| VOLUME 23 Avsrascr A method is developed for calculating stifiness influence co ‘ficients of compler sell-type structures, The object is to pro Vide w method that will yield structral data of sufficient accursey to be adequate for subsequent dynamic nnd aeroelastic analyses ‘Stifiness of the complete structure is obtained by sum stiffreses of individual wots, Stiflesses of typical structural ‘Components are derived in the paper. Basie conditions of con tinuity aad equiiirium are established at selected points (nods) in the structure. Tacreasing the number of aodes increases the accuracy of results, Any physially possible support conditions tam be taken into account, Detsils in setting up the asalyss ea be performed by nonengineering trained personnel; calculations fre’ conveniently carried out on automatic digital computing equipment, ‘Method! is ilustrated by application to « simple truss, a fat plate, and a box beam, Due to shear lag and spar web deflection, the bow beam has n 25 per cent greater deflection than predicted from beatn theory, Te is shown that the proposed meth! cor rectly accounts for these fects, ‘Considerable extension of the material presented in spose (1) Intropecrt0n Prssent conrrouearion rmavos in the design high-speed aircraft have created a number of Gifficult, fundamental structural problems for the worker in aeroelasticity and structural dynamics. The chief problem in this category is to predict, for a given clastic structure, a comprehensive set of load-deftection relations which can serve as structural basis for dynamic Toad calculations, theoretical vibration and flutter analyses, estimation of the effects of structural deflec: Received June 28, 1955, This paper is hosed on a paper presented at the Aerorlasticity Sesion, Twenty-Second Annual ‘Meeting, IAS, New York, Jansary 25-29, 154, * Structural Dynamics Unit Chief, Bocing Airplane Company, Senttle Division 4 Associate Professor of Civil Engineering, University of Cali: fornia, Berkeley ' Professor of Aeronautical Ensineering, University of Wash. iton, Seattle “+ Structures Engineer, Stractaral Dynamics Unit, Boeing Air plane Company, Wichita Division, JOURNAL OF THE AERONAUTICAL SCIENCES SEPTEMBER, 1956 NUMBER 9 Stiffness and Deflection Analysis of Complex Structures M. J. TURNER,” R. W. CLOUGH,t H. C. MARTIN, } ann L. J. TOPP** tion on static air loads, and theoretical a clastic effects on stability and control. T! 1 isa prob- n of exceptional difficulty when thin wings and tail surfaces of low aspect ratio, either swept or unswept, are involved. It is recognized that camber bending (or rib bending) is a significant feature of the vibration modes of the newer configurations, even of the low-order modes; in order to encompass these characteristics it seems likely that the load-deflection relations of a practical structure must be expressed in the form of either de- flection or stiffness influence coefficients, One ap- proach is to employ structural models and to determine the influence coefficients experimentally; it is antic pated that the experimental method will be employed extensively in the future, either in lieu of or as a final check on the result of analysis. However, elaborate models are expensive, they take a long time to build, and tend to become obsolete because of design changes, for these reasons it is considered essential that a cot tinuing research effort should be applied to the devel- opment of analytical methods. It is to be expected that modern developments in high-speed digital com- puting machines will make possible a more fundamental approach to the problems of structural analysis; we shail expect to base our analysis on a more realistic and detailed conceptual model of the real structure ‘than has been used in the past, As indicated by the title, the present paper is exclusively concerned with methods of theoretical analysis; also it is our object to outline the development of a method that is well adapted to the use of high-speed digital computing machinery. (1) Review oF Exisrinc Metnons oF SravetuRat. ANALYSIS (1) Elementary Theories of Plesure and Torsion ‘The limitations of these venerable theories are to well known to justify extensive comment. ~fB08 adequate only for low-order modes of elongated strue- tures. When the Ioading is complex (as in the case of inertia loading associated with a mode of high order) refinements are required to account for secondary effects such as shear lag and torsion-bending. (2) Wide Beam Theory: Schuerch has devised a generalized theory of com- bined flexure and torsion which is applicable to multi- spar wide beams having essentially rigid ribs. Torsion- bending effects are included but not shear lag. It is, expected that wide beam theory will be used extensively. in the solution of static aeroelastie problems (effect of air-frame flexibility on steady ait loads, stability, ete.) However, the rigid rib assumption appears to Tinit its utility rather severely for vibration and Mutter anal- ysis of thin low aspeet ratio wings, Schuerch* pel ten er, tnt ot toting toe prc waged wes oe ie Shino rac afc posit Redon Castigliano’s theorem. The method is, of course, perfectly general, However, the computational diffi tmnbe ats a cnn he pop ely Candnt sous may beatae preter a ena de fut aest geealy Showa rtunansroqed to chan a mtasay cin uy bene ton on tes teeny ie oases tere (H) Plate Methods: Fung, Reisner, Benscoter, and ‘MacNeal As the trend toward thinner sections approaches the ultimate Timit, we enter first a regime of very thick walled hollow structures, such that the flexural and torsional rigidities of the individual walls make a significant contribution to the overall stiffness of the entire wing. Finally we come to the solid plate of variable thickness. During the past few years a sub- stantial research effort has been devoted to the develop- ment of methods of deflection analysis for these stru tural types, and important contributions have been made by all of the aforementioned authors, JOURNAL OF THE AERONAUTICAL SCIENCES—SEPTEMBER, 1056 (8) Direct Stiffness Calculation: Levy, Schuerch™ In a recent paper Levy has presented a method of] ticularly suited to the use of high-speed digital com- puting machines. The structure is regarded as an assemblage of beams (ribs and spars) and interspar torque cells. The stiffness matrix for the entire struc ture is computed by simple summation of the stiff ness matrices of the elements of the structure. Fi nally, the matrix of deffection influence coefficients is, obtained by inversion of the stiffuess matrix. Schuerch hhas also presented a discussion of the problem from the point of view of determining the stiffness coefficients (ID Some Ussonven PRosuens At the presnt time, itis believed that the greatest seed is to-detive a numerical method of analysis for a class of strictures intermediate between the thin stifened shell and the solid plate. ‘These are hollow structures having a rather large share of the bending svaterial located in the skin, which is relatively thick but still thin