19.39
A19.26 ANALYSIS OF WING STRUCTURES
I A
and flange nemoera . Fig. (a) shows the bending
moment diagram due to ~he landing gear react~cn 6"
T
cr-----......
C B
alone. The internal resistance to :his ~ending 20"~
moment cannot be uni:orm on a beam section
Tn
adjacent to section A-A because of the shear
strain in the sheet panels or Nhat is called
shear lag effect. To epproximate this stringer
';
effectiveness, a shear lag trian~le of length
3b 1s assumed, and the same procedure as
discussed in the ~revlous article on cut-outs
is used in finding the longitudinal stresses.
ISO" J I I
It should be understood that the bending ~oments I I
l ! s
due to the distributed forces on the wing such
as air loads and dead weight inertia loads are I I
not included in the shear lag considerations,
only the forces tr~t are applied at concentrated I
points on the structure and must be distributed
into the beam. A side load on the gear or -I 9" !- 21" I 60"
power plant would produce a localized couple -r/ Cell
plUS an axial force besides a shear force as in
Fig. AI9.39. The resistance to this couple and l ~-.a:.t,,-II_..EO.. -=====-
axial force would likewise be based on the
effectiveness triangle in Fig. AI9.39. Fig. A19.41
A19.21 Approximation of Shear Lag Effect for Sudden intensity acting upward in the z direct10n
Change in Stringer- Area and 0.25 lb./in.a average Intiens i ty acting
Stringers of one size are otten spliced to rearward in the x direction. The center of
stringers of smaller size thus creating ~ dis- pressure for z forces is on the 25 percent
continuity because of the SUdden change in ot chord line measured f~om the leading edge
stringer area. edge and at mid-height of spar AB tor the
x air forces. Assume the 3 stringers A,
Fig. A19.40 shows the stringer arrangement 8, C develop the en~ire resistance to ex-
in a typical Sheet-stringer wing. Stringer B ternal bending moments. Find axial ~oads
1s spliced at paint indicated. The stringer 1n stringers A, 8, C and the shear :low in
area A. is decreased SUddenly by s~licing into the 3 sheet panels of cell (1) at wing
a stringer with less area A1 • stations located 50~, 100~ and 150" :ram
wing tip. Consider structure to rear of
Top Surface of Wing cell (1) as only carrying a1rloads tcnvard
to cell (1) and not resisting'wing torsion
or bendi:Ig.
Fig. Al9.42
•
I Single
-3b 3b"'" Pin
FIttings
Fig. A19.40
To approximate the shear lag effect,
assume the area of stri~er 8 at splice paint 'c
to be the average area of the two sides or I
and Wx = 5 lb./in., acting to rear and Table A gives the str~nger areas at sta-
located at mid-depth of wir~. Find re- tions 0 and 150. Assume stringers have
actions at points (a), (b) and (d). Find linear variation in area be~Neen these two
axial loads on front and rear spars. Find stations. Use 3Dt as effective skin with
primary bending moments on front spar. compression stringers.
Find shear flow on webs and walls. Neilect
structure forward of front spar and rear- Find axial loads in stringers at stations
ward of rear spar. 150 and 130 and detennlne shear flow
system at station 150.
(3) Fig. A19.43 shows a portion of a single
cell - multiple stringer cantilever wing. (4) Same as problem (3) but add an internal
T~e external air loads are: web of .04 thickness connecting stringers
(3) and (8).
Wz = 100 lb./in. acting upward and whose
center at pressure is along a y axis coin- (5) Same as problem. (4) but add a leading edge
ciding with stringer (3). cell with radius equal to one-half the
front spar depth. Take skin thickness as
Wx = 6 lb./in. aC4ing to rear and located .04 inches.
at mid-depth of wing.
TABLE A
'D' 6
J...
f-3• 3 +-" . 3j
-12
5
..i
1 4
<1-- 12 -+4 i
I I
I I
I
,
I
I
I I
I
I
I
,.032
T
~-J
T
1 ~-.050
12" .032~~ 10"
C 025
1
~5"" 52 t 5 +5j 150" I
STA.O
STA.150
Fig. A19.43
, - ,- _ .' -'.0"- - - .-. . '~"'.::1
North American Aviation FJ3 "Fury Jet". View Shows Bottom Wing Skin.
Note Integral Construction of Skin and Stringers.
North American Aviation F-1OO "Super sabre" Wing. View Shows End
of Outer Wing Panel. Note Thick Skin.
,....
Douglas DC-a Wing. View ShOWing Lower Surface of Outer Wing Panel
Between Center and Rear Spar.
CHAPTER A20
INTRODUCTION TO FUSElAGE STRESS ANALYSIS
A20.1 General. In general the pur-pose of an air- to Chapter A5, where ex&~ple calculations of
plane 1s to trans?ort a commercial payload or a fuselage shears and moments are presentee.
military useful load. The commercial payload
of a ~odern airliner may be 100 or ~ore pass- The basic fuselage structure 1s essentially
engers and their baggage. These passengers a single cell thin walled tube with many trans-
must be ~ransported safely and comfortably. verse fr&~es or rin~s and longitudinal st~ingers
For example, an airliner flies at high altitudes to prOVide a combined structure which can absorb
where temperatures may be far below zero and and transmit the many concentrated and distri-
where the air denSity 1s such as not to sustain buted applied forces safely and efficiently.
human life. These facts mean that the body The fuselage is essentially a beam structure
which car~les the passengers must be heated, subjected to bending, torsional and axial
ventilated and pressurized to prOVide the forces. The ideal fuselage structure would be
necessary safety. Air travel ~ust be acceptaJle one rree of cut-outs or dt soont.tnut ties, how-
to the passengers," thus the airplane body must ever a practical fuselage must have ~y cut-
shield the passengers from excessive noise and outs. Fig. (a) shows the basic interior fuse-
vibration. and further.nore efficient, restful lage structure of a small airplane with skin
and attractive furnishings must be provided to removed. It consists of transverse frames and
~ke travel enroute comfortable and enjoyable. longitudinal stringers. Photographs 1, 2 and 3
The portion of the airplane which houses the illustrate fuselage construction of late model
passengers on payload is referred to as the large aircraft.
fuselage. Fuselages vary greatly in size and
confi~uration. For example, the ~uselage of a
supersonic military airplane ~y house only one
passenger, the pilot, the remainder of the
fuselage interior Spdce being used to house the
,ower plant, to prOVide retracting space for
landing gear, and to house the many mechanical
and electronic installations which are ~ecessary
to :ly the airplane and carry out the various
operations for whlcn the airplane was designed
to acccmplish. Many groups of engineers with
various backgrounds of training and experience
are therefore concerned with the deSign of the
fuselage. The structures engineer plays a very
L~portant part because he is responsible :or
the strength, rigidity and light weight of the
fuselage structure. Fig. (a)
A20.1
A20.2 FUSELAGE STRESS ANALYSIS
PHOTO. NO.2
View LOOking Inside of Rear Portion of Fuse lage
of Beechcraft TWin- Bonanza Airplane.
PHOTO. NO.3
Fuselage Construction of Boeing
707 Jet Airliner.
(FOR GENERAL DETAILS OF DOUGLAS DC-a FUSELAGE CONSTRUCTION sEE PAGE A15. 32)
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A20.3
A20.3 Stress AnalySiS Methods. Effective Cross- (1) A small width of sheet w l on each side of
Section. the attachment line of skin to stringer ~s
considered as carrying the same compressive
It is common practice to use the Si~plified stress as the stringer, as was discussed in
be~ theory 1~ calculating the stresses in the Chapter A19. These effective sheet widths
skin and stringers of a ~uselage structure. If ""l are shown as the blackened elements ad-
the fuselage is pressurized, the stresses in the jacent to the stringers on the compressive
skin due to this lnterr~l pressure ~ust be added side in Fig. A20.1.
to the stresses which resist the flight loads.
In wings the skin in the ~iddle region of the (2) The remainder of the curved sheet between
airfoil 1s relatively flat and thus the skin is stringers, namely, b-(w l + w.) carries a
usually considered as made up of flat sheet maxtmum compressive stress Ocr = .3 E t/r.
panels. In fuselages, however, the skin is This value for ocr is conservative. E is
curved and curved sheet panels have a higher the modulUS of elasticity of the skin
critical compressive buckling stress tr~n flat material, t the skin thiCkness, and r the
panels of the same size and thickness. In radius of curvature of the skin. These
small airplanes, the radius of curvature of the curved sheet elements are shown by the
fuselage skin is relatively small and thus the hatched skin lengths in Fig. A20.l.
additional buckling strength due to this curva-
ture may be appreciable. A simple procedure ot Since the thin curved skin between the
approximately including the effect of sheet stringers normally buckles under a compreSSive
curvature will now be explained. stress far below the buckling strength of the
stringers, the curved sheet is treated as an
Fig. A20.l illustrates a distributed element with varying effective thickness which
stringer type of fuselage section. Assume that dependS on the ratio of the curved sheet buck-
external loads are applied which produce bending ling stress ocr to the bending stress crt exist-
of the be~ about the Y axis with compression on ing at that point for bending of the fuselage
the ~pper portion of the cell. section. Hence the effective sheet thickness
for the c~led sheet panels can be written,
'," - - - - (1)
or an effective area can be written
----(2)
...;'
x'" 4
3 2 1
to zero as the neutral axis is lpproached, ~~d
since the curved sheet Jetween the Z stringers
can carry loads up to its b~~kling stre~g~h, a
;Z 6
5 preliminary value of e r r ec t ive width 'Ii =
4:0t
will be ass~ed acting with each stringer. Thus
" total area of st~inger plus effective skin
"t: 8
7
Height", 50"
equals 0.12 ... 40 x .032:01 = 0.16 sq. 1::1. The
number of st~ingers required is theref)re 2.00/
-+,0
"
e- 11
9 Width '" 30"
0.16 = 12.
Fig. A20.2 shows how the stringers were
placed to give 12 stringers on the top hal:.
y' -'-......-a--lo-=~...J.---+------+- y' Since the skin on the lower half is in tension
and therefore fully effective, the neutral ~xis
;::=F=-::-::...:=--=::j:=N~.~A.Trial2 wtll fall below the center lir.e and thus the
N. A. Trial 1
two stringers on the center line will be con-
Z", .135" sidered as part of the requi~ed 12 stringers.
A t~selage cross-section has ~OW been obtai~ed.
15
The desired :inal result is that the naximum
16 Fig. A20.2 compressive stress will be near but nat Over
1 32000 pS1. The procedure f~om this ;oint is
18 still a trial and error ~rocess since the
19 Symmetrical about effective sheet on the compressive siGe depends
20 the Z axis
on the ma~itude of the compressive bending
21 2223 stress which in turn 1s influenced by the ~ount
Location of stringers or effective sheet ~~d the buck11ng load carried
below y' - Y' same as by the curved sheet.
above Y' - Y'.
Usin~ the preliminary stringer ~rrangement
Solution: ot Fig. A20.2, Tables A20.1 and A20.2 give the
calculation of the e~fect~ve ~ornent of inertia
The first thing to do Is to determine ap- of the section about the horizontal neutral
proximately how many Z stringers will be re- axiS, Table A20.1 deals with the stringers and
quired so that a section can be obtained to the effective sheet elements and Table A20.2
work with. Since the internal resisting ~om deals with the curved buckled sheet elements.
ent must equal the external bending moment, one
can guess at the internal resisting couple in In the trial No. 1, ~he following assump-
terms ot total compressive flange stress and an tions are made:
effective internal couple arm.
(1) A width of 30 thicknesses of skin act
For elliptical and circular sections with with each stringer en the upper or ccmpress1ve
distributed flange ~terial, the approxi~te side.
effective resistir~ ar.n of the internal couple
can be taken as 0.75 times the height ~, and (2) The area of the curved sheet betwee~
the average tension or compressive stress as the effective sheet widths as fOW1d in (1) is
2/3 the maximum stress. ~hus equating the modified to give ~~ eftective area by ~ulti
external bending ~ament to the internal resist- plying by a K ~actor of 0cr/ab. where a cr 1s
ing moment an approximate total area Ac for the the buckling compressive stress and 00 is the
compreSSive side of the fuselage section can be bending stress at the center of the curved sheet
obtained. element assuming 32000 at the extreme upper
fiber of the beam section and zero at the hori-
=: Ac (.67 0b)(.75 h), Whence, zontal center line, With linear variation in
between these points.
= 160000 • 2.0 sq In
.75x50x .67 x 32000 ... NOTE: Since the entire skin on ~he lower half
t s e r rec t tve , a mor-e logical aasumpt t on '.... cu.Ld
Part of this total area is pr~vided by the be to guess at the l~cation of the neutral axis
effective skin area. The ef~ect~ve width to and use a variation of ab between the neutral
use with each rivet line equals w = Ct ,;E/crst. axis and the extreme fiber. ~hiS approaCh will
We will take C = 1.7 which is a commonly used not be used in this example.
va Lue ,
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A20.5
TABLE A20.1
1 2 3 • 5
! , 7
• 9 10 11 12
-.
E:<planatory Notes for Table Ala.l
.-.
•, •
;1
, ::: 0 :;~
'N
N =~.o ,. • . I TRIAL NO. 1
••
Area I
az· 2 , ~, •0 ,- .,
- ·.,-- ·..-.. ••.. •• " I "'
••
•• ., I
,z2
Col. 1 For DUlllbering of sUff-
iiiei='S'iDd sh.ee't elellMlDts, .ee
.. '....'. '. .
-, a
aq.i:n.
A~
z'
."
, . >. Fig. xao.a.
.~
• ," b• • Col. 2 Stiffener area .12 •
" 30 • • 032 x .032 .18
re. s'ti!feners
, ,
I •"
. :1e .15
• lS
.15
24..21
22'°1
18.2
3.631
3.30
2.73
87.'
72.6 I 25.38
27.58 _30020 1.0
_27700
4.9.71 21.58 -23550
1.03
1.12 I
.032
.033
.036
.152
.153
.156
4..19
3.88
3.36
115.4.
98.5
72.4.
2.
' 10 • Below
the centerline each
010
• .15
.15
13.3
6.9
1, 99
1. 04
26.4 16.68 _18200
7.2 10.28 -11200
1.28
1.66
.04.1
.053
.161
.173
2.69
1.78
44.7
18.3
s'titfener co~idel"ed
aC'ting sepal"ate1y. The '"
.:: 12
~ 13
.12 I ° I ° ° 3.38
2.3 -0.18
3680 - 2.88 • 092 .212
.226
0.72
-_0.42
.04
2.'
0. T
entire skin be'tween
conllider_
." I
.224
.1201
-3.2
_6.9
-0.72
-0.83 5.7 _3.52 c •
••
.120 1.5
s'tifteners
ed as uui't. , "
::: 15
-16
.,7
.224.
.12
.160
-10.1
-13.3
-15.8
_2.28
_1.60
_2.53
23.0 _6.72
21. 3 -9.92
40.0 ! 12.42
••
• ·,•
•
.226
.12
.160
-1, 52
_L19
_L99
10.2
11.8
24.7
CoL J All z' m
~ e' to horizontal cea-,
,~
terlille aJtis.
llIeaaur-
~
·• -"••
~
.16
.12
.18
.12
_2.70
_2.24
_3.25
-2.50
45.6
41.6
66.0
52.1
TRIAL NO. 2
~ z s dis'tallce 'cc neutral
aXis ,.
found fro. l"l!-
suits of Trial No. 1.
.23 .088 24.91 _2. 19 1 54.5 -21. 52 • • .088 _1.89 4.0.7 Col. 7
Col. 8
~h • 1600000 z/1470
~ effective widtb based
aUK. 2.4.9 -12.23 714..3j 2.627 _2.86 .73 .n S'treS8 1111 Col. 7
TABLE A20.2
TRIAL NO TRIAL NO 2
1
;
-.••.."•,
I 2 3
• 5
. .-••• 6
·
7 9 10 11 12 13
- .- • •- • ..• --"...
,
• •-
"~ 15 16
·•
17 18
--. -•
0 'b
-•••• ~
•
, • •
oOM"
• • <
0
• N
_30800
>"
ON
• >< ,
Exp taoato rv Notes to r Table Azu.z Explanatory Notes for Table Azu.z
I Trial xo.
1 Trial xe, 2
Col. 1, 2, 3.
Col. 5
Col. 6
• (see Fig. Ala.l for mean1ng of terms)
.E • 10,300,000 for alW11inwa alloys
Cb val"ieS as :II. s'tl"a1ght line fl"Olll 32000
top of cell to Z8l"O at centel"11n8
=
distance frOlll c81l'tro1d of element to
., Col. 12
Col. 13
Col. 14
z - distance to ~eu'tl"ll.l ax1.
as found ill results of Trial
110• 1
Ob ; 1600000 z/1470
based on s'tress (1) of Col. 13
Col. 9 z'
centerline ax1s of cell
i
Considering resuLts of both Tables and multiplying by l since Total effective al"ea (2.627 + .183)2 • 5.62
only one half of cell was considered:
~az. (_0I.86 + 3.24.)2.0.76
ia total effec'tive al"ea • (2.4.9 + .266)2
Z ... 76/5.62 • • 135" above N.A. of trial 1,
_ 5.51 sq. in.
rNA .. 2(673 + 71.3) - 5.62 Jt .135' ~ 14.89
£,az, • (_12.23 + 2.92)2. _18.82
B1e_llt xo,
TABLE A20. 3
(&,
(5)
.039(27.08
.153(25.38
.022(23.68
_
_
_
.13)
.U)
.13)
· 1.05
3.86
0.52
strain curve ~or the three stringer t~?es loaded
in compression and with a col~, ler~th equal to
.156(::1.58 _ .13) 3.3<
(7) .015(19.18 _ .13) 0.28
the fuselage frame spacing. Fig. A20.5 also
'8l
(9)
• (la)
.161(16.68
.031(13. <18
_"
_
.13)
.13) · 2.66
0.<11
shoWS a tension stress-strain clagram for the
material which 1s aluminum alloy (2024). The
..
.173(l0.28 • .13) 1.75
(U)
(12)
.055(6.58
.212(3.38
-- .13)
Total
.13) · 0.35
0.69
19 67
ultimate bending strength will be calculated
~or bending which places the upper portion in
compression.
Substituting in equation (3)
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A20.7
TABLE A20. 4
1 2 3 4 5 6 7 8 9 10 11 12 13
Effect- Total I
stringer Stringer Linear Arm Unit Stress
tve Skin Stringer Strain K=o/cr1. KAZ' KAZ'
Area Stress Z' ATE. KA (Z = Z'I
Ar-ea Area
No. Type A, a'b A
(in) E a
Ae
1 S, 0.135 -36500 .032 .167 35. 7 -.00600 -36500 1. 00 . 167 5.96 213 I
I
2 S, 0.135 -34700 .034 .169 33.8 -.0057 -36500 1. 05 .178 6.00 203
3 S. 0.180 -31000 .036 .216 30.3 -.0051 I -39100 1. 26 .272 8.25 250 I
I
5 S, 0.080 -20500 .044 .124 20. 1 ·.0034 -31500 1. 54 . 191 3.85 I 77 I
I I
6 S, 0.080 -13400 .054 .134 13.7 -.0023 -24000 1. 79 .240
I 3.30 45 I
increasing strain, but stringer (3) which has to Z. The effective moment of inertia is there
not reached its maxi~um strength of 39000 tore twi ce the sum of Column (13) or 3252.
continues to take increasing load.
Calculation of Ultimate Resisting M~ent.
Since we wish to use the beam formula CO =
MyZ/I x in computing stresses, we must modify the The maximum stress at the most remote
stringer areas to give a linear stress variation stringer which is number (1-i' is 36500. From
since the formula is based on a linear stress the beam formula,
variation. The stringer modification factor K
equals the ratio of the true stress in column
(9 ) of Table to linear stress value in column
Mx ="b1x/Z
(4) or K = a/a 1.. The results are recorded in = (36500 x 3252)/30.7 + 0.7
column (10) • The modified stringer areas are = 3,260,000 in. lb.
then equal to KA and are recorded in column
(ll) • Column (12) gives the first moment of This bending strength when compared to any des1 gn
the modified areas about the assumed neutral bending moment about the X axis Nould give the
axiS, giving a total value ot -3.16. margin of safety relative to benci~g strength.
The distance Z fram the assumed neutral If the moment of inertia had been computed
axis to the true neutral axis is thus, without regard to non-linear stress variation,
or in other wordS, using K equal 1 for all
Z=Z KAZ' (ZKA stringers the neutral axis would have come out
_ -3.16 _ 4.9 inches below the centerline axis and the
- 4.159 - - 0.76" moment of inertia would have calculated to be
2382 in. • The reSisting moment developed
TIle true N.A. would fall about .70 inches would then be (36500 x 2382 )/33. 6 2,600,000 =
below assumed pOSition. The effect on total in.lbs. Thus the true strength is 25 percent
sum at Column (13) would be negligible, thus greater than the strength for linear stress
Table A20.4 will not be revised. variation. This result explains why such
structures test overstrength if deSigned on
Column (13) gives the calculaticn of the linear stress variation basis.
er.rect r ve moment of inertia with Z 1. being equal Atter stringer stresses are obtained
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A20.9
uSing the modified areas of Table A20.4, the I y = (15:1 X .Lx z ) + (13.86 0 1 1 + 1 0 . 6 1 : 1 + 5 . 7 4 011).1
true stringer areas ~ust be used ~o tind the
true stringer loads, which must be used in the
x 4 = 180 in.4.
shear flow analysis.
Table A20.5 gives the necessary calculations
A20.5 Shear Flow Analysis for Fuselage Structures for determining the flange bending stresses and
the net total shear load to be taken by the cell
The shear flow analysis can be made once skin. Since the cell 1s tapered, the stringers
the effective cross-sections of the fuselage have a z component, thus the stringer axial
are obtained. The procedure is the same as was loads help resist the external shear load. The
illustrated tor Wing structures in Chapter A19. summation of column (8) ot Table AZO.5 gives
To illustrate, two example problems will be -333.4 lb. for a summation for halt the fuselage
presented. section.
Example Problem 1. Symmetrical Tapered Section. Hence, net web shear at station 0 equals:
Fig. A20.6 shows a portion of a tapered VWeb == Vext • + Vtlange = 2000 + (2 x -333.4)
circular shaped fuselage structure that might
be representative of the rear portion of a -= 1333.2 lb.
fuselage for a small airplane. Since this
example is only for the purpose of illustrating The results in this particular problem show
shear flow analysis, it will be assumed that that at station 0 the flange stringer system re-
the 16 stringers are the only effective mater- sists one third of the external shear load. At
ial. In an actual stress analysiS, the effect- station 150 the web system will resist the en-
ive cross-section would have to be used as tire external shear load at 2000 lb. since the
illustrated in previOUS articles A20.3 to A20.5. load in the stringers is zero.
The problem will be to deternine the In actual deSign the net Neb shear should
stringer stresses and the skin shear flow stress be used since in many cases it will decrease
system at Station (0) under a given load system the sheet thiCkness reqUired one or more gauges.
at Station (150) as shown in Fig. A20.S. Calculation of Flexural Shear Flow.
Solution No. 1 - Solution by Considering Beam Properties
at Only One Section. VZ(web) Z a Z 1333.2
q = qo - ly
= qo - 180 Z az
If the Change in longitUdinal stringer or
flange material is fairly unitor.n this method = qo - 7.40 Z az -- (A)--
can be used nith little error i~ the resulting
shear tlow stresses. Due to symmetry ot the sectIon about the Z
axiS, the flexural shear tlow in the web at the
Moment of inertia of section at station (0) center line is zero. Therefore, qo will be taken
about centroidal Y axis: as zero and the summation in equation (A) will
15 1
\
14 _stiffeners
13
12
11
10 S
Section at Station
Section at Station 0
"
STA.O
150"
Length not to scale
.\
STA.150
150
Fig. A20.5
A20.10 FUSELAGE STRESS ANALYSIS
NOTES:
Col. 4 ab'" -Mz/l y '" -2000 x 150 x z/lBO = -1667z
TABLE A20. 6
1 2 3 4 5 6 7 6 9 10 11 12 13
NOTES:
Col. 10 Change in axial load in each stringer between stations 0 and 30 divided by distance between Stations. This
result represents the average shear flow induced by the loading up of each stringer between stations 0 and 30.
Col. 11 The width of a skin panel at Station 0 is 5.88 inches and 5. 5 inches at Station 30. The shear flow on the edge
of the panels at Station 0 equals (5.5/5. a8) AP/30. (See Art. AlS. 18 of Chapter A15 for explanation). This
refinement is usually neglected and the average values as given in Col. 10 are used which are conservative.
Col. 13 Due to symmetry of structure, the shear flow is zero on z axis. Thus shear flow at any station equals the
progressive summation of the shear flow values in Col. 12.
second method is recommended tor practical will assume the stringers are the only effective
analysis procedure. material. In actual design practice the effect-
iveness of the skin and each stringer would have
Since the section is symmetrical, there are to be considered as explained in Articles A20.4
no moments i~Guced by the in-plane components of and 5.
the stringer forces at station O.
The problem will be to determine the
The torsional sh~ar flow forces are the stringer stresses and the skin shear flow val-
same as in solution ~ethod No. 1 and these are ues at station (0) due to the given external
added to the values ot column 13 of Table A20.6 loads ot ?z = 4000 lb •• ?y = 1000 lb. and Px =
and give a pattern similar to ~ig. A20.7. 1500 acting at station (150) as .snown in Ff.g ,
A20.9.
A20.7 Example Problem. Tapered Circular Fuselage
with Unsymmetrical Stringer Areas. SOLUTION:
Fuselage cross-sections are seldom all sym- Since Ne choose to use the ~ method in
metlical relative to stringer and skin areas finding the shear flow system at station (0).
because the practical fuselage has cut-outs such we will find the stringer loads at two stations,
as dcors, etC. To illustrate the unsymmetrical namely, station (0) and station (30). The first
case a simplified case will be presented. steo is to find the moment of inertia of each
fuselage section about centroidal Z and y axes
Fig. A20.9 shows a portion of a ~apered and the product of inertia about these axes.
~~se1ag9. The stringer a~eas are such as to Table A20.7 (Col~~s 1 to 11) gives the calcU-
make the cross-sections unsymmetrical ~elatlve lations of ~he section properties for station
to oending material. Again tor Simplicity, we
(l
. ~.>~
I (J
-~
A20.12 FUSELAGE STRESS ANALYSIS
b
1- 8 .4 8 --I,
.1
Fig. A20.9
d
(.1)
.1)1 12"
'i
,c
8.48
a (.6) 1
Cep-erUpe
d 150
(.2)
(.2)
.2)
• L- x i
Z' Z Sta. 0
The skin stringers are located symmetrical:y with res~ect ~8 :he
Section at Station 0 centerline axes, however' the stringer areas as g tven 1::1.( ) or.
the figure are not symmetr~cal With these axes. It is assumed in this ;roblem that
the stringers taper uniformly between the values as given for station Gane 150.
The cell would of course have interior transverse ~ra~es which are not shown on the
figure.
TABLE A20. 7
1 2 3 4 5 8 7 a 9 10 11 12 13 14
a .80 10.5 -12.00 6.30 66.20 -7.20 86.50 -75.50 11.36 -14.46 -15080 -441 -9312
b .10 18.98 - 8.48 1. 90 36.00 -0.85 7.20 -16.10 19.84 -10.94 -21960 -441 -2239
c
d
.10
.10
22.50
18.98
0
8.48
2.25
1. 90
50.80
36.00
-0
0.85
0
7.20
0 23.36
16.10 19.84
- 2.46
6.02
-22600
-16742
-441
-441
-2304
-1719
e .80 10.5 12.00 8.40 88.10 9.60 115.10 100.80 11. 36 9.54 - 7692 -441 -6506
f .80 -10. 5 12.00 -8.40 88.10 9.60 115. 10 -100.80 - 9.65 9. 54 11958 -441 9216
•h .20 -18.98
-22.50
8.48 -3.80
-4.50
72.00 1. 70 14.40 -32.30 -18.13 6.02 18798 -441
-4'11
3673
3,0,
.20 0 1~1. ,0 0 -21.60
i;;~~
u 0 - 2.4'
i .20 -18.98 - 8.48 -3.80 72.00 -1. 70 14.40 32.30 -18.13 -10.94 -441 2634
j .30 -10.50 -12.00 -3. 15 33. 10 -3.60 43.25 37.80 - 9. 65 -14.46 4592 -441 1246
Reference Axes Z' and y' are taken as the centerline axes. General Notes:
(see FIg. A20.9)
Col. 12 0b'" 307. Or - 936. Lz
Location of centroid and transfer of properties to
centroidal axes. Col. 14 . Since the total tensile stresses equal to I
total compressive stresses in bending,
Z '" -2.90/3.40 '" -. 855" the sum of Col. 14 should equal the ex-
ternal applied normal load.
y '" 8.40/3.40 '" 2.46"
TABLE A20.B
1 2 3 4 5 6 7 8 9 10 11 12 13 14
z: y: °c"'F;Za .Ps =
Stringer Area Arm Arm az' az,2 ay' ay,2 az'y' °b
No. a z· y' a' - Z y' - y =-1500/2.98 a(Ob + Gel
a .50 9.80 -11.2 4.90 48.1 -5.60 62.8 -54.9 10.90 -13.31 -14800 -503 -7651
b .10 17.72 - 7.92 1. 77 31. 4 -0.79 6.3 -14. a 18.82 -10.03 -21090 -503 -2159
c .10 21. 00 0 2.10 44.1 0 0 0 22.10 - 2.11 -21447 -503 -2195
d . 10 17.72 7.92 1. 77 31. 4 0.79 6.3 14.0 18.82 5.81 -15647 -503 -1615
e .66 9.8 11.20 6.47 63.2 7.40 83.0 72.5 10.90 9.09 - 7088 ·503 -5011
f .66 - 9.8 11. 20 -6.47 63.2 7.40 83.0 -72.5 - 8.7 9.09 -11282 -503 7100
•
h
.20
.20
-17.72
-21. 00
7.92
0
-3.54
-4.20
62.8
,".2
1. 58
0
12.6
0
-28.0
0
-18.60
-19.90
5.81
- '.11
-17583
17::143
-503
-503
3416
,.""
i .20 -17.72 - 7.92 -3.54 62.8 -1. 58 12.6 28.0 -16.60 -10.02 12135 -503 2327
j .26 - 9.a -11.20 -2 55 25.0 - 2. 92 32.6 28.6 - a.70 -13.31 3570 -503 797
~
N Z'
~.:ic.
Reference Z' and Y' axes are taken as the centerline axes.
b I:~ 1. 10 d
Z = -3.29/2.98 = -1.10" .10 .1~.?
7.9 2'[ 2'
Y= 6.28/2.98 = 2.11 a
.sot- Z .66 e Section at
Station 30
x 1. 10 2 ::516. 6
9.a z'"
ly = 520.2 - 2.98
If.~l=-) i ~ l:;.:,{
Iz = 299.2 - 2.98 x 2. 112 :: 286.0 ;1.- tr y - y
,~l. Z2.11
I zy = -26.3 - 2.98 x 2.11 x -1.10:: -19.4 j
.26 • ll.~~ I
.20 .20 Fig.A20.10
•
(0) and the similar columns of ~ab1e A20.8 gives My = 4000 x 120 + 1500 X 8.10 = 492150 in. lb.
the calculations for station (30).
Mz = -1000x120+1500x2.11 = 116820 tn Lb , s
Before the bending and shear stresses can Pn = -1500 lb., Vz = 4000 lb., Vy = -1000
be calcalated, the external bending moments, lb.
shears and normal forces at stations (0) and (30)
must be known. Calculation of Bencting Stresses.
Station (0):
At station (0): -
Substituting K values in equation for ab: ~cments in :he all ~n~e~nal anc ex-
)l~ne, Q~
ternal ~orces ~us~ be zero. Col~~ (7) of Table
r.00262 x-1453l0 - (-.0001248 x A20.9 ~iJes :~e ~oment of ~he flex~al shear
about tn; s potnt , (See notes and 3"ig. be Low
61.1800)J y - [00156 x 5E200 -
~ '" expjanat i on. )
':'able ""'''''.~
~
(-.0001248 x -145310)J z
TABLE A20. 9
whence SHEAR FLOW CALCULATIONS
Station (30): q e
Stringer Px at P x at .6.Px Z~ m mq ql
No. Sta. 0 su. 30 30
K• =-19.4/(516.6 x 286 - 19.4') ,JU sqc m,
whence NOTES;
2 3 4 5 6 7 6
Py = Mo' Pz = M o '"
Stringer Px dv p2Y. dz Px..!!I.
No, (lbs. ) OX xdx P y z' dx dx P z y'
NOTES:
---Col. (2) from Table A20. 7
Col. (3) equals the slope Fig. A20.11
of stringers
inyandz
directions. Shear flow distribution.
(see Fig. A20. 9)
Col. (5) Values of z' where large concentrated loads are applied can
and (8) and y' from be determined by the procedure given in Articles
Table A20. 7. 18 to 20 of Chapter Al9. A more rigorous
Fig. b shows a~~lysiS can be made by the application of the
the P y and Pz basic theory as ~iven in Chapter AS.
components from
Cols. (4) and (7).
The problem of shell stresses due to in-
Total moment about O' '" ternal pressures 1s presented in Chapter A16.
-1634 + 1612 '" -22". The strength deSign or the fuselage skin in-
volves a question of combined stresses. The
broad Droblem of the strength design of struc-
92670 due to shear flow q tural elements and their connections under all
-22 due to in plane components or types of stress conditions 15 covered in Volume
stringers. II.
7000 due to the external loads.
A20.9 Problems.
Total= 99648 in. lb.
(1 )
Therefore :or equilibrium a moment of
-99646 is required which can be provided by a ~'S'
~.
+
q, 2A lb./tn. (957
.. 2 x 957
enclosed area or cell)
,
1/
i,
Skin .035
.~
Three types of string-
J' S, ers are used, namelY,
.Ok S l' S.lI and 5:5' Fig .
"" S, AZO.14 shows the
ult~te compressive
s, stress-strain curve
/032 for each ot the three
stringer types and
S:.. s~ s, also the tension
Fig.A20.13 stress-strain curve
ot the material.
Determine the Ultimate bending resistance (4) Sa~e as Problem (3) but change area of
ot the P~elage section about the horizontal stringer no. (2) to 0.3 sq. in., thus makfng an
neutral axis it the maximum unit compressive l~symmetrical section.
strain is limited to .008. Reter to Art. A20.5
for method ot solution. A20. 10 Secondary Stresses in Fuseiage stringers and Rings.
ADDITIONAL DATA. Area s'tz-tnger- S l :: .12 sq .fn, ; The stresses that are found in ~he
S.lI :: .25 sq.in.; Sa :: .08 sq.i~. E = 10,500,000 stringers or longerons at a typical fuselage by
pat •• use of the modified beam theory or by the more
rigorous theory of Chapter AS, are referred to
(3) Fig. A20.1S shows a tapered circular as primary stresses. Because of the necessity
fuselage with 8 stringers. The area of each ot weight saving, most fuselage structures are
stringer is 0.1 sq. in. A$sume stringers develop designed to permit skin buoxj.tng , which means
entire bending rBsistance. Find the axial load ttat shear loads in the skin are carried by
in stringers at station (110) due to Pz and Px diagonal semi-tension field action. This
loads at station (0). Also find shear flow diagonal tension in the skin panels produces
system at station 110 USing ~P method. Use additional stresses in the stringers and also
properties at station (90) IN OBTAINING AVERAGE in the fuselage rings. These resulting stresses
SHEAR FLOwE. are referred to as secondary stresses and must
be properly added to the primary stresses in
the strength desi~ of the indiVidual stringer
or ring. Chapter CII covers the subject at
these secondary stresses due to diagonal semi-
7 tension field action in skin panels. It is
suggested to the stUdent that atter st~dying
.110 Chapters Al9 and A20, that Chapters ClO and Cll
be referred to in order to obtain a complete
Fig. A20.15
stress picture for skin covered structures.
CHAPTER A21
LOADS AND STRESSES ON RIBS AND FRAMES
A21. 1 Introduction. For eer-ocvnamt c reasons the loads into the fusela~e shell such as those
wing c~ntou~ in the chord direction must be from landin~ gear reactions, wing reactions,
~~lntaln~d ~lt~c~t appr~clable distortion. tail reactions, power slant reactions; etc.
Unless t~e ~ing sk1n 1s qUite thick, spanwlse The dead wei~ht of all the payload and fixed
str~nge~s ~ust be ~ttached to the skin in order equipment inside the fuselaroe must be carried
to Inc~ease :he bencins efficiency of the wing. to frames by other structure such as the
Therefore to hold the Skin-stringer wing surface fusela~e floor system and then transmitted to
to c~ntour sha~e and also to 11~it the length the fuselage shell structure. Since the dead
of strln~ers to an efficient column compressive weight must be multiplied by the design accel-
strength, 1~t2~~~1 support or brace units are eration factors, these internal loads become
reqUired. ~hese str~ctural units are referred qUite large in ~gnitude.
to as wtnz ribs. The ribs also have anot ner-
major pur~cse, na~elYl to act as a transfer or Another important purpose or action of ribs
distribution unit. All the loads applied to and frames is to redistribute the shear at dis-
tne '.'ling are reacted at the wtng sucportlng continuities and practical wings and fuselages
pOints, thus these a~pll~d loads must be trans- contain many cut-outs and openings and thus
rarre~ into the Nlng cellular stru:ture ~om- , discontinuities in the basic structural layout.
posee o~ skin, strin~ers, spars, etc., and then
reacted at the Wing support ?o!nts. The aJvlled A21. 2 Types of Wing Rib Construction.
loads nay be only the distributed surface ai~
leads which require relat~vely light internal Figs. A2l.1 to 6 illustrate the co~non
ribs to ~ravide th~s carry throush or tr~nsfer types of win~ construction. Fig. 1 illustrates
~equlr9ment, to ~ther rug~ed or heavy ribs a sheet metal channel fer a leading edge 3
which ~ust absorb and transmit large concen-
trated applied loads such as those from landing Skin
gear reactions, power ?lant reac~lons anc. ~~se flange
Iage reactions. In between these two extreaes
of applied load ~a~nit~des are such loads as
.
reactions at sup~orting points for ailerons,
flaps, leading edge high li:t uni~s and the
many internal ceaa .vet ght leads such as fuel
and military anmament and other installations.
Th~s ribs can 'ffiry from a very light structure
OiO Beam
Web
,-
o
- ....
-
Ribs in 3 Spar Wing.
o 0 0
o o Removable
o Leading Edge
Portion.
O~~~~~~~
o 0 000
Fig. A21. 6
stringer, single spar~ single cell wing were used to transmit exter~al loads into the
structure. The rib 1s riveted, or soot-welded, wing cellular beam structure. Concentrated
or glued to the skin along it bo~~dary. Fig. external loads must be distributed to the rib
2 shows the same leading edge cell but with before the rib can tra~sf~r the load to ~he
spanwise corrugations on the top skin and wing beam st~~cture. In other wordS, a con-
stringers on the bottom. On the top the rib centrated load applied cirectly to the edg~ 0:
flange rests below the corrugations, whereas a thin sheet would cause sheet to buckle or
the stringers on the bottom pass through cut- cripple under the localized stress. Thus a
outs in the rib. Fig. 3 illustrates the gen- str~ctural element ~sually called a web stiff-
eral type at sheet metal rib that can be ener or a 'Neb r lange Ls fastened to the 'Neb and
qUiCkly made by use at large presses and rubber the concentrated load goes into t~e stl:fe~er
dies. Figs. 4 and 5 illustrate rib types for which in turn transfers the load to the web.
middle portion ot wing section. The rib To ~et the load into the stiffener usually re-
flanges may rest below stringers or be notched quires an end fitting. In general the c isur i ,
~or allowing stringers to pass through. Ribs buted air loads on the wing surface are usually
that are SUbjected to considerable torsional of such ~gnitude that the loads can be distri-
torces in the ?lane of the rib should have some buted to rib web by direct bearing of flange
shear ties to the skin. For ribs that rest normal to edge of rib web without causing local
belaw~trlngers thiS shear tie can be made by buckling, thus st1f~eners are usually not
a rew sheet metal angle clips as illustrated in needed to transfer air pressures to wing ribs.
