UNCLASSIFIED

AD NUMBER
ADB805294
CLASSIFICATION CHANGES
TO: unclassified
FROM: restricted
LIMITATION CHANGES
TO:
Approved for public release; distribution is
unlimited.

FROM:
Distribution authorized to DoD only;
Administrative/Operational Use; FEB 1944. Other
requests shall be referred to National
Aeronautics and Space Administration,
Washington, DC. Pre-dates formal DoD
distribution statements. Treat as DoD only.

AUTHORITY
NACA List dtd 28 Sep 1945; NASA TR Server
website

THIS PAGE IS UNCLASSIFIED

 \-
%
TEC/HNICALNOTES
NATIONAL ADVISORY COMMITTEMFOR AERONAUTICS
“%

No. 927

_“_ .f-
....---=
-.
-.___
.-
DETERMINATIOli
OJ?STRXSS-STRAINRZLATIO”NSFROM

—
By H, N. Hill
Aluminum Company of America

.-

CLASSIFIED
~CUMENT .
sdooumntoonbaina a~aaoified
in mation&ffootingthp National
%
Def8nQ of bhaUnitedS@iaswithin
L ing of theEspionage
thsme Aot,
USC50t31 ●ndS2, I Ktranmnission
..- orthere ation 2
in anyma% r ta-L %M%
pereonieprob tedbylaw,Infor-
mationeoolta iedmaybeimpart-
+ edonlyto %‘rsa inthemilitary
andnavalervio s of theUnited Washington
f\
::~?ekZe~;$ti~’~~~~~~~~~‘e’r=r~1944
Governj4nt wholuwe legitimate
inter~at therein,and-to United
Stah6a oiti;ensof kno loyalty
apddiaoretion who of
\ oessity
mustbe informed thereof.’~
●
 NAT10NAL ADV.’I
SORl?COMMITTEE-
F“OR‘iLERdN&.TT
IC)
S .----

l~03?FSETt~
YIEIJD
.’ . STR~NGTH VALUES’
By H, N. .Eill . ,.
. -,.
“ 4
,7.

The shape of-the s~ress-strai~‘c~rvefor””amaterial

. mes of considerable.interest”’to
is sorneti. the designing
engineer. This is particularlytrue when he is dealing “-
with members or elements subjectedto compressionswhich
become unstable at stresses-beyotid the el@6tic ra”ng~. ” It
is o%vious that the two compressiveproperties comm~nly
recorded (modulus” of elasticityand yield strength)are
insufficientto define the shape of the stress-strain — ““‘-
curve.
,. .. —
.. ..
Ramberg and Osga@~, .(r,eferarice 1-)ha%e tikec!l the fol-
lowing tkr8e-paramete’r~ equation for “expr@’ssing the “
relation between stress and strain for styessesup to a
value slightly great-~,:,,tha:~ the,yield strength.-. of the
material: ....’ ---- ,--—
— ----
T ...
.’
,
. , ..;: i.”.
.L.. . , “,. +
---,?:/.-.;, ~-=:.: - :.
.”... —.
..
,-’ .“,. ..-.. . .
..--..
—--..
....
..... . ..=.-:
+.. .“”---- --
..:.
}-, .. .. ‘, ..—
--.-.+-..
.:=.=.--A- ..-—
‘6 n
e=-+lf~ (1)
E .’-. :0 E
.,.. . ,,, - ---- _____
●
where ,.
..,! -.
e unit strain ., -...
. .- ,’ ----
s unit stress —. ...--.
..L. . .
,. .-
E young}.smQdulus .: . .’,. ... . . . .. ,. :.: - --
. -— ..>—_

mld .-
.,, ., ,, _.,
~ ~an~i”)n .,,. ‘.
-_.
A__~,-. _
constants”‘,for,a
given curve ,.. ._
!.. ——.
Ramberg and Osgood have-evaluated the corista’nts ~ ~tid---n —.—.
. in terms of two secant yield strength, yalues determined
-
for s~o.pes.of017E &nd..0~85E’~’i‘~:.’.
... ‘.. “,-’~“”~
“’~i=
“-,--;~- ~:~:=~.: .----
~=,: .. .. . .
.:.L.:..::.
L-,5.+:
:“--;
, - ., :“.,,*“g—-,. .=-=.-- ----...
Since yield st.ren@hvalues deterpinned6Y the off- =
set method are’”mtich
. more commonly,.
.,,, use~ tha-nthe ‘secant
.. ..>.
.. . “. .. . -- ..-
----- .-
.,... . . . ...=.
 *

—-
... .—

8.

L<
....—.—
and
—

. —.
—.