enough so that we may safely neglect its plate bending stiffues. In order to cope with this class of structures successfully, we must. base our Analysis upon a structural idealization that is suf cently realistic to encompass a firly general two- dimensional stress distribution in the cover plates; and our method of analysis must yield the load-delee- tion relations associated with such stresses. It is char- acteristic ofthese problems that the diestions of pi ‘Spal stresses in certain critical parts of the structure eaimot ‘be determined by inspection, Hence, the faniliar methods of structural analysis based upon the concepts of axial load carrying members, joined by membranes carying pure shear, are not satistactory, ‘even if we employ effective width concepts to account for the bending resistance ofthe skin. We should Tike to include shear lag, torsion-bending, and Poisson's ratio effects to a sufcient approximation for reliable prediction of vibration modes and natural frequencies Of muvlerate order. Als, we shoul Tike to avoid any assumptions of closely spaced rigid diaphragms or of orthotropic cover plates, which have been introduced in many. papers on advanced structural analysis. The actual rib spacing and finite rib stiffuesses should be accounted for in a realistic fashion. In summary, what is required is an approximate iumerieal method of analysis. which avoids drastic modifeation of the geometry of the stracture or artificial constraints ofits tlastic elements. This is indeed a very large order. Homever, modem developments in high-speed digital computing machines offer considerable hope that these objectives can be attained (GY) Mernop or Dixecr SriprNess CaLcuLATION, For a given idealized structure, the analysis of stresses and deflections due to a given system of loads is a purely mathematical problem. Two conditionssatisied in the analysis: (1) the forces de- in the members must be in equilibrium and (2) mations of the members must be compatible— tent with each other and with the boundary tions, In addition, the forces and deflections in ‘member must be related in accordance with the strain relationship assumed for the material rhe analysis may be approached from two different ints of view. In one case, the forces acting in the tbers of the structure are considered as unknown Fquantities. In a statically indeterminate structure, an infinite number of such force systems exist which will satisiy the equations of equilibrium. The correct force system is then selected by satis'ying the condi- tions of compatible deformations in the members. ‘This approach has been widely used for the analysis of all types of jndeterminate structures but is, as already noted, particularly advantageous for structures. that are not highly redundant. In the other approach, the displacements of the joints in the structure are considered as unknown ‘quantities. An infinite number of systems of mutually compatible deformations in the members are possible; the correct pattern of displacements is the one for which the equations of equilibrium are satisfied, The con- cept of static determinateness or indeterminateness is irrelevant when the analysis is considered from this viewpoint, This approach is the basis for many re- Taxation type analyses (suck as moment distribution) and has been applied to the analysis of complex aircraft structures by Levy in the aforementioned paper. This will be called the method of direct stiffness calculation hereafter. ‘Aiter reviewing the various methods available to the dynamics engineer for computing load-deflection rela- tions of elastic structures, itis concluded that the most promising approach to our present difficulties is to ex tend further the method of direct stiffness calculation, ‘The remainder of this paper is concerned with methods bby which that extension may be accomplished. (V) Supe EXAMPLes oF Sripeness IneLueNce ‘CoErricieNTs (1) Blastic Spring If an elastic spring deflects an amount 5 under axial loud F, Hooke’s Law applies and Few a Here & can be regarded as the force required to produce fa unit deflection; hence it can be considered to be a stifiness influence coefficient Eq. (1) can also be written as 8 = (VA)F = oF 2 where ¢ is the deflection due to a unit force (deflection influence coefficient) (2) Two-Dimensional Elastic Body Extending the above relations to the two-dimensional body is most conveniently accomplished by introducing STIFFNESS AND DEFLECTION ANALYSIS sor L*LENGTH A=AREA E =MODULUS (a) (b) Fic, 1, Typical pin-ended truss member jotation. Eqs. (1) and (2) become, respectively, tA} = [x1 fat 3) fa} = te {A} = (cl fF} a Here [Kis the matrix of stiffness influence coefficients A typical clement of [K] is kg! = force required at in the ¢-irection, to support a unit displacement at j in the r-direction. If fand always refer to the same direction, we can use the simpler form Ay. In either cease an element of [X], and also of {C], must obey the wwell-knowa reciprocal relations. Tn other words, the IK] and [C] matrices are symmetric, provided they are referred to orthogonal coordinate systems. As will be seen later, the symmetry condition does not apply if oblique coordinates are used. (8) Truss Member Fig, 1(@) shows a typical pin ended truss member. We wish to determine its matrix of stiffness influence coefficients. Loads may be applied at points (nodes) 1 and 2, Each node can experience two components of displacement, Therefore, prior to introducing boundary conditions (supports), [&] for this member will be of order 4X 4, ‘To develop one column of [A], subject the member toms Om = 0. Thea AL = us 6080, = 13 ‘The axial force needed to produce ALis P = (AB/D)AL = (AB/LIN ws ‘The components of Pat node? are Fy, = P 008 8, = (AB/L) Nt Py = Pcosdy = (AB/L) us rium gives the forces at node I as Pus Pa Fu = —Fu 4. (8) for thismerber then takes the form By -s ts F.| _ AE x A (| as |08. ‘The other elements in [A] are found in a similar manner, We get ” Az] x) - yt ame, FL ue at © Ne mut ut As given in Eq. (6), [Kis singular—that is its deter- inant vanishes and its inverse does not exist. This is overcome by supplying boundary conditions or sup- ports for the bar sufficient to prevent it from moving as a rigid body. For example, we may choose 1 4 = m2 = 0,1 # 0. Node 1 is then fixed, while node 2s provided with a roller in the y-direction. The only force component now capable of straining the har is F,,. The force in the bar and the reactions are given by Eqs. (5) and (6), Any other physically correct boundary conditions can be imposed. In other words, once [K] has been determined, a satution can be found for any set of sup. port conditions. The only requirement is that the structure be fixed against rigid body displace:ent { = AE Fe | Fa} Fy or {F) = ixI fa} (7) In this second column the y-components of force are given by the g* terms in Table 1; the x-com- ponents of force are given by the Xp terms. Thus FF, is the sum of a* for members 1-2 and 1-3 since these are strained due to displacement #1, Also Fy, is a ‘The for member 