Fig. 5. Fig. A21.7 shows an artist's drawing
at the wing structure or the Beechcraft Bonanza EXAMPLE PROBLEM ILLUSTRA.TING ':'RJ.NSFER OF CON-
commercial airplane. It should be noticed that CENTRATSD LOAD TO SHSST PANEL.
various types and shapes of ribs and formers
are required in airplane design. Photographs Fig. AZl.8 shows a cantilever beam com-
A2l.l to 3 illustrate t~~ical rib construction posed at 2 flanges and a web. A concentrated
in various type aircraft, both large and smaLl , load of 1000 lb. is applied at paint (A) in the
Since ribs compose an appreciable part at the direction shown. Another concentrated load of
wing structural weight, it is important that 1000 lb. is applied at paint (E) as shown.
they be made as light as safety permits and
also be'efficient relative to cost of fabrica- To distribute the load ot 1000 lb. at (A)J
tion and assembly. Rib development and deSign a horizontal stiffener (AB) and a vertical
involves considerable static testing to verify stiffener (CAD) are added as sho'ND. A t1ttlng
and assist the theoretical analysts and deSign. would be reqUired at (A) which would be attached
to both stiffeners. The horizontal component
A21. 3 Distribution of Concentrated Loads to Thin of the 1000 lb. load which equals 300 Ib is
Sheet Panels. taken by the sti~fener (AB) anc the ver~ cal
component which equals 600 lb. is taken y the
In Art. AZl.l it was brought out that ribs
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A21. 3
d b
- L-------'"'
Fig. A21. 7 General Structural Details of Wing for Beechcraft "Bonanza" Commercial Airplane.
vertical stl:fener CD. The vertical load at E Figs. A2l.9 and A2l.l0 show free bodies of that
would be transferred to stiffener SF through portion including web panels (1) and (2) and
fitting at E. The problem is to find the shear stiffeners CAD and AS and ~he external load at
tlows in the web panels, the stiffener loads (A). In rig. A21.9 the shear flows q J, and q ,
and the beam flange loads.
s,
10- 10"-+-'0"--f-- 10"-r'O"i I t~q, --- D
q':~'1I
G ~lange F D
D 9" q t (1)
'f
(I) :/'ti!fener 1 B ~800 B A
TW-
9"
qll~
/
, Web (3) /'
, 3" q I (21 'e IY (2) 1,600 lb.
/ - !3"
f- ~800 lb. E---C
, II (2)
l~ 1000 lb.
s,
10"
I
---l
,"-x q.
/H -, flan g e E 6001 b. Fig. A21. 10
r
1000 lb.
Fig. A21. 9
Fig. A21. 8
()1_~
:v"'0 "C
A21. " AND STRESSES ON RIBS AND FRAMES
PHOTO. A21.1 Type of Wing Ribs Used in Cessna 180 Model Airplane, a " Place Commercial Airplane .
..
• .. ~
,
..........
~ • '''#o>..-e~
'j, •,.
'~'--- •
.~~ v
>
J.
PHOTO. A21. 2 Rib Type Used in Outer Panel-Fuel Tank Section- of Douglas DC-8 Commercial Jet Airliner.
,
,..,
;
.,-
.,.,. -
PHOTO. A21. 3
, "
Rib Construction and Arrangement in High Speed, Swept Wing, Fighter Type of Aircraft.
North American Aviation - Navy Fury - Jet Airplane.
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A21. 5
ZFx = -50 x 10 ... lOqa = 0, whence qa = The shear flow qa in web panel (3) is ob-
50 lb./in. tained by considering stiftener EBF as a free
body, see Fig. A21.16.
Combining the two shear flows for the two loads,
ZFY :: 12q;s'" 10 x 3 - 9 x 70
q~ = 20 + 50 =70 Ib./in. -1000 = 0
F
50 = 10 Ib./in.
T II
t Iq
whence, qa :: 133.33 lb./in.
Fig. A21.11 shows the results. Fig. 9"
q,
, ",70 The shear flow q~
A21.12 shows stiffener AB as a tree body, and
Fig. A2l.l3 the axial load diagram on stiffener
AB, ~hich comes directly from Fig. A2l.11 by
1 B
,\0
could also be found by
treating entire beam to
starting at one end and adding the shear flows.
3" Cl~ t:. right of section through
panel (3). For this free
T E
body,
1000
BOO#( tension) Fig. A21.16 ZFy =-600 - 1000 + 12q, =0
B --------=1 Fig. A21. 13
whence, q~ = 133.33
q.. =70
B =qa - = 10
-=::"A Fig. A2l.17 shows diagram of axial load in
stiffener EF as determined from Fig. A21.16 by
starting at one end and adding up the forces to
Fig. A21. 12 any section.
Fig. A21.l4 shows a free body of the After the web shear
F tlows have been detennlned
~
vertical stiffener CAD. and Fig. A21.l5 the i g' A2 1. 1 7
axial load diagram for the stiffener. the axial loads in the
beam flanges follow as
B _ 570 lb. the algebraic sum of the
o shear flows. Fig. AZI.18
Tq ,
A~
I E 1000 lb. shows the shear flows
9" ='0, (tension)
+_A'10 630> (tension)
along each beam flange
as previously found. The
qa 600
3" I
upper and lower beam flange loads are indicated
I =10.
...L.. C C by the diagrams adjacent to each flange •
Fig. A21. 14 Fig. A21. 15
4700
700 lb.
Tension
'rhe shear q. and qa could of course
~lows
be dete~ined USing both components of forces
at (A) acting Simultaneously. For exa~ple,
consider tree body in Fig. A21.ISa.
; 133.33 ---
133.33 13333 70
/
~
J- -r -;
t-- 10"-1 _800
q I 133.33 133.3=L 133.33 10
'1000
- 600
,
100* tension
<'
~ig'A2L18
Fig. A21. lSa
L CIa t'IBx ~ 600
&800<1
lb.
-+
3" 3900* compr essron
E- - - C .....L...
q.
In this example problem the apolled extern-
ZFx :: 800 - 10q. (1 ) al load at point (A) was acting in the plane at
the beam web, thus eNO stiffeners were suffi-
ZFy = -600 + 9q. (2) cient to take care of its two components. Often
A21. 6 LOADS AND STRESSES ON .RIBS AND FRAMES
tLx y
s,
with q known the shear and bending ~cment
Fig. A21. 19 at variouS sections along the rib can be dete~
~ined. For example, consider the section at 9-3
in Fig. A21.20. Fig. A21.2l shows a free body
and Px are distributed to the web panels by of the portion rorNard of this section.
us ing three stiffeners S J.' S a and S,3 inter-
secting at (0). The bending ~oment at section 8-B equals:
In cases where a load must be applied MB =2qA 1. where A 1. is the area a f the
normal to the Neb panel, the stiffener must be shaded portion.
designed strong enough or tranSfer the load in
bending to adjacent webs. Let Fx equal the horizontal component of
the flange load at this section.
In this chapter, the webs are assumed to
resist pure shear along their bouncarles. In
most practical thin web structures, the webs
will buckle ~~der the compreSSive stresses due
to shear stresses and thus produce tensile
field stresses in addition to the shear
stresses. The subject or tension field beams Fig. A21. 21
is discussed in detail in Volume II. In gen-
eral the additional stresses due to tension
fIeld action can be superimposed on those
found tor the non-buckling case as explained
in this chapter.
Solution: 's
w = 8/F/in. I F
The total air load aft of beam = 8 x 4-0/2 =
160 lb. The a~. to its e.g. location fr8~ the
b=a~ equals 40/3 = l3.33 •
fl
Hence the react~ons
-rf- 30°
Iv:e:_l_
at the beam flange paints due to the loads on Area of Shaded
the trailing eGge ~ortion equals: 7.67" Portion = 15.4 sq.tn.
~\11
'"' ~
Area
A '" 139.3 sq. in.
j
I
(=L
r
10"
to lett of line be.
Fig.A21.25
/' t./
» > -20~ ~/i
I
~
"'=:: _ _ _ (l)
160
P =213. 2 lb.
d£:------f1 5°O I
- I 1" I
Fig. A21. 23 Area:=160"'A2 t 1~2 I
~qcb ,/.r
9.3" /
Bending moment at section B-B = To find Gacc' take moments about point (b)
8x2.5x1.25+4.42x2xI5.4 = 161 Inv Ib ,
A21. 8 LOADS AND STRESSES ON RIBS AND FRAMES
ZMe = -2(A1. + A.) Qadc .;. 5000 x 15 - 500 To find web shear
~d take ZFz = 0
=0
x 11.5
=-2(60 + 160) QadC + 75000 - 5750 =0 215B-Ib-t
d
74.611,
T'
9.3"
= 5000 - 9.3 x 74.6
- 9.3 q~d =0
QadC = 157.3 Ib./in. I I"ad
whence,
assumed.
with sense as
2034.....J i.- c,l
3000 whence, %d =463 1b./
'C
5000 in.
To tind qcb take ZFz = 0
Fig.A21. 27
To find r Lange
ZFz = 5000 + 200 - 157.3 x 11.5 - 11.5 load C' take ZFx 0, =
<lob =0 conSidering joint (a) as a tree body,
I~T'
I
I' 9.3'"
l~
~""';=="'t3000 C'
To tind tlange load T take moments about 5000
point (a), Fig. A21. 28
I Fig. A21.30
t- ~:.~.
a skin
8500 b
t
~t ]- Fg~-
6'1'50 ~ Web 8.40"
7500 CQ I
iI
d , .....i...-4500#
z 30" skin c
L. 5000 2~0
Fig.A21. 29
SOL1JrION:
To tind tlange load C take moments about The total shear load on the wing in the Z
pOint (b). direction equals Vz -6000 - 5000 + 2000 = =
-9000 lb. and Vx =-8500 + 7500 - 4000 + 4500 =
l:IIb =-157.3x2 (160+60) + 5000 x 15 --500 lb.
-11.5 C = 0 The boundary torces on the rib will be
equal to the shear flow torce system on the
whence, C = 500 lb. cell walls due to the given external torce
system.
To tind tlange load T take ZFx = a
ZFx = 3000-157.3x15-15x42.7-500+T = 0 From Chapter Al4, page Al4.8, equation (14)
the expreSSion tor shear flow is,
whence, T = 500 lb.
qy = -(K15Vx-K1.Vz) .r.xA- (K.Vz -KJ, 11x)
To find ~b take ZFz = 0 ZZA - - - - - - - - - - - - - - - (1)
IT z =5000 -157.3 x 11.5 -11.5 q~b =0 The constants K depend on the section
properties at the wing cross-section. Table
whence, qbb = 278 lb./in. A2l.l gives the calculation of the moment ot
inertia and product ot inertia about centraidal
The ~bove results could have been obtained Z and X. axes. In this example the 4 stringers
with less numerical work by considering the a, b, c and d have been considered as the entire
forces to right of section cb in Fig. AZI.29. effective material in resisting wing bending
stresses.
A21. 6 Stress Analysts of Rib for Single Cell
Multiple Stringer Wing. TABLE A.21.1
The moments ot the forces in the plane of ZMo = -11000 x U.8 - 3500 x 5.36 - 7500 x 5 .64 + 436
the rib will now be calculated:
x 2 x 36 + 880 x 2 x 70.8 + 419 x 2 x 38.3 =
Taking moments about the e.g. at the beam -38200 1n.lb.
cross section (See Fig. AZl.31):
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A21. 11
The resultant external shear force along the boundary forces on a rib located adjacent
the section B-B equals the summation of the z to a c~t-out is to find the applied shear flows
components of all the forces. in the wing on two sections, one on each side
of the rib. Then the algebraic sum of these
'f = ZFz = -11000 + 12 x 3eO - 436 x 0.36 + 419 two shear flows will give the rib boundary
forces. Aith the boundary forces known the rib
x 0.96 = -195 lb. web and flange stresses can be found as pre-
The resultant load normal to the section viOUSly illustrated. The procedure can best
be illustrated by example problems.
8-8 equals the summation of the force compon-
ents in the x direct1on. A21. 8 Example Problem. Wing with Cut-Out
Subjected to Torsion.
H = ZFx = -8500 + 7500 + (436 - 419) 11.8
Fig. A21.35 shows a rectangular Single
= -800 lb. cell wing be~ with tour stringers or flanges
located at the four corners. The upper surface
Fig. A21.34 shows these resultant forces
skin is discontinued in the center bay (2).
referred to Doint (0) at the cross-section. If The wing is subjected to a torsional moment ot
we assume that the rib flanges develoD the
entire resistance to normal sta-esses., we can
find flange ~oads by simple statics. STA.O
T 'I _l~a ,,
4Q'i: -
c,
I IRib,lF' "000,,
To find upper flange load Fu take moments ~ 'I , AI
about lower flange point.
zrI
whence,
= 12.6
~u
Fu-38200-800x6.6
~
4.~o
'
II,I!
21)0/
jJ'
1'7/
J, Rib
-----.1-:1--4000
Rib (A) /,1
~~~;:c:~-
moved in
lb. Bay (2)
~ ~ I ! . I
2643 lb. compression. ,q r Yr--- ---1.-
ST~O ' ;, ,
The shear flew on web equals V/l2.6 =
195/12.6 = 15.5 lb./in. This result neglects ::;1 (Rib ~o in. lb. Fig.A21.35
effect of fl~~ges not being no~l to section
8-8, which inclination is negligible in this I\ 40" ----i
case.
80000 in.lb. at Station (70) and a couple force
I~the entire cross-section of rib is at Station (50) as shown in Fig. A2l.35. The
ef!ective in bending, then the web thickness problem will be to determine the applied torces
and flange sizes of the rib would be needed on rib (A).
to obtain the section ~oment of inertia which
Is necessary in the beam equation for bending SOLUTION:
stresses. The forces at (0) would then be
referred to neutral axis of section before The applied shear flow on the cell walls
bending and shear 3tresses on the rib section will be found for two cross-sections of the
could be calculated. wing, one on each side at r tb (A).
To obtain a complete ~icture of the web In bay (1) the torsional moment M is 80000
and flange forces, seve~a1 sections along the in.lb. The applied shear flew en a cross-
~ib span should be analyzed as illustrated for section of t~e wing in bay (1) thus equals,
section B-B.
A21. 7 Rib Loads Due to Discontinuities in Wing
q = ul'! = 2 x30000
10 x 4,0
= 100 lb .ym •
Skin Covering.
This shear flow system is shown on Fig ..
As referred to before, rib~ i~ acdition to ..;21.36 wntc a is a rree body of rib (A). In
transmitting external loads to wing cell bay (Z), since the top skin is removed. the
structure are also a means of re-distrlbuting tcrsional moment must be taken by the tront and
the shear forces at a ~iscontinuity, the ~ost rear vertical webs, since any shear flow in the
comnon discontinuity being a cut-out in or-e ar bottom skin could ~ot be balanced.
acr-e of tne Nebs or 'Nalls of the .vt ng beam
cross section. ~he usual ~rocedure in f1nding The torsional ~oment in bay (2) is,
9 c, .»
-"..:-"<.. C
A21. 12 LOADS AND STRESSES ON RIBS AND FRAMES
Bay (2)
!,1.q'",300 ..."..Rib A Bay II) ---rl----- L. 30"-----!--_ Bay (3)
t~;;:::;;;:::;~~~;;::;;;1lt-4000#
l ,~~-_ ..9. ':....lQ..o _ -"'1 t q':300 T
h .110" P-.i~;::='=,=,=,=,=,=,===~
p Vl3000 Fig.a
·~:t=;:~~===:=]lrv:iooo
q:: 100 '1
t I lUGl+ J -4000Jil
I _p
q:: 10
Fig. A21. 37
q:: 100 flange loads at the bea~ ends are P = 45000/10
:. 4500 lb. (See Fig. a).
-=
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A2L 13-
:lange member. ~o !ind the shear ~lcw on ~~e bay (1) as shown. The areas of corner stringers
cross-section the frent web is tirst assumed a, 0, c and dare shown in ( ) adjacent to each
cut, and thus the static shear floW qs = Z6P stringer.
from cut face where q~ is zero. Fig. A21.40
shows th1s static shear flow.
B800 lb.
No Skin on
Bottom of r-- 15,,--t
Bay (2).
112.5
I· 30"
='" - -
312.5 tl ____
- - -I'
212. 5
-4000*
t312.5
'_4000*
is similar to the previous example involVing
win~ torsion only.
:-t
aBOO/!'
known, the rib fl~nge and web stres5es can be z
found as previously explained. L. -,-
ho:-,-'5"
a
1
I
(0.5)
AZl. 9 Example Problem. Wing with Cut-Out Subjected 6" b 4,,1
to Bending and Torsional Loads. I ~
TO l!--~------""ro:sr
2400 4'"
-r- 6"
* ...L:.-
~ . . g. A
-. " s ..ows
21 .~ " a pertl on 0 f 3. 4 str 1nger-1 3" d
single cell cantilever beam composed of 3 bays _ I (0.5
~o~ed by the four ribs. The loads on the
structure consist of loads applied to end of
~x-~
~10" z 30" ----i Fig. A2l. 44
o ,' . . /
'--"--I
A21. 14 LOADS AND STRESSES ON RIBS AND FRAMES
Qcb = 640 -100 (-4) (0.5) - 4 (20) 0.5 = 600 ZFz = 8300 - 9 x 640 + 2 x 50 - 12 Sac == C
whence Qad 320
qba = 800-100 (4)(0.5) -4 (20) 0.5 = 560 :=
l~~.~
ential bending at beams in bay (2), thus equal
the beam shear times halt the span at bay (2)
600 diVided by the beam depth.
+d R::8800
For tront beam P = 3841 x 12.5/12 = 4000 lb.
Fig. A21. 49
For rear beam P = 3841 x 12.518 = 6000 Lb ,
Vertical position of shear center: -
Fig. A21.52 shows these.flange loads
Assume section bends about centroidal z applied to bay (1). These loads are dissipated
axis without twist under a load of Vx = 2400 lb.
q = _.'!x.
Iz
ZXA =_ 4
'<cb. -4x20xO.5
ZXA
=- 40 lb./ln.
~
7°:--- 1 _
"1/
qba:: -40-4x20xO.5 = - 80
~
/
unifOrmly 1n bay (1) over a distance of 30
inches, or the shear flow per inch prOduced by
c these flange loads equals tJ' :: P/30, whence
Fig. A21. 50
e.c. 8800
.-::r-l i,.9..,,~
2400
~{H- k'" lb Fig.A21. 51
The moment of this shear flow system about
point (d) = 133.33x30x12-66.7x8x30 = 31980.
c For ZM = 0, we need a constant shear floW q =
- 31980/2 x 300 = -53.3 lb./ln. Adding this
constant shear flow to that or Fig. AZl.53
Thi~ torsional moment must be resisted by
gives the shear flow system at FIg. A21.54.
front and rear beams. Hence shear load on These results represent the effect on bay (1)
each beam = 115240/30 =3841 lb.
Fig. A21.54
f 120
I
I
'01.3 j
_
-- ,192
Fig.A21.55
• 232
I I
or in other words the bending moment at mid-
point of bay 1s zero. The flange loads at '0 ..- 272-
paints a, b, c ~~d d on bay (1) from the d1fter-
A21.16 LOADS AND STRESSES ON RIBS AND FRAMES
of removing the botto~ skin tn bay (2). Adding 192 with sense ~s shown. Since this s~lrr is
the shear flo~s of Fi~. A21.5~ to ~h1se c:' 5"1 ~. nt satng »e reverse this shear' r lc.v and rmc tne
A21'~J ~e obtain :~= final 3~ea~ ~lcws in ':lay reSisting shear flows )n t~e ~ther ~~:se siees
(1) as sho~n in fiS. A2l.55. of tr:e bay crOss-sectiJn. Fi~. ~21.5S shows
the sec t i on, with the 3 unknown shear ~1:J'I'./s
aOmJD.;;RY LOi..OS 0;' 813 (A) Qab, qbc and Qa1'
181.3r~~~ 272
I: 408 320 ;LI_ _- - ; ; - - - ] : 640
o Fig. A21. 59
Fig. A21. 57
, sults check the results in Fig. A21.48 obtained
With the rib boundarJ loads known, the in solution method (1).
stresses in the rib can be found as previously
illustrated in this chapter.
Solution No.2
Double angles at
3,3',9,9'.
200~O 11.5 ,e-i
1 I'
2000
11. 5 ' J
Fig. A2l.62 shows the frame with its
balanced load system. The internal stresses
(check)
2 , bulb angle
stringers can' now be found ~y the methods of Chapters
AS to All.
2000 2000
!
t
6' Y
11 11'
Fig. A2l. 61
I'~
..
sh3ar load Vz = 4000 produces the same shear h f
(3)
flo~ pattern as Fig. A21.62. To balance ~he /
(1) e 500
moment of -11500, a constant shear r Low q :I. /
/
(2) b'
around the fra~e is necessary. / web
(4)
- M- 11500 a = 5 .6-
Q-2i-2xnxlS lb • /1 n ,
/
;;.
t'
I
0 /
600
I" 500
I" 1000
(A = area of fuselage cross-sectton)
Fig. A21. 66
Adding this constant force system to that
ot Fig. A2l.62 J gives the final boundary (2)
supporting forces on the frame as illustrated
in Fig. A2l.65. The elastic stress analysis
+500
of the t~~e can now proceed.
2000
1500 2500 c b
a
!
" 2000
i I fX-1 10"
·1
I. \
\,
Fig. A2l.67 shows a wing rib inserted in a
Fig. A21. 63 Fig.A21. 64 3 flange single cell wing.beam J which is sub-
jected to the external loads as shown.
(3 )
l OOa
t
~
d
web
1
(2) Fig. A21. 68
web
-'
I
r
7 1000
600
to
t
500
b
I
I
(4) 1000 Fig. A21.69 (8) Sa~e as preblem (7) but with top skin re-
shows a 2 stringer, moved instead of lower ski~.
2 cell wing beam.
A ~ib is inserted (9) S&~e as (5) but with read spar web removed
to transfer 1000 lb. instead of bottom skin.
load to beam struc-
ture. (10) Same as problem (7) but with rear spar 'Neb
Fig. A21. 59 b removed instead of bottom skin.
r---10"-j Find shear flow
in rib web in each (11) In Fig.
cell aojacsnt to AZl.7l the external
line abo Also rib bulkhead leads P
flange loads ad- and P2 equal 4000
jacent to pOints lb. each and ?~
(a) and (b). equals zero. The
fuselage stringer I \.~"
~terial consists
_ JL'_
,
of tour omega.
T
~71:
sections with an
area of .25 sq, is-
in. each. Deter-
4' ~ " • 1
I
mine the skin re-
~ :-/(----=--
Sisting forces or.
the bulkhead in
ba lanc tng the
iOi': .... I RibB above loads. Fig. A21. 71
~ ~ 0(;' :1 Neglect any effec-
4' '"r,,'_JL.- -:P -
-/1_1_ .J-S:=;-'(~ P,
tive skin in this
problem.
'+" r--
oS' __
".f(
a
IjRibA
I
I
1
-'-P,
Corner stringer Areas:-
and Pa = 6000.
(13)
(14)
Same as problem (12) 'but add
In a water
p~ .- 30CO lb.
:: j r\
c {a} • I sq. in.
I
landing condition
-l d
'--,'-r-- P2
r-
(h) •. 8 sq. in.
(c) "'.5 sq. in.
(d) '" 0.4 sq. in.
the hull frame
of FIg. A2L 72
L
I
I s"-1 b
40"
I
~,
J I
*2 / w3
Ft!C. A22.1
Fig. A22.2 Idealization of the half panel by
use of substitute stringers
The sheet-stringer panel may, in general,
contain a large number of longitudinal ele- Fig. A22.3 gives the geometry of the ideal-
ments (stringers). The labor involved in ized panel.
treating this multi-element structure in detail A:,'" .703 in
2
in
2
.2
plex, multi-stringer structure is replaced by +U F=H==;,------ AR'" 0.25
one ~ving but three stringers. As indicated ,__ __lL__ t:, '" 0.0331 in
in Fig. A22.2, these stringers are:
~
~
~
-'--
t2 '" 0.0331 in
:-b. -ba ... :L
*1, a substitute stringer having for its
area all the effective area of the fully con- o~ _0 f---, b:, ,.5.96 in
A:l., Alii A 3
tinuous members to one side of the "combing - t:, - t a ba =7.56 In
stringer" (the stringer bordering the cutout)
and placed at the centroid of the area of Fig. A22.3 L:15.0 In
material for which it substitutes. The stress
which this stringer develops is then the SOLUTION:
average stress tor the ~terial it replaces.
Fig. A22.4 is an exploded view of the half-
'2, the combing stringer, being simply the panel Showing the placement and numbering of the
main continuous stringer bordering the cutout. internal generalized torces (Art. A7.9, Chapter
A7) and the external loading. Note that the
'3, another substitute stringer, this applied axial stresses were assumed to be con-
one replacing allot the effective material stant chordwise, giving stringer leads ~ro
made discontinuous by the cutout. It is located portional to the stringer areas; their sum is
at the centroid of the material it replaces, P JiI , one ot external loads.
and its stress is the average stress for this
same material." The applied edge shear flOWS, coming fr~
the spar web, were assumed constant spanwise,
The sheet thicknesses used are the saQe as from a constant shear load. Other load
as those of the actual structure ..... distributions may be tandled by allowing these
applied shears to vary trom ~anel to panel.
For very extreme load variations additional
.. An alternate id.ealization, in which stringers iH and #3 are transverse members could be inserted to create
located along the lines AB and CD, respectively (Fig. A22.2), more spanwise panels allowing a better fit to
was used in Reference (2) for a box beam loaded in torsion.
the spar shear variation. The applied shear
.... When the longitudJ.na1 members themselves contribute to flows were considered as the other external
the shear stillness of the cover (as is the case for "hat" section load and deSignated P~.
stringers riveted to the skin so as to form small closed
cross sections), an effective thickness must be used. This ?anels on the centerline have zero shear
point is discussed in Reference (3). In this source, however,
the increase in shear stiffness is accounted for, not by
cue to symmetrJ (Fig. A22.3).
increasing skin thickness, but by decreasing the panel width •
an equivalent procedure.
--_._-_
-- ..
"-'--'--
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A22.3
~~ 1 2 6 8
I.
;a
~ 1 2 3 • 5 6 7 6 • 10 11 12 13 l' 15 16 17 16
I.
17
18
3.56
11. 79
The "off dtagonal" values have negative sign because the sense of those internal generalized forces haVing subscripts (14). (15).
7.11
23.58
'.7
(17) and (18) was taken OPPOsite to that used in the derivation in Art. A7.10. A change in sense requires a. change in sign In
off-diagonal coefficients only.)
A22 4. ANALYSIS 0 F SPECIAL WING ,
PROBLEMS
,
The following matrices Nere rorned:
[-ern]
Per eq. (17) of Chapter A8:
= C"5206
-6633
-262.'0
-262.8
d('"
.93 •'
~", .•
J' I 1111'c?1 Y 7
>i'
."
~. I
, .
.~
rc r s]
= [392.1
293.5
293·U
392.1 II i .5
i
The inverse or thIs last was found; CALCULATED MEASURED
Fig. A22.5 Comparison between calculated and
~rs -~=10-[5.802
-4.343
-4.343J measured stresses (psi) on the
5.802 hall-panel.
~ 1 2
1 -10.93 -.65
~ 5 6 7 8
5 18.03 5.04
~i~
2 -18.34 -.65
.. - 3 - 4.08 -.351
= 5 5.04 10.08
4 -11.66 -.351 7 18.03 5.04
5 1.40 .115 8 5.04 10.08
6 1.40 .383 The problem. was solVed retracing the same
7 3.79 .115 steps as before, but us Ing rthe mOd1fIedf=i~
matrix, to yield the stresses:
8 15.88 .383
[G1m] = 9 - .270 .019 X 1 2
10 - .510 0 1 -11.11 -.621
11 - 1.15 -.019 2 -16.83 -.621
12 0 .0355 3 - 3.90 -.379
13 - 1.60 -.0355 4 -13.17 -.379
14 0 .359 5 2.14 .104
15 0 .108 6 2.1'4 .372
16 0 .533 7 2.52 .104
~ImJ FLEXIBLE
17 16.16 .359 8 14.61 .372
=
18 4.86 .108
RIBS
9 - .26 .017
19 23.98 .359 10 - .62 0
The above analytical results are compa.ed ~ 11 - 1.05 -.017
with NACA test data (1) in Fig. A22.5 for the 12 0 .036
loading P II = 1, P:l. = O. Agreement is seen to
be good. 13 - 1.60 -.036
14 0 .359
EFFECT OF RIB FLEXIBILITY:
15 0 .108
To investigate the influence of rib 16 0 .533
fleXibility, the problem was reworked assuming
aluminum rib caps, of constant area A s .25 in., 17 16.11 .359
as the transverse members bordering the cutout. 18 4.86 .108
The appropriate member flexibility coefficients
19 23.98 .3<i9
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRU.CTURES A22. s
Comparison of this result with the pre- shear. The action may be comprehended readily
vious one for the rigid ribs reveals that the by visualizing an extreme case: a large degree
most important effect of rib flexibility was of ~lagff would occur if the load transferring
to increase the concentration of stresses in skin were made of a highly flexible material
the combing stringer bordering the cutout. It such as a plastic sheet or even rubber. In such
should be noted, however, that for this sym- a case the inside stringers would be out of
metric panel, the use of a very fleXible rib action almost entirely! With the inside stringer
as compared With a rigid rib led to stress in- stresses lagging, the outSide stringers and spar
creases of the order of only 10% in the combing caps must carry an over-stress to maintain
stringer. Thus, the "rule of thumb" that trans- equilibrium (~actual~ in the figure).
verse tlexibilities may be neglected in sum-
metric panelS is re-affirmed. Fig. A22.7 shows the beam analyzed herein.
9 ,..--,
'0.J ..)
A22.6 ANALYSIS OF SPECIAL WING PROBLEMS
the structure for a single transverse (vertical) to zero, the stress distribution due ~o a one-
tip load symmetrically placed. In that case, half pound load at the tip ( a unit load
because or symmetry, it was necessary to treat divided equally between beam halves) was readily
only one-half of the structure. In addition, computed.
no shear flows could appear in the middle .07692
panels. Further, it is known that t~e influence
of rib fleXibility on shear lag is slight for 0
symmetric systems, so that the ribs were con- -.9230
sidered rigid in their own planes; hence no
generalized forces were needed on the ribs to .9220
describe their strain energies. 0
Fig. A22.8 shows the placement and number- .07692
ing ot the generalized forces on the halt-be~~. 0
-1.8461
1.8461
0
~lmJ = .07692
0
-2.769
2.769
0
.07692
0
-3.692
3.692
0
fleXibility coefficients were
Memb~r \.
computed with the fonnulas of Chapter A7 and
arranged in a matrix. Next, the unit redundant stress distribu-
tion was computed. Fig. A22.9 illustrates a
The shear flows q., q, q:La and q u were typical calculation, showing the stresses in the
selected as redundants. Setting these equal t t p bay for qll = 1, q"l' = q1.ll = qu = O.
I}; 1 2 3 4 5 6 7 6 9 10 11 12 13 14 15 16 17 18 19 20
1 2'71
2 16 6"10
3 1 .74 2.684
4 0.0 4. 724
5 17.99 4.202
6 3271
7 16 640
9 2.684 8.939 1. 785
8 4.724 17.84 4.255
1 4.202 15.78 3.708
11 32 1
12 16 640
1 1. 785 7.142 I. 785
14 4.255 16.084 3.840
1 3.708 13.89 3.324
16 3271
17 16 540
16 1.785 3.571
19 3.840 7.518
20 3.324 6.439
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A22.7
rrj = r 175
-1848
-1260
- 460.7}
16,810 9,536
3'~~
r- =
[,210
16,810
9,536
3,041
32,160
9,536
3,041
9,536
25,030
3,041
3,041
3,041
18,650
- ,1m
~
' 285 -L505 - ,656
r:". -~ = 10-' -1.505
L.:: s.J - ,656
4.219
-1.000
-1.000
4.687
- .279
- .495
Fig. A22.9 Application of a sell-equilibrating unit - .183 - .279 - .495 5.519
redwl.dant stress qll.
and, finally, per eq. (23) of Chapter AS,
16 a a a a CALCULATED
17 a a a 1 3.416
a 2.364
18 a a a
1.278
19 -12 -12 -12 -12
20 12 12 12 12 5, 1
A22.4 Stress Analysis of a Box Beam With a Cutout. T~~ calculat10n :or ste~ (2) above, will
new be carried out oy conslder1ng a unit shear
In Article A22.2, one technique was em- flew, q12 = 1, applied ~hile all other loads
ployed for computing the stresses around a cut- are zero. Under this loading con~iticn the
out. In that analysis the effect of the cutout relative displacements at rec~~dant c~ts 2, 7
was presumed to have been localized about the and 17 are equal to =<2,1'1' 0<::", 1aand OO:::l",,~a
cutout region; consequently, the problem was respectively, where these coef:!cle~ts have
treated by isolating the affected panel. QUite been computed pr'evtoi.rs Ly 1.:1 Ar t i c Ie ";22.3.
often, when the cutout is placed well inboard
on the wing, its influence on the root stresses To restore continuity (reduce the relative
is appreciable. Therefore, it is desirable to displacements to zero) th~ee red~~dant ~0r:es
be able to consider the overall ;roblem of the are applied, one eaCh, at the cuts 2, 7 ~nQ 17.
box beam with a cutout for such cases. The appropriate equations spec:~y1ng cont~nu1ty
are
nThe most convenient and the most rapid
method of analyzing structures with cutouts is
the indirect, or inverse, method. The analysis
by the indirect method is made in two steps.
First, the structure is analyzed for the basic
condition that exists before the cutout is
made. The results of this basic analysis are
used to calculate the internal forces that Note that the~e equations say si~?ly that
exist along the boundary of the proposed cutout. the deflections at the reduncant cuts due to
External forces equal and OPPosite to these the (unkncwn ) r'ecundant forces must be equal
internal forces are then introduced; these and oppoat t e to the deflections due to q::.:a = 1.
external forces reduce the stresses to zero
along the boundary of the proposed cutout, and All coefficients in the above equation
consequently the cutout can new be made without were computed in Art. A22.3. SpeCifically,
disturbing the stresses. n (3)-
• The procedure described here is quite generally useful for •• See Appendix for a method of "extracting" the inverse of
studies of the effect of remoVing one or more members; this matrix from that previously found for the complete
such might be required for an analySiS of the effects at
structural damage. ~rJ matrix.
-
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A22.9
Note that q1~ is now zero and the cutout A22.5 Analysis of a Swept Box Beam.
panel may be "lifted out w•
WExperi~ental investigations of swept box
In the case of a structure under a variety beams have shown that the stresses and distor-
of external loadings (m = 1, 2, 3 ---), the tions in a swept wing can be appreciably dit~er
~ore general equation, carresponding to the
ent in charac~er from those that would exist if
the ~oot were ~o~l to the wing axis. The
above, i s principle ef~ect of sweepback on the stresses
IG l = IG l _fG} L G ,,,,,J occurs under bending loads and consists in a
concent~atlon ot bending stress and vertical
L imJcU'l'OUT li;nJBASIC 'liJl.ll
A22.10 ANALYSIS OF SPECIAL WING PROBLEMS
shear in the rear spar near the fuselage. With found in reference (8). One method of ~n~lyz
regard to distortions, the effect of sweep is ing such wings is given in Chapter AZ3.
to produce some twist under loads that would
produce only bending at an unswept WIng and THE STRUCTURE:
some bending under loads that would produce
only twist of an unswept wir~.~ (6) The structure shown is F:g. A22.12 is a~
idealization at the NACA test beam of refer-
In the following example a swept box beam ences (6) and (9), in which a Single substitute
is analyzed by the matrix methods of Chapter AS stringer has been employed along the cover
and, in particular, by the specific techniques sheet ta allow for the ant:tl~ated shear lag
of reference (7). The method accounts for the effect. The figure shows only one-half or tte
interaction between the swept cover panels and complete unit, which was built symnetrically
the longitudinal members. It is this action about the axis corresponding to the langitudln~:
that is responsible for the distinctive axis at the airplane.
structural craracteristics of the swept box
been , Only tip loads were to be applied (at
points A and B). The outer section of the be~
Again, we emphaSize that the method used was assumed to carry stresses which could be
here 15 strictly applicable to thln-sklnne~ calculated reasonably well by the engineeri~g
Wings at beam-like proportions only. Consider- theory at bending (E.T.B.). For this pur,ose
ing the wide variety at structural layouts it was jUdged satisfactory to consider the
which may be employed in swept w1ng configura- outer 66~ at the Je~ as a Single bay (A-B-D-CJ.
tions, a comprehensive treatment carillot be given If loads were to r4ve been a~pl:ed inboard of
here. An excellent review at methods better the tip, it would :iave been necessary to con-
adapted to thick-skinned construction and to steer additional bay dtvf s i ons between .;'-3 and
"plate-like" (very thin, Wide) wings, may be C-D (that is, insert additional ribs at statlcGs
f
/"-.
52"
.050 .050
--i.e H
,..A-.539 t,,-
Section C-C
.050 .050 (Steel) :;--t=. 125
66"
.078 t-,
A A
t·l +
J A=l. 966 A=I.430
of load application). Rib C-D was located at affects total deflections only and can be
one of the actual rib locations in the NACA omitted in a stress analyses where ceriect icns
test specimen and was assumed rigid in its own ar-e not sought.
;>lane.
Since only symmetric loadings were con-
The choice of bay C-D-F-E as a single bay sidered in this analYSis no shear was trans-
was somewhat arbitrary. For improved accuracy, mitted by the carry-through bay and hence no
additional ribs inboard of C-D could have been shear flows were shown in that portion.
used. Note that any ribs placed inboard or
point F will produce triangular skin panels in Sets at additional axial forces (q18
the cover sheets. Examples of treatments for through q1e:) were applied to the ends or the
such panels may be found in references {9}, flanges and stringers adjacent to the Obliquely
(10) and (11). cut ends of the cover sheet panels in bay
C-D-F-E. These forces are necessary to account
Rib E-F was considered flexible in its own for the interaction between the swept covers
plane, it being known that the flexibility of a and the longitudlnals. As shown in Fig. A22.14,
rib is important at a locati0~ where a structure the pure shear flow on the oblique edge is ob-
changes direction.* Note that this rib was tained by superposing onto the panel a zero-
made at steel in the test speCimen. resultant system consisting of a uniform tensile
stress of intenSity 2q plus a pair of concen-
Effective areas of longitudlnals as shown trated balancing loadS. The balancing loads
in Fig. A22.12 were computed by considering all must be contributed by the bordering longitudl-
of the skin to be effective. The spar cap nals and hence react on these as tensile loads
areas are equal to the sum of the areas of the (Fig. A22.l4C). The balancing loads applied to
angle member at the cap location, plus one- the stringers are shown dashed since they are
half of the effective area of material be~Neen internal forces within the bay and are not to
the cap and the substitute stringer (this area be entered into the equilibrium equations for
i~cludes several stringers as well as Skin) the structure.
plus one-sixth of the attached spar web area!*
The substitute stringer area was collected in
like ma~~er from the half-panelS to either side. P=qL
Fig. A22.13 shows the choice and numbering From an energy Viewpoint, these dashed
of the generalized forces. forces account for the additional strain energy
stored by the axial components or shear flows
The beam was rigidly supported at points in the non-rectangular panelS. This energy is
E, F and at the two corresponding points on the stored in the cover panels themselves (and is
other beam half. These might correspond to the accounted for in this manner since the longl-
:uselage ring attach points in an airplane. tudinals contain the effective area irom the
The vertical end caps on rib E-F were considered cover sheet·** ) and in the longitudinals Which
rigid ~xially, so that no fleXibility co- react against these components.
efficients were associated with the reactions
qal5 and qae:' Flexibility of these members "Dashed loads 11 are applied to the longi-
tUdlnals adjalsnt to any obliquely cut panel
end. Similar dashed loads would be applied to
• The effect of neglecting this rib's flexibility is demon-
strated later in this example.
-** This much of the energy could be accounted for in another
** The factor of 1/6 is used so that the effective area con- fashion by modifying the member flexib1llty coeUlcient for
tribution of the web results in a structure having the same the sheet panel. See Reference 12. where this was done.
moment of inertia about a horizontal ax:I.s as the original. However. that referenc e incorrectly neglects the additional
Some of the problems of idealization are discussed in energy stored in the longitud1nals, as was demonstrated in
reference (2), p. 16. Reference 13.
,_. - ~ ---
the outboard ends of the panels in bay C-D-F-E Stresses in inner bays:
if they too were cut obliquely. Such ~anel
configurations arise often in swept wing con- According to the discussi~n of Art. A8.12,
struction having ribs parallel to the air- Chapter AS, the deter.ninate stress d~stributicn,
stream. Formulae for ~ore general quadrilateral
panelS are given in Reference - §tmJ, may be any stress distribution in 8qU~:'
ibrlum With the applied loads, and preferably
'!'HE STRESS DISTR!BUT I O,:S one close to the final true stress distribution.
The magnitUde of the redundant fcrces is re-
For the symmet.r-t c rcadmcs considered duced by use ~f ~ satisfactory estiillate of the
here the structur e was indeterminate only two true stresses.
times since the auter bay was assumed to be
determinate by the S.T.B. The stresses in ~he two ~nner bays were
determined for both gim and glr si~ultaneously.
Stresses in outer bay by S.T.B.: Since this inner portion of the structure is
two times indeterminate we can estimate two
Flange stres ses at rib C-D (for both loads. For this purpose the two flange loads
P1 = 1 and Pg = 1 ) q 111 and q lo4 were written as
2.924
Joint F
E'% qb= .O2713+~
q., + q •• = 0
... 07143 Ibs q• .3 + qlla = 1.414 q16
Til
= 5.513 P~ + 5.513 PlI + 1.414 q~"
The unt t load was shifted 15" to either side
by application a r a torque, T = 15 in.lbs. Joint E
The unt rcra shear flow superposed was
q.. + q •• = 0
q=~=2X
15
210 = •
03571 Ibs
in
q•• q.. = 1.414 q..