., .
,. - -—
a.
.l@g,-d ,., .
.= log .&.-+
,.
., ..,,,-’-
.
.-., ; .,”n
...~o~
.... 0.I~. ... ..7
:. :- .(f2a),
E ,-’0 “‘“ .+ #~-...-
.
., ... ,,. --- . .-—. --
,. .,, .- . .“.”..,.. ---”- ‘“-”” ..
Subst~”tutfngS1 ‘“a%d dl, “-an”d
.,-. ;Tg2 ,..
-&:.“-
;“:<
~ -l’nt’o
:*q”;;a:;-: ““w
tion (2a) gives two simultaneousequationsin K“’and n;
which when solved for. n ,yfeld the relation
....
‘..
,.
.- -, .... . .
“~lo g da
()~ & .
n=
s~ ,,... ,.-, :.
log —
()SL
.. . .
From equation (2), K can be expre~~ed-.i,,
y...-y.:=,. ,:..
‘m-:
.!

K=kor X=dl
Sa ~! ,.i~:;-,.:&:$ “:”.; ,..”. .(4) -’.
’....
e
()F ()T .
.,
,.-
.’:
”,:”:
.
.. ,.
, . ,,..,
.. :., ....
.;...:---.-,-..
:.?,,:
..:,... -,
: ... ... .... :::>
L:
-;
.. ; >...
*-T--: A .,:-...
:,>..
: : .
. -..-:
*. -.
-
Su%stitti”l%’n”g
“equa~~o’n (4) “in-t
0.~,eq,tiiktoi”..(l~:)..
giveff:.f.or
,“t
W“ ?
equation of the stress-straincurve
.“ J.
,. ..;, .-.. ..-,!
“ . :“-;”!.
i: : :.;.
,.,
., ~~’::.:4,
., ..-, -.

~.:.‘: fi ‘d’;’-”;
,:.

(;’)-
.. . .
~-”~,”+
.“..’
e
.E 2
:.”-.
,~’u,~ ‘;__d
“E 1
()
<“
~la)
......
. . ..
 .,.
the valuefbr ,,. ~ ‘~af,~g’gi~en,
. ,,.. .. by””e~uation (3).,
.. ,,” ,,, ..
:A.’l.ogical’lo
ffset‘valueto.ugefor,det&rmlning yield
stren-gtha in a!d~it’ion ‘to”
‘“the
commonlyused Value of
0.002 (da), would be half this value or 0.001 (all).This
offset value will locate a point on the ‘s~ress-str~fn
curve,.betweeh” the elastic range-’arid the conventiontil
yield strengthvalue, which is in the region of the curve
which is of greatest significancein plastic buckling.
Substitutingvalues of 0.001 an~ 0.002 for al and dz,
re~pectively,the stress-stra”in curve can be exyressed

in which ~
E m
e =-s + ().002 -% n
s~
. . .
.
0,301 .—
●
n= ., (3a)
4 ,t”” SZ ““ “ ~ -~--’>.
. log — ._
() S1
.,,
RAT!IIS
Ol?~!I!RESS
TO ~YTX”CTIvEMODULUS
.:
., . . .- . .“---
..In problems dealing w~t~ &ckling” &~’”st:r6s8es beyond:-
the elastic range”of the”material, the siqnifica.nt feature
of”the stress-straincurve is the r?lation between the
stressand-the “slopeof the durbe at that.stiess. The -
S1O e of the curve is commonly-called the t,.angedt ’modulus
(ETT . ..—
,,.
.. . . . .
Formula’sthat give“critical%uckl-ing- S~;=~9=-?Or-- Y
elastic action are a-ppl~cablefl
to:buckli’rig in the plastic
range of the materialj if the You-ng~smo”duluste”rti (E) is
renlaced by the p~oyer effectivemodulus term (EF).
Theoretically,the value for reduced or effectivemodulue
(EF)will always be greater than the corresponding. value
fOr tangent modulus (ET). (See reference 2j) Experience“
has indicated (reference3) that:the use of the tangent
modulus in the Euler equati’onfor, col~mns gives calculated
values for critical stress that are in reasonable agree-
ment with test results. It is thereforeprobably somewhat
conservativeand not illogicalt’oassume, for design .
purposes, that the effectivemodulus can be represented