1-2, and F,, is —A? for member 1-3, j JOURNAL OF THE AERONAUTICAL SCIENCES SEPTEMBER, 19 (VP) Stirpxess Awatysis of Snerce TRUSS Once stiffness matrices for the various component units of a structure have been determined, the next step of finding the stiffness of the composite structure may be taken, The procedure for doing this is essen- tially independent of the complexity of the structure. AS a result, it will be illustrated for a simple truss as shown in Pig. 2 ‘The stiffness of any one member of the truss is given by Eq. (6). Since length varies for the trass members, this term should be brought inside the mateix. It is then convenient to call the elements of the stiffness matrix 3? = M/length, ete. Then X, a, and Kg repre- sent the essential terms defining the stiffness of the separate truss members. These are conveniently cal. culated by setting up Table 1 From the last three columns of Table 1 the truss stiffness matrix can be written directly. This is best * seen by forming the truss equation [q, (7a) analogous to Bq, (3) for the single member. ‘ree ‘The formation of all cohunns in Eq. (7a) can be ex- plained by considering any one of them as an example. The second column will be chosen, It represents the case for which ty » 0, all other node displacements (va) signs follow from the basic stiffness matrix given in Eq, (6). Since equilibrium must hold, the sum of these reomponents of force must vanish, Similarly, F,, is the sum of the Ka terms for members 12 and 1-8. Likewise, F,, is the negative value of Xa for member 1-2, Finally F, is —\a for member 1-8, These forces must also sun to zero if equilibrium isto hole,STIFE This proce all possible node displacement components are taken into account. In each case the displacements are com: patible ones for all members of the truss. is repeated for all columns. In this way A structure having various kinds of structural com- ponents~beams as well as axially loaded members, for example—would be treated in the same manner. However, the basic stiffness matrix for each type of member would have to be known. Deriving these for units of interest in aircraft design represents a major part of this paper. ‘The matrix of Hq, (Va) is singular, This is altered by providing supports for the truss sulicient to prevent it from displacing as.a rigid body when loads are applied, Any sufficient set of supports may be imposed; here wwe choose to put morn In other words, nodes 1 and 2 are fixed, while 3 is left free. L 1 Pa 1+ - - I 2v2 avez L L _ 3y2 22 | 1 1 r ne _ 23 We P, t anaes 22 V8 Pa eee 0 Fa ee 4 Af the partitioned square (stiffness) matrix is designated by [a= Boa) Bie | Dost ‘expanding Eq. (Te) leads to the following two sets of equations: {fowl and Fay Fa y Stat [fe = [BY Vos f (Sb) Fa iq, (Sa) gives unknown node displacements in terms of applied forces, us F, fash tay {fot (a) while Bg, (8), together with ag, (Qa), gives unknown Feactions in terms of applied forces, NESS AND DEFLECTION ANALYSIS 300 ee I t A,E (SAME FOR ALL MEMBERS) L EL —-—e xu 2 L 3 Fre. 2. Simple tras I is now convenient to rewrite Bq. (Za) and simul tancously partition it as shown by the broken lines in Ba. (7). 22 Foe 3 L : wi "BVI si 1 1 ove | Taye 0 ° ° = F, i = wy ay fh (ob) Py In dynamic analyses of aircraft structures it is ordi- narily sufficient to determine [A]-'. This is the exibility matrix, Tt is interesting to note that [4] can be found from the complete [X] matrix by merely striking out columns and rows corresponding to zero displacements as preseribed by the support conditions A complete stress analysis leading to the truss mem- ber forces ean also be carried out. It is merely neces sary to know the force-deflection relations for the individual members, or components, of the structure, ‘This is a straightforward problem for the truss and, therefore, will not be discussed further in this paper. It is worth while to notice Hat once the stiffness matrix has been written, the solution follows by a series of routine matrix calculations. These are rapidly carried out on automatic digital computing. ‘equipment. Changes in design are taken care of by properly modifying the stifiness matrix. This cutsFLANGE AREAS Fig, 8. Wing structure breakdown, analysis time to a minimum, since development of the stiffness matrix is also a routine procedure. In fact, it may also be programmed for the digital computing machine, (VII) Suwsary—MetHon oF Direct CaLcuLarion (2) A complex structure must first be replaced by an equivalent idealized structure consisting of basie stric- tural parts that are connected to each other at selected node points. (2) Stiffness matrices must be either known or de- termined for each basic structural unit appearing in the idealized structure. (3) While all other nodes are held fixed, a given node is displaced in one of the chosen coordinate direc tions. The forces required to do this and the reactions set up at neighboring nodes are then known from the various individual member stiffness matrices. ‘These forces and reactions determine one column in the overall stiffness matrix. When all components of displacement at all nodes have been considered in this manner, the complete stiffness matrix will have been developed. in the general case, this matrix will be of order 3m X 3, where n equals the number of nodes. The stiffness matrix so developed will be singular (4) Desired support conditions can be imposed by striking out columns and corresponding rows, in the stiffness matrix, for which zero displacements have been specified, “This reduces the order of the stiffness matrix and renders it nonsingular. (6) For any given set of external forces at the nodes, ‘matrix calculations applied to the stiffness matrix then yield all components of node displacement plus the external reactions. S-SEPTEMBER, 1955 (G) Forces in the internal members can be found by applying the appropriate force-deflection relations. ‘The primary functions of the engineer will be to provide the information required in steps (1) and (2) above and to provide the individual member force- deflection relations if a stress analysis is to be carried ‘out, Steps (2) throngh (6) can be performed by non. engineering trained personnel. Changes in design can be taken into account by correcting local stifiness con- tributions to K, Node densities can be increased in regions of masimum complexity and importance. If vertical deflections only are required, as in the case of the aireraft wing problem, the 3n X 3n matrix for K can be reduced to order 2 X m by a sequence of matrix calculations, Physically, continuity of displacements in three directions at cach node will still be maintained. wim In carrying the above procedure over to stiffened shell structures, it is first necessary to perform steps (1) and (2) of the previous outline For a wing structure the idealization will be made by replacing the actual structure by an assemblage cof spar segments, rib segments, stiffeners, and cover plate elements, joined together at selected nodes. Fig. 3 shows the proposed idealized