= 7.392 P, + 7.392 p. + 1.414 q~2
I1Z 1 2
1 .1071 .03572
2 .06284 - .00858
= 3 .00858 - .06284
4 .03572 .1071
... This is a rather crude way to estimate these loads and is
5 2.924 2.924 used here only for simplicity. The analyst is generally better
advised to exercise a little more ingenuity in making these
6 3.581 3.581 estimates, even to the extent of being guided by other swept
7 2.924 wing solutions.
2.924
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A22.13
E 1 0 0 0 -1 0 0 q.
~ 1 -.7115 0 0 0 0 0 q.
G
• G
•
"
01'
0 0 21.2 0 21.2 0 1
l~ iN==.
F ,;
~
~
~.
0 0 1,682 1 0 0 0 q" =
~
0 0 0 1 1 0 0 q ..
C H ~
Cap EX}
Atter inverting the matrix ot coefficients
on the lett hand side at this last equation,
and multiplying through thereby, the stresses
were obtained as
Joint G
c, .1006 .0295 .00676 -.00676
.707 q:l,::) q •• =0 q. .0791 - .0208 -.01753 _.00950
.707 q1,,) q .. + q .. = 0
q" .0013 - .0411 .00401 .0230
Cap GF q" = .0422 .1135 -.00675 .00675
q.. .2409 .2409 .00676 -.00676
qll1, 3.087 .964 -.9355 -.3444
Rib Vertical at F q.. -2.379 -1.473 -.2284 .3625
+ .04545 q~40
The fi~st fIve of the above equations
were readily solved by substitution, y~elding:
- ~!
-437-
~'"'"
A22 14 ANALYSIS OF SPECIAL WING PROBLEMS
52.3 1-
q.. = = 2
~Pn~
X (- ) q , = -12.5 q,
1"'( 1 2 12" 14" X '25
• - x
.0 :::I
1 .1071 .03572 0 0
= -.989 P lo • .260 Fa + .2191 q~a • .1168 q~
•
2 .06284 - .00858 0 0 q.. = ~Pclt = 73.5
2 X 125.8 '5
X ~ X I-I qe = -17.5 q,
3 .00858 - .06284 0 0
4 .03572 .1071 0 a
'"' -1.384 Plo • .360 r, • .3068 q;a • .1662 q"
Similarly, for the panel HGFD;
5 2.924 2.924 0 0
6 3.581 3.581 0 0 qlo7 • -.0146 Plo •• .773
.460 P a -.0449 q~a - .2576 q~
..iiiiiiiiiiiiiiiiiiiiiiiiii
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A22.15
.,
MATRlX OF MEMBER FLEXIBIUTY COEFflCIENT-'· (E = 1) {FormulaB from Chapter A7:lnd Re£erenee 7)
1 , 3 4 5 , 7
• s 10 11
" ia 14 15
" 17
" 19 ac 21 ! 22 " ,.
" "
15.400
1 2 60
1 40
70.170 1 .46 15.46
50.0 8.984 8.984 8.984
52.33 6.542 6.542
112 130
• 400
46000
,
5 0
15.46 30.92 30.9Z
.984 17.97 17.97 17.97
.42 .08 13.08
15.46 30.92 30.92
8.984 17.97 17 97 17.97
8.984
6.542 ." 13.08
17.97 7.97
13.08
2130
9.046 4. 2
, 4 '" 04 4.523
4.523 9.046
s.
4
•
19.58
CO!lPARISON WITH TEST DATA: requirement. More details concerning the test-
ing teChniques and method at data presentation
5.2 would probably resolve this conflict. Both test
/
~.6
2~
3.8
Root
2.6
2.6
-,
118)......
2.8
3.2
f\
2.'
2.'
~.6
2.'
and calculated values clearly exhibit the
characteristic build-up at stresses in the rear
spar of a swept Wing.
RIB E-G-F RIGID:
ConSidering that rib SGF was relatively (7) weme , L. B., and Lans tng , 'A •• A Metccd ~Gr
rigid to begin with - being nade of heavy Reducing the AnalYSiS ~! Com~lex Retun1~nt
gage st3s1 - it may be seen that neglect of Str~ctures to a Routine Procec~re, J8urnal
the flexibility of a corresponding all-al;~inum Q: Aero. SCi., 19, Oct. 1952
rib could lead to serious arrors.
(S) 'williams, M. L., A Review of Certain
REFERENCES Analysis Meth~ds for Swe~t Wing Structures,
Journ. of Aero. SCi., 19, p. 61~, lS52
(1) Kuhn, P., Duoer-g , J. E. , and Diskin, J. H.,
Stresses Arolli,d Rectan~ular Cut-Outs in (9) ~e1denfels, R., Zender, G., and Llbove, C.,
Skin-8tringer Panels Under ft~lal Load - II, Stress ~d Distortion Ar~lysis of a Swe~t
NACA 1rJR w6B (ARq 3J02), Oct. 1943 Box Beam Having Bulkheads Perpe~c~cular ~o
the Spars, NACA TN 2232 . •
(2) Rosecrans, R., A Method for Calculating
Stresses in Torsion-Box Covers with Cut- (10) Bispl1nghoff, R., and Lang, A., An In-
outs, NACA TN 2290, Feb. 1951 vestigation of Defo~tlons and Stresses
in Sweptback and Tapered ~ings with Dis-
(3) Kuhn, P., and Chiarito, P. T., Shear L3.g continuities, Mass. Inst. of Tech. Rept.,
in Box Beams: Methods of Analysis and July, 1949
Experimental Investigations, NACA TR 739, (See also, Journ. of Aero. ScL. ..la, p , 705,
1942 1951)
(4) Haft, N. J., and Libby, P. A" Recommenda- (11) Denke, :., The Matric Solution of Certain
tions tor Numerical Solutions of Reinforced Non-Linear Problems in Structural Analysis,
Panel and Fuselage-Ring Problems, NACA TR .jour-n, of Aero. se i .• ~' 1956
934, 1949
(12) Levy, S., Compirtat Ion of Influence Co-
(5) Kuhn, P., Approximate Stress Analysis of efficients z'or Aircratt Struc-:ures witt'..
Mu1ti-3tringer Beams with Shear Defor.natlan Discontinuities and Sweepback , Journ. of
of the Flanges, NACA TR 636, 1938 Aero. SC1., 14, Oct. 1947
(sa) Kuhn, P., Deformation Analysis of ~lng (13) Is Lfnger , J. S., Stress Analysis and
Structures, NACA TN 1361, July 1947 Stress Measurements for a Swept Back Wing
Having ~ibs P~ra11el to the Airstream,
(6) Zender, G., and Llbove, C., Stress and McDonnell Aircraft Corp. Report 1127, April
Distortion Measurements in a 45° Sweot Box 1949.
Beam Subjected to Bending and Torsion,
NACA TN 1525 ..
i
CHAPTER A23
STRUCTURAL ANALYSIS OF A DELTA WING
ANALYSIS BY THE "METHOD OF DISPLACEMENTS"
ALFRED F. SCHMlTI'
quadrilateral "torque boxes" connecting ~olnts are written for ~he individual elements of the
ot intersection of the spars and ribs. This st~ucture, they ~re Simply ~dQed to g~ve the
stiffness ~trix of the composite. Finally,
the resulting stif~r.ess ~trix ~zy be inv8:ted
to yield the matrix of ~lexibility influence
/ Rib coefficients (Eq. 2 "in r-everse"}.
We note here, ~s an aSide, ~hat t~e Monent
Distribution Method of Chapter All, the Slope
Detlec~ion Method of Chapter A12 and the~od
of Successive Corrections of Cr~pters A6 and
A15 are all ex~~ples ot the "Displacement
Method" o~ structural analysis. In each of
these methods the displacements are taken as
the independent variables and these are adjusted
to achieve equilibrium of the loaded str~cture.
The nadjUBtment n may appear as a systematic
relaxation or artificial constraints (~oment
Distribution and Successive Corrections) or it
Fig. A23.1 Idealized Delta Wing Structure may be dane mathematically !~ one stroke by the
solution or a set at simultaneous equations
(Slope Deflection Method). The latter approach -
idealization is discussed in more detail in the solut10n of a set of simultaneous equations - 1s
example. Assume that the points of inter- essentially that :'ol1owed herein, the "so Iut t on"
section have been numbered as in Fig. A23.2. being effected by ~trix inversion.
-
, ~ :,_~. _~ . ~'._" ··<'~.'-~~'.'k"". >_~,>,: ••_ .~._': . -~-'.. :'--;e/'--~':~'·'>--·' ,c . . ·,·.{::·.1.. 'T'... ~,_.<.;_.":~:_,_:...,.,,::..:;-.:::
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A23.3
this phase of the problem are too great to per- Beam Element Prooerties
mit an expansive treatnent here.· Briefly,
Levy's recommendations are; (Moments of inertia, in (inches"), are
assumed to varJ linearly between numbered
(a) all skin may be considered effective points)
between spanwlse spars when computing the cap
areas of such spars. This assumption is sub- li2 Fuselage Beam (1-2-3-4) Spar 2-5
ject to modification, of course, if spanwise
stresses are anticipated which will buckle the
3@48"· -48"--
skin.
(b) for streamwise ribs an effective width
==
1= 2000
"II
2700
-- !
3400
I
2200
of O.362L, where L is the rib length, may be
taken as acting with each rib cap (Fig. A23.3a).
(e) for the leading edge spar an effective Rib 5-6-7 8par 3-6-8
width of skin of .181 of the span between span-
wise spars is taken as acting with the cap of
each such spar segment (Fig. A23.3b) - 48" - - 48"- _48"_48"_
, '
1 r ! !
I = 9.24 20.79 3.59 47.23 40.82 12.33
_ _·_3@48"
-
'V'"
I (b)
-
[
1= 6.23
48"-
!
2.87 5.09
[
4.78
£
3.82
1 !
0.28
Fig. A23.3 Effective Widths of Cover Sheet for
Rib and Oblique Spar Caps (After
Levy, Ref. 2)
Leading Edge Spar (1-5-8-10)
The properties of the structure (Fig.
-----3@87.8~'l_-----
A23.1) after idealization, are summarized as
fol1 ....S: f ,
I '" 16.02 12.29 8.30 O. 26
Net Point Beam Depth
(see Fig. A23.2) (inches)
1 - - - - - 5.56 SIGN CONVENTION Al'Jll NGrATION:
2 - 9.12
The Sign convention and notation adopted
3 7.80 in conjunction with the grid numbering scheme
4 - - - - 3.16 of Fig. A23.2 is as follows:
5 4.88
Forces:
6 7.26
Pm - transverse torce at jOint m, ~ UP
7 - 3.06
8 - 4.02
Mm - moment at joint ~ acting about a
pitching axis, + NOSE UP
2.74
9 -
Nm - moment at joint m acting about a
10 - 0.72 rolling axis, + ~IGHT ~ING UP
All cover skin t = .051" Qm - moment at joint m acting in the plane
of an obl1que beam member (a :nember
neither parallel nor normal to the
st'reamwtse direction), + CLOCKWISE
',/HEN THE MEMBER IS VIE'';ED ',-lITH ITS
• A rational, systematic means of treating cover sheet JOINT NUMBERS INCREASINJ LEFl' TO
panels is given in Reference (4). RIGHT •
A23.4 STRUCTURAL ANALYSIS OF A DELTA WING
~imJ =
culations of the above type appear in this 2 .567 -16 -8
prOblem, an alternate (but equivalent) pro-
cedure was developed for systematically treat- 3 .333 -8 -16
ing these members. This procedure 1s detailed
4 0 0 0
in the follOWing section.
Beam Element Influence 80efficients Finally, the following matriX triple pro-
duct is formed to give the i~:luence coefficients
The ~ethod em~layed to ~btain the beam
(ct. Eq. 24, Chapter A7).
A23.6 STRUCTURAL ANALYSIS OF A DELTA WING
= [1~~::0
-119.9
-141.6
27715
2566
- 1 .J.O:I.::'
0 __
2 ". . ;::.;::.
3104
'J the 1eflections relati,e to the beam element
ends to the loads a?plied to the bean element,
end reactions exclUded. It 1s desired next to
obtain the "compjete" beam element stiffness
matrix in which ~bsol~te beam deflect~cns are
related to all the loads applied to the bean
o~J written out as an equation; element incl~ding the end reactions.
_1l9'J~{N.}
{
~ .} llo.20 -141.6 Fig. A23.3a to the absol~te deflecticns of ?ig.
A23.8b.
5~ = -141.6 2776 2566 ?~ -(3)
(al
L
f
cient matrix.
be of the form
.
element stiffness matrices tor relative deflec-
tions by invert~ng the above influence coeffi-
-- The results after inverting will
From the geometry
+
'c'!ll+s - ~
L
c~ the figure, one r2S
6m+r = -
L - kn+r
L
_ iL + iI,n+. - iI,n
<Im+. - p+s L
where ~kJ may be called the Wrelative stiffness
matrix. Specifically, for Spar 4-7-9-10
For comparison with the matrix form to follow,
N .) [,3554
~:~~~:~~{::)
.02303 these equations are rewritten as
r , = .02303 .003013
{P" - .005306 -.001600 00143~
• 6. -T
1 +
T
1
~+s
I...m+r - L I...m+r
• We note that the inlluence coeUicient matrix to be inverted o",+r = + D.:n+r ~+s
does not appear to be "well conditioned" in the sense of Art. L L
AS. 12 (see p. AB.28). The situation is more apparent than
real, however, and arises because of a difference in units 1 1
among the coefficients of Eq. (3). Thus, the appearance is <Im+. = - L +
L ",,+.
readily altered by "scaling" (factoring out appropriate con-
stants). For instance, we may write
+ ~+.
31.0~
so as to prOVide a transformation fram absolute
6./,0 l1.96 25.66 lOP. to relative displacements. In matrix f0~ the
index (subscript) is understood to inc~ease
Scaling in this fashion does not actually enhance the con- monotonically down the column, and inasmuch as
dition of the matrix (which is basically related to the size
of the determinant of the matrix) but it does permit one to
the grid numbering scheme gives this same
assay the problem better. property to the joint numbers on an element one
has:
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES A23.7
where
- - - - - - - - - - (5)
~ Nm~
Pi:!
r J where
1
1 -L"
J
I l2:----~ : T1 0
------1-----7------
r I
1 ! ~------:£J 0
0
r
r
r !.m-L
r
I
r
I
r
I
I
Q
r
I
t: ~lm - L)/.!J
-----------,--
I J
t
[TJ = I
r
I
I
-L- I
I
I
I
[I] I
I
I
-!.m
L
I
r
I
,
I
o
J
I
J
I
I
J
I
I
0
I
I
0
r r
I
I
I
J
r
0 [sJ = I
I
I
J
[I] I
I
I
I
---- -- ----1-- ----- J r
I
I
I
I
I
I
I
1 1
0 -T ~---~ I
I T 1 a
I
r t
I
0
I
__ ~ 1. _
Pm+ s = -?-- t Z
a-r P!l1+r • Lm+r + N."L·+ s
is the relation be~Neen loads and relative
deflections at a beam element. Substituting
from eqs , (5) and (6), into (4), one' obtains
In matrix form, the general expression
which introduces the end reactions in te~s ot
the other applied loads is
~i = [sJ [kJ[TJ
WITH END
tll ----- ABSOLUTl:
(7)
Eq, (7) is the relation between absolute de- For Spar 2-5
flections and the co~plete loading on the ele-
n[= L~'~n
ment. ve let -.1218 ,1218
Pa -.1218 .004208 -.004208 -.08054 2:.",
PI - .1218 -.004208 .004208 .02054 ~~
be the "complete" ele~ent stiffness ~trix.
.K" 1.555 -.08054 .0605? 2.317 ~"
Note that [SJ is the transpose of G'J, as
For Spar 3-6-8
it must be if [KJ Is to be s~~etric.
To continue the illustration, the and
[sJ matrices for Spar 4-7-9-10 are now written.
For this member the rotation at station 10 Is
not considered and the couple at that end is
zero, hence the last row and column of [TJ and
[TJ .'
p.
p.
P,
:;
.
.I " ~~
-.03423
-.2534
.1055
_.02::..29
-.02423
.00275';'
_.::;0-'1.1:2
.:J01:X:S
.01725
.:265
_.00.:;.1:2
. '''''',,-<>-
J,.....
_. ::;03066
_.05.03...
_.02228
.J013~3
_ ,'O:J3CC'6
.OOl~C8
. :)3908
OJ '~1"''''''
':;':;: t::
.,,3,,0~
1.33
I·
~.
1.
n
Equation (5) written for Spar 4-7-9-10:
P, -4.312 .1557 -.2385 .09993 -.017l8 6 ,
~, P, = 5.560 -.2385 .4343 -.2701 .07429 -,
-,
0,. -
[_oo,~
0 _.667
a
1.0
a
a -.333
OOOO~ il,
il,
.,
D -1.533 .09993 -.2701 .0:::;............... -.1027
" ........ .04560
I c.~
,,,
t
rJ 0
Equation (6) , written for Spar 4-7-9-10:
'~::~rl
3 -.G£253 .0275
r- a- .0,536:0 _.00:2:>. .v,"'" v _. I
M. La a '. '" -.02253
·:o:;;~ :: J
? .0276
:'
_ .OC12C4 .CClS3Z o ·c·' I '
rJ IT'
p. -.01606 .0~,;g52 -.001637 .0,6252 .0.:65
?
p.
= .02303
-.006306
-.001637
.0,:5252
.ooaoia
-.00160::J
_.001600
.OOH38
.C~2238
-.0~46~
:: j I
.3>178 -.01846 I .C1:;'~ J I .-.l~O I'~ c I
p,. -.001622 .0.:65 0,.:22:3:0 _ .::J,'i-B3Z .::J,2229 :: 0
P, -.01946 .C,~';'90 -.O,d901 ::J -.':;1756) c I
p. ,01".;..s -.0,.54,,,0 .c.saes -.0,394:1 .0030<34 -.01.0.5" o!
In the manner of the above illustrative P.
case one goes through the calculations for the .';'1&0 -.01766 .OC3~24 .21...07 l.~';' I .3C16 ,0 i
remaining beam elements to obtain:
::J J - .01259 • JO~ 1001 •.3016 I --q, . cc:;:;sz I
o
• The notation . ON means there are N zeroes following the .. Because of its Size, this matrix relation (and some to
decimal point and preceeding the first significant figure, follow) is written in a condensed tabular form. The correct
e. g••• 0.. 276 '" . 0000276. interpretation should be obvious.
M
~ransfo~~tiJn of Matrices for Oblique 8ea~s: Next, consider resolution or the in-plane
T~e complete stiffness matrix of any beam
bea~ element cQuyles. Counles Q~, Q.8 and Qa
element not either ~arallel to or nor.nal to the have components in the pitching and rolling
strea~ise direction, such as the leading edge
directions. ~or exa~ple, at joint 5
spar in the present example, requires an addi-
tional modification to yield expressions relat-
I1 8 = sin J\.. Q.. .707 Q.. =
and
ing couples and rotations about pitching and
rolling axes, rather than in the plane of the Na = cos A Q,s = .707 Q,s
member. Such a modification is readily made.
Similar relations for the other couples along
In the present example, a pitching rotation this member lead to the matrix transformation
of joint 5 (9 s) results in a rotation of jOint 5 for this element's loads:
in the plane of the leading edge spar given by
~~ = sin A 9, • •707 9" 11, .707 0 0 0 0 0 0
where A- is the sweep angle of the spar. Like- N, .707 a 0 0 0 0 0
wise, a rolling rotation of j~::nt 5 (~,) has a Q,
rotatlor~l component in the plane of the leading
P, 0 1 0 0 0 0 0
edge spar given by P, 0 0 1 0 0 0 0 P,
T:
= cos .'- 9, = .707 ~, P, 0 0 0 1 0 0 0
P,
Tten, when joint 5 experiences both pitching and 11, = 0 0 0 0 .707 0 0 P,
rolling rotations, the total rotation in the Q,
plane of the leading edge spar is N, 0 0 0 0 .707 0 0
11, 0 0 0 0 0 .707 0 Q,
rn
J. II = .7079 11 + .707 ~,
N, 0 0 0 0 0 .707 0 T"
This last equation, and s~ilar ones for joints
1 and 8, when put in matrix fOIiJ, yield the P" 0 0 0 0 0 0 1
~atrix transformation for the displacements.
~l
0 1 0 C 0 C 0 equation for the leading edge spar as given in
,, 0 Q .7C17 .707 0 0 0 ", tabular form in the last sectIon. The opera-
". o
0 c 0 0 0 o
c
•707 .tsn 0
'. t10n involves premultiplying by one transforma-
, .i,o) l- 0 0 0 0 0 1 I. tion matrix and postmultiplying by the other.
The result is:
"
I-
'"::'0
.
Leading Edg~ . SDar cv
namDlete _.,
Stl~Pness ~atr1x . . ,
(~rans~ormed)
A23.10
Re-solvlng this, we write
(r~ Ln - - - : GJ Lo
T ' '<n) Pm = Lm(bn + bml (t-nom + Lmbnl Cn>rCTAL
"'T L / 2 - - L / 2 ' --,-
_ _ _ .. _ _ .Iil - +bn
This last expression relates the load at
I
~s,
t~Axia
:
\-----.lbm
(m)
point m to the detlection at that point, with
the ather three corners undetlected. 7he re-
L...... \ ....... I
actions developed at these Jther corners are
L
,........, w" '
now found in terns of Pm USing the equations of
Equivalent Representative statics (s~ation ot forces and summation ot
, Root Section moments- about two axes). The r-esuj t , expressed
in matriX fo~ is
Fig. A.23. 10
A, = Lnbm + ~bn +
Ln(b r +
bs (Lm - Lnl
os)
An (assumed) average rate of twist, Q, Is com-
puted approximately by using the GJ at a repre- -(Lnbm + Lmbnl + br (lm - Lnl
sentative section halt way along the box.* A.
" Ln (or + b s )
Q = O~ = Pmbm O~ Pnbn It now the corner pOints have absolute
displacements Om, dn, dr and tis, one can find
Then the deflection On and ~ are given by from the geometry of the unit that the total
relative displacement of point m may be writ~en
= QLnbn = l'mbm OJ- Pnbn -nbn
On I.
Cn>rarAL = L 1 A,
GJ = [KJ =0.000482
-1
_1
1 1 -1
1 1 -1
1 -1 -1 1
Here A is the area enclosed within the tube
cross section by the median line of the tube For box 6-7-8-9
wall and the integration is carried out around
the tube perimeter (index s gives distance along 1 -1 -1. 1
the perimeter). For the torque boxes en-
countered in delta wings it is probably satis- -1 1 1 -1
factory to neglect the ds/t contribution from [KJ = 0.000310 1 1 -1
-1
the vertical webs, it being small compared with
the corresponding contribution from the cover 1 -1 -1 1
sheets.
A23.5 Complete Wing Stiffness Matrix
In the example wing three boxes (2-3-5-6,
3-4-6-7 and 6-7-8-9) are to be used. These The stiffness matrix tor the composite
boxes are each 48 inches square in plan and wing now may be obtained by forming the sum or
have average depthS (assumed here to oe the the complete stiffness matrices tor the beam
unlfor.n depthS ot the representative sections) elements and the torque boxes. For this pur-
at 7.26~, 5.32" and 4.27~, respectively. Fig. pose a large matrix table Is laid out and en-
A23.ll shows the assumed representative cross tries fram the individual stiffness matrices
section of box 2-3-5-6 and its OJ calculation. are transferred into the appropriate locations.
Wherever multiple entries occur in a box these
are summed.
ie
,
l_ t
48"
= . 051 ·,
_ Betore the large wing mata-tx is laid out
it is necessary to observe first that the
A: 7.26 x 48 =348 in .. matriX which would be obtained as just indicated
Note: would be singular, i.e., its determinant would
f t:= 2 x. 051 " 1. 88 x 10
~ ~ ~ E be zero, and hence it could not be inverted to
G := 2:6
yield the flexibility influence coefficients
4(348)8. 1 . .. (see Appendix A). This condition arises due to
GJE=I" 1. 88 x 10~x 2.6" 99.3 lb.fn. the tact that the equations represented (19 in
(ds/t contribution of vertical webs neglected) number in the example problem) are not inde-
pendent; three of these equations can be ob-
Fig. A23. 11 Calculation of GJ for the Representative tained as linear combinations ot the others.
Section of Box 2-3-5·8
That there are three such interdependencies
In similar fashion one finds: may be seen from the existence ot the three
equations of statics which may be applied to
For box 3-4-6-7 the Wing (summation at no~l forces and sum-
mation at moments in two vertical planes):
hence three ot the reactions expressed by the
structural equations in the matrix may be found
A23.12 STRUCTURAL ANALYSIS OF A DELTA WING
tram the others by the equations of statics. singular. However, if this ste) ~s carr~ed au:
To remove the "Singularity" rr-cm the stiffness one fines that :he resultar.t inf:~e~ce co-
matrix it is only necessary to droD out three efficients are those for- onjv t.ne "':e.l.::'-wl:-,:
equations - achieved by removing :hree rows acting alone and supported by ccr.s:r~in:s as
and corresponding columns (so as to retain a ass~ed above. To account for the ~resence o~
symmetric stiffness matrix). the other ::al! of the vnng , i t is necessar-y to
speCify additional geo~etr~c con1:tlons along
The act ot removing the three equations the airplane centerline. This ste? 15 accon-
selected is also equivalent to assuming the plished by assumir~ the following deflect lens
corresponding deflections to be zero. In this zero (ellmir~ting their corres~ondlng rows and
way a reference base for the deflections 1s columns from the matrix):
also established. The choice of reference base
is somewhat arbitrary, but, following a sug- for symmetric loadings (zero
gestion of Williams (7), a triangular base will lateral slope or rolling
be employed as shown in Fig. A23.l2. rotation along the alr,lane
centerline)
0.01070
from
from
':'OTAL
box 2-3-5-6
box 3-4-6-7
[:.::nnJ ' . . ing Stiffness Matrix, Ar,tisymmetrlc Case (Note: Voids denote zeroes)
, I,
", '. ",
'. '" " " I I, j, I. I. I.
P, .01070 -.GOl<;ZO -.003376 .O~310 .02760 ! .1065 -.05634-
p. .001420 .004066 .0~310 _.00191 .0~2238 -.0~066i .02303
p. .003376 .G~310 .002604 -.0~4537 -.0~4998 .009947!-.o,441 -.02228 .009947 .045.:::>5
p. .0,310 -.00191 -.0,4537 .001892 -.0~4632 .006910 -.005306
,
."
0
.0,2238 -.0~~998 -.0~4632 -.0~2729 .002391 -.001662
,
.002391
M, .02760 - .OOS06E .009947 1.555 .1507 .2094- .7168 .1507
M. -.0~441 .006910 .002391 .1507 .7211 .1507 .3889
N. .2094- .4487 .2094
N, 4.302 1.5s.:;
N, .1065 -.02223 3.256 -.2334
N. .02303 _.005306 -.001662 .3554
N, .009947 .7168 .1507 .2094 1.555 3.0338 .1507
N, .05634 .04555 .002391 .1507 .3889 -.2834 .1507 1.719
The Wing flexibility influence coefficients are now found by forming the inverse of the above
s~ifrness matrix (see Sq. 2). An automatic digital computing machine is the essential tool for
this step. The details of the procedure and techniques employed in forming such inverses are not
of concern here and only the result 1s presented. It is assumed that this phase at the work can
be handled with dispatch by experienced computer personnel.·
..
0 P, p. P, P ~o n. M, N, N, N, N. N. N.
73';'.3 se8.7 1386 :,';64- 2"'19 -17.14 -14.S9 7.750 _.113...0 -15.77 -2.3.52 .5092 -14.22
"
.>, 988.7 2'~03 2118 4030 537<5 -16.32 -38.23 a.610 .04SS1 -19.96 _70.~ -.1342 -24.2.2
'. 1386 2118 4001 4751 7692 -36.09 -25.10 17.57 •.5620 -23.72 -30.20 -1.555 -66.22
'.,
-~Q
15G4
Z"'lS
~30
53,6
4"'61
7692
5-558 13,610
13,510 2.5,670 -;:>a.69
-~a.09 -32.32
-131.0
1 18 • 01
27.75
.1::.26
.2836
-34.28
-~7 .C4
-6:5.70
-26.56
-.5052
-.7S<06
-67.49
-123.3
-17.14 -18.32 -36.09 -33.09 -58.69 1.<25 .09551 _.~77g .07'273 .3438 -.201.2 .4227
"•• -IL5!? -33.231-25.10 -32.32 -1.'31.0 .09.551 .2701 -.00806':;' .02523
.0.'303
.2976 .6358 _.07823 -.2044
t 7.'7"60 8.510 17 .57 13.01 2"'.75 -.47-;;) -.00a06... 2.,513 .~73a -.1527 -.1592 -.131: _.2198
I, -. ::'~40 .0';"':':>1 .,')620
.1825 .2836 _.0-2'73 .02~25 .04-38 .2S54 .80834-7 .0)9086 -.17.;.;3 -.01744
t _1."3.7"7 -19.96 -2.3.72 -34.28 4 7.04 .;503 .2[;76 -.1.':27 .00834'::' .;:---'"
.~, .'::614- .02309 I .1::;28
I ~. -2.3.;:;2 _70.:'0 -30.':00 _65.70 -25.56 .-34:38 .6338 -.1592 'OJ sees .-~61'" '3.271 1-. ':]0251-+ .)1...20
I, .--~O;2 -.1342 -l.~55 - .50.'52 -.-846 -.2012 -.072.23 - .1311 -.17.;.3 - .-::'<:309 -.0025H . ...a21
". ~~ ,
.V"...:>.::....
I, -H-.22 -24.22 _'36.22 -'57.49 -123.3 ..;.2"- -.2Q44 -.2198 -.017;;.4 . 1~2S -.014,,0 (.04,82... 2.079
" In spite of the exercise of great care and much ingenuity, problems of such great size and such a nature will arise occasion-
ally as to defy satisfactory inversion. For these problems such clever physical concepts as "block" solutions of portions of
the structure are available. The interested reader is referred to a discussion of Ref. 4 and to techniques of block and group
relaxations in the literature on Relaxation Methods (e. g. Ref. 8).
A23.14 STRUCTURAL ANALYSIS OF A DELTA WING
A23. 6 Wing Internal Stresses Equation (10) provides t~e cesired relation
between ~xternal loads applied tc the wing and
following t~e calcul~tion of the Ning t~e forces acting on a speCi~lc eleme~t. These
influence coefficients, a relationShip is avail- ~crces on the element might be locked upon as
able between the a~plied wing loads (assumed those necessary to hold the element in the de-
given) a~d the wing qefo~tlJns. In a sJ1nbolic flected srape conforming to that of the wing.
fashion, Of course, with these element forces known the
finer details of the stress distribution within
the element are readily found by standard
techniques.
~~}
consideration.
.8mlm }
--- ---
={.1m}
8m
~. .::., o.
". ,
, "
b~ EL=rr ~ W!m
II,
p.
.3554
-.0160(·
-.016013
.O~~952
.0230:3 -.005306 -. ;:·01562
Hence, the previous equations may be combined [KJ: , p. .023C .3 -.001637 .003013 -.J01600 .0.2238
immediately to y1eld
P, - .oor .306 .0.5252 -.C016CO .001438 -.:J.4632
P., -.u....J,662 a . --
• • ':""Q I •0• 2-->8
;,;:... -.0.';'632 • J~2229
.
.
Expanding (and rearranging to agree in numerical order with the wing matrix),
A, AT A, A. A" e, 9, ~. ~. ~, ~. ~, ~.
N. 0 .0230:3 0 -.005306 -.001662 0 C 0 0 0 .3554 0 0
p. 0 -.001637 0 .°36252 .0,,165 a 0 0 0 0 -.01606 a I 0
rKJ
J - E p. 0 .003013 0 -.001600 .°3223 8 0 0 0 a 0 .0230:3 0 0
L £LL'"1'!ENT
p. 0 -.001600 0 .001438
-'°3 4632 0 0 0 0 0 - .005306 0 a
P" a '°3 2238 a -'°3 4632 '°3 2229 0 0 0 0 0 -.001662 0 0
L
Then multiplying out with [~~WING per Eq. (10l, we get the five forces en spar 4-7-9-10 (see
p.
P,
P,
p.
-.C~367 -.0017'7 .o, 794 .0,304 .002694 .0.1736 .0.3834 .0.2706 0 -.0.8n 1.00 0 -.0.555 p ..
N.}
p.
p.
-.0304
.06294
-.1945
.3735
•1212 .2190
,2098 -.'0191
.4985
-.2915
-.0,320 -.001257 .0,180
.0,5408 .001850
.0.24-9
-.0,330 -.0.625
.00145 -.02696 -.0.687 -.004:34
M.
-.00299 .03148 M,
.0,1705 .007071
p. -.00375 -.1650 .05466 .3787 _.9174- -.0,107 .0.975 .0,1133 .0.486 .001451 _.00281 -.0,135 -.001107 N,
p. .002905 -.01364 .03347 -.07,,65 .7092 -.0,111 -.0,682 .0.348 -.0.1202 -.0.5815 -.001665 .0.3325 -.001618 ,N,
:::u:s:1. N.
N.
N,
N,
WHIG
Each column ot torces in the above matrix 4) Turner, l1. J., Clough, R. 1,./., Martin, H. C.,
must satisty the equations of statics on the and Topp, L. J., Stiffness and Deflection
element. A variety of checks on the accuracy AnalysiS of Complex Structures, Journal or the
of such a result are thus available. Aeronautical Structures, 23, September, 1956.
The student will find it instructive to
study carefully this last result to observe in 5) Kroll, W." Effect ot Rib Flexl'bili ty on the
what manner a load applied to the wing appears Vibration Modes or a Delta Wing Aircraft, Insti-
on this spar element. For instance, one sees tute or Aeronautical Sciences, ?reprint No. 585,
that of the load p~ ~ 1 applied to the wing, 1956,
.3735 goes onto spar 4-7-9-10. Examination ot
Fig. A23.l reveals that the remainder, 1.00 -
.3735 = .6265 must be taken up by rib 5-6-7 and 6) Wooley, Ruth, Check at Method ror Computing
the torque boxes 3-4-7-6 and 6-7-8-9. Influence Coefficients of Delta and Other Wings,
National Bureau of Standards Report 3655, 1954.
REFERENC!l3 (Avallab1e as ASTIA No. AD40866).
CONVAIR
SUPERSONIC F-I02 INTERCEPTOR
CONVAIR
SUPERSONIC B-58 BOMBER
",,-"
~;::~~~~~,]t4?~f~"'",.
~".~ ~~
COMMERCIAL TRANSPORT FOR THE 1970s? -- This is an artist's conception of a passenger transport of the future -- cruis-
ing at three to five times the speed of sound at 60,000 feet or higher. This is one of hundreds of configurations considered by
Convair Division of General Dynamics Corporation at San Diego, Calif., in its studies of supersonic airliners.
applied. Relative to the length or time in thus repeated loads are important In design of
applying the load to a member, two broad structures.
classitications appear logical, namely,
(1) Static loading and (2) Dynamic loading. Dynamic or Impact Loadlng. A dynamlc or impact
For purposes or explanation and general loading when applied to a member produces
discussion, these two broad classitications appreclable shock or Vibration. To produce
will be further broken down as tollows:- such action, the load must be applied tar more
rapid than in a static loadlng. This rapid
Cont i nuous Loading. application at the load causes the stresses
Gradually or slowly applied In the member to be momentarily greater than
Static Loading loading. it the same magnltude or load was applied
Repeated gradually applied statically, that 1s slowly applied. For
{ example, it a welght ot magnitUde W is
loading.
gradually placed on the end of a cantilever
nnpact or rapidly applied beam, the beam will bend and gradually reach
Dynamic Loading loading. a maximum end deflection. However i: thls
{ Repeated impact loading. same weight of magnitUde W 1s dropped on the
end at the beam from even such a small helght
Statio Continuous Loading. A continuous load as one toot, the maximum end detlection will
ts a load that reaatne on the member ror a long be several times that under the same statlc
period or time. The most common example is the load W. The beam will vibrate and finally
dead weight or the member or the structure 1t- come to rest with the same end detlection as
selt. When an airplane becomes airborne, the under the static load W. In bringing the
weight ot the wing and its contents 1s a con- dynamic load to rest, the beam must absorb
tinuous load on the Wing. A tank subjected to energy equal to the change in potentlal energy
an internal pressure tor a considerable period or the falling load W, and thus dynamic loads
of time is a continuous load. Since a contin- are otten reterred to as energy loads.
uous load is applied tor a long time, 1t 1s a
type ot loading that provides tavorable con- From the basic laws ot PhySlcs, torce
ditions tor creep, a term to be explained later. equalS mass times acceleration (F = Ma) and
Far airplanes, continuous loadings are usually acceleration equals time rate at change at
associated With other loads acting simultan- velocity. Thus it the velocity or a bOdy
eously. such as an airplane or missile Is changed In
magnitUde, or the directlon of the velocity
Static Gradually or Slowly Applied Loads. A of the vehicle 15 changed, the vehicle 1s
static gradually applied load Is one that accelerated which means torces are applied to
slowly buildS up or increases to its maxunum the vehicle. In severe flight alrplane
value without causing appreciable shock or maneuvers like pulling out ot a dive rr-on
vibration.. The time ot loading may be a high speedS or In striking a severe trans-
matter at aeccncs or even hours. The stresses verse air gust when flying at high speed; or
in the member increases as the load is in- in landing the airplane on ground or water,
creased and remains constant when the load the torces acting externally on the airplane
becomes constant. As an example, an airplane are applled rather rapldly and are classed
which 1s climbing with a pressurized tuselage, as dynamlc loads. Chapter A4 discusses the
the internal pressure loading on the fuselage subject ot airplane loads relative to whether
structure 1s gradually increasing as the they can be classed as static or dynamic and
difference in air pressure between the inside how they are treated relative to deslgn at
and outside ot the tuselage gradually increases aircraft structures.
as the airplane climbs to higher altitudes.
B1. 4 The Static Tension Stress-Strain Diagram.
Static Repeated Gradually Applied Loads. It a
gradually applied load Is applied a large num- The intormation tor plotting a tenslon
ber of times to a member It 15 referred to as a stress-strain diagram ot a material is ob-
repeated Load, The load may be ot such nature talned by loading a test specimen In axial
as to repeat a cycle causing the stress in the tension and measuring the load with corres-
member to go to a maximum value and then back ponding elongation over a given length, as
to zero stress, or tram a maximum tenslle the speCimen 1s loaded statically (gradually
stress to a max~ compreSSive stress, etc. applied) tram zero to the falling load. To
The situation envolving repeated loading Is standardize results standard size test
important because it can cause tallure under specimens are specified by the (ASTM) Amerlcan
a stress in a member which would be perfectly Society For Testing Materials. The speed of
sate, it the load was applied only once or a the testing machine cross-head should not
small number or times. Repeated loads usually exceed 1/16 Inch per Inch ot gage length per
cause tailure by fracturing without warning, minute up to the yield point at the material
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES Bl.3
The clad aluminun alloys rave two E values proportional limit stress.
as indicated in the lower diagram of Fig. Bl.l.
The initial modulus is the same as for other TenSile Yield StreSS (Fty)' In referring
aluminum alloys, but holds only up to the pro- to the upper diagram in Fig. 81.1, we find
portional limit stress ot the sott pure that some materials show a erarp break at a
aluminum coating ~terial. Immediately above stress considerably below the ultimate stress
this paint there is a short transition stage and that the material elongates considerably
and the material then exhibits a secondary with little or no increase in load. The
modulus of Elasticity up to the proportional stress at which this takes place is called
limit stress of the stronger core material. the yield point or yield stress. However many
This second modulus is the slope of the second materials and most flight vehicle materials do
straight line in ~he diagram. Both modulus not show this sharp break, but yield more
values are based on a stress using the gross gradually as illustrated in the middle diagram
area wntch includes both c are and covering of Fig. Bl.l, and thus there is no definite
material. yield point as described above. Since
permanent deformations ot any appreciable
Tensile Proportional Limit StresS. (Ftb)' The amount are undesirable in most s~ructures or
proportional 11mit stress is that stress which machines, it is normal practice to adopt an
exists when the stress strain curve departs arbitrary amount of permanent strain that is
tram the initial straight line portion by a considered admissible for deSign purposes.
unit strain of 0.0001. In general the pro- Test authorities have established this value
portional limit stress gives a practical of permanent strain or set as 0.002 and the
dividing line between the elastic and lnelastic stress which e xi st.ec to cause this permanent
range of the ~terial. Th~ modulus of strain when released from the material is
elastiCity 1s considered constant up to the called the yield stress. Fig. Bl.I shows how
B1.4 BEHAVIOR OF MATERIALS AND THEIR PROPE RTIES
it is determined graphically by drawing a line Figs. Bl.3 and B1.4 compare the shapes of
fram the 0.002 point parallel to the straight the tension stress-strain curves for some
portion of the stress-strain curve, and where cammon aircratt materials.
this line intersects the stress-strain curve
represents the yield strength or y1eld stress. Bl.5 The Static Compression Stress-Strain Diagram.