. “:---— —–
 ,

4 Technical
l?AC:~ ~ote No. 92? -

by the correspond
ing~~lue ‘oftiangentmodulus.’” ‘Ifthis
assumptionis made, the,comparativebackli~g stren@h of
various a“luminum
alloys in the plastic range can’ba
readily determinedfrom.the equationsfbr
,,, their itress-
strain curves. <’.
In general,any.buckling’equationcan be expressed
in the fern .~.
. .
s“
-=CD (5)
Ey’ .. . ---

where ... —-
s critical etress
.
E~ effectivemodulus correspondingto the strese S
c coefficientdependingbn the type of member and on
the nature of the loads and restraintsacting onq.
the member or element
. . . ...
and
D a functionof the dimefi’sions
of the piece
—.
In determiningcritical stressesin the plastic range.it
Is thereforeconvenientto have a curve for the material,
showing stress (S) plotted against-bhe ratio of stregs to .,

effectivemodbIus $’) on the basis of the assump-

( ~)’ or’
tion previously stated,against the ratio of stress to ‘
tangerit’
modulus (’)
s
~1“
Tha comparativebuckling strength
of d!.fferent
maberialscan be determined‘by“a comparison
of such curves.
The ratio of stress iio’t,angent
modulus can be ex- .
pressed . ..

(6)
de

;,’
 ●

NACA TechnicalNote No. 927 5

The term de/dS can 6 ayaluated~y differentiatingequa-

tion (lb); Equati6fi16) then becomes . -!
-.. .
s =—
fiE
s + 0.002 n
(“)n—s’
s~
(6a)

In order to demoristratethe effect of the shape of

the .stress-stirain
~curveon the buckling strength in the
p“lapticrange, consider two alloys having the properties
—.-
shown in the followingtable: ..
.
Yield strength Yield strength s~
Alloy at 0.002 off- at 0.001 off- n
set, Sa eet, S1 ~
(lb/sq in.) (lb/sq in.) .-
A. 50,000 43,000 1.163 4.59
B 50,000 48,000 1.042 16.8 I
A

-.+ ,
Both alloys have a modulus of elasticityof 10,000,000
pounds per square inch. Figure 1 shows stress-strain
curves and curves of S against S/ET for these--two ..
.-
alloys. .4comparison of the curves indicatesthat for
stresses below about 20,0”00pounds per square inch the
behavior of both materials is essentiallyelastic. P
1 Essentially elastic action continues in alloy B up to a -
stress of about 35,000 pounds per square inch. For
stressesbetween 20,000 pounds per equare inch and 45~OO!
pounds per square inch alloy B has I.owezvalues of S/ET
and consequentlygreater buckling strengththan alloy A.
Abov~ 45,000 Pounds per square inch, howeverl the buckling
strength of alloy A is greater than that of alloy B. .
It is evident that whereas a knowledge of the yield
etrength (stressat 0.002 offset) of a material is in- —
adequate to define the shape of the stress-etraincurve,
the determinationof an additionalyield strengthvalue
correspondingtb some other offeet value, together with
the Youngis modulus of the material may provide sufficient
additionalinformationfor evaluatingthe stress-strain
relation and consequentlyfor determiningthe buckling
resistance of the material in the plaetic range.

Aluminum Researoh Laboratories,

Aluminum Company of America,
New Kensington,Pa., October 14, ~943. ._

--
6 NACA ‘T’ekhnic,al
Note No.“927
. . ‘r*
.,>.
r.7-e.
. . ,.
”..
REFERENCES .“-’:2’f
‘ -..,.
..
.’ :, ,,,.
,. ,.,-
1. Ramberg, Walter.,and Osgood, WilliaFR.: De60ripti. on
of Stress-StrainCurves by Threy..~ameters.
-.! T.N.
No-.9i32,$TACA,1943. ..-. ::..
2. Timoshenko,S.: Theory of Elastic.Stability.
,McGraw-HillBook Co., Inc., pp. 1L$6,.
.-. .384.
,,
3. Templin, R. L., Sturm,R. G., Hartman~,~~.C., afidHolt,
M* Column Strength of”Vari.ousA1.urninum
A11oYs. . . .
T::h. Paper No. I.,Aluminum Res. Lab.’.,
Aluminum CO. .
of Am., 1938.
,.

&

..

●
--

.-
. I
 1
. . ,h

(1 block= 10diviaiona onl/30H mgr. ecale)

70, CF30 g

i
80,0W I E
Alloy A Uloy B A11oY A Allo YE ~
‘, 1. C+
$ !D
/ — - ‘i
I \ ~
~ —
So,ooo
- /
I — - m
/ - al
-r
d
4
0
<40,000
S
/’ /
z 4 I

: Scl,
ooo
0 {
; /
.

ao,olm /
I

10,OOC

(a) I (b)

o .003 .004 .OLM .006 .010 .Olao .01 .oa .03 .04 .05 .00
8trbin (0), In.lin. W*

Yigura 1.- 8tree.6-@train mnd atree.v- S/ET ourvee! for two alloym having the same wdulum and the same yield mtrength
offirat), tut with
(o.ooa different atrem-mtrtin 0UW38.
-
G
P