structure. The decomposition of the structure can be carried further swith some increase in accuracy (for example, by de- composing spar segments into spar caps and shear webs), or it ean be simplified by treating the structure as an assemblage of spars and torque boxes. The degree of breakdown should be consistent with the complexity of structural deformations required by the problem at hand. (In a vibration analysis the order of the highest mode is a determining factor.) In light ‘of the proposed idealization, it is necessary that stiffness matrices be developed for the following components: beam segments consisting of anges joined by thin webs, and plate elements of arbitrary shape. In addition, provision must be made for taking stiffeners into account and possibly for including the effect of sandwich type skin panels, In the general case, spars will be swept, nonparaltel, and not necessarily orthogonal to ribs, It will generally be convenient to transfer stiffuess values for any given member to a fixed set of reference axes. These refer- ence axes will be chosen as rectangular Cartesian (x, y, 2) in order to preserve symmetry in the total K-matrix. ‘An outline of the determination of member stiffness for simple steuctural elements is given in the paper. Further details are presented in Appendixes. Deriva. tion of stiffness matrices for more complex clements, can be accomplished in a straightforward manner. ‘However, in the analysis of an actual structure, it will bbe necessary to weigh the relative advantages of em- ploying a small number of darge complex elements against the advantages of using a larger number of small elements for which simple stiffness coefficients Suirrenrp Suut StRUCTURES may be in resol essing arid me formu Fir sunifo aomay be employed. ‘The main criterion to be observed in resolving this issue is that the problem must be pro- grammed so that as much as possible of the data proc- essing is performed automatically by the computer and not by human operators substituting in complex formulas. (UX) Spars ano Riss First we consider the untapered beam segment of iniform cross section shown in Fig. 4. Its stiffness matrix will be determined by application of beam theory, which is extended, however, to inchide shear web flexibility. Nodes, 1, 1, 2, and 2! are established as shown in Fig. 4. The following notation is used: moment of inertia of beam section about neutral (3) axis f= thickness of shear web E, = modulus of elasticity of flange material G = modulus of rigidity of shear web material 9 = Poisson ratio ad Displacements are assumed such as to be compatible with elementary beam theory. In other words, ays) (+) 0 -—oet__| —te/L) 1K) = Fa +a) 2/3) (1 ~ 20) 0 WL where n= 36/6) [1/(HL4)) cu) Contribution of shear web deformation to the above liffness matrix is indicated by values of > 0; for a rigid shear web 2 = 0. ‘As a simple example of the use of the beam stiffness ‘matrix, we consider a cantilever of length L and loaded by force P at the free end (nodes 1 and 1’). Putting n= Oand applying Bq, (11a) gives: Fro. 4 STIFFNESS AND DEFLECTION ANALYSIS si eos Fro. 5, Rectangles Cartesian axes systems, me wel w= mu a0) Stiffness in the y-direction is assumed negligible. ‘An outline of the derivation of the stiffness matrix for the above beam segment is given in Appendix (A). It is shown to be of the form aye (ay G/L) (4/3) A + 9) 0 oO = (he/L4) Wk ORL {Fal _ GEC > eye (2) Fg, (12) may be inverted to yield tip displacements tm and t in terms of applied load P (F., = 0, P/2), The results are uy = —(PLY/2E1) (h/2), ta = PLA/SEL which agree with known results, Tn an actual wing structure, spar and rib segments will be more or less randomly oriented with respect to a set of standard reference axes. As a result, trans- formation of stiffness matrices for these anembers to the ‘standard set of axes will generally be necessary. ‘The basis for such transformations is given below. Let the direction cosines of 2, y, a-axes with respect to standard £, 5, Zaxes, Fig. 3, begit JOURNAL OF THE AERONAUTICAL Fre: 6._ Stifened cover skin element | ‘Simple geometrical considerations then give the follow- ing equation for relating forces in the x, y, # system to forces in the £, 5, Zsystem sy» * vob fee outer te | Fe) Tu 0 0 0 Fi) las ty me 0 0 0 Fal =| % % 0 0 0 7 Ft [0 0 0 HM ve: 0 0 0 be My we Pry [0 0 0 % me or, {F) = (elle (3b) acemments are vectors similarly related to the coordinate systems as forces and hence transform under G rotation of axes in the same manner. Consequently, fa} = (el toh ay ‘s where fa = fp ete te From the above and Eq, (3) it follows that, IR] = 18] 1 fet = (el) IX} Fey) 18) where [KJ is the stiffness matrix referred to the stand- ard Z, 5, # set of axes. Beam segments encountered in the analysis of real structures will be tapered in depth, and flange areas will be variable; generally the segments ‘will be taken short enough so that the variation in depth may be assumed linear. Derivation of stiffness Snatrices for elements of this kind is straightforward, fand details will not be included in the present paper. SCIENCES SEPTEMBER, 1958 (2X) StIFFENAD PLATES: ) Stiffeners A plan view of a typical portion of stiffened cover skin structure is shown in Fig. 6. Nodes are initially established at points 1, 2, 3, and 4. The included ‘structure then consists of spar segments (1-2 and 3-4), fib segments (1-3 and 2-4), and stiffened plate element {-2-3-4. _Stiffeners may be conveniently lumped with spar caps and, if desired, into one oF more equivalent stiffeners located between spars. In this latter event additional nodes must be established, as at the inter- sections of these equivalent stiffeners with the ribs. ‘The stiffness matrix for a lumped stiffener of constant area A, length L, and modulus is, A ele a | ry ‘Derivation of a similar matrix for a tapered member is straightforward; the area A is replaced by a suitable fmean vale, The influence of shear Tag effects on Toad-deflection relations for the panel and stiffeners: can only be incltided if odes are established at inter- ‘mediate points on the ribs, between spars. @) Plate Stifiness ‘The quadrilateral plate element 123-4 of Fig. 6 js assumed to possess in-plane stiffness only. Since ‘two independent displacement components can occur at each node, the order of the K-matrix for this plate Glennent will be § X 8. The problem of ealeutating K is not an easy one, and the solution offered here is felt to have potential usefulness for finding approximate solutions to many two-dimensional problems ia eles- ticity. Relore proceeding with the method developed for calotlating K of the plate element, it is pointed out that a so-called framework analogy" exists, whieh per- mits one to replace the elastic plate with a lattice of lastic bars. Under certain conditions the framework then deforms as docs the plate and hence can be used to calculate the plate stifiness. The determination of a lattice representation for a rectangular plate is rela tively straightforward; however, plate elements of non- rectangular