Ult1mate Tensile Stress (Ftu)' The ultimate Because satety and light structural weight
tensile stress is that stress under the maxi- are so tmportant in flight vehicle structural
mum load carried by the test specimen. It design, the engineer must consider the entire
should be realized that the stresses are based stress-strain picture through both the tension
on the original cross-sect1onal area or the and compressive stress range. This is due to
test speclmen without regard to the lateral the fact that buckling, oath primary and local,
contraction of the specimen during the test, is a common type at failure in rll6ht vehicle
thus the actual or true stresses are greater structures and failure may occur under stresses
than those plotted in the conventional stress- in either the elastic or plastic range. In
strain curve. Fig. B1.2 shows the general general the shape or the stress-strain curve
relationship between actual and the apparent as it departs away fram the initial straight
stress as plotted in stress-strain curves. line portion, is ditterent under compressive
The dit!erence is not appreciable until the stresses than when under tensile stresses.
higher regions at the plastic range are Furthermore, the various tlight vehicle
reached.
.,....-------...,
~ ACtualStre~/
materials have dirferent Shapes far the region
of the stress-strain curve adjacent to the
straight portion. Since light structural
weight is so tmpartant, considerable etfort is
made in design to develop high allowable
it~
..
~j
Appar...
compressive stresses, and in many flight
vehicle structural units, these allowable
Stress ulttmate design campressive stresses tall in
~ '---::-,..-,::--,---- Fig. Bl.2
the inelastic or plastic zone.
Fig. Bl.5 shows a comparison or the
stress-strain curves In tension and compres-
sion tor four widely used aluminum alloys.
Below the proportional limit stress the
modulus at elastiCity is the same under both
tension and campresslve stresses. The Yield
g
-" stress in compreSSion Is determined in the
same manner as explained for tension.
Compressive Ulttmate Stress (Fcu)' Under a
static tension stress, the ultimate tensile
stresS of a member made trom a given material
is not influenced appreciably by the shape of
the cross-section or the length of the member,
however under a compreSSive stress the
Ultimate compreSsive strength at a member Is
greatly intluenced by both cross-sectional,
shape and length at the member. Any member,
unress very short and ccecect , tends to
buckle laterally as a whole or to buckle
laterally or cripple locally when under
compreSSive stress. It a member Is qUite
0 short or restrained against lateral buckling,
-"
:;: 180 then tailure for some materials such as stone,
wood and a rew metals will be by detinite
.
53
,;
120 tracture, thus giVing a definite value tor
the ult1mate compreSSive stress. Most air-
craft materials are so-ductile that no fracture
80
•• 1s encountered in compress1on, but the materIal
53" 40
yieldS and swells out so that the increaSing
cross-sectional area tends to carry increaSing
load. It Is therefore practically impOSSible
to .select a value at the UltImate cccpresetve
stress or ductile materials Without having
Fig. Bl.4
-
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES Bl 5
'0
:!.4S-T3 SH[ET 24S-T4 [XT"USIONS
/
.0
TllleM"!!:$$ .. Q,:2~O
.L X-CO"'''~l!~OH\
-IN
.0
TIUCIl.Jrl(SS" 0.250-11'1.
,/ "
, ""oTiNSION 1'[1'1'10"",,\ f, - - - - -
.7',,
-~~
;_1 I
~-T!NSIO!'!:_. "- ......... I~sslo"
-- ,,,
W'COt,lPIl£SSION
'0 I
'0 •• -::s
• I I'-. ,,
·•
'0
/ /
'0 '0 Fig. Bl.6
'0
/ /
/ '°11 The slope of this tangent gives the local rate
0'0 e • • • ot change of stress with strain. The secant
I! ;( 101 IN/IN E X I~ PSI
'0 00
• x
a • • •
10 1 IN/IN E X 10-· PSI
'0 modulus Es Is deter.mlned by drawing a secant
(atr-at ght line) tram the origin to the point
.0,- I COMPllU:U01. in question. This modulus measures the ratio
.'N~
'l'ENSIO'" I I
between stress and actual strain. Curves
,of-....
I·COMPIIl(SSlOIll .0
w-ttNIIOH
r-- t-- ..J po which show how the tangent modulus varies
~~ .. -t("'IO
m wt th stress are referred to as tangent modulus
curves. Fig. 81.5 illustrates such curves
'0f-
I ~~ .0
/
"- tor tour different aluminum alloys. It should
.0
•<
/ ~ '0
;;
• / be noted that the tangent modulus Is the same
as the modulus or elasticity in the elastic
··•• / V
<0 • range and gets smaller in magnitude as the
· I '0 stress gets higher in the plastic r.ange.
30
·V
'0f-
~
'0 · II B1.7 Elastic ~ Jnela.stic Aettoll.
I
'0 It a member is subjected to a certain
'0 L f.'LCt. ...a 7SS-U SHEET
.....D "LATt / stress, the member undergoes a certain strain.
It this strain v.anlshes upon the removal or
TH'CII:7SS 0jIS-O'jS.'l"
'0/ 755-T. !XTI'USIONS the stress, the action is called elastic.
•
a
x 10'
• • •
I"""
'0
E x 10-- PSI
0
0 z
nUCIlNIE:SS.. 3-tN
• • •
Generally_speaking, tor practical purposes,
x_(',.._ • 1I to. INn" e x 10'"
'0
PSI " a material Is considered elastic under stresses
Up to the proportional l1mi t stress as
Fig. Bl. IS preViously detined. Fig. 81.7 illustrates
elastic action. However, it when the stress
is removed, a residual strain remains, the
sane arbitrary measure or crlteron. For action Is generally reteITed to as inelastic
wrought materials It is normally assumed that or plastic. Fig. BI.S illustrates inelastic
Fcu equals Ftu. For brittle materialS, that action.
are relatively weak In tension, an Fcu higher
t::- i~
than Ftu can be obtained by compressIve tests Elastic Action
or short compact speCimens and this ulttrnate
compressive stress is generally reterred to
as the block compressive stress. ••
•" Permanent
S1.6 Tangent Modulus. Secant MOdulus. In Q)
straiD
Modern structural theory tor calculating Sir.... "t j- straiD
the compressive strength ot structural members
Fig. Bl. 7 Fig. Bl.8
as covered in detaIl in other chapters of this
book, makes use of two additional terms or
values which measure the stiffneSS ot a member Bl.6 Ductility.
when the compressive stresses in the member
tall in the inelastic range. These terms are The term ductility tram an engineering
tangent modulus or elasticity (Et) and secant standpoint indicates a large capacity or a
modulus ot elastic! ty (Es). These two modi- material for inelastic (plastic) deformation
fications of the modulus ot elastiCity (E) in tension or shear without rupture, as
apply in the plastic range and are illustrated contrasted with the term brittleness which
in Fig. 81.6. The tangent modulUS Et Is indicates little capacity for plastiC de-
determined by drawing a tangent to the stress- formation without failure. From. a phys t ca.L
strain diagram at the pOint under consideration. standpOint, ductility is a term which measures
Bl.6 BEHAVIOR OF MATERIALS AND THEIR PROPERTIES
~I
elongation of a tensile test specimen arter
failure, tor a specified gage length, and 1s ••• fp
A
;;"
~'"
usually an accurate enough value to compare
materials. ~
~
:>
La - L ,) dE
Percent elongation = ( L Q 100 = measure of
I
ductlli tv.
E C H I K
where LQ =original gage length and La = gage Unit Strain (e)
length after fracture. In referring to Fig. Bl.9
ductility in terms of percent alangation, it
is Unportant that the gage length be stated, high and the strain at this stress is high,
since the percent elongation will vary Nith or from equation (2), when the proportional
gage length, because a large part of the total limit stress is high and modulus of elastici~J
strain occurs in the necked down portion of the is low.
gage length just before fracture.
In Fig. 81.9 if the stress is released
SI.9 Capaci.ty to Absorb Energy. Resilience. Toughness. trom point D in the plastic range, the recovery
diagram will be approximately a straight line
Resilience. The capacity of a material to D& parallel to AO, and the area CDE represents
absorb energy in the elastic range is referred the energy released, and often referred to as
to as its resilience. For measure of hyper-elastic resilience.
resilience we have the te~ modulus of
reSilience, which is defined as the maximum Toughness. Tougr~ess of a material can be
amount of energy per unit volume which can be defined as its ability to absorb energy when
stored in the material by stressing it and stressed in the plastic range. Sir-ce the
then c~pletely recovered when the stress is term energy is involved, another definition
removed. The maximum stress for elastic would be the capacity of a material for
action for computing the modulus of resilience reSisting fracture under a dynamic load.
is usually taken as the proportional limit Toughness is usually measured by the term
stress. Therefore for a unit volume of mOdulus of Toughness which is the amount of
material (1 cu. in.) the work done in stressing strain energy absorbed per unit volume when
a ~terial up to its proportional limit stress stressed to the ultimate strength value.
would equal the average stress f' p / 2 times the
elongation (eo) in one inch. If we let U In Fig. 81.9, let t equal the ~verage
represent mOdUlus at reSilience, then stress over the unit strain distance d~ from
F to G. Then work done per unit volume in
stressing F to G 1s fd~ which is represented
u • Gp) Ep ---------- ----- (1) by the area FGRI. The total work done in
stressing to the ultimate stress f u would
then equal I ok fde, which 1s the area under
But e p = tplE, hence
the entire stress-strain curve up to the
U =(t;) (t,:) =!p"/2E (2)
Ultimate stress point, or the area 0 A J K 0
in Fig. 81.9 and the units are in. lb. per
cu. inch. Strictly speaklng it should not
include the elastic resilience or the ener~J
absorbed in the elastic ~~ge, but since this
Under a condition of axial loading, the modulus area is small compared to the area under the
at resilience can be found as the area under curve in the plastic range it 1s usually
the stress-strain curve up to the proportional included in tougr~ess ~easurements.
limit streSS. Thus in Fig. 81.9, the area OAB
represents the energy absorbed in stressing It should be noted trat the capacity at
the material tram zero to the proportional a ~ember for resisting an axially aDplied
limit stress. dynamiC load is increased by increasing the
length at a member, because the volume is
High reSilience is desired in members increased directly with length. However, the
SUbjected to shOCk, such as springs. From ultimate strength remair.s the same since it
equation (1), a high value ot resilience is is a function of cross-sectional area and not
obtained when the proportional 11mit stress is of volume of the material.
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES 81.7
When a material is stressed, it will (2) Locate any other point on the typical
derorm in the direction at the stress and also curve such as point C, and draw line
at right angles to it. For axial loading and rr-ca 0 through C.
for stress below the proportional ltmlt stress,
the ratio at the unit strains at right angles (3) Locate point D on line 0 C by the follow-
to the stress, to the unit strain in the ing rat.t o r-,
direction of the stress is called poisson's
ratio. It is determined by direct measurement OD=~XOC
in a tensile or compressive test at a speCimen,
and is approximately equal to 0.3 tor steel (4) Repeat step 3 to obtain a number or
and 0.33 tor non-terrous materials. In many pOints as shawn by dots on Fig. BI.II,
structures there are members which are Subjected and draw smooth curve through these points
to stresses in more than one direction, say to obtain desireu stress-strain curve.
along all three coordinate axes. Poisson's
ratio is used to determine the resultant stress
and detormation in the various directiOns.
solution ot such problems requires information on the stress-strain curve. Table 81.1 gives
given by the compressive stress-strain curve. the values of F o . ,., to.ee. n , etc .• for many
Since flight vehicles make use of many differ- flight vehicle materials. Notice that the
ent materials, and each material usually has shape parameter varies widely tor materials,
ma.ny different states of manufacture which being as low as 4 and as high as 90.
give different mechanical properties, the
question at time required to obtain certain HI. 13 Influence of Temperature on Mater1al Properties.
design intormation from stress-strain curves
becomes important. For example, in the Before the advent of ~~e supersonic air-
aluminum alloys alone there are about 100 plane or the long range miSSile. the aero-
different alloys, and when elevated temperatures nautical structures engineer could design the
at various time exposures are added, the number airframe ot aircraft using the normal static
at stress-strain curves required is turther mechanical propertles ot materials, sInce the
greatly increased. temperature rise encountered by such aircratt
had practically no effect on the material
Fortunately, this time consuming work was strength properties. The development of the
greatly lessened when Ramsberg and Osgood turbine Jet and rocket Jet power plants pro-
(Ref. 1) proposed an equation to describe the vided the means ot opening up the whole new
stress-strain curve in the yield range. Their fIeld of supersonic and space flight. The
proposed equation specifies the stress-strain flight environmental conditions were now
curve by the use of three parameters, the greatly expanded. the major change being that
aonnua at elasticity E, the secant yield aerodynamic heating caused by high speeds in
stress F O. '1" which is taken as the Una at the atmosphere caused surface temperatures
slope 0.7E drawn trom origin (see Fig. Bl.12), on the airframe which would greatly etfect
and a parameter n which describes the Shape the normal static material strength properties
of the stress-atrain curve in the yield region. and thus temperature and time became important
In order to evaluate the term. n, another stress in the structural deSign ot certaIn types of
Yo •• is needed. which is the intersection of tlight vehicles.
th9 curve by a line or slope of O.SSE through
tho origin (s•• Fig. Bl.12). Bl. i4 Creep of Materials.
E/l Stress-Strain
a material because If the temperature is high
enough the material will melt or tlow and thus
have no load carrying capec t tv as a structural
member. When a stressed member Is subjected
to temperature, it undergoes a change at Shape
c.....
in addition to that of the well known thermal
expanSion. The tenn creep is used to describe
thiS general intluence at temperature and time
on a stressed material. Creep is detined in
general as the progressive, relatively slow
change in shape under stress when subjected
to an elevated temperature. A simple illustra-
tion ot creep Is a person standing on a
bituminous road surface on a very hot summer
day. The longer he stands on the same spot
_ ¥o-cJ .....Jl... FIg. 81.12 the deeper the shoe soles settle into the road
).......... ~ The Ramaberg and Osgood proposed three surface, whereas on a cold winter day the same
parameter representation of stress-strain time ot standing on one spot would produce no
relations in the inelastic range ta r-, noticeable penetration of the road surface.
.. Ee t 3(
Fo.'I'=FO'.~~Fo.?
t )n. --- - - - - - (3)
High temperature. whe..l used in reference
to creep, has ditferent temperature values for
different materials for the same amount of
The sQuation for n Is, creep. For example, mercury, which melts at
-38OF. may creep a certain amount at -75 0F,
n = 1 =. log. (17/7)/log.(F •••IF •••• ) (4) whereas tungsten, whf en me 1 ts at 6170OF, :nay
not creep as much at 200QoF as the mercury
Fig. Bl.13 Is a plot of .quation (4). The uncer -75OF. All materials cr••p under-
quantities EeIF-o . ? are non-dimensional and may conditions Of temperature, stress and time ot
be used in determining the non-dtnenslonal stress application. The simplest manner in
curves of Fig. Bl.14. E. n, and F 0.'1' must be which to obtain the etrects at creep Is to
mown to use these curves 'n obtaining values study its ettect on the static stress-strain
... ~t..-
diagram tor the material •
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES Bl.9
Fig. B1. 13
50
.0 ""'-
30
r-,
25
-,
20
-,
~
15
-, toe. (17fT)
11-1+
~. (FO.7"'0'.86)
10
••
1
• <,
•
5
I'....
<,
• r--,
3
t"-
•
1. 5
1. 0
1.01 1.02 1.03 1.D41.051.08 1.08 1.1 1.15 1.2 1.3 L'" 1.51.8 L8 %.0
"0. T/FO• as
Fig. Bl.14
1.'
,
1.' ,I
»<
, ~ /"
f:?
1.0 "
0.' h -a
~~
f
FQ.";
0.'
~~ ,
0 .•
1/ 1
0.'
o
1/0.' 0.' o•• 0.' 1.0 1.' 1•• l.S '.0
C / c'"
-::..- t' ~,
,- .,:
Table 81. 1 Values of Ftu> 1""",. Eo:. 1"0.7. F 0.85, 'to or v arlOUS !Iota
. rer ia1s
,
Unci ., Roo m &. El evat odTempera ures (F rom R .. t 6)
MATERIAL
Temp.
E"". Temp. I
,
e, I"tu".
-. 'r I" :Y:' ., E, 1"0.7 I 1"0.as ,
i ,,i '"
OF ~. n
H,. I koSi 106pl!l i I kosi I kosi . I
STAINLESS STEEL
AISI 301 114 Hard Sheet
Tr:uuJ'ferse Compresllion
1/, I, RT zs 125 ao 27. a 73 I, sa 5. ,
"
45.5
83.5
J7
as
I
6.0
4.4
4.7
1/2 '00 no 52.8 ac. 1 44 31 3. s
1/, 1000 ae 45.2 15.6 40 30.5 4.3
AlSI 301 3/4 Hard Sheet 1/, RT 12 <7, 180 27.0 163.5 151. 5 13.2
Trana'l'eMie Compresaton 1/,
1/'
<00
600 I '"
138 '"
138
24.1
22.4
153 I 142.5
140
13.2
I
Lofli1,tu4lna1 Compresllion
11'
,/2
1000
RT 12
11J
175
ua
78
16.9
26.0
'"to
127 I 1Jl
61.5
11.2
19.2
7.5
I
1/,
1/'
400
60Q '"
"a
71
70.3
23.3
21. 5
65
65.5 I se
56.5
I '.6
e. a
1/, 1000 na 59.3 18.2 ss 46 5. S
AISJ 301 Full &rei Sbeet 11' RT 6 165 179 27.0 <83 172 <8
Tranuerse COl'nprepwll. 1/,
1/,
.00
600
168
15'
"a
"a
25.1
23. a
17.
<7,
18. I ie
18
Lou¢tudln.a1 Comprnllwn
11'
1/,
1000
RT a
131
185 Ii(
130
65.
21.6
26.0
141. 5
77.5
'"ea
135.5 21.5
5.2
1/, 400 "a 60. a 24.2 74 59.5 5
11'
1/'
'00
1000 '"
131
79.9
66.3
22.9
20.6
74
ea
sa
42.5 , •••s
3.
l'l"'" PH B&r II. Fol"l1nP 1/, RT e 165 27.5 168 <80 J4
1/, .00
700
180,
'"as
146 '"
105.5
25.3 "7 129
er
I <8
n
17.7 PH (THtOSO) Sheet, Strtp II. Plate.
1/'
1/'
1/,
1000
RT 180
62.6
23.1
21.2
29.0
'"60 sa
145 II 7. 1
7.4
t '" .010 tD .1251D.. 1/, .00
'"H.ea
18'
H. 27.11 '"n7
'146 s. a
28.0
65
171
74
134
7.2
22.5
t ... 020to.187ir1.
PHU-7Mc (RH950) Sheet &: Strip,
t '" . 020 tD .187 in.
11' RT II 4 aae '00 28.0 are 189 '-'
i
L.OW CARBON &: AL.LOY STEELS ,
AJSI 1023 1I 1025 Tube. Sheet" Bar,
Cold Flniab.ed RT aa ee ae 29.0 32. '7 31. 5 J4
AJSl4130 NQrmaJUed, t ~ .188 In. 1/2 RT J3 90 70 29.0 61. 5 53 e.•
125
I,
61. s
48.2
30.8
113
27.3
23. a
40.8
29.0
ss
40
aa
III
48
32.5
aa
102
I
7.3
5.'
4.7
10.9 ,
1/'
11'
500
.60
113
SO I 98.3
58.9
27.3
23.2
se
66.5
SO
61. 5
10.9
I
AISI 4130, 4140. 4340 Heat Treated
1/,
1/'
1000
RT 18.5
34
150
49. '7
14'
20.8
29.0
45.5
145
41
140
"ass.a I
11' '00 135 27.3 rae as I
1/' 550 105 '"88.5 23.2 58 '"
63.5 18. 5
AISI 4130. 4140, 4340 Heat Truted.
1/'
11'
11'
1/2
1/,
1000
RT
500
650
100'
,
I
15
76
180
'"
'"sa
63.8
17'
15'
109.3
20.6
29.0
27.3
23.2
20.6
"
17'
15'
109.4
57
17'
153
10'
.
10.9
eo
aa
s.e
AIB14130. 4140. 4340 Heat Treated 11'
1/,
RT
500 I 13.5 200
180
"
19' 29.0
27.3
"
19'
58
19' JO
17' 172..5
1/, '50 140 23.2
16' .46
as II
A Y
11' 1000 r 104 '"87.1 , 20.6
121.5
57
117
83 19
I
A-2M (AMS m!5Aj Sheet, ?Iate
I,
."..., 11'
1/2
RT
600
1000
15 140
'"sa
95
88.4
29.0
24.4
93
57
er
57
ai
14
13.5
11'
1/, I 1400
115 81. '7
50.3
19.8
14.2 50
ts
47
12• .5
15. 3
K-MONEL Sbeet, Age Hardened
MONEL Sheet:. Cold Rolled '" Annealed
1/,
1/2 I RT
RT
15
35
12'
70
JO
ae
Z6.0
26.0
ee
20
ea
17
13. 5
'.4
.
tNCONEL-X 1/, RT 20 155 10' 31.0 10. 100 23.5
1/2 .00 162 95.6 28.9 94 89 17
11' '00 141 90. a 28.4 a8.6 18. 5
1/, 1<:00 10. aa 23.2 78.8 ar
"
I
I, I
i I,
I
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES m. n
1'lOble al 1 Valu
" .
of Fw I''Y' Ee, FO 7 FO 85
()fL. ~L
n, [or V;u1.01lB MaterwlI Under Roo m &: Ele v aled l'om", r:;ltl1rell (FrQ m Ret 6j (C0 nUnu ed)
MATERIAL
Temp.
""''"'.. Temp.
'F ,
o. F"" FCY'
"
I 106pSl -. .
Fo• 7. F O•8S
I ,
ALUMINUM ALLOYS "" "" Ul'"
'"
2014~T8 ~Naiona 2 RT 7 60 ea 10.7 53 50.3 18.5
t ::!' O. 499 in. 2 300 42.5 10.2
2
2
420
600
"28to 21 3.2
4.1.5
20.5 "
19.5
,. S
"25s.,
8.0 7.' S••
1/2 300 Sl 4:1.5 10.2 44.0 42.5 25
1/2 420 31 28 9.2 ae 25.2 29
2014-1'6 Forpngs 2 RT 7 ea sa 10.7 52. :1 SO 20
t ::!' 4 In. 2
2
2
1/2
1/2
300
'50
600
300
..0
sa
29
10
53
32
41
22
43
7••
25.5
10.2
9.2
7. ,
10.2
9.2
40.5
,
21.5
42.5
280
.. 38.5
20
3.0
"21.
19
12. e
3.2
15.8
"6
2024-1'3 Sheet & Plate, 2 RT 12 65.'~Q 10.7 .a-c 36.
.
*"
11.' ,
Heat Treated. t :: . 250 In.. -: 2 300. 10.3 35.7 • 33.5 15 ...
I
I 2024-1'4 Sheet &. Plate.
2
2
2 RT
SOO
700
12
"
28
36
7.•
8. ,
8. ,
1.7
2U
6.2
31!.7 •
22.8
s.s
34. s.,
10.9 j
8.2 '
ia e
Heat Treated, t ~ O. SO In. 2 300 34
2
2
'00
700
24
7
10.3
8.' """
22
30.5
21
14. e
10.2
2024-1'3 Clad Sheet &. Plate,
Heat Treated, t = .020 to .062 In.
2
2
RT
300
12 60 27
34
6.'
10.7
,0. 3
60
35.7·-
25
"
25
30.3
18.5
12
11
... .. .
2 SOO 24.5 B., 21.7 20 7.9
2024-1'6 Clad Sheet &: Plate,
Heat Treate<l, t ? O. 063 In.
2
2
2
700
RT
300
8 ea
B.S 6. ,
10.7
10.3
'.8
44.3
•••
40.7
l}5
11
2 SOO 22 8.' 31.5 28 8.3
2 700 6 8.' 7.0 6.0 6.8
2024-1'8 Clad Sheet &. Plate, 2 RT 8 60 47 10.7 47 43 10.6
Heat Treated, t < O. 063 In. 2 300 43.2 10.3 42.3 2L7 10.8
2 SOO 31 8.' 29.s 28 7.8
2 700 6 6.' '.00 '.0 U
2024-1'81 Clad Sheet, Heat Treated, 2 RT s ee
tcO.064 In. 2 300 " "50.s 10.7
10.3 51. :I
51.8
46.5
11.2
io
2
a
6061-1'6 Sheet, Heat Treated &: Aced, 1/2 RT 42 ae
t<O.25In. 1/2 300 " "
29.5
10.t 34 31
1/2 '50 ,11.5
3••
B•• "
19.3
28
17.7
28
10.9
7075-1'6 Bare Sheet &. Plate,
t~0.501n.
1/2
2
a
RT
BOO
300
7 76 .....
54
7.5
.. ,
7.0
10.5
6.6
to. .
55.8
6.2
B3
52.5
15. :I
9.2
15.8
7075-~ Extrusions,
t~0.251n.
a
2
1/2
2
2
...
425
'00
RT
300
7 73
25.5
8
30
70
54
8.1
'.2
.. ,
B.l
10.5
25.4-
7.2
34 .s
72
22••
'.2
32. e
"54.'
12.1
3.7
re
12.8
58.5 11'
2 '50 22•• 7.8 21.3 IL. 7.2
2 .00 8 '.3
,..s
6•• '.3 3.2
7075-1'8 Die ForginjJ:',
t ~2 ill.
1/2
2
2
2
2
"0
RT
, .
300
600
7 71
25
"
47.S
18.5
7.0
7.B
10.5
8••
7.B
'.3
53
47.8
17.3
'.0
.
"
.Ll
te
3.7
B.8
15.2
1e.e
12
19
1/2 22
7Q75- 1'6 HaDd Forginp.
Area :s 16 sq. In.
2
2
RT
"0
300 • 72 B3
51.6
7.8
10.5
••• "
63.8
~1.2
22
61.5
60
10.9
25
21.5
2 '50 7.6 20.3 13.7
2'"
"'.0
7075- re Clad Sheet &. Plate,
t ~O. 50 In.
2
1/2
2
2
a
RT
600
.5O
300
'50
B 70 .7.6
24
60
:l0.5
'.3
7••
10.5
•••
7.8
....
6.0
se.e
54
19.7
25.3
er.e
51.7
17.5
•••
19.5
19.5
20
<6
2 600 7.7 '.3 7.7 S.s 3.6
7079_ TlS HaPd Fof"iinP.
</2
</2 RT
"0
, 67
23
59
7••
to. 5
27.2
59.5
.'"
25.3
57.5
12.4
,.
23
t ~ 6.0 In. 1/2
1/2
300
'600
W ,
67
21
•••
7.8
4S.5
20 12
</2 7.0 '.3 e., 3.s '.0
MACNES M .... LLO
AZ61A Extrusions,
t~0.249il1.
RT • ae 14 6.3 12.9 12. J is
HK31A-0 Sheet </2 RT , 12 30 12 6. S 10 8. , 6
a.s e.s •.s
I
t • 0.016 to O. 250 in. 1/2 300 20 11.1 6.16
</2 .00 is '.3 4.94 7.s '.6 ••a
</2 '00 10 '.9 3.77 3.3 I .• 2.2
KK31A-1i24 Sheet, 1/2 RT
• 34 19 6.s 17.3 14.6 6.2
t-0.250ill.
THMn ~eaJ.~
</2
1/2
</2
300
.00
.00 I
22
17
11
17.7
14. a
7.a
L2
•••
3.8
IL 6
13. 1
'.7
12.6
10.5
S.2
I
I
e.,
,
'.9
..
Plate &. Strip
n-ao-w Annealed
&. Sheet, t ~.
Sheet,
Bar
167 ill.
1000
i/a
1/2
</2
RT
RT
.00
600
10
10
120
130
10.
sa
~
I '"
123
se
... s
15.5
16.0
14.1
t3.0
119.5
127
67
85.5
I 102
124.5
83
aa
I
."
13.7
22·
I
</2
</2
800
1000
67
70 t
". ,
SO .s
11.8
'.7
80.5
er
I
I
-n
59.5
21. 5
"
0,......
~
.....---
Bl.12 BEHAVIOR OF MATERIALS AND THEIR PROPERTIES
Fig. Bl.lS (curves A and B) show the B1.15. The General Creep Pattern.
stress-strain curve for a material. Curve (A)
is tor a low elevated temperature condition A typical manner of plotting creep-rupture
and curve (B) that ot a high elevated test data is illustrated in Fig. Bl.1S. For
temperature condition. The results were ob- metals tested at high value of stress or
tained by a normal testing ~chine procedure temperature, three stages in the creep-time
requiring a short time test period, hence the relation can be observed as shown in Fig. Bl.lS.
results can be considered as independent ot The initial stage, otten called the stage ot
time. primary creep, includes the elastic deformation
--
and that region where the rate ot creep de-
~;":'t p',,- pc ! tormation decreases rather rapidly with time,
~!:'!-:;,
I:';",:
'.rr
"r:~":::'
'';::~~~,
::p- ~r-
.
i
I'
I
' J..- i
;
...
,~, ...~l..._-L_
.... .... __..... ---
...f --
HMr'-....
--
which no doubt Indicates an influence ot strain
hardening. The second stage, otten referred
to as the secondary creep stage, represents a
//c ., .-
,,~, 'c_
-'0'
, j ;
...««:
0
:~;i .- -7' f
:.::.;,z-_~_ ; stage where the rate ot strain has decreased
W,.~~ ~
~:
-. , .-- I. DIQlIt1'-'tWE TIJeOlIl
to a constant value (except tor high stress)
. i--·- tor a conSiderable time period, and this stage
~~~~i~ """~ ,<" ,. , H' ! represents the periOd ot minimum creep rate.
.
~~.
I
..
,
.~ ,
~~ ~.?
,."., ..
t"
...• ~,
e"
~;"
, / r-
-. ,- I
,
' .
....
. , reduction in cross-sectional area leads to a
higher stress, a greater creep rate and tinally
rupture •
f.'.! L". .' II I' :" FO-" !&~
~-::~-. . , Transition Point. The inflection ~oint between
.....
"
r.S
i;.'!c ,•.
--
..•. r..
"y ',"
,i'.~, ?:".i' , - the constant creep rate ot the second stage and
J
the 1ncreasing rate ot the third stage is
""""'- reterred to as the transition paint. Failure
Fig. Bl. 15 (Ref. 2) - Effect of temperature and time on the generally occurs in a relatively short time
strength cba.raetertst1cs of metals. attar the tranSition point. Transition points
may not occur at very low stresses and ~y
The tigure shows that the higher also not be definable at very high stresses.
temperature (curve B) reduces the ultimate Minimum creep rate is that indicated in the
strength, yield strength 'and modulus ot second stage, where the creep rate is practi-
elasticity or the material as compared to cally constant.
curve (A) which is a test at a lower temper-
ature.
Initial j Second Stage Third
EFFEX:T OF TIllE (Approx. Constant) Stage
stage :
Rupture
I Point
It temperature and stress are ot such
combination as to prOduce appreciable creep, ,,
I
I
1
i
o
"'f---+-+--+-+-~~+--+-
~'C-I- , - - - , - - - - , - - , - , - - - - ,
I
Fig. Bl.22 - TensUe rupture stress of 2024-T3 aluminum Fig. 1.23 (Ref. 3) - Effect of Temperature on tntimate
alloy for temperature rates from O. ~F to l00 0F per second strength (Ftu) of 20l4-T 6 Aluminum Alloy.
and ultimate tensile stress of stress·strain tests for 1/2-
hour exposure.
vehicles. Fatigue failure is failure due to
strength respectively of aluminum alloy as being stressed a number of times. For example,
compared to values when loaded after the a beam may be designed to sately and efficient-
material has been exposed 1/2 hour at constant ly carry a deSign static load and it will carry
temperature. this static load indefinitely without failure.
However, if this load is repeated a large
B1. 19 General Effect of Low Temperatures Upon Matertal enough number of times, it will fail under thiS
Properties. static deSign load. The higher the beam stress
under ~he static deSign lead, the less the
The development or the missile and the number of repeated loadings to cause failure.
space vehicle brought another factor into the
ever increasing number ot environmental con- To date no adequate theory has been
ditions that etfect structural design, namely, developed to clearly explain the fatigue
extremely low tetnperatures. For example, in failure of materials. Fatigue failure appears
space the shady side of the night vehicle is to begin with a crack starting at a point of
subjected to very low temperatures. Missiles weakness in the material and progreSSing along
carry rueta and oxidizers such as liqUid crystal boundaries. A microscopiC examination
hydrogen and oxygen which boil at -423 and of metals indicates there are many small
_297oF respectively. In general, low crackS scattered throughout a material. Under
temperatures increase the strength and stiff- the action ot repeated stress these small
ness at materials. This etfect tends to cracks open and close during the stress cycle.
decrease the ductility or the material or, in The cracks cause higher stress to exist at
other WOrdS, produce brittleness, a property the base ot the crack as compared to the
that is not desirable in structures because stress it there were no crack. Under this
of the possibility ot a catastrophiC failure. repeated concentration of stress, the cracks
In general, the hexagonal closely packed will gradually extend across the section or
crystalline structures are best suIted for the member and finally causing complete
giVing the best service under low temperatures. failure of the member.
The most important at such materials are
aluminum, titanium, and nickel-base alloys. Fatigue testing consists of 3 types:-
Fig. 81.23 shows the effect of both elevated (1) the testing of material crystals, (2) the
and low temperatures 0n the ultimate tensile testing of small structural test specimens,
strength ot 2014-T6 aluminum alloy under and (3) the testing or complete compOSite
various exposure times. structures. A tremendous amount or test
information is available for the second type
B1.20 fatigue of Materials. of tes~1ng. More and ~ore attention is be1~~
given to the third type of testing. For
DeSigning structures to provide satety example, a complete airplane wing or fuselage
against what is called fatigue failure is one is often subjected to elaborate fatigue
of the most important and dlfttcult problema testing in order to insure the safe deSign
rac ing the structural designer at flight lite of the airplane.
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES B1. 15
!
'Jl 10 X 10"
-, of Mater1als.
8 x 10 •
'"r--... at impact tests upon the stress-strain curve
as compared to the static stress-strain
diagram (Re t . 5).
•
7 x 10 10 •
! I I j
10" 10"
Number of Cycles for Failure, logH-
10 'r looj : :
I '
i
, ,
, ,i
I :r
"'
•
, II i
'
04 f.'jsec
i
A '.±t.'
50
Fig. B1.2. i~i
~' I~ 1 : i
, ,
'I'at:e B1.2 snows additional impact testing
I I I , ,
I I
so I I Slotic v '" /5C f:'sC'cl
results as compared to the static test results.
~40 Ih
c- , , , ]V '" co rl/s9Ci I , i Table B1. 2
~30
,
I , I 'I
, , , I , Comparison of Strengths, Ductility and
I
,i
,
§ , , I , I
I Energy Absorption Under Impact
, ,
·••
'.20
~ 101
I
I
, I
,
,
, ,
,
I
,
i
, I
I I
I and Static Loads. xxx
,
i!OO Impact Loading" by D. S. Clark & G. Datwyler,
'/ ~ 80 frl$f1C I
Proe. A.S. T. M. 1938. Vol. 38.
--<.C:-t-- : ,
,,
s" I
I
I I
I
I
I !:;/~1I,1 r\ "-.'
I I I i \ , ,I I
REFERENCES
~I I'/ ~
.,-, !
I
I
I
,
I I I I ,
JSO fI/s.cr"-:- I
!
I :
,,
i
Ret. 1. Ramberg & Osgood. Description of
Stres~-Strain Curves of 3 Parameters.
NACA, Tech. Note 902.
"
~
,
I
I I ,
,
,
:
I
I
I
I Ret. 2. T1me and Temperature Greml1ns of
, Dest·.llctlon. By L. A. 'rerxorvicn,
'" i I
,
, I
Correll Aero. Lab. Research Trends.
I I I i I I I I I I i Se"'; • 1956.
0 ce 04 06 co .,0 /2 .14
(.Inil ef0l'l90hen. '''',lln. Ret. 3. Military Handbook (~IL-HDBK-5)
Aug. 1962.
Ref. 4. NACA Technical Note 3462.
Fig. Bl.27 - Stress-strain curves, SAE 6140. drawn lO20oP. Ret. 5. NACA Technical Note 868.
Ref. 6. From Structures Manual, Convair
Astronautics.
CHAPTER B2
MECHANICAL AND PHYSICAL PROPERTIES OF METALUC
MATERIALS FOR FUGHT VEHICLE STRUCTURES
General Explanation. It would require several hundred pages to list the properties or the many
materials used in flight vehicle structur~l design. The metallic materials presented in this
chapter are those most widely used and should be SUfficient for the use of the student in his
structural analysis and design problems. All Tables and Charts in this chapter are taken tram
the government publication '~111tary Handbook, MIL-HDBK-5, August, 1962. Metallic Materials and
Elements for Flight Vehicle Structures~. This publication 1s for sale by the Supt. or Public
Doc~~entsJ Washington 25, D.C. The properties given in the various tables are for a static
loading condition under room temperature. The effect of temperature upon the machan1cal
properties 1s given in the various graphs.
I
Alloy .... AISI 4130. 8630, I AISI 4130. 4140. 4340. i
4140 AISI
and 8735 8630,8735, and 8740 4340 4340
8740
I I
Form ....... ........... . Sheet, strip, All wrought forms
plate. tubing
I
Condition .. .. . N Heat treated (quenched and tempered) to
I obtain F I .. indicated
I
Basis .. ....... . ..... . .... . . ... ... . ..... .. . .. . .. ... ........ . .. .. ... ........I (0)
Mechanical properties:
F h • ksi.. ... 96 90 125 150 180 200 260
F ,,,' ksi. .. .......... 75 70 103 132 I, 163 176 217
F.", ksi .. ... . ..
' 75 70 113 145 179 198 242
F ,.., kai . . . . ...... ... I 55 55 82 95 109 119 149
F h ... ksi:
(eID~1.5) . .. . I .. ...... 194 219 250 272 347
(ejD=2.0) ....
, 140 140 251 287 326 355 440
.!
Ph". ksi
(e(D=1.5) . .. . i . ... 151 189 230 255 312
(eID~2.0) . .. ..
,
... I, 180 218 I 256 280 346
.,,I
, i
" ,
e, percent. ~ L 10
See table See table 2.3.1.l(c)
2.3.1.l(b) •T 3
E, 103 psi. .
i---'---------'------
i 29.0
E., 103 psi. 29.0
,
G. 10e psi ..... -! 11.0
--------1-------------------
Physical properties: 1
82.1
: :t ..
-"'.. - ~ '....J
B2.2 MECHANICAL AND PHYSICAL PROPERTIES OF METALLIC MATERIALS FOR FLIGHT VEHICLE STRUCTURES
A:SI ~~LOY S~2SLS (Cc~t.)
'.,~~~=._._._.
,~140
-
e
~
E
,
Slrenq1h Of temperature
E~posure up to
' --'
1/2 hr.
o
~120
E
o
o
cr ,
'"
-e Fig. 82.2. Effect of temperature on the compressive
•
o , E. yield strength (Fey) of heat-treated AISI alloy steels.
'" 80 roo
•I ..
e
oe 60
,
FiT Ft. I .
~
0
.'
..
eo
,
!
40 ,
,
, ,
,
,
" 600 ecc 1000
Temperal"llI, F
IlOO I-lOO 600
·, .cc
o -400
o
,
~
l
$00 soc 1000 ;lOO 1400 1600
T_ral......, F
i. --=-=-..o.
~
0 . ~==SE=~~~l~~~~~
._;-;;:.:.~=
,J
ec
"
" '0 400 600 ecc '000
r''''p''QfUI'e, F
'200 1400 1600
Table B2. 2 Design Mechanical and Physical Properties of 5Cr- Mo- V Aircraft Steel
I
Mechanical properties: I
F h • ksi ...... ........ 240 260 280
I
FIll. ksi .... ..... ......I 200 220 240
F. II ! ksi ..... · . . . . . . . . I 220 240 280
,
F .... ksi ...... · . . .... 145 155 170
F~ .... ksi:
(e/D-1.5) ..
(eID~2.0) . . . ... .
F ••II , ksi:
(e/D-1.5) ..
... .... .. .... .
400
. ......... ... .
.. . ...... . ... .
435
.. . . ...... ... .
....
..
. . ....
465
. . .....
-,
Fig. 32. 9 Effect 01 temperature on
(e/D-2.0) . . . 315 340 365 the compressive yield strength (Fey)
01 5 Cr-Mo- V aircraft steels.
e, percent:
Bar, in 4D ... .... . 9 8 7
Sheet, in 2 in. Cb} . 5 5 4 !
Sheet. in 1 in ...... 8 7 5 i.
1
E. 10' psi .. .. . ...... 30.0 •.;
E •• 10 8 psi .. ' 30.0
G, 10' psi. . . .... 11.0
Physical properties:
«I.Ibjin. 3 . . . . .
I 0.281. Fig. B2. 10 Effect of temperature on
C, Btul(lb)(F) . 0.11(') (32" Fl. the ultimate shear strength (Fsa) of.
K, Btul[(hr) (It') (F)lltl. 16.6 (400° to 1,100" F). 5 Cr-Mo-V aircraft steels.
a, 10-~ in./in./F ..... 7.1 (80 0 to 800 0 F); 7.4 (80 0 - 1 . 2 0 0 0 F).
j
I
•
.'
-,
Fig. 92. 6 Effect of temperature on
-,
Fig. 92.7 Effect 01 temperature on Fig. H2. 12 EUeet of temperature on
the bearing yield strength lFbry} of 5
the ultimate tensile strength (Ftu) of the tensile yield strength {Fty} of. 5
cr- Mo- V aircraft steels. Cr- Mo- V aircraft steels. ce- Mo- V aircraft steels.
82.4 MECHANICAL AND PHYSICAL PROPERTIES OF METALLIC MATERL\LS FOR FLIGHT VEHICLE STRUCTURES
Alloy. 17-7 PH
Mechanical properties-
Ft•• ksi . 180 210 200 170 i 185
F,•• ksi . 150 190 180 140 ! 150
F ••• ksi . 158 200 189 147 ' 158
F ••, ksi. . 117 136 130 111 120
F~ ••• ksi:
(eID-1.5) . 2'97 346 380 280 305
(e/D~2.0) . 360 420 400 340 370
F~.." ksi:
(eID-1.5) . 225 I 285 270 210 225
(elD~2.0) . 247 I 313 297 231 247
8, percent .... See table 2.7.2.1(b) 6 5
Physical properties:
tal. lb/in.~ . 0.276.
C, Btu/Clh) (F) . 0.11. (~)
K, Btullihr) (ft')(FJ/flj. 9.75 (at 300 0 F).
a, 10-4 in.lin./F. 6.3 (70 0 to 600 e F) for TH 1050.
6.8 (70 0 to 600 0 F) for RH 950 .
• Test d.1reetJon longitudina.l for widths less than 9 tn.: transverse for short transverse (tbtcknesal direction.
widths 9 in. and over. • Vendors guaranteed minimums for F,., Fro, and e.
• Test dlneUon longitudinal; these properties ncr eppucabre to the • Calculated value.
I
j
J
J
•
~
•
Fig. B2. 13. Effect of temperature on the ultimate Fig. B2. 14. Effect of temperature on the tensile yield
tensile strength (Fm) of 17-7 PH (THI050) stainless steel. strength (Ftyl of 17-7 PH (TH1050) stainless sreet.
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES 82.5
roc
''''
.....
.~
! ..
l
Fig. 82. 15. Effect 01 temperature on the compresl!Iive Fig. 82. 18. Effect 01 temperature on the ultimate
yield strength (Fey) 01 17-7 PH (THI050) stainless steel. bearing strength {FbruJ of 17-7 PH (THI050) stainless steel.
IJ
•
.:
Fig. B2. 16. Effect of temperature on the ultimate Fig. B2. 19. EHed of temperature on the bearing yield
shear strength (Fsn) of 17-7 PH (THI050) stainless steel. atrength (i'bry) 0117-7 PH (THI050) stainless steel.
IJ
•,J
1
•
! e &CO 100 IOOCI 1200 l..co I.cG
l
-.'
Fig. 82. 20. EUect of temperature on the tensile and
compressive modulus (E and Ed of 17-7 PH (THI0SO)
stainless steel.
•
82.5 MECHANICAL AND PHYSICAL PROPERTIES OF METALLIC MATERIALS FOR FLIGHT VEIITCLE STRUCTURES
,
Alloy, . .... 17--4 PH
i
l
Form .. Plate Bars and forf;ings(b)
Fig. 82.23 Effect of temperature on
Condition .. ... H 900 H 900 the compressive yield strength (Fey)
of 17-4 PH (H900) stainless steer.
II
e'/D-1.5) ...... 255 255
e'/D=2.0) ...... 280 280
e, percent:
In 2 in............ 10
,
104.D .......... ... 10
)
l
l
E, 10 psi ..
8 29.0
E •• 10' psi . 30.0
G, 10' psi . 11.0 Fig. 82.25 Effect of temperature on
the ultimate bearing strength (FOru)
of 17-4 PH (H900) stainless steel.
PhysioJ properties:
w, lb/in.3 .•.. 0.282.
C, Btu/(1b)(F) 0.11 (32" to 212" F).
K, Btu/[(hr)(ft') (FJlrtl. 10.3 (at 300" F); 11.2 (at 500" F); 13.1 (at 900" F). iI
a, 10- 1 in. lin. IF. 6.0 (70" to 200" F); 6.1 (70 0 to 400 0 F); 6.5 (70 0 to
900 0 F).
,
"
;
• Vendon guaranteed minimums (or F, •• F, .. and (. l
• Test direetion longitud.in.a.l; these properties not apptlceble to the shon transverse (thiclme5S) direction.
iI
"
l
l
Fig. B2.21 Effect of temperature on Fig. 82.22 Effect of temperature on Fig. 82. 27 Effect of temperature on
the ultimate tensile strength (Ftu) of the tensile yield strength (Fty) of the tensile and compressive modulus
17-4 PH (H900) stainless steel. 17-4 PH (WOO) stainless steel. (E and E c > of 17-4 PH (H900)
stainless steel.
A::fALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES 82.7
Table 82.5 Design Mechanical and Physical Properties of AM-350 Stainless Steel
Alloy, AM-350
Condition .. DA seT
Thickness, in ... 0.187 and under 0.187 and under
Basis. s s
Fig. 82.30 Effect of temperature on
the compressive yield strength. (F cyl of
Mechanical properties: AM~350 stainless steel (double-aged).
F,~, ksi . 165 185
F!I/, ksi . 135 150
F CI/' ksi . 142 158
F.~, ksi. . 107 120
F"r~. ksi:
(eiD = 1.5) .
(eiD=2.0) ...
272
330
305
370
j
F"rl/' ksi: I
(e/D=1.5) . 202 225
-s
(eiD=2.0) , 223 247 I
e; percent. 10 10
!
l
E, 10 psi.
6 29.0
Fig. 82.31 Effeet of temperature on
E" 10 5 psi., 30.0 the tensile and compressive modulus
G, 106 psi. 11.0 (E and Ed of AM~350 stainless steel
(double-aged).
Physical properties:
w, lb/in. 3 . 0.282.
C, Btu!(lb) (F) . 0.12 (32 0 to 212 0 F).
K, BtUl[(hrj(W) (FHtj .. 8.4 (at 100 0 F); 11.7 (at 800 0 F). j
a, 1O-~ in.iin./F . 6.3 (70 0 to 212" F); 7.2 (70" to 932 0 F).
!
I
• Test direction longitudinal for widths less than 9 m.; transverse for widths 9 in. and over. •
.'
!
l
j
!
l
Fig. 82.28 Effect of temperature on Fig. 82.29 Effect of temperature on ~ 19. 82.33 Effect of temperature on
the ultimate tensile strength (Ftu) of the tensile yield strength (Fty) of the tensile yield strength (Fty) of
AM-350 stainless steel (double-aged). A11-350 stainless steel (double-aged). AM-350 stainless steel (SeT).
I
!
•~ ""
J •~ '"
J J '"
•.r ;
'"
1 "'
• ~
! l e
l e = sec ecc
"""F
"'"TtrroerQrun, "00 «cc ,"00
Fig. B2.34 Effect of temperature on the tensile and Fig. B2.36 Effect of temperature on the compressive
compressive modulus (E and Ed of AM-350 etatnless yield strength (F cy) of AM-350 stainless steel (SCT).
steel (SeT).
!J j
;
J ;
E J
•
! ~
l l
Fig. B2.35 Effect of temperature on the bearing FIg. B2. 37 Effect of temperature on the ultimate
yield strength (Fbry) of AM·350 stainless steel bearing strength (Fbru) of AM-350 stainless steel
(SCT). (SCT).
Table 82.6 Design Mechanical and Physical Properties of AISI 301 Stainless Steel
Mechanical properties:
• F I ,.. ksi:
L .... ....... .. . 75 125 150 175 185
T ... . . ...... . ... 75 125 150 175 185
Fill' ksi:
L ...... · . ...... .. 30 75 110 135 140
T ... ..... ....... 30 75 110 135 140
F.", ksi:
L ........... ... . 35 43 58 76 85
T .. . . .... . . .... 35 80 118 160 179
F .., kai . .. .. . ..... 40 67.5 80 95 100
F br . , ksi:
('/D-1.5) ...... ........ .. . ....... ......... .. . . ...... . . . .......
(,/D=2.0) . . ' . .. . 150 250 300 350 370
F~.". ksi:
(,/D=1.5) .. . . ... .. . ......... ...... ....... . .. . . . . . .. .. .
(,/D-2.0) . .. . 50 140 200 240 270
e, percent ... ...... . . ( .) (.) ( .) (.) ( .)
E, lOG psi:
L .... ... . . .... 29.0 27.0 26.0 26.0 26.0
T ..... . .. 29.0 28.0 28.0 28.0 28.0
E., lOG psi:
L .... .. . ......... 28.0 26.0 26.0 26.0 26.0
T .. ....... . ..... 28.0 27.0 27.0 27.0 27.0
G. 10' psi. .. ...... ... 12.5 12.0 11.5 11. 0 11.0
Physical properties:
Cot, Ib/in.3.... 0.286.
C. Btu/(lb) (F) . .. . 0.108 (at 32' F) .
K, Btu/[(h.) (ft')(F)/ft). 7.74 (at 32' F).
a, 10- 6 in./in./F ... 9.2 (700 to 2000 F).
82.10 MECHANICAL AND PHYSICAL PROPERTIES OF METALLIC MATERrALS FOR FLIGHT VEHICLE STRUCTURES
,J
Fig. 82.39 Effect of temperature on
ii Fig. 82.48 Effect of temperature on
the ultimate tensile strength (Ftu) of the compressive yield strength (Fey)
AISI 301 (half-hard) stainless steel. .' of AISI 301 (full-hard) stainless steel.
J
l
.
J
J
1
J
J,
Fig. 82.45 Effect of temperature on J
the tensile and compressive modulus
(E and Ed of AISI 301 (half-hard)
stainless steel.
ii
ii •
~
Ii
s• ,J s
,J ,!
Fig. 82.46 Effect of temperature on
the ultimate tensile strength (F tu ) of
Fig. 82.42 Effect of temperature on AISI 301 (full-hard) stainless steel. Fig. 82. 51 Effect of temperature on
the ultimate shear strength (Fsu) of the bearing yield strength (F'bry) of
AISI 301 (half-hard) stainless steel. AISI 301 (full-hard) statntess steel.
ii
•
J
,
J
Fig. 82.43 Effect of temperature on Fig. 82.47 Effect of temperature on Fig. 82. 52 Effect of temperature on
the ultimate bearing. strength (Fbru) the tensile yield strength (Fty) of AISI the tensile and compressive modulus
of AISI 301 (half-hard) stainless steel. 301 (full-hard) stainless steel. (E and Ed of AISI 301 (full-hard)
stainless steel.
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES 82. 11
Table 82.7 Design Mechanical and Physical Properties of 2014 Aluminum Alloy (Sheet and Plate)
Alloy. 2014
Condition. -T6l1
Basis. A B A B A B A B A A A
I - --- - - - --- - - - - - -
Mechanical properties:
Flit, ksi
L. 65 67 68 70 68 70 67 68 65 63 59
T 64 66 67 69 67 69 67 6B 65 63 59
ST . 58 54
FIll, ksi
L. 58 60 60 62 60 62 59 62 59 .'57 55
T. 57 59 59 61 59 61 59 62 59 57 55
ST 53 51
Feu, ksi
D. 60 62 60 62 61 64 61 59 57
T. 61 63 61 63 61 64 61 59 57
ST 59 57
F, ..,ksi. 41 42 41 42 41 41 40 39 37
Fbr.., ksi c
(../])=1.5). 102 105 102 105 101 102 98 89 88
(e/D=2.0) 129 133 129 133 127 129 124 129 112
Fbrr/' f.:lliC
(r/D=1.5) 84 87 84 87 83 87 83 80 ii
(el D =2.0) 96 99 96 99 94 99 94 91 88
e, percent
L.
7' 6
8
8
6
6
..
,6 "3 4
2
3
1.5
ST 1
E, 10· psi. 10.5
E c, 10' psi. 10.7
G, 10. psi -.l.()
Physical properties:
1oI,lb/in. l 0.101
C, Btu!Ob)(FI 0.23 (at 212°F)
K, Btu/[lh,)(ft')(FI!ftj. !)O (at i7°F)
<J:, 10-- in./in./F 12.5 (68° to 212°F)
B2.12 MECHANICAL AND PHYSICAL PROPERTIES OF METALLIC MATERIALS FOR FLIGHT VElllCLE STRUCTURES
Table 82.8 Design Mechanical and Physical Properties of Clad 2014 Aluminum Alloy (Clad Sheet and Plate)
Mechanical prcperdee:
PtMo Jt;.n
L .. ..... .. .. ......... 64 64 65 67 65 67 64 66
T ....... ............. 63 63 64 66 64 66 M 65
ST. . . ... .. . ...... ... .. .. ..
F trh /en
L .. ... .. . ... ...... . ... 56 56 58 60 58 60 57 60
T .. .... . .. ........... 55 55 57 59 57 59 57 60 .'
ST .. .. . .. . .. . ........ .. .. ..
p~ k.ri
L .. ........... ...... . 56 56 58 60 58 60 59 62
T .. .. .. ........ .. . . .. 57 57 59 61 59 61 59 62
ST .... ... .... .. . . ... ..
Ft. .. len .. .. . ...... . ... 39 39 39 40 39 40 39 40
P"".,/eno
(./D-1.5) .. 96 96 98 101 98 101 96 98
(./D-2.0). .... 122 122 124 127 124 127 122 124
p.........
(./D-l.5). ... .. ... 78 78 81 84 81 84 80 84
(./D-2.0). ......... ... 90 90 93 96 93 96 91 96
e, percent
L .... . .. . ............. 7 8 6 6
T ....... .. .......... . 7 8 6 4
E, 10' pn.... .. . ... . ..... .
10.5
Ee, 10' pai. ... ....... .. ....
10.7
G, 10' "... .... . ..... . .. . ...
4.0
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES B2.13
Table B2. 9 Design Mechanical and Phystcal Properties of 2014 Aluminum Alloy (Extrusions)
Alloy. 2014
Heat u-ted
andapd
by"'"
Thickness, in. 1I ....... 0.125-0.499 0.500-0.749 0.750-1.499 1.500--2.999 3.llOO--l.499 ;;::0.750 0.125"'-'99
Basis .. ......... A B A B A B A B A B A A
Mechanical properties:
F tvo kri
L. 60 61 64 68 68 73 68 73 68 73 68 eo
T ................ eo 61 64 67 63 66 61 63 58 61 se se
F'I1I, kn
L. ...... 53 57 58 62 eo 6.5 eo 6.5 eo 65 58 53
T .. ... ... . .. . 53 57 55 59 54 58 52 55 49 53 47 47
PCfI1lui
L. . ........... ...... 55 59 eo 64 62 67 62 67 62 67 .. 53
T. 53 56 58' 62 57 61 57 61 57 61 .. 48
F N. kn.. ......... 35 35 37 39 39 42 39 42 39 42 39 35
FM, hi
(,ID-1.5). 90 92 96 102 88 95 88 95 88 95 .. ..
(,ID-2.0). 114 116 122 129 109 117 109 117 109 117 ..
P/wv.Jcn
(,ID-L5). ....... 74 80 81 87 78 85 78 85 78 85
(,ID-2,O) .. .... . 85 91 93 99 84 91 84 91 84 91 ..
I, percent
L .. ... .. .... 7 7 .. 7 .. 7 .. 7 .. 6 7
Td .. .. ... . ... . 5 5 2 2 .. 1 1
E, 10- ]Mi.. 10.5
E e, 10- pn . . ..... 10.7
G, 10- pn.. 4.0
., '-I ,~
" ..;.. \
-,.,..-, ~
B2.14 MECHANICAL AND PHYSICAL PROPERTIES OF METALLIC MATERIALS FOR FLIGHT VElfiCLE STRUCTURES
Table 82.10 Design Mechanical and Physical Properties of 2014 Aluminum Alloy (Forgings)
Alloy 2014
ilasis.. ........ A A A A A A A A A A
---
Mechanical properties:
Ftv, k&i
L. ..... .... . . .. 55 65 65 65 62 60 65 65 62 60
T . .. ... ..... .... 52 62 63 6. 59 58 63 63 59 58
ST ..... 60 60 56 55 60 60 56 55
FIV,bi.
L. .... .. . ....... . . .. 30 55 55 53 53 "0
0- 55 53 53 52
T. ....... 28 52 55 53 52 50 55 53 52 50
S7' .. . . .. .. .... 55 53 52 50 55 53 52 50
F CVo in
L. ...... .. 30 55 55 53 53 .')2 55 53 .j3 52
T•. .. ..... .... 28 52 55 53 52 50 55 53
ST ... ..
[l ft> m .. .... 34 39 40 40 38 37 40 4Q
F",.,.,k.ri
(t:/D-1.5). .... 91 91 87 84 91 91 87 84
(,ID-2.0). 117 117 112 108 117 117 112 108
F'In-v,1csi.
(,ID-1.5). 77 74 74 73 77 74 74 73
(t:/D-2.0). 88 85 85 83 88 85 ss 8.
e, percent
L. .... . ....... 11 7 10 9 7 5 10 9 7 5
T. ............ 3 6 5 3 2 4 3 2.5 .,
ST. 3 2 1 1 2 2 1 1
E, 10' ]Mi. 10.5
E e• 10' psi. 10.7
G, 10' psi .. 4.0
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES B2.15
"'J
"'0
-
~ -
- ~
- ~
-
,ec - - -
- ~
~
- ~--
- STUtC.'t
(J.'osvu
n
"
TC"~(.UII.[
r e '0 000 .. .-
- - -
- ---
- -- -- -- -. -
- ~ ~
-- -
--~
0:=
--
- -
...,...... -=-==-
- - - - --
'/2
'0
'00 - --
- -
'000
'0 000
...-
--
-- - -
- -
- - --
- -- - -
'"
-,==:== --- - -
~=
-- -- - - ---
o '''00 -lOO 0 200 )00
-'00 '00 JOO '00 500 lO
T04"E:FlATtRE. .".
Fig. B2.53 Effect of temperature on the ultimate strength (Ftu) of 2014·T6 aluminum alloy (bare and clad sheet 0.020-0.039
in. thick; bare and clad plate 1. 501-4. 000 in, thick; rolled bar, rod and shapes; hand and die forgings; extruded bar, rod and
shapes O. 125-0.749 in, thick with cross-sectional area ~ 25 sq. in.).
,'" ,
-, -.
-'':L -~
- '.......--..
: -'-'-t+.....-
':1. __ ~~
I - ."-"._;"
~_~_tE:'
,,:~- - 'J ,.,
'''' '-- -
•.
-+-l-~.; Snuu"
:-7 AI H .. ~[tUvU
,,.... El'OSIII[ v~ 10 10 000 u
'00
80
=
-..;::: "
,
- --'/2'0 .., ......
zw - tc '000 ..
--~.,
'00
V 40 000
•
" -
~
.--..., - ,.--
,---'
-
I ·:.::2E?~-
:~ :-- ~-'-.,~~: rr;
-"-
-
~
--- -
- c= t=;:-=
-lOO -200 -'00 o 200 JOO '00 500 lO 700
•
Fig. 82.54 Effect of temperature on the tensile yield strength (Fty) of 2014-T6 aluminum alloy (bare and clad plate
3.001-4. 000 in. thick; rolled bar, rod and shapes; hand and die forgings; extruded bar, rod and shapes 0.125-0.499
in. thick with cross-sectional area a 25 sq. in.).
B2. HI MECHANICAL AND PHYSICAL PROPERTIES OF METALLIC MATERIALS FOR FLIGHT VEHICLE STRUCTURES
j
I
.'
,!
Fig. 82.55 Effect of temperature on
the compressive yield strength (Fey)
of 2014-T6 aluminum alloy (all
iI Fig. 82. 58 Effect of temperature on
the bearing yield strength (F'bry) of
2014-T6 aluminum alloy (all products
products except thick extrusions). except thick e..x trastcns).
J
!
,r"
I
2024 ALUMlNUl1. ALLOY (BARE SHEE:I' & PLATE, EXTRUSIONS, BAR, ROD & '"lIRE)
Table B2. 11 Deatgn Mechanical and PhySical Properties of 2024 Aluminum Alloy (Sheet and Plate)
ThieJmes (in.}.
0 : = 0.501 1.001- 2.(lOI- 0.010- 0.250- 0.501- I 1.001-1 2.001-
O.OZQ-O.500 0.012-Q.128
<0.250 0.500 1.000 2.000 3.000
sw... A A
0.249 0.500
"""AT ,.\ B A B ------12...
1.000 2.000 3.000
Meo:h&nica..t properUe8:
A A A
B_~_+.~- _A_I_8_ ------
A 8
lI,., .en:
L .. se ee es ra
62
""
62 60 68
" 63
" eo " 61 70 62 66
.
lily> .q,.'
L. 4() 38 .18 36 36 sr ... ... " ... 46 .. 60 ea 4() 41
I"C'It
T
.en
L.
T.
4()
4()
38
.18
36
38
36
38
38
36
..
43
..
4()
38 42
4()
42
4() 52
49
..,
54
51
se
4()
4()
'I
41
. 4() 36 38 36 38 46
" 42 38 43 .7 42 56
... 4() 41
.
(eID_2.0) 118 122 118 II. 196 129 12. 121 120 129 118 122 133 137 118 128
P~bi·
(eID_l.5).
(eID_Z.O)
pell:ent
56
"
53
61
53
81
53
61
53
er
sa n
19 62 "
14
69
78 ;: I:; eo ee
ea
"
. .
96
88
100
se
"
st
ee
• I
f,
s 10'
T
,.. '" 12 8
'" • '" 12 8 ,_: (d) , i', i"
10.5
Eo, 10' pn'. 10.7
G, 10' pM. '.0
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES B2.17
2024 ALU11IN1JM ALLOY (BARE SHEEr & PLATE, E:XTRUSI ONS, BAR, ROD & '.IRE) (Cant.)
Table B2. 12 Design Mechanical and Physical Properties of 2024 Aluminum Alloy (Extrusions)
Alloy.................... · . 2024
Heat Heat
treated treated
Heat treated by user· cold
worked
and aged
Thickness, in ... ............ 0:050- 0.251}- 0.501}- 0.750- 1.500- 3.000:;; All thick- :<0.250
0.249 0.499 0.749 1.499 2.999 ness
:;;::2:5, :;;::32
·
Cross-sectional area, in. l .... :<25 =<32
I
Basis ........ ........ ... . .. A B A B A B A B A B A B A A
I-
MechanicaJ. properties:
Fl•. k&i
L ... .... .. . ...... 57 61 60 62 60 62 65 70 70 74 70 74 57 64
T ... 57 61 60 62 60 62 ;8 61 ;4 57 ;0 ;3 SO ..........
F~. k&i
L .. ........... .. 42 47 44. 47 44 47 46 54 ;2 ;4 52 64 38 56
T. ...... 42 46 43 46 42 45 41 44 38 41 36 39 36 ........ ..
F••• k" ·..... . ...
L. .. . ....... 38 41 39 42 39 42 44 ;2 SO ;2 SO' ;2 38
T ....... .. . 38 41 39 42 39 42 42 48 42 44 42 44 38 ..........
F••. kJli ... ...... . 30 32 32 33 32 33 34 38 38 40 38 40 30 · .. . . . . . . .
F m. kai
('ID~U) . ... ... .. 85 91 85 91 85 91 85 91 85 91 85 91 85 ..........
(e(D=2.0l. .. . 108 114 108 114 108 114 108 114 108 114 108 114 108 .. ........
F..... lcai
(,/D=U). .. . ;9 66 60 66 60 66 61 66 62 66 62 66 53 ..........
(,/D~2.0) . ..... 67 75 69 75 69 75 71 75 73 75 73 75 61 .. ........
e, percent
L .. ....... ... 12 12 12 ... . 10 10 . .. 10 12 ..... .. . ..
To .... .. 6 ... . 6 .. 5 .. . 2 . ... 2 ... . .. ...... · . . .. . . . . .
E, 10- pai ... . . .. ..... 10.5
E e. 10' pai.... 1O.i
G, lOt psi. 4.0
82.18 MECHANICAL AND PHYSICAL PROPERTIES OF METALLIC MATERIALS FOR FLIGHT VEHICLE STRUCTURES
2024 ALUI1INl.J11 ALLOY (3 ..e....p .E SHE2I' "'= ?L{'~S, ~T:\USIC:NS, SAH, ROD & '~'::\.:::) (Corrt . )
Table B2. 13 Design Mechanical and Physical Properties of 2024 Aluminum Alloy (Bar, Rod, and Wire);
Rolled, Drawn or Cold Finished; Rolled Tubing
Alley. 1 mu
I--------~----
Form .. .. 1 Bar. rod and wire; rolled. drawn or cold- finished Tubing
, - - - - - - - - - - - - - - - - -----,.------
Heat 1 Heat
Condition .. I -T4 or -T351 Heat treated I
treated
by
i treated
cold
user' worked
~36
1 1 :
I '1
: - - 1
'-
Mechanical properties: I ' '1 I
r.; ksi 1
L. 62 i 82 62 62 I 62! 62 1 62 64 70 I 62 68
T ... 61 59 57 55 i 54 521· 1 j.
r.; ksi
Li 40 ~ ~ ~ ~ ~I ~ a' 46 1 40 ' 60
T. '1 40 39 38 37 37 36 . I
i
1
F~II' ksi
L .. .. .. .. "I 32 32 32 32 32 32
1
42 46 40
T... '1' . . . . .I.
F .... ksi.. .i 37 37 37 37 37 37 37 1 39 42 39
r,.•. ksi. I
(e/D-1.5) " . " .. 93 93 93 93 93 93 I
93 105 96
(erD-2.0) " . '1 118 118 118 118 118 118 118 133 122
F br•• ksi ,
1
(eID=1.5). L 56 , 56 56 56 56 56 I 59 , 64 58
(e(D=2.0)...... ' 64 I 64 64 64 64 64 !. 67 ,
74 64
e, percent
10'1 101
1
L. " " 10 10 10 10 10
T. " ". 10 ,
1
8 6 4 2 2
1
1°'1 !
Physical properties:
J
0.100
w,
C, Ib(in.
Btu/ilb) IF). '.1'.
0.23 (at 212" F)
K. Btul{(hr) (ft 2HF)/ftj 70 (at 77° F) •
a, 10-s in.iin./F. 12.6 (68 0 to 212 0 F)
Table 82.14 Design Mechanical and Physical Properties of Clad 2024 Aluminum Alloy (Sheet, Plate, and Coiled Sheet)
,
"·orlll ", , -_ ..... - Slwct llnd Illate Coiled shed 811eet and Illato
-~-
lIellt
lleat trented IIeat trellted • lIellt treated b)" user' lIellt treated lIellt treated, cold worked,llIloJ aged
end fvlled rreared "d
IIl:ed
----- >-
Ooudntcn. -1'3 -'fJ6 -T< -1'42 -T< -TO -Tltt -T'" Z
--
'">-
Oasis .. ----.- A A
" " " "
A A A A A A A A A A A A
-- A - " "
" " -- A
" " " -- " "A A A A
" " - -" - -
A A A
-- -- -'- - - - -
..
-- -- -- -- -- -- --~
--, -- -- N
0
:6 '" " es
ae fll; III
T :1" I~ S2 IJ~ ali ati a,
,'" '" " " " a, '" ae '" " '" as sa "- O
31;
'" a, "3. " '" " :I."'" ,,' "" " r,; '"" " "'" '" "Ii," "
J.
r
a, J:, a,
n
u;
"
Jij,
~'l
ae ali " as
a,
" '" '" " W
C
" '" "a:, " "" "" "' " " " "" '" " I, " "" " '"
ta C-
.,
."
,.·•• ,hj ,
" 3, a; a,
" a,
" J.~
'" '" " " " " a~ 3\1
~
'" " " .
:l~ :I~ 4~
.,
~
11M In I:?
'" .. l'i
(fiD)-2d 12, 1:t1 12, lJl 12ll12~ 116121 112 JIll lOll 110 112 111I iii
.
1I~ l~" 1211<'11 12,~
f'" •. .\.Ii" '" Ii, ., tJ
Cl
", ...'"
l" " '" " '" lUI lU"TU;" """ " "
(f!D)_U ll~
'" " "
" "" ""' "'"'. "" " "" ,II 6. ~:I'" " '" ""
"~:I ", '"'" '" "" " "'" " ", 1:/111
III
,:I "
Iii
"' "
T'
(,'DJ~2.U 102 1Il4
1O~r:~;
,1\ 511
, "
~
'/' , . 4 ...
'" " " "
(t). ,
'" "
,
"J • (oJ
E, lUlllli
"Ii IO~
'"
~
"'''H~I " I
n t<
K,.,·
,-:" lilli'll
lJ.~
!I I
,
I
~ .•'
! ! •
I I
Fl.g• .82.60 Elfect of temperature on Fig. 82. 64 Effect of exposure at Fig. 82.68 Effect of temperature on
the ultimate tensile strength (Ftu) of elevated temperatures on the room- the ultimate bearing strength (FOro)
2024-T3 and 2024-T4 aluminum alloy temperature ultimate tensile strength of clad 2024-T3 and clad 2024-T4
(all products except extrusions). (Ftu) of 2024·T3 and 2024-T4 aluminum aluminum alloy (sheet).
a.1loy (all products except thick
extrusions).
II
!
I
FIg. 82.61 EIfect ot temperature on Fig. 82.65 Effect of temperature on Fig. 82.69 Effect of temperature on
the ultimate tensile strength. (Ftu) of the compressive yield strength (F cy) the bearing yield strength (Fbry) of
2024-T3 and 2024·T4 aluminum alloy of clad 2024-T3 and clad 2024-T4 clad 2024-T3 and etas 2024-T4
(extrusiOns). aluminum alloy (sheet). aluminum alloy (sheet).
!I iI
,•
!
'-....- - 1
__ 100 __
f _ ",,,oODQO'"
'-'
Fig. 82. 62 Effect of temperature on Fig. 82.66 Effect of exposure at Fig. 82.70 Effect of temperature on
the tensile yield strength (Fty) of elevated temperatures on the room- the tensile and compressive modulus
2024·T3 and 2024-T4 aluminum alloy temperature tensile yield strength (E and Ed of 2024 aluminum alloy.
(all products except extrusions). (Fty) of 2024-T3 and 2024-T4
aluminum alloy (all products except
thick extrusions).
j
!
I
FIg. 82. 63 Effect of temperature on Fig. 82. 67 Effect of temperature on Fi.g. 82.71 Effect of temperature on
the tensile yield strength (Fty) of the ultimate shear strength (F su) of the elongation of 202C-T3 and 2024-T4
2024-T3 and 20Z4-T4 aluminum alloy clad 2024-T3 and clad 2024-T4: alumtnum alloy (all products except
(extrusions). aluminum alloy (sheet). thick extrusions).
;or-_
Table 82. 15 De.sign Mechallical and Physical proper-t ins of 7075 Aluminum Alloy (Sheet and Plate)
., "''"-
/o'jll.' klJi ~
L. 76 78 17 76 17 76 76 82 78 80 73 75 70 72 68 66
T 76 78 17 76 17 76 17 80 17 70 73 75 70 72 68 66 '3en
ST .. 67 69 64 66 64 62 :-
Z
/0"/1' hi >-
1> 66 69 67 70 67 60 66 72 68 7l 62 65 60 62 58 56 t: "
7'
3T
It'e/l, klSi
65 68 66 66 66 68 66 66 66 66 62
61
65
63
60
56
62
60
58
56
56
64 "~
H "
-
t'l
~
C>
I.. 67 70 68 7l 60 7l 66 72 68 7l 65 67 63 65 62 60 >- z
T
37'.
70 73 7l H 69 71 66 72
40I
68
..
71 65
64
67
66
63
62
65
63
62
62
60
60
b
Sl ....
0
r,« ksi
r-; bi"
46 47 46 47 '6 47 47 46 47 43 45 '\ 43 49 30
~ r-
(,jD ~ 1.5). 114 117 116 119 . 108 110 110 115 109 112 102 105 68 10\ 65 6'
[>J
(,/D~2.0)
P,,'1I' kISt"
144 148 146 150 139 142 142 147 140 144 13\ 135 126 130 \22 110 '""
(e/ D = 1.5) 62 67 64 68 87 00 00 64 88 62 81 78 81 75 73
5:
(,/D~2.0) 106 110 107 H2 100 104 104 108 102 106 63
8'
67 00 03 87 8'
'"
rn
e, percent ~
L. .... 7 .. 8 8 .. 6 5 .. 5 .. 5 .. 5 tJ
T 7 8 8 6 .. 4 3 3 3 2
37' . ... .. .. .. .. .. .. .. \ 1 1 '"
"U
E, 10' pili. 10.3 5:-3
e; 10- psi. ........ 10.5 to
0, 10' llsi .... _.. 3.9
Physical properties:
w,lb/in. '. 0.101
C. B'u/(lb)W). 0.23 (at 212°F)
K. B'u/«I,,)(f")(F)/f,1 76 (at 77"F)
II, 10-' in./in 'I". 12.0 (68" to 212"F) ..
(A - ' _ , )
1.001-1.500
1.101 -a 000
"8M T..L.. '.1.1.1.1.
L
L
..
0.15"'1.000 L ..
i
II Rl
I
,
Table B2. 16 Des ign Mechanka.l and Physical Properties of Clad 7075 Aluminum Alloy (Sheet and Plate)
w
w
3:
Alloy. .- ..... 'o"
~
Clad 7075
Form ...
Coudiuou
Sheet anti plate
T6' ,.
n
r-
0.015 0.040-- 0.063- 0.188- 0.250- 0.600- 1.001- 2.501
2.001 3.001 3.501
~
'I'hioknesa, in. . . ..... . ... 0.039 0.062 O.IM7 0.249 0.499 1.000' 2.000' 3.000b
'
2.500~ 3.500' 4.000"
Beeia. A u n n n ---
A A
-- c-- -- - - -- -- - - - -A A n A n A n A n A n A -o
~-
-- -- -- A 0 '"
~
Mechanical properties: -o
Pill, k8i~L c»
70 73 72 74 73 75 75
T. 70 73 72 74 73 75 75
77
77
75
75
77
77
77
75
79
77
76
75
78
77
71
71
73
73
68
68
70
70
66
66
64
64
>-
r' ,.r-o
~
SOl' .. .. 67 69 64 66 64 62 ~
PIli' bi-L. 61 64 63 65 64 66 65
H -o
67 65 67 68 66 69
66
ill
,..r-~ '"'"
60 62 58 60 56
sr
T. 60 63 62 64 63
..
65 64 66 64 66 64 66 64 67 60
61
62
63
58
59
60
60
56
56
"
54
54
/I'CII. bi-·L. 62 65 64 66 65 67 66 68 66 60 66 69 66 69 62 64 61 62 60 58 r- ;'j
T 64 67 66 68 67 69 68 70 66 60 66 69 66 69 62 64 61 62 60 58
0 OJ
sr
1" ..., ksi c . 42 44
.. 64 66 62 63 62 60 "',.rn 0'"
i'fn", ksi--(el D = 1.5) 105 110 lOB
43 44
III
44
110
45
112
45
112
46
116
45
105
46
108
46
108
47
\11
45
\06
46
109
42
09
43
102
40
95
41
08
30
02
38
90
c-,tn 3:'"
(e/D=2.0). 133 139 137 \41 139 142 142 146 135 130 139 142 137 140 128 131 122 \26 119 115
f:> ...,.'"
1<'/)'1/, k:siC-(e/ D = 1.5). 85 90 88 91 92
(elf) =2.0). 08 102 101 104
90
102 106
91
104
94
107
84
98
87
100
86
90
88
102
86 90 78 81 75 78 73 70 "~ r-
r-
e. pcrcent-c-L 7 .. 8 8 8 8 6
99 104 90 93 87 90 84 81 u n
5 5 5 5 en
...OJ~
7' . 7 8 ..
81'
8 8 8 6 4 3 3 3 2
'"
L-'
1 1 1
..
~-J
.
l
Sec. 111.:1 10.3 H1.:l
-u r-
s; 10' lJ8i-l'ri. U.5 !I.8 10.0
~: '"
a, 10" }lsi
::il~C .. 10.5
!1.7
10.5
IIU)
I tJ.5
." '"0
L';
'"r-"
It).:!
Physical properties: -~
n
0
w,lb/ill. 1
C, Btu/(1L)(F)
K,lItu/{(hr)(ft»(l")/ft]
Ie, 10-' in./ill./l<'
..
0.101
0.23 (at 212"1<')
7li (al 77° F)
~
" "q
<:
OJ
12.U (tiB" ttl 212"1")
a 10'01' the ..tr"", reli"ved !.emper -T66J, aU ...alu~ lor the -Tll
II Tb....e valuell except In lhe ST direction have been adju.ted to include the IntlueD"",
"'
o
r-
"I;""ption 01 ,.'~ •. Alllllic.aLle ,..'~. valu"" ar.. q 'ollow.:
Thoi(/,;lI"" (i-n.J l>iu"li..... 01 ted
of the 1 1/2% per .Ide nomInal claddlDlJ thickneae. '"
...'"
0.260-· 0.499
0.600-2.000
2.001-2.600
2.601-8.000
L
I.
I.
L
"'ell (A tl4l"e.,
ea
eo
. "See TaLIe 8.1.1.1.1.
n"
ci
-s
" ""
OJ
'"
Table B2. 17 Des igu Mechanical ~U1d phystcat pr-oper-ties of 7075 Aluminum Alloy (Extrusions)
----~---------- - - - - - - - - - - - - - ; : ; = - -
7075
Alloy - - - - - - ---------._---------
Furm . Extrusions (rod, harH, /Iud shnpes}
Condltioll
. -_'1~"c:;. ==-----
- -
Cro:;:;-bl~d,i(Jllulllrcu, iu. 2 <20 ._- <20 I
>2t1, <32 . <32 >-
"' -- --0
Z
4,5UO~5.0
ThidlJu~HS ~, in. !J 0,250- O.4ml 0.5000.749 O.750-1.4!l9
- --
1.500-2 .H!)!)
-
3.000-4AWJ 0 >-
...e-
--0
A II A II A II A II A II .. A
---_ II A II en
Habb.
--- .•.. - - ~~- -_._-------- ----._.- - --- - - - _._-_.- ---- - --- ---- ---- - - - - - - 00
".
Meehani!",ll pruplJrtie::;: r- ~
00
p ..., k,;j §
L 78 82 81 85 81 85 81 85 81 85 81 84 78 81 78 81 H
>-
l' 76 78 77 70 73 75 72 74 eo es 62 U4 62 tH UO U2
~ Z
"
F,~, kl>i ".
us
L
1'.
70
U4
7 -l.
U·7
73
UU
77
U8
72
U3
7U
66
72
U2
7U
U4
73
5U
7U
58
71
54
74
5U
70
54
73
5U 53
71
55
r::
0
"'"
00
r.; ksi 7U 72 75 72 75 U8 72
"' ~
o
z
71 75 74 78 73 77 72
1'.
L.
71 75 74 78 73 77 73 75 60 72 U5 U8 !:1
45 45 47 45 4U 43 H 43 H
;J 0
45 47 47 47 c:
/t'n, ksi
F~, .., k,;i
43 45
" en
H
'"
(1:/»= 1.5). 101 JO 7 07 102 U7 102 !J7 102 07 102 07 101 04 07 lH 07 0
Z
en
'"
r-
~
(1:/»=2.0) . 125 13I 130 t3li 130 13U 130 l3l1 130 136 122 12ti 117 125 117 J 25 o
0:
F ~'"' J,;,;j ~I
(t'/J)= 1.5). 01 oU 80 85 70
101
84
IOU
70
101
M
JOI
70 84
IOU
78
02
81
96
77
01
80
U5
75
88
78
02
&J
0 "'
e,
(el J)= 2.~) ,
pl"J'(~I"IIL
U8 104 102 108 106 H
~,
8
''""
0:
L 7 8 7 8 7 8 7 8 7 8 7 8 U U H
4 3 I I I I o
1'. 5 5 - _._,,-- - r-
- - - - - - '-'-'-'-'-'-'-'--'-' ----
B, loa -psi .. 10.3 ~ '"
H~, I (Ja IMi, . 10.5 00
on
--(}!. _!~~~~--'-'-''-'-''-'-':.'-'.'-- ~- - - - - - - - - -
Phyaieul Ilr()II.~rlil"::;:
----
3.9 0
"
"''c:"
0,101 o
w, Ih/iu,3 .. s:
'" "''"
H
0, HLlI/(Ib)(J<'). 0,23 (lit 212°]<') c:
1\., HI u/I{hr} (fLl) (F)/fLl 76 (lit 77° [") "l
~.~.~~~~_~0)_;'_'-I_F '_'_''_'_''_'_'". .~"--- 12.9 (68° to ZI2°F)
~~~~~~~'---
n1d11~n Ii".) Aria ('f, in.l
_
l'irldlon 0/ 1'1" F..CA oalu~l)
'" 00
• I"ur tI ... stress rcll"""l! 1"rHl'n~ -'j'li.'t1O ..Ill! -'1'6:i1l. ull vuhir~ Jar the -'1'6 temper llflVly, wllh
llo~ """'1'1["10 of V.... I. A1JI,lkllbh~ I.... \'llllll'~ erettstco uetow: l.lillll---2.llllU <'" I. 10
ee
}.,. (A ""lUI,) 3.00lH.tll1l <32 I.