form present basic difficulties. For ex- ‘ample, if one attempts to apply the rectangular grid work to a nonrectangular plate, difficulties arise in Attempting to satisly boundary conditions. On the “ther hand, if one goes to nonrectanguilar lattice forms, difficulties arise when attempting to satisfy the stress: ‘Strain relations in the interior of the plate. Consider- ‘such as these led to eventual abandonment of this approach. ‘The concept finally employed for determining plate stiffness is based on approximating actual plate strains by a restricted strain representation. In other words, no natter what the actual strains in the plate may be, these will be approximated by a superposition of several simple strain states. The method for doing atiosis and the accuracy of results based on such a repre~ ntation form an important portion of this paper. ito give an initial illustration, the actual strain tribution in a rectangular plate element can be proximated by superimposing the strains that espond to each of the simple external load states pwn in Fig. 7. ‘These load states are seen to repre- pt uniform and linearly varying stresses plus constant far, along the plate edges. Later it will be seen 4 the number of load states must be 2x — 3, where = number of nodes, HBefore commenting further on the scheme suggested bre for analyzing plate elements, the method will be plied to the triangular plate of Fig. 8. The triangle rot only simpler to handle than the rectangle but it will be used as the basic “building block” for ating stiffness matrices for plates of arbitrary ‘We start by assuming constant strains, or a = (1/E) (aq — 9m) = u/2x | b= (VE) (ey — v0.) = dr/dy_— | (7a) 6 = (/G)rm = Qu /dy) + n/N) er it will be pointed out why we are restricted in the hoice of strain expressions. Integrating we find the placements to be wnactayt+ By a v= byt tc A + Cf A re, A, B, and C are constants of integration which fine rigid body translation and rotation of the tri- Lew STIFFNESS AND DEFLECTION ANALYSIS i 3 (xy4s) 1 104.4) 2(%20¥2) Fic.8. Node designation for triangular plate element, angle. Hence the triangle can displace as a rigid body in its own plane and undergo uniform straining accord- ing to Bq. (17a) ‘Displacements at the nodes can be determined by inserting applicable node coordinates into Eq, (17D) In this way six equations occur which are just sufficient ior uniquely determining the six constants of Eq, (17). ‘As a result the constants become known in terms of node displacements and coordinates. It is this part of the solution which determines the number of terms ‘which must be chosen in the strain expressions or alter- natively the number of applied edge stres states which must be used, The number i always twice the number fof nodes minus three, Wlenee, for the triangle we re- quire three termus and five for the reetangle (or quadri- lateral). To proceed with the solution, we solve directly for stresses in terms of node displacements 1, ti, ty ete If xq & x — ) and dy = (1 — »)/2, this gives Log ET fa m {ty 1 | Jas tee (18a) | J a ys o” to} = ts) fa (sb) ‘The next step is to obtain the concentrated forces at the nodes which are statically equivalent to the applied constant edge stresses, The procedure for doing this “will be briefly illustrated for the case of the shear stress Fig. 9(a) shows the shear stresses on the circum- seribed rectangular element, and Fig. 9b) shows the corresponding edge shear forces on the triangle. As before «x, ys tefer to coordinates of node points Forces on any edge are equally distributed between nodes lying on that edge. For the forces as given in Fig. 9(), thisleads to Fy oo = 1 72 = Hiall/2) toy | { ° : 2 ‘gl): JOURNAL OF THE AERONAUTICAL SCIENCES—SEPTEMBER, 1956 where the superscript refers to case 3 (that of shear stress). This procedure is repeated for the two normal stresses, Superimposing results for these three cases then leads to the following system of equations for node forces in terms of applied edge stresses: Fl -3% Fry 0 Fal ft) 98 Ff ~ 2) 0 Ba 9 Fa] 0 Ix] cereal An alternative approach to the above method for calculating the plate stiffness matrix is to calculate the Strain energy in the plate due to the assumed strain Gistribution and to then apply Castigliano’s Theorem for finding the node forces. This procedure can also tbe conveniently carried out in terms of matrix oper- ations; details will not be included here, however, since tthe result is the same as that already obtained. ‘Stiffness matrices for plates having four and more nodes have been derived and studied. The advantage in introducing additional nodes lies in the fact that a more general strain expression may then be employed ~ for equivalently additional load states as illustrated by Fig. 7 may be used for the plate. As a result a choice ‘etween two points of view may be adopted; first, the simplest or triangular plate stiffness matrix may be used fand the desired accuracy obtained by using a sufficient inuriber of subclements, or second, a more general plate stiffness matrix may be wsed with fewer subelements. Experience to date indicates that satisfactory results can be obtained using the triangular plate stiffness matrix. ‘Some additional plate stiffness matrices are given in Appendix (B). ‘To summarize briefly the meaning and significance of the plate stiffness matrix, itis first pointed out that this matrix relates node forces to node displacements. ‘Asa result the plate stillness can be immediately added {FA = Wile} (206) ‘Substituting Eq. (186) into Eq. (20), tA) = ITLIS 3} (200) Comparing this last equation with Eq. (3) shows that Ik] = ITVS) en Carrying out the indicated matrix multiplication and patting by = (1+ 9/2 gives: to spar, rib, ete, stiffnesses which are also given for specified nodal points. However, the plate node forces are statically equivalent to certain plate edge stresses, Furthermore, these edge stresses will tend to, ‘approach actual edge stresses, even of a complex nature, if sufficient subelements are used. A result of these ‘equivalent edge stresses is that continuity will tend to be approximately maintained along common edges of subelements, between nodes. In other words, we are fassuming that a plate under complex strains will deform = ju a manner that can be approximated by relatively simple strains acting on subelements into which the larger plate has been divided, The accuracy of this representation should increase as the number of sub- ‘clements increases. (3) Quadrilateral Plates In the analysis of wings and tail surfaces it is generally convenient to employ a subdivision of cover plates such that most elements are of quadsilateral shape. ‘The stiffness matrix for such elements ean then be de rived in one of two ways: (a) the previous solution demonstrated for the triangle can be extended to in- clude the quadrilateral and (b) the quadrilateral can tbe subdivided into triangles and its stiffness matrix determined by superposition of the stiffnesses of the individual triangles. In this section the latter pro- cedure will be