Thitt,u.t (f".) Au.. (,q. h•.) mUd/ono/lnl
10 t.WO-5.j)OO :<32 I. 0'
<u.2lU <20 I.
I .'or eltmslon~ with olllsllllhHIII: leila, Llle load clirrylnlllil>lIlty 0'
such ICl:stilllill he dt'II'OIl!l,C<\
O.25IlU.HU
07W·I,tUll
:<20
<20
J.
I. "
11 011 the bll.!ils ot the 'lro,wrlle:l ot ure llj'I'Wl'rllllo eclurun (:()(resl>ou<lhl~ 10 Ihe Irg Ihlckllt,s.:J
B2.24 MECHANICAL AND PHYSICAL PROPERTIES OF METALLIC MATERIALS FOR FLIGHT VEHICLE STRUCTURES
Table B2. 18 Design Mechanical and Physical Properties of 7075 Aluminum Alloy
(Hand Forgings and Die Forgings)
Mechanical properties:
Ft»o .bi
L .. . . . . . . .. .. . · . 75 73 71 75 73 71 75
T ........ ....... 75 71 69 73 71 69 71
ST. , ... .. . .... · . 72 68 66 70 68 66
r,., .en:
L .. ....... ..... 64 61 60 63 60 59 es
T ....... . . ..... 63 60 58 61 59 57 62
ST. ... . .. .. ..... 63 60 58 61 59 57
F~Ui
L .. .. . ......... 64 61 60 63 60 59 6.i
T ... . . .. ......
' 63 60 58 61 59 57 58
ST. . . ... . .. .. ..
F... Ui. ........ . ... 45 44 43 45 44 43 45
" .... Iui
(.ID-l.5). ... 97 95 ss 97 95 85
(.ID-2.0) . ...... 13.5 124 114 135 124 114
" .... hi
(.ID-I.S). .... · . 90 79 78 88 78 77
(e/D-2.01. ..... 96 91 90 94 90 88
t, percent
L. ... .. ...... 9
.,7 4
",., 7 , 7
T. ...... .. . 4 2 3 2 3
ST .. .. . ........ 2 2 1 I 2 I
E, 10- pIi .. ....... 10.3
Eo 10- pri . . . .. . . . .. 10.5
G, 10- pn. ....... 3.9
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES B2.25
Table B2. 19 Design Mechanical and Phys icnl Properties of 7075 Aluminum Alloy
(Bar, Rod, Wire and Shapes; Rolled, Drawn or Cold-Finished)
Form. . . ........... Bar, rod, wire and shapes, rolled. drawn or cold-finished
Mechanical properties:
F l .. , ksi:
L ...... . . .... 77 77 77 77
LT .. . . ... 77 75 72 69
r.; ksi:
L ... .. . ..... 66 66 66 66
LT .......... 66 66 63 60
F••, ksi:
L ............ 64
LT . . . . . . . . . . . ...... ...... .......
64 I
..... ...... ......
64 64
. ... . ..
F.... :':~i. .... ... 46 46 46 46
F hn .. ksi:
(,/D= 1.5) .... 100 100 100 100
(,/D=2.0) .... 123 123 123 123
F~ ..., ksi:
('ID=I.S) .... 86 86 86 86
(,/D=2.0) .... 92 92 92 92
e. percent:
L .. .. .. .. .. . '7 7 7 7
LT . .. .. . .. . . 4 3 2 1
j
I
Fig. 82.75 Effect of exposure at
elevated temperatures on the r-oom-
temperature tensne yield strength
!I
(Fty) of 7075-T6 aluminum alloy
(all prccucts}. .'
,! • ,!
Fig. B2.72 Effect of temperature on Fig. 82.79 Effect of temperature on
the ultimate tensile strength (Fm) of
7075-T6 aluminum alloy (all products) J the bearing yield strength (Fbryl of
7075-T6 aluminum alloy (all products).
I
,!
j
Fig. 82.76 Effect of temperature on
the compressive yield strength (Fey)
of 7075-T6 aluminum alloy (all products).
,
a
..: .0
,~
,
,-",--.'
"'" 0.00
,f
FIg. B2.73 Effect of temperature on Fig. 82. 80 Effect of temperature on
the tensile yield strength (Fty) of the tensile and compressive modulus
7075~T6 aluminum alloy (all products). (E and Ee) of 7075-T6 aluminum alloy.
I
.•' •
!
Alloy . AZ31B
-0 ~H26
Condition ...•.
Thickness (in.) . 0.016- 0.061- 0.250- 0.501- 0.016- 0.250-- 0.375- 0.501- 1.001- 0.250- 0.375- 0.439- 0.501- 0.751- 1.001- 1.501-
0.060 0.249 0.500 2.000 0.249 0.374 0.600 1.000 2.000 0.374 0.438 0.600 0.760 1.000 1.600 2.000
I-~-I-~-I-- --I--,---f---I-- - - - - - - - - - - - - - - - -
,.
Z
A B
BnsiB A B A B
------- -- ---------------- -------------- --- '"
Mecha.nical properties: ''""
~
1",.., kai. ~
L . 32 36 32 36 32 30 39 40 38 37 36 34 39 38 38 37 37 35 35 c
Z
T . ...... 40 39 38 37 36 40 39 39 38 38 36 36
(.~:~;/':!i;
'{,': 18
19 16 19 16 ~ 30 26
''29
24
27
22
26
20
23
27
30
26
29
26
29
26
28
23
26
22
25
21
24
r.; kai. 18 17 16 16 14
L 12 13 12 10 10 24 26 20 16 13 10 22 21
T- .
F kai .. .. . . 17 23 17 23 17 18 26 18 18 . 18 18 18
F_ klli.
(,/D~ 1.6)... 60 62 60 62 60 68 60 66 64 58 66 56 .
(,/D~2.0)........... 60 61 60 61 60 68 70 66 63 ... 68 65 66
1"6"" kai.
(,/D~1.6).....
(,/D~2.0)
e, percent
29
29
30
30
29
29
31
31
27
27
43
43
44
44
38
'38
34
34
. 40
40
39
39
36
36 . .
~
L
T
. ............
.
12 18 12 19 12 10 6 8
8 .... 10
8 8
10 10
6.~
8 8
10
6
8
6
8
6
8
6
8
6
8
6
8
6
8 .'"
c;
o
c:
6.6
2.4 ''""
~
Physlcel properties:
w. Ib/in.' . 0.0639
C. BTU/(Ib)(F) . 0.25 (a.t 78 0 F)·
K. BTU/(I,,)(lL')(FJ/fl . 66 1212 to 672" F)
_. 10·' in./in./F . 14 (6f; to 212 0 F)
• Ihthll.tL..... • Trll.llsverllll F •• tdlowtlbles ere 6fJUlli Io or cruUt than tb, 1lJbIttud.lna1 POI .UO....blu.
(C')
".;
iJ\,
B2.28 MECHANICAL AND PHYSICAL PROPERTIES OF METALLIC MATERIALS FOR FLIGHT VEmCLE STRUCTURES
j /
.I
.,
1 ..
.I
• Fig. 82. 87 Effect of temperature on the ultimate bear-
ing strength (Fbrn) of AZ31B~H24 magnesium alloy.
!
l
j
.I
.!-. Fig. 82.8a Effect of temperature on the bearing yield
strength (Fbry) of AZ318-H24 magnesium alloy.
I
.I
•
!
l
. ... HK31A
Alloy . .
Z
"~ t:l
A B A B (.) t:l
Buala . . . . . ............ A B ~
- §
,. '"
W
~
o
Mechentce! properties: f: z
PI., kai . ..... ... . .....
F,u. kai .....
30
18
32
19
30
l'
30
16
29
14
34
20
30
28
34
23
36
24
34
21
34
18
33
16
33
15
27
13
...
0
0
43
22
40
22
43 . .......
22 . .......
..... ...
23
49
23
52
23
49
23
51
23
49 . ....
23 .. . . ....
0
e, percent. . ... " . 12 20 12 12 12 4 0 4 8 4 10 10 8 4 ~
E, 10° pai . . . . ........ . . 6.5 0 '"
w
. . ..... •. 5 1.;
'"Z "''0""
EI, 10' pai.
G, to' p.ti. .. . ... . ... . ... 2. 4 H
g
"''""
Physical properuea:
w, lb./in.'. .... . . ..... 0,0647
C, BTlJ/Ob.llF) ... , ... 0.25 (320 to 2120 F)
K, BTU/(h,.)(ft.')(F) ft. 60.0 (at 68° Ji') '"
W
• f'rop"rU~s tor 'I.heet lIud pllliu IlJ'(l lako:!D plUlIllello lile direction of rollin.. Tl'anIVCraoproPOrtIall • Mechanl(llli proponlea are ba.sod UJ!'llD tho ruaranteod te.D.1lI1o propertlea from IIllplmllely-(:ll4t
are equal to or Irellter than tho Irmll;IUldlulll propertloa. tlll& ban. 1'he IIWIchanlcal propenle, Ilf ban cut from eNUnl' U1aY be .. klw as 16 porocnl of the
• RekreliOO III10uld be lIllOdll to tllll q>eclll(l roqulremt:lIl1 of tho procur1P, or oarllt\cllllu, qeDey tabulated values .
with rellUd to tho USil of lhe abovo value.lll tho delll,u eaatlDp.ct'
I
• J
82.30 MECHANICAL AND PHYSICAL PROPERTIES OF METALLIC MATERIALS FOR FLIGIIT VEmCLE STRUCTURES
"'" ·
•f '"'" •,te '"
so
~ •
• ~ •s- ~
.'
• '" , '"
0
~ ,
0
T _ '..... F
Fig. 82.90 Effect of temperature on the ultimate Fig. B2.93 Effect of exposure at elevated
tensile strength (Fta) of HK31A~H24 magneeuim temperatures on the room-temperature tensile
alloy. yield strength {Ftyl of HK31A~H24 magnes iurn
alloy.
T~t_.F'
Fig. B2.91 Effect of temperature on the tensile Fig. B2.94 Effect of temperature on the ultimate
yield strength (Fty) of HK31A-a24 magnesium tensile strength (Fta) of HK31A~T6 magnesium
alloy. alloy (sand casting).
i
•
~
•
Fig. B2.92 Effect of exposure at elevated Fig. B2.95 Effect of temperature on the tensile
temperatures on the room-temperature ultimate yield strength (Ftyl of HK31A-T6 magnes ium
tensile strength (F tu ) of HK31A- H24 magnesium alloy (sand casting).
alloy.
Ll
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES 82.31
Table 82.22 Design Mechanical and Physical Properties of AZ61Aa Magnesium Alloy
(Extz-ustcns and Forgings)
Alloy. AZ6IA
Mechanical properties:
FhI , kn
L. 38 39 36 36 38
T. .. ..
r.; hi
L .. 21 24 16 16 22
T .. ..
F¢11. k.ri
L. ..... ..... I' I' 11 11 I'..
T. . . . .. ... .......
F,," m. 19 19 ·. 19
FIIrv, k$i
(,/D -1.5). ......... '5 45 50
(,jD-2.0). 55 55 ·. 60
F.... kM
(,j D -1.S). . .. · . 28 28 28
(,jD-2.0). 32 32 ·. 32
e, percent ..
E, 10- psi.
8 9 I 7 7 6
6.3
Ee, 10' p"':. ·. 6.3
G, 10' p.ri. 2.4
Physical properties:
w,lb/in. t . 0.1)647
C, Btuj(lb)(F). 0.25 (at is''F) Ir
K, Btuj[Cb,)(fto)(F)/ftj. 46 (212" to 5i2"F)
~, 10-' ir-../in./F 14 (65" to 212"F)
<J PToperti~s '-or extruded b&l'S, rods. shapes, tubes. and fol'ltings are during fabrication,
tll.ll:~n oaralI~1 to the dlrecttcn of ~xtru.sion or maximum ffil'tAI new ~ Estimated.
B2.32 MECHANICAL AND PHYSICAL PROPERTmS OF METALLIC MATERIALS FOR FLIGHT VEffiCLE STRUCTURES
Table B2.23 Destgn Mechanical and Physical Properties of AZ80Aa Magnesium Alley
(Extrusions and Forgings)
Alley. .U80.'
Thickness, in .... .... ,., ... <'0.249 0.2.10- 1.5~ <'0.249 0.250- 1.500-
1.499 2.499 1.499 2.499
Meehaniea1 properties:
F ta • kn
L .. . ........ ... ....... 43 43 43 47 48 48 42 42
T. ............... . . ... ..
Fly, k.ri
L .. ............ . .. .... 28 28 28 30 33 33 2" 28
T ... • • • • . .. . .
• • • ' •••• 4 •
·. .. ..
JI~ kn
L .. . . ... .. . . .. . . .. . . . .. 17 17 .. 28 27 18 25
T ....... ....... . .. .. ·. ..
p... hi. ......... ....... 19 19 19 20 20 20 20 ,0
PIw.,kn
(.ID-1.5). ... ...... ' 48 48 '8 50
(.ID-2.0). ........ 56 56 56 .. 70
F.... kri
(.ID-1.5) .. ......... , 36 36 36 42
(.ID-2.0) .. ........ 40 40 40 ·. ..
e, percent...... ..... ........ 9 8 6 4 4 4 5 2
E, 10' pIi......... .... 6.5
B" 10' pn. ..... ... . ..... 6.5
G, 10 1 p:ft... ........... .. . 2.4
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES B2.33
.. _~
Table B2. 24 Material Specifications for AZ63A
Magnesium Alloy
'''- A_r,n,il. 1I
e.com_ _ Il
-,train
-ttrQIn
C. ren ..l. lan~l' ~ ..l_
o'C_'~lft04IIl",
Specification Type of product
s.
TABU: 4.2.3.0<"). Denp MccA4nical aM PhyftctU
P-ropmiu of AZS3A Mag7Ulnum- Allov (elUting_)
Thickness (in.) .. ..
'-'"
A.r.....II. ttnn-,traln
Basi8~ .. ...... . ...... e.c~" ...-~n
C.Ttnlil, ~ lr'OClul...
Do~"",~
Mechanical properties:
F tur lui .... . . .. . . ... 24 34 24 34
Phi. k..n.. .. . - ........ 10 10 11 16 1
Ft:fI, Jeri .. ... . .. . ." . 10 10 11 16 §
F,.." kai ... 16 17 19
F/wv., kn
(,/D-L5). 36 36 50
!
(,/D-2.0) .. 50 50 .. 65
Fl1rT/.kn
(,/D-L5). 28 32 36
(,/D-2.0) . ..... 30 36 45
'.
t, percent .. 4 7 2 3
E, 10- pn.. 6.5
Ee• 10- pn... 6.5 • •
$traWl,QOOI inA\
•
G, 10· pori. 2.4 TGftqenf Mocl"~ 10· ""
j ..
~
w
Alloy 8Mn
J ~o
J J
Form Sheet, plate, llud strip
• ~n
>-
! ! t-
Coudiuon AlIllealed
1 1
1000 1200 1400 1600 ~
'~"
Fig. 82,101 Effect of temperature on the ultimate 00
Thickllt'::lll, ill Fig. 82,98 Effect of temperature Oil the ultimate ~
tensile strength (Ftu) of BMn annealed tttantum shear strength (Fau) of 8 Mil annealed titanium o
>-
alloy. allOy.
~ r-
'o
ljasis A o-J
H '"
0
'00 o-J
~ ''"..."
s,- 'i'l"
H
Mccllll.uieal properties:
1",.., k~i
L I~O r- '"
0
r 120
r-
0 '"
;:
1"'11' bi
I- flO
'"
tn
...'>-"
r r-
110 ~ t-
1"~II' kei ~ n
L
~...
110 200 400 600 lIOO lCOO 1200 1400 1600 200 400 600 eoc ICOO 1200 1400 1600
~
Fig, 82.99 Effect of temperatur-e Oil the tensile Fig. 2.102 Effect of temperatur-e all uie ultimate to
Ftm.. /.:~i
yield strength (Fty) of 8 Mn annealed titanium hear-lug st rength (to'bru) of 8 Mil annealed titanium
(e/l) = 1.5)
(el D =2.0)
110 alloy. alloy. "'co r-'"
n.; k.~i "3
'"0
(el/) = 1.5) 130 '" '"
H
'0
,00
'"
e,
(el D = 2.0)
Iwr"l~nt 10
!
j eo
I,
I
!ii::~
r-
Cl
E, 10 0 pei
Re , 10" JI$i
.. 15.5
ro.u
J !
< eo
,Ijjijjjilf
' ,,:I ~ IJ '
;'j
'"
M
G, HI" JI$i J •
.:
, m
n
• I
eo .@!JW
J r-
\I'Iii .nun.,.
,:!,!I!:; M
Physical properties: " eo ..•' ...
'"
,S
,~"
w,lll/ill.' 0.171
lillirii '"
C, Btu/(Ib)(r').
1\,IHu/{(hr)(ft')(I<')/ftI
0.118 (ilL 68"1<')
e.a
1000 1200 1400 1600
0
0 '00 ,00 "00 "00 "00 ...o"ct
a, 10- 0 ill./iu./F. 4.8 (ilL 2IXjOJ")
Fig. 82.100 Effect of temperature 011 the Fig. 2.103 Effect of temperature Oil the tensile '"
M
compressive yield strength (l<'cyl of 8 Mu and compressive modulus (~ and Ed of 8Mn '"
annealed trtanium alloy. tltaniutu alloy.
,/
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES B2.35
IlaoUo ........ ... , . .... A A Fig. B2.104 Effect of temperature on Ftuo Fty and E of
6Al-4V annealed titanium alloy (sheet and bar)
-
~Ieebanical pecperties:
fi't.. hi
L. ... 130 130
T. .. .... ... . .... .. 130
F'll, hi
L .. .. .... . ... 120 120
T. ...... . .... .. 120
r.; kJti
L.. .... 126 126
T. ..... . ... . ..... 126
1"... ksi. ....... . .... 80 7.
p"",, k.si
(,!D-1.5I. 196 191 Fig. 82. 105 Effect of temperature on the compressive
(,!D-2.0) ..... .. 248 yield strength (Fey) of 6Al-4V annealed titanium alloy
244 (sheet and bar)•
p.... kn
(,I D -1.51. ... 174 163
(,ID-2.0) .. .... 205 198
t, percent .. ........... 10 10
E, 10- pn
L .. .... . .. ... 16.0 15.4
T. . . ..... .. ..... .. 16.4
E(, 10- pn
L .... .. . ... . .. 16.4 16.0
T . . ..... . .... 16.9
G, 10- pai.. . . ..... .... 6.2 ..
Physical properties:
lIf, Ib/in.'. .......
C, Btu/llb)(F) ...
K, Btu/[(h,)(lt')(FI/ltj.
0.160
0.13S (at. 68°F)
3.8 (at 63°F)
alloy (sheet and bar).
-'
Fig. B2. 106 Effect of temperature on the ultimate
bearing strength (Fbru) of 6Al~4V annealed titanium
Jj
•
~1'::t
:><... ~"---
B-2. 36 MECHANlCAL AND PHYSICAL PROPERTIES OF METALLIC MATERIALS FOR FLIGHT VEHICLE STRUCTURES
Alloy. Inconel X
Form. Sheet
~Ieehanical properties:
P,... hi
L. ...... . ..... 155
T. . . . .. . . ..... 155
FtJIokri Fig. 82. 109 Effect of temperature an
L ... ... . . . .. . .. the tensile yield strength (Fty) of
100
precipitation heat treated Inconel X
T. ....... . 100 nickel alloy•
fi'~/IJ hi
L ... ..... 105
T. ... 105
F,.. ksi..... lOS
Fbm. kin
(t>ID-1.5).
(,fD-2.0). .. 286
F1)r", kri
(t'/D-1.5).
(,/D-2.0) 186 Fig. 82. 110 Effect of temperature an
e, peeeent. 20 the compressive yield strength (Fcyl of
E, 10- pri.. ... . ....... 31.0 precipitation heat treated Inconel X
E~ 10-pn. .... 31.0 nickel allay.
G, 10- pn.. ........ .
Phy&ical properties:
w, lb/in.e. 0.304
C, Blu/ObHF). 0.109
K, Blu/[(lu-)(fl')(F)/ft] 8.7 (80° to 212°F)
z, 10.- in./in./F ... 6.4 (100° to 200°F)
'_f
Fig. 82. III Effect of temperature an
the ultimate shear strength (F su) of
precipitation heat treated Inconel X
nickel alloy.
'-'
Fig. 82. 112 Effect of temperature on
the tensile modulus (E) of Inconel X
-,
Fig. 82. 113 Effect of temperature on
the ultimate bearing strength (FOru)
,-,
Fig. 82. 114 Effect of temperature an
the bearing yield strength (FbryJ of
nickel alloy. of precipitation heat treated Inconel X precipitation heat treated Inconel X
nickel alloy. nickel alloy.
PART C
PRACTICAL STRENGTH ANALYSIS &
DESIGN OF STRUCTURAL COMPONENTS
CHAPTER C1
COMBINED STRESSES. THEORY OF YIELD AND ULTIMATE FAILURE.
Ct. 1 Uniform Stress Condition ~l ~l
..+
~llD[J \4~O
Aircraft structures are subjected to many
l~l
types ot external loadings. These loads otten
cause axial, bending and shearing stresses
acting simultaneously. It structures are to be ,+
deSigned satisfactorily, combined stress re-
lationships must be known. Although in practi-
cal structures uniform stress distribution 1s
- ~l
g;:.~~
o'
0
Fig. Ct. 3 Fig. Ct.4
not common, still sufficient accuracy tor
design practice 1s provided by using the stress stresses.
relationships based on uniform stress assump-
tions. In deriving these stress relationships, Fig. Cl.4 shows a tree body atter the
the Greek letter sigma (a) will represent a block has been cut along a diagonal section.
stress intenSity normal to the surface and thus
a tensile or compressive stress and the Greek For eqUilibrium the sum of the forces along
letter tau (~) will represent a stress intenSity the x-x axis equals zero.
parallel to the surface and thus a shearing
stress.
0(1) + 2 (~l cos 45 0) =0
cos 45
Cl.2 Shearing Stresses on Planes at Right Angles.
1 cos 45 0) cos 450 _
hence, 0 =2 ('t"
i - -r (2 )
z dydx
I
/
/
2, I
I • <Judy
, I
'I
, ~
g "
:. "'o&~
/
/ ~
---"-,--
/ /
, , / ,,/ ,,
2 Fig, CL 7
a z dydx I
Ftg. cr.s
an dudy
dzdy _ dydx _
But dudy - cos Q and dudy - e in Q, whence
an = Ox cosa'1 + Oz s1n ag - - - - - - - - - (2a)
,,
Q/ / / '!- ~;;;"--
The normal stress an at a po1nt 1s always // / -, --
/ I
less than the ~tmum princ1pal stress Ox or
o"z at the po irrt ,
ZFx = 0
hence,
'" =Oz sin Q cos Q - ax cos Q sin Q
an dudy cos dudy sin Q - ax dzdy - 't'xzdxdy
'" = (Oz
or, Q -+ '!
- ox) sin Q cos Q
=0 ----- --------(4)
or,
' " = (1/2) (az - ax) sin 2 Q, where Oz Is
ZFz = 0
maximum principal stress and ax Is :nin!::!l.lm
principal stress.
an dudy sin
Q - ~ dudy cos Q a z dydx ~xzdzdY
Since sin 2 Q is maximum when Q :: 45 0 , =0---- - - - - - - - -(5)
'" = (amax - amin )/2 - - - - - - (3 ) By dividing each equation by du and noting that
(an - azl sin g - (T - TxZ) cos g = a - - -(7) Cl. 7 Mohr's Circle for Determination of PrinCipal
Stresses.
The maximum normal stress an will be maximum It is sometimes convenient to solve
when Q equals such angle Q' as to make ~ = zero. graphically tor the principal stresses and the
=
Thus if ~ = a and Q Q' in equations (6) and maximum shear stress. Mohr's circle furnishes
a graphical solution. (Fig. C1.9a). In the
(7), we obtain,
Mohr method, two rectangular axes x and are z
(an ax l cos g' - 'txz sin g' =a - - - - -(8)
chosen to represent the normal and Shearing
stresses respectively. Taking point a as the
origin layoff to scale the normal stresses ax
(an azl sin g' TxZ cos g' =a - - - - -.(9)
and Oz equal to OB and OA r-espeot t vely. It ten-
In equations (8) and (9) an represents- £he sion, they are laid ott to right of point a and
principal stress. Dividing one equation by to the lett if compression. From B the shear
another to eliminate Q' , stress ~xz is laid ott parallel to Oz and with
the sense of the shear stress on the face DC of
Fig. Cl.9b, thus locating pOint C. ~ith point
an-ax: "t'xz whence, E the midpoint of AS as the center and with
"xa on - Oz radius EC describe a circle cutting OB at F and
On
=ax + Oz + ~ '(ax
2 --V ZOZ )
.+'txz·- - (10)
G. AD will equal Be and will represent the
shear on face AS of Fib. b. It can be proven
that OF and OG are the principal stresses Omax.
and Omin respectively and EC 15 the maximum
shear stress "t'max.' The prinCipal stresses occur
In equation (10), tensile normal stress is on planes that are parallel to CF and CG. (See
plus and compression minus. For maximum an use Figs. c and d). The maximum shear stress occurs
plus Sign before radical and minus sign tor on two sections parallel to CH and CI where HEI
minimum. on. is perpendicular to OB. Ifox should equal zero
then a would coincide with A.
To tind the plane or the principal stress-
es, the value of Q' may be solved for trom
equations (8) and (9), which gives: °n(max)
TcZT
T
J;foX: R
.-Lax-az
,---
which °acts and the 45 0 plane. As stated 2
(zj
b
Fig. Cl.12
z
0x-!-- ! / -r-·o-x ,'-
. ,J~
"g ,
--- g
I
Fig. C1.10
I
0-.
crz. ax are principal
stresses.
z
On on plane (defc) ---j I
I
Fig. C1. 13
.___.. I
I
0n(min) :;.
Ox • 10000
+ - an(m a x)=120 7
-jLJ
: -2070 Z
A D
B
I Substituting values,
I
I I an = 1000~ + 0 ~ V(1000~ + O)~ . . 5000" = 5000
---+1OX20,~ ... 7070 hence, 0n(max.) : 5000 ... 7070 : 12070 pSi
On(mln.) = 5000 7070 = - 2070 pSI
- ~ 't'xz = 12000
I Fig. C1. 14
O'z .. 20000
--
--------------_.._--.. I. _-~
and O'z = -20000 are lald off equal to 08 and OA -r YZ _Yx:y2)a - (OXOyOz + 2YyzYxzYXY - 0XYyz-;J
respect 1 ve Ly .. TXZ equal to 12000 is lald off - OyYxz• - OzYXya =0 - - - - - - - - - ~ -(14)
?arallel to OZ at A locating C. rlith ~ the
mtdpo tn t of AS as the center of a circle of
radius EC a circle 1s drawn which cuts the ox
axis at F and D. The maximum no~al and shear
stresses are indicated on the rlgura.
Fig. cr.is
/
y Fig. Cl. 16
--.i..
Q''''180~15'''' _ Fig. Cl.l? shows the principal stress
o B F system which replaces the system of Fig. Cl.16.
It can be shown that the maximum shear stress
~. is one of the following values.
I I -r:nax.=::t(OL all)
- - - - - -(15)
~ ~ ;:::to",,",]
+ 1
or r: max. = -"2 CO. - 0'3)
+ 1
I " •• = -"2 (0, - OL)
"n(mini1420~
cr t max,
ron(max)' -2420
The planes an which these shear stresses
act are indicated by the dashed lines in Fig.
Algebraic Solution Cl.IS J namelYJ adhp., bdge and dcef. The
largest or the shear stresses in equations (15)
depends on the magnl~ude and Signs or the
=ax + Oz
2
- az)" + -rxz ..
2 principal stresses J rememberin~ that tension
is plus and compression is ~inus when making
the SUbstitution in equations (15).
= 10000 - 20000 :!: .,j(1000o-~-20000)) ', 12000'
2
.:~
5000 t 19200 b C
Cl. 10 Triaxial or Three Dimensional Stresses The strains under combined stresses O3,':'"e
usually expressed as strains in the direction
For bodies which are stressed in three or the prIncipal stresses. Consider a c?se or
directions, the state or stress can be defined Simple tension as illustrated in Fig. Cl.l9.
completely by the six stress components as The stress 0 1 causes a lengthening unit strain
illustrated in FIg. Cl.16. USing the same e in the direction or the stress OLJ and a
procedure as was carried out for a two-di~en Shortening unit strain ~' in a direction at
sional stress system, it can be shown that there right angles to the stress OL.
are three prinCipal stresses O'LJ a. and 0.,
whose values are the three roots of 0 in the The ratio of e' to e is called Poisson's
following cubic equation. ratio and is usually given the symbol~. thuS,
IJ. = ~I /e
Cl. 6 COMBINED STRESSES. THEORY OF YIELD AND ULTIMATE FAILURE.
,
cr, ,
)-
-i--cr, (19), which gives,
,
E' , E' /
~
/
2" ,:-- 2"
cr,
/
Fig. Cl. 20 U =..l.. + 0:+ 0: - 2iJ.(01.0.... G,GIS + 0,,01.)
:",-r--" lcr, 2E
01. Fig. Cl. 19 - - - - - - - - - - - - - - - - - -(20)
Yield Factor at Safety. This term is detined incapable of performing its required function.
as the ratio of the yield strength of the Failure may be due to rupture or cOllapse or
structure to the limit load. due to excessive deflection or distortion.
Ultimate Factor of Safety. This term is Cl. 15 Determination of the Ultimate Strength of a Structural
defined as the ratio of the ultimate strength Member Under a Combined Load System. Stress
at the structure to the limit load. Rat1o·lnteraction Curve Method.
Yield Load. This term Is defined as the limit Since the structural designer or flight
load multiplied by the yield factor of satety. vehicles must insure that the ultimate loads
can be carried by the structure without failure,
Ultimate Load. This term can be detined as the it is necessary that reliable methods be USed
limit load multiplied by the ultimate factor to determine the ultimate strength of a
of safety. This resulting load is often structure. Structural theory as developed to
referred to by engineers as the design load, date is in general SUfficiently developed to
Nhich is misleading because the flight vehicle accurately determine the ultimate strength ot
structure must be designed to satisty both a structural member under a single type at
yield and ultimate failure and either one may loading, such as axial tension or ccmpression,
be critical. pure bending or pure torsion. However, many
of the members which compose the structure at
Yield Margin or Safety. This term usually a flight vehicle are SUbjected Simultaneously
expressed in percent represents the additional to various combinations at axial, bending and
yield strength at the structure over that torsional load systems and thus a method must
strength required to carry the limit loadS. be available to determine the ultimate strength
ot a structure under combined load systems.
Yield Margin or Safety = Yield Strength _ 1 A strictly theoretical approach appears too
L1m1t Load difficult for solution since failure may be
due to overall elastic or inelastic buckling,
Ultimate Margin or Sarety. This term usually or the local elastic or inelastic Instability.
expressed in percent represents the additional
ultimate strength of the structure over that The most satisfactory method developed to
strength required to carry the ultimate loads. date is the so-called stress ratio, inter-
action curve methOd, originally developed and
_ Ultimate Strength presented by Shanley. In this method the
Ultimate Margin at Satety - Ultimate Load 1 stress conditions on the structure are repre-
sented by stress ratios, which can be con-
Cl. 14 Required Strength of Flight Structures. sidered as non-dimentional coefficients
denoting the fraction of the allowable stress
Under Limit Loads:- or strength for the member which can be
The flight vehicle structure shall be developed under the given conditions ot CQm-
,designed to have surrtc ient strength to carry bined loading.
simultaneously the limit loads and other
accompanying enVironmental phenomena tor each For a single Simple stress, tt~ stress
design condition without undergoing excessive ratio can be expressed as,
elastic or plastic deformation. Since most
materials have no definite yield stress, it is
crnmnon practice to use the unit stress where a R = stress ratio • f -(22)
.002 inches per inch permanent set eXists as
the yield strength of the material, and in where r is the applied stress and F the
general this yield strength stress can be used allowable stress. The margin of satety in
as the maximum stress under the limit loads terms or the stress ratio R can be written,
unless definitely otherNise specified.
1
Under Ultimate Loads:- M,S·:il-l. O - - - - - - - - - -(23)
The flight vehicle structure shall be
designed to withstand simultaneously the Load ratios can be used instead of stress
ultimate loads and other accompanyIng environ- ratios and is often more convenient.
mental phenomena without failure. In general
no factor of safety is applied to the environ- For example ror axial loading,
mental phenomena but only to the limit loads.
R • P/P a , where P = applIed axial load
Failure of a Structure:- and Fa the allowable load.
This term in general refers to a state or
condition of the structure which renders it
;, o ~-;.
",,-..•.1 ...;
ci. 8 COMBINED STRESSES. THEORY OF YIELD AND ULTIMATE FAILURE.
For some ot the stmpler combined load Since this theory gives the same results
systems, the exponents of the stress ratios in as the well known energy of distortion method
equation (24) can be determined by the various it is often referred to as the Equivalent
well known theories of yield and ra tjur-e that Stress Theory. The octahedral shear stress
have been developed. However, in many cases theory may be stated as follows:- In elastic
or combined loading and tor particular types action at any point in a bOdy under combined
or structures the exponents In equation (24) stress action begins only when the octahedral
must be determined by making actual failure shearing stress becomes equal to 0.47 fe'
tests of combined load systems. where f e is the tensile elastiC strength of
the material as determined from a standard
SInce the stress ratio method 'NaS pre- tension test. Since the elastic tensile
sented by Shanley many years ago, much testing strength is somewhat indefInite, it Is common
has been done and as a result reliable inter- practice to use the engineering yield strength
actiop equations with known exponents have Ft y• In this theory it is assumed that the
been obtained for many types or structural tensile and compressive yield strengths are
members under the various combined load the same. -
systems. In a number ot the tollowing chapters,
the interaction equations which apply will be Figs. Cl.21 and Cl.22 illustrate the
used in determining the ulttmate strength conditions ot equilibrium involving the octa-
deSign of structural members. hedral shear stress. In Fig. Cl.21, the cube
Is subjected to the 3 principal stresses as
C1. 16 Determination of Yield Strength of a. Structural shown. A tetrahedron is cut irom the cube
Member Under a Combined. Load System. and shown in Fig. Cl.22. Three ot the sides
of this tetrahedron are parallel to the
As explained in Art. Cl.14, the flight
vehicle structure must carry the ltmit loads 3
Without yielding1 which in general means the 3
Yield strength ot the material cannot be
exceeded when the structure is subjected to
the limit loads. In some parts of a flight
I I
vehicle structure tnvoLvtng compact unf t or
pressure vessels, biaxial or triaxial stress
condit~ons are often prOduced and It is I, - ,,
,
-+--1, I,
-n~-
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STH CTURES C1.9
prinCipal axes, while the n~rmal to the fourth For a triaxial stress system J
side makes equal angles with ~he principal
axes. The octahedral shear and normal stresses
are the resulting stresses on the fourth side.
The equation for the value of the normal
octahedral stress 15, For a biaxial stress system, f y• fS yz J
t s yx = 0
_ 1
foct-3'(f1.+fa+f,,) ------(25) j' = ,; f~ + f~ - fxt z + 3f sxz
ll
- - -(32)
vhence
SOLUTION: Condition (3) (42000) •
Since a torsional
shear stress has now f SoIvtng , p = 970 psi.
tx-ottx
been added, the new ,z f s_=8000
! ......... ---
stresS is still two PROBLEMS
dimentional, however
the given tension (1) The combined stress loading at a point
stresses are nat } f~ =:8000 in a structure is as to11ows:- f z =
principal stresses f
z
-xa -1000, t x =
-2500, t s =
2000. Determine
due to the addition the magnitUde and direction ot the
at the torsional shear principal stresses. Determine the
stress. maximum shearing stress. Solve both
analytically and graphically.
!x=41300 psI. !z=8600 ps f , !s=8000 ps i ,
(2) Same as Problem 1, but change t z to 4000
Instead at
tinding the principal stresses and t x to -3000 and t s to 2500.
and USing Eq. (30), we will use the t x and t z
stresses and use Eq. (32) (3) A solid circular shatt is subjected to a
limit bending moment at 122000 inch
poundS and a torsional moment of 250,000
inch pounds. It di~eter is 4 inches
and the yield tensile stress is 42,000,
=" 41300 1I+ 8600·- 41300 x 8600 + 3 x 8COO·
what is yield Margin or Satety.
? = 40200 pst • x.a, =~
40200
- 1 =-.01 (4) A thin walled cylinder ot diameter 6
inches is subjected to an axial tensile
Thus yield is indicated since M.S. Is
load ot 15,000 pounds, and a torsional
negative. moment or 12,000 inch pounds , l..lhat
should be the wall thickness it the
Example Problem 2.
permissible yield stress is 30,000 psi.
-Q-t,
I
d = diameter = 100w Nadat , Theory or r'tcw and Fracture or Solids.
C2. 1 Methods of Column Fatlure. Column Equations. The range AS in Fig. C2.1 Is tor a range
or Lip values of below 20 to 25, and repre-
In Chapter Ale, the theory ot the elastic sents a range where failure Is due to plastic
and inelastic instability of the column was crusntng of the column. In other words, the
presented. The equations tram Chapter Ala column Is too short to buckle or bow under
tor a pin end support condition are:- end load but crushes under the high stresses.
This column range or stresses 1s usually
For elastic primary failure, referred to as the block compression strength.
nil::; A column, however, may fail by local
Fe = --"--=-- - - - - - - - - - (1) bUCkling or crippling due to distortion of
(Lip) ,
the column cross-section in its own plane.
For inelastiC primary failure, The horizontal dashed line in Fig. C2.1
represents the condition where the prtmary
column strength is limited by the local weak-
Fe = - - - - - - - - - - - - (2) ness. This line moves up or down aCCOrding
to the value or the local weakness. The
deten:nination ot the column strength when
Where Fe = compressive unit st;~ss at failure 1s due to local weakness is covered
failure = PIA stress. in another chapter.
E = Young's modulus C2.2 Column End Restraint. Fixity Coefficients.
Et = tangent modulus Column Effective Length.
L = column length
The column strength is influenced by the
p = radius or gyration of cross- end support restraint against rotation and by
section any lateral supports between the column ends.
The letter C is commonly used to indicate the
Fig. C2.1 shows a typical plot or Fe end fixity coeffiCient, and C = 1.0 for zero
versus Lip. It the column dimensions are such end restraint against rotation, which can be
as to cause it to fail in range CD in Fig. prOduced mechanically by a pin or ball and
C2.1, the prtmary failure Is due to elastic socket end support !1ttlng. Thus including
instability and equation (1) holds. This the end restraint effect equations (1) and (2)
range of Lip values is ofteL referred to by can be written,
engineers as the long col~~ range.
A B c D
FC = , Fe = CTt.lEt
- - - - - (3)
(Lti> )" (Lip)"
I I I I
Local Crippling Limit Let L' = effective length of the column
which equals the length between inflection
points or the deflected column under load.
Fe
Then L' = Live (4 )
20,000 29 x 10 5 119. 56
30,000 29 x 10 5 97.62
40,000 29 xlO 5 84.54
50,000 29 x 10 " 75.62
60, 000 29 x 10 e 69.03
70,000 29 xl0'" 63.91
80,000 29 x 10 " 59.78
89, 600 29 x 10 os 56.49
96,000 27.55xlO" 53. 19
102,000 26. 10 x 10 " 50.22
107,400 24.65 x 10 5 47.57
112,200 23.20 x 10 e 45.15
117,000 21. 75 x 10 " 42.81
121, 500 20.30 x 10 e 40.58
125, 800 18.85 x 10 e 38.43
Fig. C2.2 130,000 17.40x10" 36.32
134, 000 15.95xl0 5 34.25
C2.3 Design Column Curves for Various Materials. 137,700 14.50 x 10 e 32.22
147,000 11.60 x 10 " 27.89
For routine design purposes it 1s con- 158,100 8.70 x 10 e 23.29
venient to have column curves ot allowable 167, 600 5.80 x 10 " 18.47
tailing column stress Fe versus the effective 173, aoo 2.90x10" 12.82
slenderness ratio Il/p. In equation (5) we
will assume values ot Fe' then find the This equation is plotted in Fig. C2.16.
tangent modulus Et corresponding to this For a given material, n,.F o . ? and S ~ust be
stress and then solve for the term L' /p. known. Then assuming values of F, we can
Tabla C2.1 shows the calculations for find corresponding values of Et/E from Fig.
17.7 PH (TRIOSO) stainless steel sheet at C2.16. For values of Z, Fa.? and n refer to
roam temperature. The reSUlts are then plotted Table Bl.1 in Chapter Bl.
in F1g. C2.7 to give the column strength curve.
Similar data was calCUlated tor the material C2.5 Non-Dimensional Con... .m Curves.
under certain exposure time to different
elevated temperatures and the results are also QUite useful non-dimentional column
plotted in Fig. C2.7. Figs. C2.3 to C2.l5 curves have been derived by Cozzone and
gi ve column curves tor other materials under Melcon (See Ret. 3).
various temperature conditions. Use ot these
curves will be made in example problems later The E:uler column equation is
in this chapter. The horizontal dashed line F = n a£t / (L1 / p ) a , which can be written,
is the compressive yIeld stress. Values
above these cut-ott lines should be sub- (L'/p)'
stantiated by tests.