adopted. See ee eC eee ee‘Two simple subdivisions of the quadrilateral into jangles are shown in Figs. 10(a) and 10()). ‘These to different stiffaess matrices for the quadrilateral ‘nique result is obtained by using the subelements swn in Pig. 10(e). ‘The interior node will be located the centroid, although any other choice could be used. For the general quadrilateral plate it has proved to be ferable to program the calculation of the stiffness trix for high-speed computing equipment. In the fase of the rectangle, however, an explicit derivation, Jean be readily carried out. The necessary calculations, included below, are given here, since the end result is useful and since these calculations serve to illustrate a step of some importance in carrying out the analysis fof a more complete structure—for example, a wing or tail surface. ‘The rectangle and its four triangular subelements, with interior node number 5 at the centroid, is shown in Fig, 11. Stiffness matrices for the triangles can be caleulated from Eq, (22), or more conveniently from Bg, (B-3) of Appendix (B). In determining K of the rectangle, superposition in the following form is used: K Kit Kut Kin + Ki Since five nodes have been established, K for the rec~ tangle will initially be of order 10 X 10. This will later be reduced to order 8 X 8 to give a result con- sistent with the choice of four external nodes; only at these external nodes is contact implied with adjoining structure. The immediate point is, however, that K for each triangle must be increased to order 10 X 10. before superposition is carried out. This is accom- plished in the usual way—that is, by introducing appro- priate rows and columns of zero elements. In order to simplify the expressions for elements appearing in the stiffness matrices the derivation of K for the rectangle will be restricted to » = 1/3, ‘On superimposing stiffnesses for the component tri- angles of Fig. 11 it becomes possible to express Eq. (3) in the form (Fa) cs r Fa Py mL Fo 4) (23) | Fa Fa) UA ‘Since forces are to be applied to the rectangle by stresses, Jlent to forces acting at nodes 1, 2, 3, and 4, the can be applied to Eq. (23). Doing this results in the vo sets of equations written below: STIFFNESS AND DEPLECTION ANALYSIS |= Falta) Twat ® @) o 10.9, Shear loading on ogular plate clement. @) Fre. 10, fe) Decomposition of quadrilateral plate into triangular sstbelements xu 2x22 “Triangular subslements for rectangular plate AUxps:) Fra 1L I~ 6 (UNIFORM) + =0.050 IN. £ =10.5X10°PSI v eV TOTAL LOAD = 2LBS. 0,12, Clamped rectangular pat subjected fo nim in,si) JOURNAL OF THE AERONAUTIC {F {a UR \ we a ta Ae iE wr yeh em teh (2 Fn =| mloOS Fa * (ems) : anh ‘ eeiyeen ots cmb tm [ame om 4% f Sots Bena am Lowe oe eS If the order of r-terms in the above equations are re- ‘atranged from ptt, £4 10 2, Pe Un Bey it will be dis- covered that Ke equals Ky: provided we replace mit. Ku everywhere by L/m, ‘The corresponding form for Ke may be written without difficulty. It is again pointed out that the above plate stifiness matrix is based on v = 1/3, “The process of eliminating displacements at node 5 is similar to the situation that arises when only w dis: AL SCIBNCES—SEPTEMBBR, 1958 Solving Eq. (246) for displacements at node 5 and sub- stituting the result into Bq. (24a), = ayer 1294" f 2) 4, Comparing Bq. (25) with Bq. (@) IK] = [Al = [al ich BY 26) Carrying out the calculations required by sults in the following rectangular plate stiffness matrix: (2! where, when m = (ve — x)/( = 9 ser 16 (K) placements are to be retained in a wing analysis. In {his latter problem it then becomes necessary to elimi hate all wand v components of displacement. The procedure for doing this is the same as that used in ‘eliminating 15 and % from the above problem of the rectangular plate. (a) Example It is of interest to carry out calculations on a simple example and compare results obtained by applying the plate stiffiess matrix with values that can be regarded ascorrect, ‘For this purpose the plate of Fig. 12 is analyzed using several different methods. Defleetions at several points ‘due to the indicated loading will be calculated. Since ‘an exact solution is not available, correct displacementsSTIFFNE Method Fig i Relaxation 1B 2 Simpietncory 1S 5 Plate Komatex 1 4 135 5 Ve 6 tie will be taken as those calculated by applying the re- lasation method to the fundamental equations govern ing this problem. Although details of these caleula- tions are not presented, results are listed in Table 2 ‘The problem is interesting for at least two reasons. First, the accuracy obtainable using various numbers of subelements ca be observed, and second, the effect of using random orientation of subelements—vith respect to the plate edges—can be observed. Results of all calculations are summarized in Table Node locations and subelements are illustrated in Fig 13, In Table 2 the solution based on simple theory was obtained from u = PL/AE and, = — «. Ibis observed that on this basis both mw; and oy agree quite vell with the relaxation solution ‘The erudest plate matrix solution i Fisted in Table 2 4s Solution No, 8. It was obtained by considering the plate as a single clement whose stiffness is given by Bq, (27). The results for u and t are seen to be Freasonably good. Solution No. 4 considers the plate 3s consisting of four rectangular subelements as shown fin Fig. 13(b). Again the stiffness matrix was obtained by using Bq. (27), this time for cach subelement. greement with relaxation results is seen to be satis- factory, particularly in regard to 1. Also the dit ences between 1) and 1: are approximated aceu- tely by this solution. It is to be remembered that ‘actual strain distribution in the plate is complex nature ty all wales by 10-€ 130 12k Oa vaso 13a teas 70 1355 11998) 688 os Each subquadrilateral was considered as consisting of four triangles in a manner analogous to the treatment. described previously for the rectangle of Fig. 1. In Solution No, 5 we note that w: and wy are not equal, a ‘consequence of the random nature of orientation of the subelements. By increasing the number of random. subelements as in Solution No. 6, this lack of symmetry, in results is virtually removed. Comparison with relaxation values is seen to be very good for bath Solu tions 5 and 6. A more comprehensive example is given in the next seetion of the paper. (XI) Awatysis oF Box Beast As a final example, the box beam of Fig. 14 will be analyzed for deflections, using the stiffness matrices previously derived. ‘The box is uniform in section, unswept, and contains a rib at the unsupported end. The following di sions apply: a/6 = 1, 2b/h = 10, t, = ty in, Ap = bt/2, @ = 400 in. As the simplest possible breakdown, we consider the box to consist of two spars, one sib, and two cover skins, ‘The nodes are then as shown in Fig. 15. Forces may be applied at the nodes at the free end, Two cases will be investigated: (1) up loads at each spar (bending) and (2) up load on one spar and a down oad at the other spar (twisting), Solutions 5 and 6 in Table 2 were carried out ia a ‘The spar matrix is given by Eq, (Ia). Calculation et of minutes on a high-speed digital computer, showsit to be 4 oF H WhOr Wy My Ory OF Hey 1.13903 | 0.05227 0.00333 (2s) 2| 0.50308 0.05: 1.13903 0.05227 0.00333 —0.05227 0.00333, iven by Eq. (27a) and for this case hecomes 90878 ] 37500 1.39778 | 19329 0 0.90879 1.15928 0.7500 1.39778 en 31018 0 0.39634 0.37500 0.90879 0.87109 —0.37500 0.60959 0.37900 1.89778 634 0.37500 ~0.31916 0 0.19329 0 0.90879 500 —0.60959 0 0.371090 =1.15928 0.37: L078,sis JOURNAL OF THE ABRONAUTICAL SCIENCES ‘The rib has not been defined as yet. case, the ri for the rib: 0.13086 [K] = Et} —0.00976 re | 0.06413, — 0.00076 —| In the second ease, the rib is treated as a flat plate. rectangular flat plate is of order 8 X 8. ‘Two possible rib configurations will be analyzed in this paper. is considered as a beam identical in section to the spar. -SEPTEMSER, 196 In the first ‘This leads to the following stiffness matrix 0.00098, 0.00076 0.18086 0.00098 0.00976 0.00088, (308) ‘The general stiffness matrix which has been derived for a However, in the present instance, the following conditions must be intro~ ‘Guced to insure compatibility with the other portions of the structure (see Fig. 15 for subscript locations): my “and and, likewise, for the forces Fa = Fae Fy = Fw and Fa = Fev Pay = Py “Treating the rib asa flat plate (F = 0.050 in.) and apply ing the above conditions leads to the following rib stiffness matrix: 5.65088 1x) = 2] 0.87500 0.05764 Ki= 7) 1issisi —0.37500 5.65088 0.87500 0.03754 0.37500 0.03754 (30b) It is anticipated that the choice of rib will have little ‘effect on deflections due to the bending-type loading and a more pronounced effect on the twisting-type Toading, 2.04782 =0.37500 1.52864 z1| —0.05227 —0.00076 (KI) ="2| 019829 0 ° 1.09515 = ° 0.00078 = ‘Phe inverse of this matrix is the flexibility matrix, Pa, Fa 0.81646 0.22705, 2 | -10,47344 0.20384 0.08123, = 5.55027 1.66224 2.72985, 0.08123 1.26026 5.01982 From the flexibility matrix, deflections due to applied loadings we find the following (rib treated as beam). ‘Using the same technique as described for the simple truss, it is now a straightforward matter to form the stifiuess matrix for the complete box. Advantage can be taken of the following: (1) structural symmetry that exists for the box with respect to the sy-midplane and (2) restriction in this problem to loads that act normal to this plane, Under these conditions each pair of upper and lower surface nodes will experience, jn addition to equal vertical deflections, equal but ‘opposite displacements with respect to the xy-midplane. In other words, the box will deflect in the sense of a conventional beam. ‘The spar and rib stiffness ma: trices already provide for such elastic behavior. The plate stiffness matrices make no distinction, other than jn the sign of the node forces, for a reversal in direction ‘of node displacement. Consequently, if the normal loading is carried equally by upper and lower nodes, only the upper set will need be considered when forming the box stiffness matrix, Due to the division of load. ing, correct deflections will result. In this manner the stiffness matrix for the box is found to be [Eq. (20a) used for rib stiffness] yoo wo 3s momo Fa 409, 29098 5.55027 5.01982, 142, 67781 0.81646 = 0.22705 10,4734 1.6622: = 2.72085 409. 39998, toads can be found at once. For the two cases of applied Case Fofirst Case 1 (bending): Forces of 1 1b. acti w= 10N.85/E = ty = NONS5/E th Case 2 (twisting) BBM M/E th - =-5,3045/E Similar results may be calculated for the case when the rib is assumed as a plate, Complete details are not given. In bending we get w: = 10,888.12/E, ay = ~310.56/E, and 1 = —1825/E. Twisting results are tu = 3615,72/E, t = —25.84/2, and 0 349.52/E. It is now advisable to select additional nodes and recalculate the previous deflection data, When added nodes have little effect on results, the process can be considered to have converged. Whether convergence be to the correct values requires additional information, ‘These questions are now examined. First, solutions are found for the node patterns shown jn Fig, 16, Vertical deflections at node 1 for Dending-type loading are as follows: Fig. 16(0 say = 8558.0/E Fig. 1606) Fig. 1646) , toy = S591,9/B S.A/E tis scen that the change in w in going from the node pattern of Fig, 16(b) to 16(¢) is about 1/2 per cent Consequently convergence can de asstimed to have teen attained with the solution found from Fig. 16(b). Obviously the frst solution, based on Fig. 15, is in considerable error. This is due to the poor tie between Spurs and cover plate, Fig. 16(a) introduces an addi onal tie between these two components. ‘The de creased value of t for this case therefore reflects the added stiffuess duc to including the two nodes at the rmid-span location. ‘An tinexpected result is the close agreement between the solutions based ont Figs. 16(a) and 16(b). In fact it would seem reasonable to expect Fig. 16(b) to lead to a smaller value for tw than that given by Fig. 16(a) Careful scrutiny, however, indicates that these results are quite reasonable. Whereas the node pattern of Fig. 16(b) accontts for shear lag in the cover plate this is not the case with Fig. 16(a). As a result, the added stifiuess in Pig. 16(b}, due to te additional nodes ‘connecting spars and ‘cover skins, is offset by the ‘xdded flexibility introduced by shear lag in cover skins ‘The results indicate these factors to be nearly equal; hence the reason for the nearly correct values given by Fig. 16(a). Fig. 16(¢) allows for shear lag and, at the same time, provides for adequate tie between spars and cover . STIFFNESS AND DEFLECTION ANALYSIS \g upward at each spar (nodes 1 and 2). 320.47/E 320.40/E Force of 1 Ib, upward at node 1 and 1 Ib. downward at node 2, =98.46/E = 154. 99/E WB 46/E a = 154. 99/E. plates. It can therefore be felt that this node pattern Will give final results which represent convergence of the method, As mentioned previously, this is substan tiated by comparison with values obtained from Fig. 164) ‘There remains the question as to what is the correct value for w; for this problem. Elementary beam Uheory gives wi — 6,900/2, and, if extended to include shear distortion of spar webs, gives t = 7,740/F. Using Reissner’s shear lag theory," the tip deflection is obtained as t4, = 7,900/E. "Finally if Reissner’s shear fag theory is modified to include spar shear web de- formation, the result is ws = §,740/E. This is the ‘most acctrate theory available, Tt agrees to approxi mately 2 per cent with the numerical sokution based on stifiness matrices “The pronounced shear lag effect in this problem and its marked influence om the vertical tip deflection are mificant. It is precisely this effect that produces a very complex stress distribution in the cover skins. Nevertheless the plate stiffness matrix developed in Bq, (27a) and based on tsiangulur subelements repre sents this stress pattern with gratifying effectiveness "The solution for