F n'
C2.4 Tangent Modulus Et from Ramberg-Osgood
Equation. The problem there~ore resolves ~tself
into obtaining and expreSSion fOT Et/F from
The baste Ramberg-Osgood relationship the non-di~ensional relationship. To do this
tor Et is given as tallows: (See Ref. 1) multioly both sides of equation (~) by Fa ?/F
and equate to B a. •
1
- - - - - - - (7)
(;~~ ~. 9 1
=-;:;---;;---;:;----:-
-L.- + ~ n (-E....-)n
Fa.? 7
= B'
Fa.?
- - (8)
120 RoT.
R. T. Flg. C.4
Fig. C1• .1
--~
110 S!JOGF'
.!lee1 AJS[ 4130,' ~l4O. 4340
-1t.U-Tl'eamlnFiia-~ t:zsOOO'"
_Sl<!eI, AlSt 4130, H40; 43olO T_lIIp. Blip'-up to 1/2 II&. •
t"'" O. las
roe -- .-------,
Temp. E:::qlo....re up to 11: IIR.
-F ...QOOO<HRo~. 1---.
m
."
·-"!'":':""..'1~---
~ ~-_.--~-' -.-"-",..""-,-,,,--,-
- I 0I)lI0'P -:-
l()ll(l9F··
I~O-----R;-'l';
rie, C%.~
Steel Als:l 4130. 4140. 4340
--.--- H:iU'1'.i'nted ToFi;u;- "-15'ססOO
120--'---- I:O~---'---~
-~~""85Q01f
aecsn
so
50'
.
10-20
--_._~_.
:, ,-f "
"",,,~'--:>K~
C2.4 STRENGTH OF COLUM!'iS WITH STABLE CROSS-SECTIONS
1B0;.,.....,.....~·-· ._~,---
170-----
RoT.
'" Ro-T.--'
150 ---- •
L 010 _ 0; 125 In.
:r"mp., Exp', up to 1/2: HR.
F - "iaooco .-
'" .. -Temp.' E:qI. ~p to 1/2 fiR.
Pm'" ISOOOO
tu
_Fer -. 182000
'" Fey-'" 1ll5QOO
140·
130 '"
'"
110 ,.
100· .-
~ sc
,"
80!--'---
"
70·
''::~~'''':',.i-.:_~ :;,
40 ;....:.: . !-
"...---.:'-',---o,,-,.,..
20
...
~,----,----.-
:~_~:::~~;"""--'----:~-_.----,.--.-'-,.-
,
.•..,.........
. ".- ...... _ ..- '.---' -
~.~i-';'- ~¥---,. "-~'';;;::'''~''-i'--''~'-i''''''-'' '. ;--,-,-,,--~o~ 50- 50 ec 'go" 'ioo
10203040 "L'/f' ., eo 90
'"
L>/f'
".' .. . "'- .,
;,-,... __ ._..,._.:,._ .
' . " "
:.~:_:. ~-"':_:::.....c.:i._
'c' . ", ... " ,., " . . ."",.
t>. :,:;, .' "" ,
...:..-i------'-•.c.:::;..:::.':.:'::.:.:,:....:,-,.c..:.:.:.,-;-.:..,_
50 ,~ .
. . . . . . ., .
._-----"----'----,-.._---"--.- .-
,
,. ,".-,'
----
·~--'i'7---.--.-----'~----.,..;-··,--'·
"-'---
, '._~-
10 ~ 30 40 50 • 70 80 90
-
100
L'IP
ac - - - - - - - _ . -_._---'-- Figure C2. 10
10--rD""30~·--~60--70 ae 90 100
Vip
Figure C2. 9
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES C2.5
,
R. T.
-
f':Ilt. C2,l1
70'l5·Te Alum. Alloy BJltl''l8iOll.
t ~ O.2Ull.
Temp. EJql. • Z HR.
.. FIi", C'.12
__ 'lO'r.!.·Te_Alum. Allay Clad ~
I:! a.50!D.
1'e~. EXp • • a 1m:
30
-'-""45Q:offi ilal,·:'-""-··
20'" -.--.--
20 -----_ .._--_._."
=:-:::::;:::~
iu ........
~ W ~ ~ ~ w ro ~ ~
L'IP
10 W ~ ~ 50 W ro ~ -~6 100
Figure C2. 12
L'IP
Figure C2. 11
Fl(; C%.IS
'l0'r5-:t'll Ahzm. All>:lT {Ole Fcc;mgat
. "-'-...
---r-~Z'!li:"----~·-'· .-.-
TflIIP---'-.'Z BR.--
__,)'__,"_71O!lO ....,. __.._
--- Fcr--saooo
"<--
40:_
30
\0 20 W ~ ~ ro ro SO ~
L'II"
Figure C2. 13
C2.6 STRENGTH OF COLUMNS WITH STABLE CROSS-SECTION
40 --- --------
'" ---_._-------~~--'--------
~3040501lO108090
L'IP ,-.-,-"20--30---40----50-60-70-80-- 90
L'!P
Figure C2. 14
Figure C2. 15
-:::;1. 00 1. 00
i'-. -c -,... r-, , I"\··n- 0 i I
z
o
n
l--.'
r-, <,
t--.
f---- <, ~"
t--.
.:'-( ""
0 I"::- .,0
\ .'0 ..n"-I i ,1
1 1 I
s-
1'- 1 1
\'r .90 .90
I'--.. t---- r-, 1'- i-, h" 1".\.9 I
I,
,
! i
,I I I
"
~
..; .80
<,
l--. 1"- 't-. r-, 7':> l\ l\ \ I 1 1 ,
.80
1'--- r-,
,
<, ~\ l\ ~\\
I \ , , 1 I
0
.o-f'- ~ k 3 R: ,\ ,\ : ,
,
,,
I I .70
I 1'.5p:;; 1:::-\- i\\ ~\\
"
,
I -r-,
1
,
, ,
,
,
,
, I
.e 0 I ! t"'" ~ \'\: 1
, I
I I I · .0
!
,
, 1 !"":ile.\\
i, ,
, I 1 i ,
,
E, , ,, W- t-- r--: , 1 1 I !
E .5 0 ,I , n ~ 2. 9 .50
1 !~ I'-
, ,
I ,,
t--.. 1
,
,
i
I
I
\'-- ",N
I I I I-----:....J. , I : i I
.4 0 I" ,Ii:'- ,
~I
1
.40
1 1 ! i I ,
1 i-----. '-1 1
.I
.3 0
i
E,
I 1
, 1 I 1
,
,
! I~
[:\ - ,
1'--.'
'i-..... 1
i 'i---J.
~:
I --j
.30
,
,
E =1 .! n (..,..L..tn- ~ ,
I i 1 ,"-. 1 I .1
, ,
,
+7 F o ... 1 1\\ ~. <, ]"'--..., N 1
,
.a 0 , , , I · 'C
~ ~""'-. 'i----: I, I~
I I I I I I I
I I
I Fig. C'. I. Dimensi.onless tangent modulus stress curves. I
, ~~ 1--..---"""" -s~ I , · ,0
. 10
! I 1 I i
1 1 \\ 'R~~~ ,---"l
-!~ ~Il ,<f ~
, 1
I
,
i : , I 1 I I
o .10.20.30.40 -
.so .60 .70 .80 .90 1.001.10 1.201.30 1.40 1.50
F/F" ... Fig. C2.16 (Ref. NACA T. N.902) (1:: sis].
,.
Z
,.
r-
-,.''"'""
1i . 0 . i Z
tl
tl
-
l'l
0.9
\d'p·l~ c'"
~IJr
Z
0 .•
0
'"
Fe
0.7
ilt '"r-
-
-F- 0.8 ,"
0.7
I,
.
I'.···
;"
o
.'"
-,.
>oj
0.5
-'"
l'l
i
.70 o
r-
i i, l'l
i I
'"
>oj
l'! 'c":
0.1
.'. ! "c:
>oj
'" II
Fig. C2.17
Ee il
C2.8 STRENGTH OF COLUMNS WITH STABLE CROSS· SECTIONS
particular ~ropertles of the material EO.7/F. Thus in Fig. C2.l8, ~he deflection of
paint (0) away from tangent at nidpoi~t c
Inser-t Ing va Iue of F • Tt~t/(L' /p)a in equals unity in our assumed conditions and i~
equation (3), also equals the first ~o~ent of :he area of
the M/E! diagra~ between (0) and (C) about (0).
Et ~ ( , , . ,'\ B' (Fig. C2.I9).
( n" Et/( L '/.0 ) a) -E-J= The value of the ordinate for M/EI diagr~n
or B =.!c
n
j ,',., (£)
E \ P
----(9)
at any point x rr-cn 0 is :1 sin 1"'! LX
The total area under ~he curve is,
The use or the curves in Fig. C2.17 will
be illustrated later 1n the example problem
sotut i ons .
P
area = -,I .!
.L
"
sin rXdx II [- * cos
'
flected column had been assumed as a parabola
with unit deflection at ancpomt . Ff g . C2.20
shows the M/EI diag~~.
r
p
!
SIll.
C.UItVe:
The area or one-half the diagra~ =
1 2 P _ PL
2' ('3 LX,I) - 3EI
; [-~'I
L. I- C
Taking moments about 0 ot the area between 0
L
'~ Fig. C2.18 Fig. C2.19
and C;
5 5PL"
I ~ Yo --I=..£!:..
I 3EI 16 L = 48EI bence ,
'U
P
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES C2.9
We can now apply the same procedure to C2.6 Design Column CUrves for Columns with Non-
Uniform Cross-Section.
a colunn with non-unifonn cross-section. The
steps in this procedure for a column Figs. C2.2l and C2.22 give curves for
symmetr'ical about the center point are as rapid solutIon of two types of stepped columns.
follows:- Figs. C2.23 and C2.24 gives curves tor the
rapid solution at two forma of tapered columns.
(1) Assume a sine curve tor the deflected Use at these curves will be illustrated later
column with unit deflection a center in this chapter.
paint.
C2. 7 Column Fixity Coefficients c for Use with Columna
(2) Plot a moment of inertia (I) curve for with Elastic Side Restraints and Known End
column cross-section. Bending Restraint.
(3) Find the bending moment curve due to end Figs. C2.25 and C2.26 give curves for
load P times the lateral deflection. finding fiXity coefficient c for columns with
one and two elastic lateral restraints and
(4) Divide these moment 'falues by the EI Fig. C2.27 gives curves for finding c when
values to obtain MlEI curve. The modulus restraining moments at column ends are known.
of elasticity E is considered constant. Use at these various curves wili be
illustrated later.
(5) Find the deflected column curve due to
this M/EI loading. C2.8 Selection of Mater1al8 for Elevated Temperature
Conditions.
(6) Campare the shape of the derived column
deflection curve with that originally Light weight is an Unportant requirement
assumed as a sine curve. This can be in aerospace structural design. For columns
done by multiplying the computed de- that fail in the inelastic range of stresses,
flections by a factor that makes the a comparison of the FCY/W ratio or materials
center deflection equal to unity. Since gives a fairly good picture of the efficiency
the assumed sine curve is not the true of compreSSion members when SUbjected to
column deflection curve, the computed elevated temperature conditions. In this
deflection will di~fer somewhat fram the ratio Fey is the yield stress at the particular
sine curve. temperature and w is the weight per CU. inch
of the material. Fig. C2.28 shows a plot of
(7) With the computed deflection curve, Fc /w for temperature ranges up to 600 0 F.
modified to give unity at center point, With 1/2 hour time exposure tor several im-
repeat steps 3, 4, 5 and 6. The results portant aerospace materials.
this time will show derived deflection
curve still closer to the assumed C2. 9 Example Problems.
deflection curve.
PROBLEM 1.
(8) To obtain th~ desired accuracy, the pro-
cedure in step (7) will usually have to Fig. C2.29 shows a forged (I) section
be repeated again. member 30 inches long, which is to be used as
i
I
~~llJ
, '"' • l' I"
i " " :; ~ _. ..j .'
, .'
I~ ,i
t' . I t
>
•
1 • 't (
,I
"
••
I:il :ri, ,
" I
,
, ..
! ,n
III I, I
,
,
, .,.I , ,
,I ,I'!
'-1-, ,
,
,
. ..
.T
~
.. "'...'"o
i..
II
I
I
'
,
.1
-
o
z
r.lIo '"
, i
•, I 'I '
. . .. I '.1
.. ...
• T !
I .. 1: .. ..
• ,
! I
II II
Fig. C2.23
(QlJlnla
e"
2,0
1.8 40 i
"''"
1.8
1.' 20 ,.J ~
o
Z
i'if
2.8 ' ""j ,
1.2 . . . . .
! I ..,o
1.0 0.1 0.2 O.S 0.6
Fig. C2.25 x/L C 2. 8 1
2) , i I
';,l,'
8 I'
2.2,ill
I: i'
7
I
2.0 .' :
6 1.8
c
1.6
I'
2; 1.0
i i
1I I 1.' ,
o 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 o 10 20
Fig. C2.26 xlL Ref. Convair Structure Manu.al .l±.!o Fig. C2.2'l
EI
(j
C2.12 STRENGTH OF COLUMNS WITH STABLE CROSS~SECTIONS
[ ••
by the end restraint conditions. For failure
by bending about the x-x axis, the end restraint
against rotation is zero as the single fitting
bolt has an axis parallel to the x-x axis and
,Y thus c the fiXity coefficient is 1. For-
failure by bending about the y-y axis we have
end restraint which will depend on the rigidity
at the bolt and the adjacent fittir~ and
structure. For this example problem, this
restraint will be such as to make the end
fixity coefficient c = loS.
Fig. C2.30
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES C2.13
?or failure about x-x axiS, Thus we make use ot the curves in Fig.
C2.17.
L' : L/VC: 30/VT: 30, L'/Px: 30/.83
Case 1. Material 7079-T6 Alum. Alloy forging.
: 36
Table 81.1 of Chapter 81 s~rlzes certain
material properties. The properties needed
For failure about y-y axis, to use Fig. C2.l7 are the shape factor n, the
moduls Ec and ~he stress FQ.~o Referring to
L' : 30N'"T.5: 24.6, L' /p y : 24.6/ .60
Table Bl.l, we find that n = 26, Ec :
: 41 10,500,000 and Fo.~ = 59,500.
Therefore failure is critical for bending The horizontal scale in Fig. C2.17 in-
about y-y axiS, with trip = 41. volves the parameter,
Case 2. USing the 300 0F curve in Fig. C2.14 Using Fig. C2.17 with 1.01 on bottom
for the same L'/.o value, we read Fc = 40,400, scale and projecting vertically upward to
=
and thus P 40,400 x 4.375 =
177,000. n =26 curve and then horizontal to scale at
left side of chart we read FcIFo.'l'_ = .842.
Case 3. USing ~he 600 0r curve, Fc reads 6100
and thus P : 6100 x 4.375 = 26700 lbs. Thus Then Fc : 59,500 x .842 = 50,100, as
subjecting this member to a temperature of compared to 50,500 in the previous solution
600 0F for 1/2 hour reduces its strength tram USing Fig. C2.14.
220,000 to 26,700 Ibs., which means that
Alum. Alloy is a poor material for carrying Case 2. From Table B1.1 for this material
loads under such temperatures since the subjected to a temperature of 60QoF for 1/2
reduction in strength is quite large. hour, we tind n = 29, Fc • 9,400,000 and
:ase 4. ~~terial 17-4 PH stainless steel Fo.'T =
46,500.
forging. Fig. C2.8 gives the column curves
for this material. For L' /.0 =
41 and using Then B : ~ j 9,~O:goOO (41) : .917
the roam temperature curve we read Fc =
135,200 and thus P = 135,200 x 4.375 =
591,000 rbs • From Fig. C2.17 tor B = .917 and n = 29,
we read FcIFoo'T = .88, thus Fc = 46,500 x .88
C2.10 Solution Without Using Column Curves. = 40,900 as compared to 40,400 in the previous
solution.
vnen primary bending rat iure occurs at
stresses above the proportional limit stress, EXAMPLE PROBLEM 2.
the failing stress is given by equation (5)
which is, Fig. C2.3l shows an extrUded (I) section.
A member composed of this section is 32 inches
Fe = n'Et/(L'/p)' long. The member is
braced laterally in the I_lt~l
Since Et 15 the tangent modulus of x direction, thus
elastiCity, it varies with Fc ' and thus the failure will occur by
relation at Et to Fc ~ust be known betore the
equation can be solved. To plot column curves
bending about x-x axis.
The ~ember 1s ?in ended 1'2 x _
it - - -1t
for all materials in their ~y manufactured
forms plus the various temperature conditions
would require several hundred individual
column charts. The use or such curves can be
and thus c = 1. The
material is 7075-T6
Extrusion. The problem
is to tind the failing
1
,
,
,
"
1
-1--
avoided if we know several values or parameters stress Fc under room
regarding the material as presented by temperature conditions. Figo C2.31
Ramsburg and Osgood and expanded by Cozzone
and Melcon (see Arts. C2.4 and C2.5) for use This (I) section correspondS to Section
in colunh~ design. 15 in Table A3.15 in Chapter A3. Reference
C2.14 STRENGTH OF COLUMNS WITH STABLE CROSS-SECTIONS
to this table gives, is due to the fact that the stress existing
under a L'/p value of 51.7 ls near the pro-
A = .594 sq. in. Px = .618 portional limit stress or Et is not much
dlfferent than Ec• the elastic modulus.
L' = L, stn c = 1, L'/px = 32/.618 = 51.7
To illustrate a situatlon where the 7075
Fig. C2.1l gives the column curves for material becomes more efficlent in co~parlson
this material. For L'/p = 51.7 and room to the 2014 alloy, let us assume that our
temperature we read Fc = 38,500 pSi. ~ember has a rigid connection at its end whict
wlll develop an end restraint equivalent to a
Solution by using Fig. C2.17, fiXity coefficient c 2. =
From Table Bl.l tor this material we tind Then L' = 3ziV2""= 22.6 and L' /p =
n == 16.6, Be :. 10,500,000 and F 7 = 72,000. Q • 22.6/.618 =
36.7.
Then, B =~ / 10,~~~~go (51.7) ~ 1.36 For the 7075 material fram Fig. C2.1l, Fc =
58,300
From Fig. C2.17 tor 8 = 1.36 and n = 16.6, we
read FC/F •. 7 = .537, hence Fc :: .537 x 72,000 For the 2014 materlal, we use Fig. C2.l7
= 38,600.
_ 1 j 53,000 _
8 - n 10,700,000 (36.7) - .823.
Consider the member is subjected to a
temperature at 4500F for 1/2 hour.
From Fig. C2.17 tor B = .823 and n =
From Fig4 C2.11, Fc =21400 pSi 18.5, we read Fc /F 0 . 7 = .87, when Fc = .87 x
53,000 = 46,100 as compared to 58,300 for the
Us1ng F1g. 02.17:- 7075 material, thus 7075 materlal would permit
lighter weight of required structural materlal.
From Table 81.1, n :: 8.8, Ec = 7,800,000 The stude~t should realize that if the
and Fe/F 0 . 7 = 29,000.
stress range Is such as to make Et =
Ec' then
the bending failure is elastic lnstead of
B =1. j 2!l,000 (51.7) = 1.00 inelastic and equation (5), USing Young's
n 7,800,000
modulus of elasticlty Ec' can be solved
From Fig. C2.17 we find FC/F 0 = .74 directly Without resort to column curves or
a consideration of Et, since Et is equal to
• 7
A very cammon aluminum alloy ln aircraft The student should reallze tr~t equation
construction is 2014-T6 extrusions. Let it (5) ls for strength under primary column
be required to determine the allowable stress failure due to bending as a whole and not due
Fe for our member when ~de ot this material. to local bUCkling or crippling of the member
or by twisting tailure4 The SUbjec~ of
Since we have not presented column curves column design when local failure is involved
for this material, we will use Flg. C2.l7. is covered in a later chapter.
From Table Bl.1, tor our material, we In example problem 2, we have assumed
find n ~ 18.5, Ec = 10,700,000 and Fa 7 = that local cripplir.g ls not critical, which
53,000 . calculation will show is true as explained
and covered in a later chapter.
Th n B
e
=.1n j 53,000
10,700,000 (51. 7) = 1.16" C2.11 strength of Stepped Column.
From Fig. Cl.17 tor B 1.16 and n = = 18.5, The use of curves in Fig. C2.22 will be
illustrated by the solution for the strength
we read Fc/F o • 7 = .71, hence Fc = .71 x
53,000 =
37,600. of two stepped columns in order to lllustrate
both elastic and inelastic tallure at such
The result shows that the 2014-T6 materlal columns.
gave a failing stress of 37,600 as compared to
38,900 tor the 7075-T6 material which has a Case 1. Elastic failure.
Fcy of 70,000 as compared to Fcy = 53,000 for
the 2014-T6 material. The reaSon for the Flg. C2.32 shows a double stepped pin
7075 material not showlng much higher column ended column. The member is machined from a
failing stress Fc over that for the 2014 alloy 1 inch dlameter extruded rod made from
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES C2.15
7075-T6 material. The problem is to find the This is a relatively short column so the
maximum compreSSive load this member will failing stress should fall in the inelastio
carry. range where E is not constant, therefore the
solution is a trial and error procedure. we
Portion 2 Portion 1 will base our first ~~ess or trial on an
i ~3/4" Dia. j +1"Dia. j Portion 2
J.
averagb Lip value.
The stresses in each portion are, Referring to Fig. C2.16 and using 0.465
and n = 16.6, we read Et/E = 1.0, thus Et =
t, = 1000/0.7854 = 1280 psi E and thus no plastiCity correction for
Portion L
t. = 1000/.4418 = 2270 psi
For Portion 2, f./F o . ? =59,500/72,000
These compressive stresses are beloN the = .826
proportional limit stress of the material so
Ec 15 constant and our solution is correct. From Fig. C2.l6, we obtain Et/E = .675
whence, ~t = .675 x 10,500,000 = 7,090,000.
Case 2. Inelastic Failure.
EL 10,500,000 x .0491
The column has been shortened to the El a = 7,090,000 x .0155 = 4.7
dimensions as shown in Fig. C2.33. The
diameters and material remain the same as in
Case 1.
From Fig. C2.22 tor aiL = .5, we obtain
Portion 1
8 = 5.6.
J
OUr guessed strength was 25,300 lb. Our C 1-1/4 -.058 A 1-l/4·. 058 B 1-1/4 •. 058 D
guessed strength and calculated strength must <
be the same so we must try again. I
Trial 2. Assume a critical load P ~
23500 lb. I
t, = 235001.7854 29900 * E F
I+-- 30" - _......-
)1;
t. 23500/.4418 ~ 53100 Fig. C2.34
.Fig. C2.27 shows curves tor ~lndlng the In Fig. C2.27 we need term I..l. L/~I. The
end fiXity coefficient c for two conditions L/El refers to ~ember AB. Thus ~ LIEl =
of known end bending restraint. (258,000 x 30)/29,000,000 x .0367 = 7.28.
~e use the upper curve in Fig. C2.27
To illustrate the use or these curres,
a simple problem will be solved. since restraint at both ends of member AB
is the same. Ttus tor I..l. LIEl = 7.28, we
Fig. C2.34 shows a 3-bay welded steel read end fiXity coefficient c = 2.58.
tubular truss. The problem 1s to deter.nine
the allowable cQmpressive streSS for member Then L' = Live = 30;'; 2.58 = 18.5.
AB. This strength is influenced by the fixity
eXisting at ends A and E. The diameter and o for member = .422 inches.
wall thickness of each tube in the truss is
shown on the figure. The material is AlSI L'lp = 18.6/.422 = 44.0
steel, Ft u : 90,000, Fty = 70,000, E =
29,000,000. From column curve in F1g. C2.3, we read
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES C2.17
L! /p = 30/v'2':'8 x .422 = 42.5. Then from (b) For 1/2 exposure at 300 oF:- F O • 7 =
Fig. C2.3, Fe = 56,600 psi. Since the far 29,000, Eo : 9,500,000 and n = 26.
ends are less than fixed, the assumption that
far ends are pinned gives fairly accurate For the above two cases (a) and (b),
results. determine Et (tangent modulus values) n-om Fig.
C2.l6 and then calculate and plot column curves
In a truss structure all members are for these 2 material conditions.
carrying axial loads and axial loads effect
the ability of ~embers to resist rotation of (2) Fig. C2.36 shows
their ends. Art. A11.l2 of Chapter All the cross-section ,"t
explains how to take account of the effect of a compression i
of axial load upon the stiffness of a member member. Calculate
as required in calculating the end restraint the failing com-
coefficient ~. pressive load under
the following
C2. 13 Columns With Elastic Lateral Supports. cases:-
Then L' : L/ve: 24/1"2."92: 14.08 The pin anced single stepped column as
shown in Fig. C2.37 is made Qf AISI-4130
L' /0 : 14.08/0.125 = 113 no~allze1 steel, Ftu = 90,000, Fcy =
70,000. Determine the ~tmum compressive
Fe
3
E
= (L'n /p)3
3
:: n x 29,000,000 = 22,500 load member will carry.
(113)3 t:>-psi 1" Sq. ~r
(7) Same as Problem (5) but member Is exposed (11) The cylindrical tapered member in Fig.
1/2 hour to a temperature of SOOoF. C2.39 15 used as a compression member.
If member Is made :Tom AISI Steel 4140,
(8) Same as Problem (5) but change dimension Ftu = 125,000, what Is the failing load.
(a) to 10 inches, and L to 14.28 inches.
Find the failing compressive load for
the doubly stepped column in Fig. C2.38 L
if member is made from 7079-T6 hand
forging. -+- ae 12 -----oj<- b·9
30"
:I
Fig. C2.39
C3.1 "'AI"
~ - ----.--- -
-7'r:.~ ~. ~,: .~;_:' ';. ;" ,/;~. ;.~:,~~-,> ~~~,~:,;~' . ~. :</:/~<:.~~£~ :,~/'.:,'~ ~.;.~~:::-;.- _":. :1~.:i: ~= ';-"1,(" ~. ~,~" ~ ,:~>~ -' ; ..-, ~<~~~~~~:~~~:~:~~;t~~.?~f- :;,~~;~.
C3 2 YIELD AND ULTIMATE STRENGTH IN BENDING
strain at the -ncoo i nt of each str-t». .c t th the given bending ~oment, and al~J ~J~ teter~lnir.g
strain known on each strip, the stress existing t~e true internal resisting ~:~e;.t Jf a bea~
can be !ound by use of the stress-strain curve section, structural deSign 2~gi~eers ~ake ~se
in Fig. C3.2. The total loae on each str1;J o~ a fictitious failing ben~ir.s stress ?o'
then equals the stress times the area of the wtn cn is re re n-ec to as a cccujus 0:' ructu-e
strip. The internal resisting ~o=ent then stress in pUTS bendin~, T~3n the ultimate
equals the s~tlon of the load on each bending ~om3nt that can be developed jy a
strip times the distance from the strip to given beam cross-secticn ~~d a ~lven ~aterial
the neutral axis. is M = FbI/C. DeSign cu:ves fer finding ro'
the modulus ot rupture, are given later in
Table C3.1 shows the detail ~alculatlons. this chapter.
If the neutral axis has been selected in the
correct position, the values in col~~ (6) Since there are ~ny flight vehicle
ot the table should add up to zero since total materials and all kinds ot SD2Des used in
tension must equal the t~tal compression on structural members, the basic approach for
the beam cross-section. The small discrepancy solution as illustrated in Table C3.1 becomes
ot 740 poundS in the summation of column (6) very time consuming. DeSign engineers always
1s not enough to change the location of the search for simplified methods which give
neutral axis or the total internal resisting sufficient accuracy. Thus Cozzone (Ref. 1)
moment appreciably. Column (7) gives the has developed a Simplified procedure for
total internal resisting moment as 56735 in. finding the modulus ot yield or rupturing
Ibs. when the strain is limited to the .010 bending stress Fb' The method is Nicely used
strain as previously discussed. The stress in t~e aerospace industry in structural
at this strain tro~ Fig. C3.2 is 49000 psi. deSign.
USing this stress in the well known beam
=
fonnula M rr/e , we obtain M: = 49000 x 0.785 C3.4 The Cozzone Simplified Procedure.
= 38450, which is much less than the true
value or 56735. The Cozzone method in its si~ple5t fern
assunes a symmetrical rectangUlar bea~ section
TABLE C3.l and the same shape of the stress-strain curle
1 2 3 4 5 5 7 in both tension and compression. ?1;. C3.3
represents the true bending stress '~riat1cn
Strlp Unit Res. over the bea~ cross-section when failure
Slrlp
No.
Area y E Stress F :crA Moment occurs. Cozzone now replaces this tru~ curve
.. "A" CO M= Fr by a trapizoidal stress variation as sh0wn :n
.058 0.935 3075 2760
Fig. C3.4. The stress f o is a fictitious
1 .00867 53000
2 .102 0.840 .00773 52500 5350 4300 stress which is assumed to exist at the neutral
3 .135 0.75 .00685 52100 7025 5040 axis or at zero strain.
4 .153 0.65 .00591 51500 7870 4820
5 .165 0.55 .00494 51000 8410 4310 Fig. C3.3 Fig. C3.4 1_ Im _!
..[]IbT
6 .180 0.45 .00398 43000 7740 3200 m
[tMAX=fml~-oI~fb'"
7
8
9
10
.185
.195
.197
.200
0.35
0.25
0.15
0.05
.00302
.00205
.00108
.00012
33200
22800
12500
3200
6140
4450
2460
640
1920
945
280
10
[2
-- --- N·A.lO
IMAX=f
f
o
r i ;
L
11 .200 -0.05 -.00084 - 7250 -1450 130
12 .197 -0.15 -.00181 ~ 17800 -3510 680
13 .195 -0.25 -.00276 -29500 -5750 1650
14 .185 -0.35 -.00374 -3'5500 -6560 2540 fMAX fMAX N. A. ""-
-40000 -7200 3510 True Assumed Assumed
15 .180 -0.45 -.00470 Stress Stress Stress
16 .165 ~0.55 -', 00566 -43000 -7100 4170
17 .153 -0.65 -.00663 -44800 -6850 4710
Fig. C3.6
18 .135 -0.75 -.00759 -46000 -6210 4880
19 .102 -0.84 -.00846 -47200 -4810 4210 f MAX[ ---;-- ~
20 .058 -0.935 -.00937 -48000 -2780 2690
Total 3.140 740 56735 ~ I .....-// I
Col.
Col.
1
3
Rod divided into 20 strips. 1" thick.
y ,. distance from centerline to strip c. g.
-
•
OS
?
--- /
~Assumed Stress- I
I
Curve
-- -
_" .:~~).r,[.:~':'-~:'-"":{-~~;,~~;~'-~'.~1,:~~;'~·(~·~~~~~::":~:f~~,·L~'::··+;·~'~~;''''''5~';·'~:i··;t''':~;';~~~I.:--.~",,:~- -,. ';"~+ -~;/~~,', r-"-:"...~., '."~,'. ~,~v
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES C3.3
-I- - ,
,
V-
k Factor for Some Typical Shapes
,.
I.
u
,.,
,
Z~
--_ .-
rectangle (r) and a triangle (~) as shown in
tne r tgur-e , _.-1-
......
Let, Mb = total lnternal resisting moment.
internal m·y,nent developed by
-
_--
.._........
.. _-_.. -
,,,,._-
.._-
mr h',1
portion (r). ','rrrr-rrrrrrrrrrrrrrrrl
mb = internal mom~nt develooed by
portion (b).
. '
.. . r--
,
(5) Beam-column, curvature and shear lag to give the value of fa. The value of ~o fro:n.
effects are considered negligible. this chart o~eration ~ives f o 2 29000. T!':er.
from equation (3)
C3. 6 General Accuracy of Method.
FbYield = 65000 + 29000 (1.5 - 1) = 79500 p~i.
(1) It 1s exact for a rectangular section
under pure bending with moment vector parallel Then yield bending moment = Myp = Fb1/c
to a principal axis.
Thus Myp = 79500 x .0938 = 7460 In. lb.
(2) For double s~etric sections under
pure bending and moment vector parallel to a For finding the ultimate resisting
prinCipal axis, the accuracy should be within bending, we use Ftu which is 75000 as the
5 percent .. value of f m in equation (3). Ag~in going to
Fig. C3.17 to stress of 75000 on stress-st~ain
(3) Single symmetric sections will vary curve and the vertically dOem to to curve, we
tram practically exact to definitely uncon- obtain f o = 70500.
servative (moment vector normal to axis ot
symetry) . Then F b (Ul t )· 75000 + 70500(1.5-1)= 110250 pSi
(4) For sections subject to combined Then MUlt ': Fbllc = 110250x.0938 = 10370 in.lb.
bending and axial load, the results will vary
tram practically exact to conservative. Let us assume that 1s 1s desired to limit
the strain in the extreme tiber ~o .03 lnc~es
(5) For unsym:netr1cal bending, with and per inch. What would be the bending ~oment
without axial load, the results will vary from developed under this limitation.
practically exact to conserlative.
From Fig. C3.17 for a unit strain of .03
C3. 7 Example Problems 1n Finding Bending strength. the corresponding stress from the stress
strain curve is 74700 and the f o stress 1s
EXAMPLE PROBLEM 1. 61200.
A rectangular bean section is 0.25 inches Then Fb = 74700 + 61200 (1.5-1) = 105300
wide and 1.5 inches deep. What yield and
ultimate bending moment will the section Then M = FbIle = 105300 x .0938 = 9900 In. lb.
develop when made trem 7075-T6 extruded
alum1nu:n alloy. EXAMPLE PROBLEM 2.
Solution: The modulus ot bending stress is The symmetrical I beam section in Fig.
given by equation (3), (a) 1s subjected to an ultimate design pure
bending moment M': 14000 in. lb. ~h~t Is
Fb = t m + to (k - 1) - - - - - - (3) the margin of safety if
the beam is made' of
k _ 2Q. - 2 x 0.. 75 x .25 x .375 x = .1408 magnes i urn forging AZ6lA
.0938 which has Ft u = 38000
1.375 125
. . ..fI*.
2!JEl
- DC - (.l.) .25 x 1.5'(1/.75)
12 and Fty = 22000.
k = 1.50 Ix = 1~ x 1.375 x 2 3 _
1 .125
The value of k could also be found in Fig. 1
C3.7. 12 x 1.25 x 1. 75 3 Fig. (a)
From Fig. C3.l9, for t m = Ftu = 38000 we bendIng strength was developed. From Fig. /
tind in projecting vertically downward to to C3.l9, the unit strain when stress Is 38000
curve gives r 0 = 23700. Then subt. In 15 0.35. Then sInce plane sectIons remaIn
equatlon (3) plane atter bending, the unit straIn at· point
.50 lnch from neutral axls 15 (.5/1) (.035) =
Fb = 38000 + 23700 (1.17 - 1) = 40770 .0175. From Fig. C3.l9, the stress existing
at thIs strain is f = 31000 psi. A linear
Mult = FbI/c = 40770 x .358 : 15000 in. lb. variation ot stress as used in the flexural
equation would give halt the maxImum stress
llar.ln of Safety = M"M1t _ 1 = 15000 _ 1 = -'-(J7 or 38000/2 = 19000 psi as agaInst the true
• 14000 stress at 31000.
Assume we desire the stress inte~sity at
a point 0.5 inches trom neutral axis it full
.
50
e ..
A
10
0.112 0.0"
e ""lA
0.01 ,."
0.02 0.0" 0.011 0.08 0.10 0.12
""'lA
.. ~-=~@i:j~i4io:~£"" '-';:'::'':-o:;::~
,::-:2:~t;-'·:·~!;~:",,,~:f~~i;~'I~:;::t··· -,:··":':;:"-'=if::~~,,.='000::'::..:
E~;_:,~·.-'i:;g~":;if~?::;~:~:". ,~::.:'.::: :~'f:..- -==....:=::-::. ~-:-:d~'::i::2:J.~l:~""
'; ...~.:l
"s s
- .,:"-"01 .:..~.::-t:~~;E~E:~ii~E';ii,--'=-'::E~-§§:"o;§f~¥':'~~"=r_~"'::;,":''':-;-':
;~:" ,0-::::;i~:Ol;';:.~::f0~g'~;::·:~'-:.,~q~,i~~E"5'fE2f~ ~~f.#~",~'o-'
,." '.N 0.011 0.08 0.10 0.11 o 0.01 0.02 0.03 0.04 0.05
, ""lA e 10/. .
._....:~,~_.__ .i_ic~L~~!_~:,Al.1ar~~,PIaIe'~
. ..:..r:so-~ : ""::"""'"'- _
-c-=---c,---,.--.-:---:,:-~;~,.,.~==~~= . . ·-'-,=Cjii.f~
-----------_._.~-
;::~'.~·i·:~~"''';:,:b;~,".t<'':\~.« ,:.,: ;, j
10fE:o:"'~-' -':,~::~::::.~~~.:.:.=~~.~~_.
-.-.
o.~.;·~~.c~t~-=~;~
0.01 u.oa 0.03 0.04 0.05
e WID W"
~~-,-oo;:~...,..:.;,:.:~'----'---,---o
~;U!imi",,' Hljnd'''''''''hlliri
. :';.;." ..::.:,. :':;.u.•. Sl$'lD'-;'-"'; ., ~ ~I '
-~--...,.~--,-,---c-;;- -" ,,'~'" .""";",-,.~;-~''"~'~'-'~---.
-::~~~~-~,::::·.:~=~t~;;~=~
-,,--. -
rJ~~;~L;'7:,':--;'2:~;':";---.
,'+riL.~~J,',_
.... . ..,
,<0
- -"-"~'?:"-~--?"'=--"-----"-".- t1Iia.-~k;'. r·~-
--" --- .....---.- ----~." ~
';
".-
.
J
<0
0.02 0.03 0.008 O.012 0.018 0.020 0.024 0.028 0.032 0.035
, W .. I':. \lDcbel/lnl:/ll
.....::...-'=~ __ _ 'C--,_·-., ;,
Portion 1. Fig. (d) shows
how the lower portion (1) ~ 1-0.1"
!.' is made a symmetrical T ," , T
section about neutral axes 278,,1,1.39
by adding the dashed
portion. The internal -~r-u''-{--N.A.
1.39
bending resistance will be _ J:. \
tound tor this entire
Fig. (d)
section in Fig. (d). one
halt ot this amount will
then be the true moment developed by portion (1).
E:CAMPLE PROBLEM 3. Unsymmetrical Section.
I =121 bh ~ _ 1
- 12 X 0.1 X 2.78
a
= 0.178
Fig. (b) shows a tee beam section,
symmetrical about the vertical axis. It the
material is 17-4 PH stainless steel, what
1,/c, = 0.178/1.39 = .128
ultimate bending moment will be developed it Q, = 1.39 X 0.1 X .695 = .0965
bottom portion is the tension tlange.
2Q., = .193
--<..1_1-1/2 ---+I Ell"'. 0154
---- = 0.193 = 1 5
i Lo. 0.128 •
I"
IN.A. --
0"
----1------
iO.610
From Fig. 03.22, Ftu = 180000 which equals
y=1.39 to trom curve = 156000
1____ ___l
o. 1"~ I....
.~
E1. =.03
Then Fb, =t m + to (k - 1)
Fig. (b) Fig. (e) = 180000 + 156000 (1.5 - 1) =258000
M • (Fb, 1Jc, It =258000 X .12B X
The neutral axis will tirst be determined.
0.5 = 16550 in. lb.
3Cf.-
C3.8 YIELD AND ULTIMATE STRENGTH IN BENDING
The stress for a unit strain of .0154 As explained in solution (1), for e =
fram the stress strain curve in Fig. C3.22 is .0154, f m :: 172000 and to :: 129400.
172000, and to :: 129400.
Fbll = 172000 + 129400 (1.15-1) 191400
Then Fb, = 172000 + 129400 (1.14-1) = 190100 M, 191400 x .0842 = 16100
M, =~ (Fb,,1,/e,) = 190100 x .1704 = 16200 Mtota1 =M 1 M3 = 16620
+ + 16100 =
32720 In. lb.
Total resisting moment:: Ml. ? Mil :: 16550 ?
16200 • 32750 In. Ibs. EXA11PLE PROBL~ 4. Fig. C3.24 shows an un-
sym~etrical I beam section. The material is
Solution 2. 7079-T6 aluminum alloy die forging. The
upper portion 15 in bending compreSSion. It
Instead or making each portion a will be assumed that the compressive crippling
s~etrical section as was done in solution (1) stress for the outstanding upper legs of the
and dividing the results by two, we will tind section is ,65000 psi. (The theory and method
the internal bending resistance at each portion ot calculating crippling compressive strength
as is when bending about the neutral axis ot is given in another chapter.) ihe ultimate
the entire section. Equation (3) now becomes deSign bendIng moment is 16500 in. lb. Find
tor each portion at beam section, M.S.
Tr-;f~i
15
r, :: ~ X 1.5 x .61:5_ 1 1- • --+1 Upper portion: (See
t~ 161=c2 Fig. I)
',' 0.1
~ x 1.4 x .51:5 N.A_
_1 x 1 .~-
0.822 :.i"
Fig. (hi I NA -'3 x . 822' - '3
1
= .0514
-LN.A.-- - --N.A.
xL4 X 0.722 3 = .102 Fig. (0
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES C3.9
INA = ~x
- 31 x
0.75 x 1.178"
0.65 X 1.078
3
I!_N.A.
1.178
C3. 9 Section With One Axis of Symmetry. with Moment
Vector not Parallel to Either Axis.
INA = .137, I/c = .137/ 1 1 axis, the procedure in this case Is the same
as for the double aymnetrlc case.
1.178 " .1164 "0.75-; To. i
Fig. 0) = Rb x + Rb y
k = .1425/.1164 " 1.222
cedure can be followed I (2) Resolve the given moment M into compon-
which IS qUite con- I Fig. C3. 26 ents Mx p and Myp '
ssrvat I ve . Y
(3) Follow the sa~e procedure as betore.
(1) Resolve the given ~oment into cry,nponents
about x and y axes, or Mx and My. (4) The stress rat10 Rb = Rbxp + Rbyp
(2) USing Mx follow the procedure as given
in the exanple problems and find Rbx =
MxlMax' where Max is the internal moment
C3.10 YIELD AND ULTIMATE STRENGTH IN BENDING
C3. 11 Alternate More Exact Method for Complex a shear V = 600 Ibs. The problem is to find
Bending. the margin of safety under this combined
loading.
A beam sectIon when resl~ting a p~e
external bending moment bends about an axis For pure bending only the stress ratio is
that is called the neutral axiS. No matter
what the shape of the beam cross-section for
- M -- 15000
Rb-I1 14000 = ' 933 (the va.iue
1 crt u'5000 i s
a
any given external moment, there is an axis
obtained from example problem 2).
about which bending takes place. The general
case involves an unsymmetrical beam crass-
section and material which has different The stress ratiO in shear is Rs = fs/F au ,
where f s is the flexural shear stress and Fsu
stress-strain curves in compression and the ultimate shear stress of the material. -
tension in the inelastic range. The neutral The problem therefore is to find the value of
axis therefore does not pass through the
centroid of the cross-section and thus the ts'
method at solution is a trial and error
approach. The solution procedure is outlined The equivalent trapeZOidal and tr iangiuar-
in Chapter A19, Article Al9.l7, and there tore bending stress distribution will be determined
tor the deSign bending moment of 14000 in. lbs.
will not be repeated here. Also the chapter
dealing with the design at beams with non-
buckling webs explains and illustrates .how For a triangular stress variation,
the ultimate bending resistance at an entire Fb =Mell = 1400010.358 =39150 pSi.
beam section is determined.
From example problem (2) the shape factor
k was 1.17. On Fig. C3.l9, the curve for
C3.12 StrengtbUnder Combined Bending and k = 1.17 has been plotted. Starting with the
Flexural Shear.
Fb stress of 39150 at the left scale, ~un
The previous part at this chapter has horizontal to an intersection with the k =
dealt with the determination of the strength 1.17 curve, the prOjecting vertically down-
at a beam section in pure bending. The usual ward to intersections with the stress- strain
beam deSign problem involves tlexural shear curve and the fa curve to give 35800 and
with bending. In tlnding the true internal 19700 for fa' The stress results are shown
resisting moment, the Cozzone simplified graphically in Fig. ~~.28a and Fig. C3.28b.
method derives a trapezoidal bending stress
distribution which will produce the same t 11-- 'i--.{ 19700--+1 I 1.... 16100 /+39150..
J _:""==Tt~](
internal resisting moment as the true internal
bending stress system. A triangular stress
system is then derived which will also give
the true bending moment.
19700 x 0.875 x 0.125 = 2155 C3. 13 Strength Under Combtned Bending Flexural Shear
and Axial Compresston.
14100 x 0.5 x 0.875 x 0.125 = 771
The SUbject of the ultimate strength
Total Force 8906 Ibs , design under combined loads is treated in
detail in a later chapter.
For Case 2:-
A conservative interactIon equation tor
Portion (a) combined bending, shear and axial load is,
t
coincide with center line axiS, thus use
trial and error method. ...... 0.1
CHAPTER C4
STRENGTH AND DESIGN OF ROUND, STREAMLINE, OVAL
AND SQUARE TUBING IN TENSION, COMPRESSION, BENDING,
TORSION AND COMBINED LOADINGS.
:~:~¥~~~;;~~->~C~;2~fu~;;i~
is not an important factor. Round tub mg is
the best shape tor transmitting torsional 700
torces and thus Widely used in control systems.
Round and square 'tub Ing permit Simple con-
nection or end fitting design. The metals g 600 j;~~~~~~-"~'"~;;~=:~~~~~~~~i
~ : f~:~g~~~~;~:'~~2-~~~;:_;~:~;~rf~~~~
industry has made available a large number at
diameter and wall thicknesses and thus the
structural designer has a large number ot sizes
to select rroa. 2 ~~"c:"_j':::.::::3·':=:-=-::: .. -. . ' ·'··-'"""E--- --. :-:-"'':;:';:;'1
~ f"'"-''' ,",",.."",=, -.~.''''''',~,.,,~'''''
C4.3 Design for Compression. any tendency of the test curve to rise rapidly
tor very short lengths where failure is by
The strength of members with stable cross- block compression. Table C4.l shows the
sections when acting as columns can be cal- resulting short and long column equations
culated by Euler's equation it the bending after values of Fco and E have been substituted
failure Is elastic, or E 1s constant (eq. C4.1) in equations C4.3 and C4.4 and E in the Euler
and tor inelastic bending faIlure, Euler's equation. The column headed transitional
equation ~lth the tangent modulus Et replacing L'/p, represents the value at L' /p where failure
E (eq. C4.2) checks test results. (The student change trom inelastic to elastic failure or, in
should refer to Chapters AlB and C2 for theory other wordS, it is the dividing paint bevneen
on column strength.) the so-called long and short column range.
Thus if the equations are used, the Lr/p value
(C4.1) must be known in order to select the proper
(L'/p)"
equation.
Fe' (L'/p)" - - - - - - - - (C4.2) C4.5 Column Formulas for Aluminum Alloy Tubes.
U Is the effective length and equals Fram (Ret. 1), the basic short column
L/~ where c is the column end fiXity equations for aluminum alloys are:-
coerr icrent •
Fe • Fe o [1 - 0.385 (L' /p)/nV E/F e o] (C4.5 )
Long -and Short Columns
stress of the material. MathematIcal curves The equatior~ for determining Fco are
were then derived to tit the test results. given in Table 04.2 (from Ref. 1). The table
Engineers referred to the .columns which tailed also indicates which ot the three short column
by inelastic bending as short columns, and equations to use for the various aluminum alloy
thus reterred to the equations that tit the ma.ter-ta.Ls ,
test data as short column equations. The
columns that ~ed by elastic bending were To illustrate the use of Table C4.2, the
then referred to as long columns.· The test column formula for 2024-T3 aluminum alloy
CUI"le tor long columns would follow the Suler tUbing will be derlved:-
column equation (C4.l) and thus tests were
not necessary to establiSh allowable falling From Chapter 82, we r ind the following
stresses in the so-called long column range. strength properties for 2024-T3 tubing,
Thus over the years short column equations
based on test results have been presented by Ftu a 54000, Fey. 4200C
official government agenCies for use in
structural design. The official publication From Table C4.2, the equation tor Fco is,
tor the aerospace tield is the Military
Handbook IUL-HDBK (Rer. 1). Fco ~ Fey (1 + ,; Fey/lOCO SUbstituting,
C4. 4 Column Formulas for Round Steel Tubes. Fe o • 42000 (1 + V42/1000 • 42000 (1 + .205)
1025 .. , •....•...... 55 36 36 36,000-1. 172(L'/p)3 ", C4.3 122 266 x lO'/(L'/p)' ·... (d)
4130 , •... , ..... , .. ,
Heat-treated e alloy
95 75(0) 79.5 79,500-51. 9(L'/p) J,. 8,. C4.o4 91 286 x 10o!l/(L'/p).11 ·... (d)
steel. , , ...•• ' ...
Heat-treated alloy
125 103 113 113,000-11. 15(L'/p)' .• Co4.3 73 288 x 10'/(L'/p)' ·... (d)
steel, ..•... '" .•
Heat-treated alloy
150 132 145 145,000-18. 36(L'/p)" .. C4.3 83 286 x lOo!I/(L'/p).II ·... r')
steel ......•..••. 180 183 179 179,000-27. 95(L'/p) a •• C4.3 58 286 x 10o!l/(L'/p).11 ·... (d)
a Equatfun C4. 1 may be used in the short column range if E d Not necessary to investigate for local lnBtabUity when
is replaced by Et obtained from the combined stress- O/t 50.
strain curves for the material. e This value is applicable when the material is furnished
b L'/p '" L/py'C: L'/p shall not exceed 150 without specific in condition N (MIL·T-6736) but the yield strength is
authority from the procuring or certificating agency. reduced when normalized subsequent to welding to 60
c Transitional L/p is that above which columns are 'tong" Itsi.
and below which they are "short." These are approxtmate
values.
are given in Tables C4.3 and C4.4. or inelastic instability of the column as a
whole. As the slenderness ration L'/p gets
Table C4. 3 Column Formulas for Magnesium- smaller, the Fc stress increases. Now it the
Alloy Extruded Open Shapesa
diameter of the tube is relatively large and
GENERAL FORMULA the wall thickness relatively small or, in
other WOrdS, it the diameter/thIckness (D/t)
K(Fey) • ratio 15 large, tailure will result by local
Fe' (L',jJ)m crippling or crushing ot the tube wall and
this local tailing stress is usually repre-
(Stress values are in lesi)
sented by the symbol Fcc. The values or Fcc
in general have been determined by tests (see
Alloy K a m Max. Fe design charts for Fcc versus D/t ratio).
MIA .• , ,.. .
. '
AZ3IB, AZ61A, AZ80A.
. . 180
2,900
1/2 1.0
1.5
0.90 Fey C4.9 Design Column Cha.rtl!I•
1/4 Fc~
AZ80A-T6, ZK60A-T5 . 3,300 1/4 1.5 O. 6 Fey In design, column strength charts are a
a Formulas given above are for members that do not fail by great time-saver as compared to substituting
local buckling. In the various column equations, thus a number
at column charts are presented in this chapter
to facilitate the strength check at columns
Table C4. 4 Column Formula for AZ31B-H24 and the strength deSign of columns. Fig.
Magnesium-Alloy Sheet C4.2 is a chart of L'/p versus Fc for heat
(1.05 Fcy)'(L',p ) , treated round alloy steel tUbIng. Fig. C4.3
Fc'" 1.05 Fcy- Is a similar type or chart for aluminum alloy
4 n"E round tubing. Fig, C4.4 gives column charts
Max. Fc .. F cy for magneSium alloy materials. All three
charts are taken tram (Ret. 1). Figs. C4.5
and C4.6 represent a turther S1mplication ror
the design at steel and aluminum round tubing.
C4.7 Short Column Equations for Other Materials.
C4.10 Section Properties of Round Tubing.
For other metals tor Which short column
equations are not available, the use of Euler's Table C4.3 gives the section properties
equation, USing the tangent modulus St can be of round tubing. A tube is deSignated by
used (eq , C4.2). Refer to Chapter C2 for giving its outside wall diameter (0) and its
intornatlon on how to construct column strength wall thickness (t). Thus a 2-1/4 - .058 means
curves using this equation. a tube with 2-1/4 inch outside diameter and a
wall thickness in inches ot .058. Since a tube
C4.8 Column Failure Due to Local Failure. is symmetrIcal about any axis, the polar moment
or inertia, which Is needed in torsion problems,
The equations as presented give the equal twice the rectangular moment of inertia
allowable stress due to failure by bending of as given in Table C4.3. For weIght comparison,
the column as a whole and the action is elastic the weight at steel and aluminum tubing is
STRENGTH & DESIGN OF ROUND, STREAMLlliE. OVAL AND SQUARE TUBrnG
C4.4 IN TENSION. COMPRESSION, BENDING. TORSION AND COMBINED LOADrnGS.
5052-All Tempera
6083--All TempeJ'l!l
~AUTempera
54M-All Tempera AU Products F•• (1 +..,; F,./lOOO) Equation 1.732."/ElF.. Equation
MM-All Tempe" C4.5 C4.8
6Ofl1-T4, T"Sl, T4510,
T4511
2014-T6, T651
~T8,T81,T86,~1 Sheet and Plate*j
7075-T6, T6S1 Rolled Rod, Bar and F •• H +../ F .,/1333) Equation 1.346• ../E / F.. Equation
7173-T6, T6S1 Sha.pes; Drawn C4.6 C4.8
Tube
Rolled Rod, Bar and F ,.(1+ ,; F .,/2000) Equation 1.22. ."/E/F.. Equation
Shapes; C4.7 C4.8
6Ofl1-T6, T65l, T8StO, Drawn Tube;
T6S11 Extrusions
'"IDdudes eJad u weUu tNIre sheet and plaC.l. Equation C4. 8 may be used. In tbe 'hort uolumn rllZI.P If E' Is
TtmsttioDal L'I_ Is that above _hleh the columns .,. "1onC'" llnd l'l!plaeed by £, obtained from the eompreulva StrlSSoStraincurve for the
beknr _b1ch t.be1 ..... ''short''. matllr1&l.
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES C4.5
C4.11 Some General Facts in Tubing Design. Tension Allowables Near Welds in Steel Tubing (X-4130)
"0
... \ --
rcce at Weld Cutoffs
I
"" 1\ FlU' ISO k$1
II
I Heot Treated After Weldin
Heat Treated Cutoff,ksi Reference
10 Flu,ksi
160
i \., 180
150
144
120
REF. 1
150
\
125 95 r-
Normalized
100 76.6
Cutoff Meattreofed
95 67.5 I-
ofter welding
1\ Ftu~ISOksi
"" \~ .0 '0
-.... ...
1
·
u
eo
\1 a0
a• 20
"
~u
0
. '\
-,
<, ....
0 20 60 80 I00 I20 4
10 I60 eo
I '00 220
L'Ip
50 I I
~
• I I
Euler curve, E = 6.5 .lIOe psi I----
.5 40
AZ80A·T5 and ZK60A-T5 I
••
~ 30 _ 0 96 Fey
in
VV e.lfruded shapes (except hollow)
I I
AZ80A·F e.ltruded sncpes
I' +-
;
c
E
.2 20
o
~V .I' ~~Round tubes,
(except hollow I
I I
Fey = 10 ksi
I I
U'
"""'~ I
u.."
10 I i
.
Fey !
o I I I
.
o
~ 8
;;i
~ 6
1;'"
z 4
."'''"" 2
~
"6
(.J
(:=2
C=l
"'~
is 20
.'"
0
0 20 40 6{) 80
~ 16 COLUMN LENGTH - INCHES
~
[g
.
g 12 38
.'"
ez ~ 32
."'''""
z
8
..
'" 28
0
:Ii
"0
,.J ~ 24
(.J [g
C=2
C=1 0
0
20 40 80 80 .'"
0
20
i'l
COLUMN LENGTH - INCHES
.'"
o 18
z
z
"0 50 '".'"
.
0
'" "'
z
:Ii
12
"'Z
Q 40 "
,.J
0 8
< (.J
"'
" 30
.
g 4
'c.". 20 C=2
z C=l
."'
''""
z 10
20 40 60
COLUMN LENGTH - INCHES
80 100
:Ii
"6 Fig. C4.5
(.J STRENGTH OF CHROME MOLY ROUND STEEL TUBES
0
C=2 FTY '" 75, 000 PSI FTU • 95, 000 PSI
C=l 0 40 80 120 160
COLUMN LENGTH - mCHES
STRENGTH & DESIGN OF ROUND, STREAMLmE, OVAL AND SQUARE rusmo
C4.8 IN TENSION, COMPRESSION, BENDINCi, TORSION AND COMBlliED LOADINGS.
7000 U .r .- 'Ftt;;;q4;6--:-
~ . _.=-~~:l::::~=::.~r~~-:~~
'" 4000
0:4
.J
•. ~_.
.-':~~:~_"';-:_~~~~f=;!~ IlIiz;; .
'"
<
~ 3000
':':_:i:::;:~':;:::.t.r:~; . *'::::::::;
~-':::""'l··.":- -:-...:,o:.-:~_--: ';::
;:::~?t·':;-:~l::~:-4~J
'" ~: ~;:~ ~:i~~-~~~-~5
:.·::.-:::::'=.~f~~§
~+-:-=-"':-~::t;~:- e,:=...-; -=-·t:=:J
~1000
'" -_~~ __.~. ::-:::":'---:L:=~:~!=~:f~NW-~*~
- -=::::~I
o
120304050 60 20 40 60 80 100 120
COLUMN LENGTH· U'lCHES (C .. 1) COLUMN LENGTH - INCHES (C '" Xl
35000
14000
~
Q
S 12000
..
Q
1.. ,
30000
Q 25000
~'OOOO ~
.J
..
.J
.J 8000
~ 20000
'"<
'"
~ ~ 15000
'"<'"
.
8000
..
.J
<
~ '0000
i... 4000
;:
.J
e 5000
'"'" 2000
0 20 40 60 80 100 120
COLUMN LENGTH - INCHES (C'" 1)
35000
~t:.~:~
~
Q
30000
.''""
Q
25000
.;-wi::~-F~*~
~ ~. ~~~~g~_::
; 15000 '
..
'"
<
Eo< 10000
I
~
~
:l 5000
lii" - c:
: :.1__
a 20 40 60 80 100 120
COLUMN LENGTH· INCHES {C .. 1}
,--.-. ,~""'--._; --- ---~ :-::--._---:--~ - ..:......_- ~ -~~ ~~--:::-:-~ -= ---;-=:- -=--"-------
/'
Wilight
b
Wlltght
LIt/iOO i_. Lb/IOO iu.
Dla.
1/'
lla ••
.022
A
.OU1.
.P
.0.10
I
.000103
I/Y
.oooan
D/'
11.3. ...••• ...•••
StOilI Dural Dia.
1 1/8
Go••
.035
A
.20:13
.P
••501
I
• OU.5
I/Y
.0113•
D/.
53 ••0
et••l
5.73 ........
Dural
....
.058 .3311
.nu•
........
3/' .02' .03063 .1231 .000442 .003...6 13.3. • '8- .31 .eee • 3616 .16168 2'.80 10.47 3.7"
.035 .03731 .120' .000548 .003tU 10.12 1.08 .3. .083 .4873 .&342 .18787 .200110 22.'0 13.25 •. 73
.08018 .116. .000882 ,003836 1.85 1.43 .'1
••• •
.....7' •••
.0.1 .3003 .8800 .14288 .14288 0&0.80
.04.162 .167:1 .001180 .004 ...1 11.0 1.17 .06a .353' .8a61 .18816 .18896 30&.0&6 10.03* 3.5S-
1/'
.eas .0:1.113 .11141 .001390 .006659 14.2' 1.0&6- .0611 .3961 .614.5 .U5a .1851'" 30.75 11.11 '.00 ;>
•0.0 .0""3 .1618 .00178' . DOn..... 10.20 1.'6 .eaa .....18 ••783 .2300 .2301 2.... 10 14.18 5.06
Z
.0811 .68611 .n..... .2588 .2588 21.011 18.11 5.'18
'j' .02'
.035
.05252
.08.81
.2113
.2090
.002346
.002833
.007603
.009065
22.30
11.85
1.48
1 ••• •
.M
.e8·
."
2 1/4 .048 .3388
. s....
.7'1.3 .2052 .1824 0&5.80 '.5' 3."3
.'"
;>
e-
-
.04' .08867 .204" .003704, .011852 12.77 2.51 .051 .17153 .2.01 • 2134 31,'0 1l.SO· ".011.
'j'
.058
,028
.10331
.08351
.2018
.25U
.004185
.OOCl ..5
.0134.211
.011053
10.11
28.80
21•• 3
1,80 ...
1.05
....
.065
.0.3
.095
.4... 21
.5851
....32
• 172a
.7887
.762•
.2685
.3322
.3741
.236•
.29:13
,33211
34.80
27,15
23.70
12.84
1'.01
18.22
4.52:
5.72
8.61
>oj
g• '"
,0311
.OU
.07882
.107.1
.2631
.2485
.00503.
.008161
.OU.21
.01718a
21 ....a
16.30
2.23·
3.06 1.0' ' a 1/3
....
.0"1 .3773
.4450
.8887
.8636
.2834
.3318
.2287
.2655
61.00
43.10
10.88
12.60
3.83
4.50
•n ;>Z
,..
.. .,"'""
.0118 .1260. • aUII .001601 .02:031 12••'" 3.51 l.a.
.065 .13961 .a433 .001278 .oa201 11.63 3.'6 1.42 .oal .4.7:1 .8813 .3681 ,2960 3•••11 14.0._ 11,03* w
.083 .6302 .U60 .4801 .3888 30.10 11.86 8.38
.7' ....16 • aD. 3..
..'.," "-
7/. .038 .07"61 .2988 .006S81 .016289 31.:13 2.11 • Ot5 .7178 .8&0' • 6191 a8.30 1.27
.035 .0.238 .2913 .008181 .018853 26.00 2.82· ••4·
.0". .12716 .a925 .010882 .024&1 11.86 3.60 1.2' a 3/4 .04' .4168 .8661 • 3783 .215• 68.10 11.71 4.20 0
.058
.066
• 14881
.18Ml
.28'•
.28615
.01248"
.013853
.02853
.03UI
16.10
IS."
4.2a
•• 88
1.11
1."
....
.058
.0811
• 4901S
• 6"'83
.8964
.......
.06211
....34
.4446
..au
.'18•
.3233
.3698
.4501
47.40
42.30
33.111
13 ••0
lIS.IO.
19.70.
4 ••8
6.55-
7."
Z
1 ,036
,0...
,068
.10811
.U640
.34l"
.3387
.012388
• 018594
.019111
.02"n
.0331•
.03822
38.68
20.40
11.20 ....
3.01. 1.01.
4.15 1.48
I.U
•
.095
....
.058
.7124
.6361
••393
1.0403
.8.01
.11802
.50"
.3868
:18.85
n.70
2:1.48
18.18
8~O3
5.42 - ...
'.", ...0
>oj
-'"
.17164 .3337
.085 .190.3 .3314 .020870 .OC193 16.S8 1.41 1.93 .0811 .19'3
.7808
1.0380
1.031a
.6451
.8091
....305
•15398
4'.20
36.16
18.•6
21.66.
6.08
1.70- '" r-
1 l/B .035 .11.85 .3858 .01782 .03188 32.10 3.40· 1.2:1 .0.$ .8870 1.021. .'1118 .810" 31.61 24.66 8.78
...0 o
.049 .16564 .3808 .0:u.02 .04270 a2 ••1S 4.8'· 1.88 .120 1.0U7 1.0181 1.1278 .7ISII 25.00 30.18 11.00
.058 .184.2 .3780 .02176 .04833 11.40 6.11 1.31 >oj
.085 .21860 .31156 .03062 .01421 11.30 '.14 a.lIo 3 1/... .058
.066
.5818
.8504
1.1:187
1.1283
.1410
.8261
.41580
.11077
66.io
60.00
18 ••7
18.40
Ii. 8.
8.11 0 '" .,""
" -
1 1/4 .036 .13360 .4287 .oa481 .03.... 3!i.10 3.7'· 1.35· .083 .la51 1.laOI 1.0381 .837. 38.16 23.3" 1.35*
" '"
,049
.058
.066
.1848'
.2172
. 2420
.«250
.42lf
.411'
.03339
.03887
.G4280
.063...2
.06187
.08818
211.60
n.n
19.2a ....
6.23· 1••7·
6.111 a.20
2.46
31/2
.095
....
.120
.'418
1.1100
.1014
1.1180
1.1074
1.2141
1.1721
1 ....472
1.0348
.7211
.1808
.6114
34. aD
27.10
63.80
28.18
33.43
18,81:1
26,20
8.62
11 ••5
7."
Z
>oj .,
n
r-
-"'" '"
1 3/' ,035 .1473 .4738 .03309 .04814 31.26 •• 17 1.41 .083 .tilt 1.20U 1.3012 .7t36 42.20 9.01 d
.04.9 .2041 • 4691 .0449a .08534 28.05 6.78* 2.01 .095 1.0182 I,aO«3 1.473. .8ua 38.85 :18.70• 10. as.
.058
,065
.24.00
.2875
.4681
.4838
.011213
.06113
.011183
.08341
23.70
al.15
8.80
1.68
a.43
2.70
.120 l.a742 1,1868 1.8220 1.0"11 21.15 3.,00 la.89 '"
z >oj
3 3/4 .086 .16:15 1.3031 1.a171 •••14 67 ••0
o.
21.30 7.80 d
1 1/2 .035
.04.
.0118
.0811
.1'11
.2234
• 2628
.2.30
.6181
.5132
.1510a
.11071
.04324
.058111
.08841
.076:18
.06181
.07141
.011:11
.1007.
4a.1O
SO.80
alS.86
13.01
4.118
'.32·
7.ta
•• 30
....
1."
2. ae-
2 ••'
.083
.095
.120
.168:1
1.0908
1.3885
1.2.88
l.2.n
1. a841
1.8080
1.8228
2.21185
.151.
.8722
1.2036
46.20
38.60
31.2i6
:17.
30.84.
3'.70
'.81
11. Q.tIt
13.1.2
n
>oj
d
.'",
.0.3 .36.5 .6018 .08305 .124.07 18.08 10.47 3.14
• .065
.083
.8035
1.0214
1.3116
1.3852
I.U67
1.8517
.1779
.119'
81.60
...a.20
22,75
ai, 16
8,12
10.3a
1 5/8 .0311
.0.9
.1741
.3-&26
.5622
.111175
.05528
.015..0
.08803
.0.2:19
46.40
33.16
4.96 1.77
' ••7· a.48
.081S
.120
1.1615
1 ....21
1.3810
1.312i1l
2.2228
a.711U
1.1114
1.377'
42.10
33.33
33.n
41 ...0_
11.78
14 ••Olt
'"
,058 .a8511 .1164'" .08118 .lQ801 a8.00 •.ot 2.81
.065 .3188 .51120 .0.707 .11948 :16.00 '.06 3.:13 . 1" 1.1321 1.4551 3.873a 1.1408 31.16 48.10• 17.6S1t
• 1/...
1 3/4
.'"
.031
...021
.11185
.un
.606D
.119811
.08938
.147151
.079a7
18••1
60.00
11 ...0
6.32
•• 08
1.n
4 1/2 .15' a.1281 1.63'. 1.0282 2.2347 28.80 '0.40* n.lla-
.049 .2618 .8017 .09...78 .10832 35.70 7.42* 2.611· ... 3/.. .1" a ••••• 1.'143 1.0a13 2 ••583 26.a5 18.25_ 27.201t
.018 .3083 .6986 .11048 .1:162'" 30.20 8.73* 3.12
.0611
.083
.3441
.43..7
.11982
.1801
.12230
111138
.13.71 28.80
111.10
'.n
1:1.33
3.48
4.40
• U IIUIIJ)lIlD ",81.0
.11.:11'
STRENGTH &: DESIGN OF ROUND, STREAMLINE, OVAL A..~ SQUARE TUBING
C4. 10 lli TENSION COMPRESSION BENDllia TORSION AND COMBINED LOADINGS
Pa ~ Ftu (weld factor) (area at tube) Pa = FcA = 53000 x .1849 = 11650 lb.
= 95000 x .841 x .2628 = 21000 lbs. In this case heat treating produced
additional strength, whereas in Case 2 it did
!I.S. = (Pa / p )- l = (21000/18500) - 1 = not. The reason for this Is that failure
0.13, thus compression 1s critical. occurs in the inelastic stress range and heat
treating raises the material properties in the
PROBLErl 2 inelastic range. The end fiXity changed the
column tram a so-called long column to a short
Case 1. Tube size 1-1/4 - .049, L = 40 in. column.
c = 1
Material: Alloy steel, Ftu = 95000 The strength could be found also by sub-
Find ulttmate compressive load it will stituting in the short column equal for
caITy • 150,000 steel as given in Table C4.1,
Case 2. It tube was heat treated to Ft u = = 145OCO - 18.36 (66.8)' = 63000 psi.
150,000, what compressive load would
it carry. PROSLEM 3
Solution: Fig. C4.5 cannot be used since Case 1. Tube size 2 - .065, L = 24, c = 1.5
Ft u 2 150,000, thus we will use Fig. C4.2. Material Ftu = 95000. Welded at ends.
L' = Live = 40NT = 4. From Table C4.3, Ultimate deSign load' 25000 lbs.
p = .425 and area (A) .1849. = What is M.S.
L'/p 40/.425 = =
94. From Fig. C4.2, Solution: L' = L/ve = 24/~ = 19.7
using the 150,000 curve, we find Fe = 32500.
Then Pa = =
FeA 32500 x .1849 = 6000 lb. Thus From Fig. C4.5 for L = 19.7 on c = 1
heat treating the tube rrcm 95000 to 150,000 scale, we project upward to the 2 inch tube
for Ftu did not increase the column strength. and note that it intersects the horizontal
For a L' Jp =a 94, it is a long column and weld cut-off line which gives an allowable
failure is elastic and E is constant. column load at lett scale of Pa = 26700 lb.
Failure in this case is local cripplIng
The strength could also be calculated by adjacent to welds at the tube ends.
Euler's equation from Table C4.1.
M.S. = Pa/p = 26700/25000 - 1 = .[17
Fe =286,000,000/(L'/p)' Case 2. Assume tUbe 1s heat treated to Ftu
= 286 ,000,000/(94) a - 32500 pe i , the = 125000 atter welding. wbat Is tube
same as preViously calculated.------ strength.
Case 3·. Same as Case 1, but assume tube Is L'/p = 19.7/.6845 = 28.8
welded to several other tUbes at its
end and that the end fixity developed USing Fig. C4.2 with L'/p =
28.8 and pro-
is c • 2. jecting up to 125000 curve, we again note that
horizontal weld cut-oft line is intersected
~ - --
,~,>,. ',~-'. <, t ~:: -,'~~'.'~'>.~"_\~":~~~~~~'~'~:.>~.~~'~~'_'>.~~" ;,:.':;,~.,,~ ~:">~::';'~_:~~,~~,:"'f ':-~~_. :~ ~ .. _·,:~~::~,,~f~:~/J:~~(:
------- .. ~ - . -~-- ---------
giving Fe = 95000, whence Pa = 95000 x .3951 The results show that 1-1/4 - .035 is the
= 37500 Iba , lightest. Since there is danger in welding
.035 thickness to the other heavier tube gauges
If the tube had not been welded at ends particularly the engine ~ount ring which is
the dashed part ot the column curve could have usually relatively heavy tor this size engine,
been used, thus giving additional strength. a ~ln~um tUbe thickness or .049 will be used,
hence the 1-1/8 - .049 tube will be selected.
PR08LEM 4
Consider Member (2)
Fig. C4.7 shows a steel tubular engine
mount structure for a 1050 H.P. radial engine. Design loads 11650 tension and 4250
The ultimate design tension and compressive compression. Since the tension load appears
load in each member as determined tram a stress critical, the tUbe will be designed for the
analysts for the various flying and landing tension load and then checked for the
conditions are shown in ( ) adjacent to each compressive load. The Ftu of the material
member. The true length L ot each member Is equals 95000 psi. Since the engine mount in a
also shown. Using chrome-moly steel tUbes, welded structure, the strength at the tube ad-
Ftu = 95000, select tube sizes for the given jacent to the end welds must be reduced to .841
loads. It Is cammon practice to assume the x 95000 ~ 80000 psi (see Table C4.4).
column end fixity c =
1 tor engine mount
members, since the mount Is subjected to Hence tUbe area required = 11650/80000 =
considerable vibration and the true rigidity 0.146 aq, in. From. Table C4.3, which gives
given by the engine mount ring is difficult the section properties of round tUbes, we
to accurately determine. select the following sizes:
Case 1. Tube size 1 - .049 round. If a round tUbe is exposed to the air-
L = 24, c = 1. Material 2024-T3 stream, the air drag 15 about 15 times greater
Find failing compressive load. than if it were given a streamlined shape, thus
streamline tubes are used when the member is
solution: The column curves in Fig. C4.6 are exposed to the airstream.
slightly conservative because the equation used
was slightly different tram the equation now Streamline tubes are drawn from ro~~d
epec i ri ec In (Ref. 1). tubes. In designating a streamline tUbe, the
round tube from which it was made is used and
=
Use L 24, we read for 1-049 tUbe a then the fineness ratio is also given. The
tailing load ot 2800 lb. fineness ratio is the ratio LID, which dimen-
sions are shown in Fig. C4.8. The ~ost common
As a second solution, we will use Fig. fineness ratio used is 2.5 to 1. Table C4.4
C4..3. L' = L/ve = 24./v'"""L • 24.. L' /P = shows the section properties of streamline
241.3367 = 71.3. From Fig. C4.3, we read tUbing having a fineness ratio of 2.5 to 1.
Fc = 20000. Then Pa =
FcA =
20000 x .1464. = Figs. C4.9 and C4.l0 give curves for finding
2930 lb. the column failing stress F c and the local
crushing stress Fcc.
The answer could be obtained by substi-
tUting in equation C4.9, Fig. C4.8
From Fig. C4..3, Fc = 22S00 Case 2. Same as Case 1 but change material
to 2024-T6 aluminum alloy.
Whence Pa = 22500 x .14$4 :: 3300 lb.
For this material we use Fig. C4.l0.
Fcc tor D/t = 20.4. from Fig. C4..3 • 38500 (not For L'/p =
58.5, we read Fc = 26000 psi.
crttical). For D/t = 38.5, we read Fcc = 37500 (not
critical). Thus Pa = 26000 x .4972 = 12900 lb.
Case 3. Same as Case 2 but change material to _ C4. 16 strength of Oval and Square Shaped Tubes
magnesium alloy, Fcy :: 10,000. in Compression.
For L'/p = 58.5 and using lower curve on Tables C4.6 and C4.7 give the section
Fig. C4.4, we read Fe = 7600. Then Pa = FcA -= properties for square and oval Shaped tubes
7600 x .1464. = 1110 lb. respectively. For the design of these Shaped
ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES C4.13
Streamline ·TABLE C4.4 SECTION PROPERTIES OF STREAMLINE TUBING (Fineness Ratio 2. 5 to 1)
14116 con
,
Minot
J~
•
Minot
~"i¥
RO""G
0.0
Wail
Dec'! Ca,. Mo;ot
s.rm
A...
Mil\Ot "0
,m
..,~
z •
MalO'" M-o.,ot
,
Minot •
Miaot
"'.
~, IS \.4185 .10t8
0021 OIOl .14lS a ..
049 la .1079 }668
"'" "" ~,
" "50
."',
"51 .nu
1> on i.isco
"" ""u
.
10 .3140 0018 0111 .11S; 0118 0195 Hll 1611
sooc "'''
049
osa
18
11
1272
.1489
4113
.5061
OO}1
0041
0148 1703
0168 .1671 "" ,W,
'50'
0181
nIS
·H9S "" a
''''
7178
.m
.1815
"
."
,
ms
.cse
20
ta
I.H8S
, ...
.1Oti!
"'"
.m
.~Bl5
00"
00",
~,
0151
0100
0118
.10211
1912
1919
'm!
01)9
0278
0151
0145 .,..
ocr .,,,
.oco 1J( 049
". "
18 3.706S 1.)714
....
"" ,L""
.490'
1.414 .J'"
1'9)
"''''..
1.1170
"" "" coer 0253 2140 0341
"'" .'" 4S73 a
.1914
". ""
- "" '.m ""
~, }.l11
02n 1105 ~,
.." OOM
"'''
~,
" usn
2165 .7]~9 OIOl
00..
0320 1181
"'" "''' 4'"5
'" "
~,
11 4.04~~ 1.1141
""
'''' "'''
,m
1.0}7 2311
"" "''''
0246 2559 .o;SS .'51'\04
,." " ''' wn 1.~86
. "'"
lJ.( 0}5 20 714l
"a"
.083
ollS 1509
". ""
.18049
1111
.14ll)
618~
.US
OIl)
0145
Ol11
0406
14TI
241\ .0618
""
.0714 ""
.no
", »,
"'''
1.08S1
\.}959
''''
3.691
4.146
.3925
.4814
,,,
~,
"
Hi
Hi
015
~,
". """
~,
on
20 U'\04} 78S7 ien
,~,
aacc
.167S
..""""
8158
.""
.0118
.01P
.0181
.019$
0300 .UlS
0400
.~I
.0'50'1
.cns
.1m
1745
..719
i096
"'" ''''
~" oeec
""
cts
"'" "'''
~n om
""
,.,.
""
.S581
""os,
m
.nc
""
"a"
,.
4.3818 1.8S71
"'"
,,, .
.5816
S1~8
1.1800
Lm
2.111
50m
},Wl
4,011
""
.2961
""
"50
50"
20 1.0m 8S71 1611
"" "" "'" '"
1.5186 H6} .61~1
~,
.ota "
"" ""
.oe
"
U"
.1617
19H
50"
7S93
.m '''"
"" ""
1.1106
0239
.0316
.048} .304S
.OSSS
.0611
.on7 .29n
.)014
.1989
"'"
"'"
.nu
.08IS
''''' "'''
'''''
"",
6161
.61)1
.,,,
.6108
~,
"" ""a
""
16 4.noo 2.lXlOO .7014
89LO
1,0162.
2.38'
3-019
).4S'
.m,"
.~1}4
.. .....
.nc 1.1141 4.})1
,."" """
'" ".
om 1.64l4 m.
Hi 0)5
.oe
". ""
20 1.191)
"" '''' ""
"" ",.
814S
"
0116
051}
})6S
3314 """
use "'"
''''' "'"
. ."
.1416
1B~~
"'" on8 0448 3181
",n
.1l14 .1\6~9 Hi .065 16 '.0570 1.1418
"" USB
...,,,
" .oa
"" ""
~,
.m ,.,
1.08}
1.361 .0410
0118 m6
088} .3191
m'
".,
124S
,~,
1732
6)74
.",
.cs
.110
"na "" ,m
L"""
1.168S
l.I,1
}.708
4.6S1
~"
".
0976 .3149
" 1.~51
.", ,"
. !•• 1.7641
"''' "'50
..
116}
.on
",U ""
.0498
..,
Hi 20 1.]600 1.<:'000 )Ii}}
~,
". "rr
1MB
,""
"50 0611
ens
}583
'M' Ill}
nl1
.1116 O6S
.os}
16
14 "'''
1.0114
''''' "'""
1.170 50 .08106
.3551
lSlS
1~91
.176S
, "'"
.14U .116} .095 I} 1.16n
"""" "" .4341 1.478 0511 1042 3461 1191 .1101 .nc 11 L..,.,
L...,
"" " 49}9 1.619 'm .11'\04 HI8
"" ."''' ,.,,,
.7061 .156
.188
~
H. 1.14S1
ms Ion 6818 0515 .}902 ua
"''' ,.,,, i: .'"
"" '''"
Iii 10 1.S18S 1.0714
~, 1!11 .0411 .0719 }SSI .168S .1296 .06' '.1.~1l LUllS
""
8113
'"'
H M683 1.5113 USH
aeea ""
OSlO
". "" "'" .cs 1.;141
~,
""
.m """
lH9
j9H
,."
""
1.203
I.Hl ""
~"
r.sn ''"'
1.699
"'"
.10}4
1141
119B
.1SS6
,~
.Hl'
.}7+4 ""
.1731
.I!m
nc
!2111
8117
.SIS6
.8115
. ,.'"
''''
.,"
,us ,."
51 1.6~ll
2.1111
1.'~403
1.9411
".,
2)4 049
.osa
18
""
J-O}4} I.l.8S7
...,
.l388
J""
1.l~1
\.]~S
'm
"'''
.1144
.1l19 .14SO
.<S"
.1111 ,os.
9Jt5
sm
""
"" ""
.l8l'
.,,'" LSI7
"'"
~,
.oe
" 1.911 1161 ,," 50" .9ll)
'" "
.uc u
.6431
"''''
11116
1.7}0
.1198
.15)0
.2020
2411
,."
.,." ""H70
"""
.42"
30
I
60 40
I ~50"
r-, 50
I I i"
20 40 so 80 00
,OJ' I
30
o~ .
o 20 40 60 80 100 120 14<l
Lip
STRENGTH & DESIGN OF ROUND, STREAMLiNE, OVAL AND SQUARE TUBING
C4.14 iN TENSION, COMPRESSION, BENDING, TORSION AND COMBffiED LOADINGS.
-,
~.
... G••
," .035
'"ra 6549 oes
OW, 0881
'"
'"
'"
..
"uzo ~~
"
~"
=
om
coos
oon
oo~
~,
c..
".
.ce,
.
"" t.cu
9O'l,
1.195
.1051
"OJ
.1203
.10101
"'"
. ".
-
re oou Ute
~, ~u
"'''
~,
.081
O'l, 1} 1.i17
1436
,,.,
.1608
1915
"'.
'" "'''
."" u
.on
"
"
w ~" "''' .c-s
"'''' "'" '" "'"is
1.065 .1451 .1658
~,
"""
oon ",. .oss
.
,~
~,
"'"
.,"" oon al41
.. "
1.25'
1.402
.1691
.1877
.1932
.1146
,."
-
00" 1.774
""O'l' .23U