the node pattern of Fig. J6(6) was obtained in a few minutes by utilizing a program for a high-speed digital computer that computed individual plate and spar stiffaesses and then combined these into the stiffness matex for the complete bor. (XII) Repucrion 1 OgveR oF STIFFNESS MaTRIX 1) Eliminating Components of Node Displacement In an actual problem—as a wing analysis—the num ber of moxles to be ised can become quite large. Ti, for purposes of discussion, 80 nodes are assumed, the stifl- ness matrix becomes of order 150 X 150, By elimi nating w and veomponents of displacement at each node, the stiffness matrix can be reduced to order 50 X 50. ‘However, this reduction process [see treatment of Eq. (23), for example] can require the calculation of the inverse of a 100 X 100 matrix. Such calculations are best avoided at present. "The problet that arises in eliminating the « and components can be handled satisfactorily in any one of several ways. First, the calculation of the inverse of a large-order matrix can be avoided by eliminating @ Single component at a time. This is a practical ex pedient when automatic digital computing equipment820 JOURNAL OF THE ABRONAUTICAL SCIENCES SEPTEMBER, 1956 (ce) (d) io. 18, Nodes and supports fur clamped rectangular plate is available. Second, in some cases it may be feasible to climinate “blocks” of w and v components at a time, thereby reducing the order of matrices to be inverted at any one ime to a reasonable size (say 20 X 20). ‘Third, the analysis can be carried out for sections of the structure, taken one by one. For each section, as a spanwise portion of the wing, the complete stiffness matrix can be determined. Elimination of w and 2 components can then be carried out at any selected nodes, except those common to two distinct sections of the structure. Each section can be treated in this, manner. By properly adding the individual section stifess matrices, the total stiffness matrix can be ob- tained, Finally « and o displacements at nodes where the sections join together can be eliminated. ‘The stiff- ness matrix that remains will apply to w deflections only. From a practical standpoint, the method just de scribed has several worth-while features. For ex- ample all components of displacement at a given node may be eliminated. ‘This can be useful when addi tional nodes are felt to be necessary in order to account, properly for regions of maximum structural com- plexity. Even though eventually eliminated, these nodes will have contributed to the elements retained in the stiffness matrix. (2) Inversion of Stiffness Matrix Ordinarily, only the first few low-order vibration nodes and frequencies are required for the purpose of carrying out subsequent dynamic analyses. Using the stiffness matrix directly in the matrix iteration, ‘method leads to the highest frequency and correspond- jing mode. If the order of the stiffness matrix is high (say, 50 X 50), it becomes impractical to eliminate successively the higher modes and so eventually obtain the lowest modes, Inversion of the stiffness matrix leads to the flexi- bility mattis. This matrix used in the matrix iteration procedure yields results for the lowest mode, ‘There- fore, it is ordinarily preferable to know the flexibility mattis. If the stiffness matrix is of high order (say, 50 X 30), inverting it becomes a major problem it itself This can be overcome to some extent by employing the capabilities of present-day digital computing equip- ment. However, in many instances an alternative procedure may either be useful or necessary. Conse- quently, a possible approach to overcoming this dif culty will be outlined here. ‘The proposed method consists of converting the original stiffness matrix K into a lower order stfiness matrix K*. This is accomplished by introducing a set of generalized coordinates which are related to the original displacements (on which X is based) through a set of appropriately chosen functions. The accuracy inherent in K will have a direct influence on K* Suppose K is known for the cantilever beam of Fig. 17. The order of K is 10 X 10. Now assume a set of polynomials of the form, cover SkiN=te +0.08" SPAR WEB=t,=005" A ee ps 6365 50. Fio. 1, Cantilevered box beat, we! x Fro. 15. Simplest node pattern for box bea.Pi(x) = ant + bs? + att] Pex) = ast + beet + ox ; (33) i Pua) = aut + be + 0 Bach of these will be made to satisfy the boundary conditions of the cantilever which are, (oy = P/O) = P&M) = PME) = 0 [Applying these conditions results in PAG) Pas) ole/L)? — Ae/D + W/L) ao(e/L)? — role/L)* + G/L Py(s) = 140(2/E)* ~ 56te/LY + DN") We now introduce generalized coordinates gc which 466 Meated to the displacements 9, through the above paly- ree “Thin relationship i established through the equations an Pils) Pal) Pale) (> Palas) Par) «+ Palo Vit w eae a top= G8) ere 9s) | qs Lye) Lesieuy Pato Poo | At seen that the ten displacements Yu Io to be replaced by the five coordinates i f+» “The free vibration problem for the cantilever ca Pe ct up in terms of kinetic and potential eneriies,| 1 terms of original displacements 1 Sm these cenergies are, respectively, r= cya) st EA sand 1 V = a/2 ty’ Udit F anere (M] is the inertia (mass) matrix and [AJ the Driginal 10 X 10 stifiness matrix ‘Writing Eq. (35) as bl = Pita and substituting into Eqs. (36), T= a/2) (a PY Be ha v= (1/2) fo)’ (PI 1d PI ah from which we define IK"| {Py (KV {PIL tar’) = (PY (ay iP UE K is of order of 10 X 10 and P of order 10 X 5, K” Tek pe of order 5X 5, ‘The vibration analysis is now Wironmed using X* and M*. By inverting 1° Oe rier modes can be calculated directly. Or alterna: vary, Ke can be used and all modes and frequencies (39) en STIFFNESS AND DEFLECTION ANALYSIS 3 4 fo) we) ro) ‘pio, 16, Additional node patterns for box Dea _ 23456789 f. 10 EQUAL een) evo 1g 17, Station seletions on enntilever bea determined, starting with the highest Ths is feasible fe fer is of sufficiently low order (say, 10 X 10) “This process can be modified in several respects, and the purpose here isnot to give an exhaustive treatment rae Pather co simply point out a possible approach fe the problem. Preliminary caleulations, indice that Bee Rea may possess practical value. Extension 10 the tensioned grid can be made by generalizing the procedure suggested above. APPENDIX (A) Dearvarion oF SPAR Srav#Ness MATRIX -the structure and notation are desered in Section (6X) and Fig. 4. Trans are assumed to carry axial stresses while Ue veatenthics shear stresses. Cover plate material is not wepiaied as part of spar flanges. Derivation below i ppased on conventional beam theory’ Case vy = can! 9% 0; all other components of node ds- placement for the beam = 0. “The deflected beam and necessary forces and rea ce ee thown in Fig. A-l. Due to forces Fat the Ho a the beam deflects upstard. The F, forces vege downward defection. Beam theory, including Sfrects of uniforray distributed shear im web, gives PyhLt , 2FoL* ao Pehl, Fol w= 5 ‘BET (+n) ab Poble , Felt : o= en * eD (2 where w and # are deflection an slope atthe left end of