Analysis and Design of Flight Vehicle Structures by E F Bruhn PDF

Chapter No. al z as Ag AIO Att Al alg. Ala AIS, Als Al? alg. 24 | 25 A26 TABLE OF CONTENTS ‘The Work of the Aerospace Structures Engineer. STATICALLY DETERMINATE STRUCTURES (Loads, Reactions, Stresses, Shears, Bending Moments, Deflections) Equilibrium of Force Systems. Truss Structures. Externally Braced Wings. Landing Gear. Properties of Sections - Centroids, Moments of Inertia, ete. General Loads on Aircraft. Beams Shear and Moments. Seam - Column Moments, Torsion - Stresses and Deflections. Detlections of Structures. Castigliano’s Theorem. Virtual Work. Matrix Methods. THEORY AND METHODS FOR SOLVING STATICALLY INDETERMINATE STRUCTURES Statically indeterminate Structures. Theorem of Least Work. Virtual Work. Matrix Methods. Bending Moments in Frames and Rings by Elastic Center Method. Column Analogy Method, ‘Continuous Structures » Moment ‘Slope Deflection Method. ‘bution Method, BEAM BENDING AND SHEAR STRESSES. MEMBRANE STRESSES. COLUMN AND PLATE INSTABILITY. lid and Open Sections - Shear Center. Flow in Closed Thin-Walled Sections. Membrane Stresses in Pressure Vessels. Bending of Plates. Theory of the Instability of Columns and Thin Sheets. INTRODUCTION TO PRACTICAL AIRCRAFT STRESS ANALYSIS Introduction to Wing Stress Analysis by Modified Beam Theory. Introduction to Fuselage Stress Analysis by Modified Beam Theory. ‘Loads and Stresses on Ribs and Frames. Analysis of Special Wing Problems. Cutouts. Sheer Lag. Swept Wing. ‘Analysis by the “Method of Displacements”. THEORY OF ELASTICITY AND THERMOELASTICITY The 3-Dimensional Equations of Thermoelasticity. ‘The 2-Dimensional Equations of Elasticity and Thermoelasticity. Selected Problems in Elasticity and Thermoelasticity.TABLE OF CONTENTS Continued Chapter No. FLIGHT VEHICLE MATERIALS AND THEIR PROPERTIES: 81 Basic Principles and Definitions. 82 Mechanical and Physieat Properties of Metallic Materials for Flight Vehicle Structures, geese Bsoae 10 on 12 13 D1 D2 D3 STRENGTH OF STRUCTURAL ELEMENTS AND COMPOSITE STRUCTURES Combined Stresses, Theory of Yield and Ultimate Failure. ‘Steength of Columns with Stable Cross-Sections. Yield and Ultimate Strength in Bending. ‘Strength and Design of Round, Stream Torsion and Combined Loadings. Buckling Strength of Fiat Sheet in Compression, Shear, Bending and Under Combined Stress Systems. Local Buckling Stress for Composite Shapes. rippling Strength of Composite Shapes and Sheet-Stiffener Panels in Compression. Column Strength. Buckling Strength of Monocoque Cylinders. Buckling Strength of Curved Sheet Panels and Spherical Plates, Ultimate Strength of ‘Stiffened Curved Sheet Structures. Design of Metal Beams. Web Shear Resistant (Non-Buckting) Type. Part 1. Flat Sheet Web with Vertical Stiffeners, Part 2, Other Types of Non-Buckling Webs. Diagonal Semi-Tension Field Design. Part 1. Beams with Flat Webs. Part 2, Curved Web System. ‘Sandwich Construction and Design, Fatigue. (Oval and Square Tubing in Tension, Compression, Bending, CONNECTIONS AND DESIGN DETAILS Fittings and Connections. Bolted and Riveted. Welded Connections. Some important Details in Structural Design. Appendix A Elementary Arithmetical Rules of MatricesAccelerated Motion of Rigid Airplane ee AMD sireratt Bolte. 110122) DL? Sirerait Sure. 102202 Dna Alzeralt Wing Sections - Ai ‘Types sees aircraft Wing Structure = Tress Ty air Fores : Allowable stresses (and titeractions) 1.38 Analysis of Frame with Pinned Supports. ARI Angle Method ct pplication of Htatrve Methods fo Various Structures... AT.22 applied baad sss ALL Axis of Symmetry | 112... AR4 Beaded Webs... .. . . - 10.16 ‘Beam Design - special Cases. 3.10 ‘Beam Fised End Moments by ‘Method af Area Moments. AT.32 Beam Rivet Design 10.3 ‘Beam Shear and Bending ‘Moment ASL ‘aeame = Forces ata Seccion | AS.7 Beams - Moment Diagrams. | A8.8 Beains with Non-Parallel Flanges - one Beams - Shear and Moment Dageams AB Beams ~ statically Deter~ tminate & Indeterminate... AS.1 Banging and Compression of Columns se es ss Aa Bending Moments © Elastic Center Methods = 2... ADT ending of Rectangular Plates alata Bending Strength ~ Baste ‘Approach. 3.1 Bending Strength - Example” Problems see 3.4 Pending Strain of Rowse : chs Bending Stengih Soli ‘Round Bar ees ot Bending Strasses |... At Bending Stresses - Curved Beams ALLIS Bending Stresses - Elastic | Raage = pees ALAS Bending Stresses ~ Non ‘omogeneous Sections . . . . ALG. IL Benaing Stresses About Principal Axes... Aa. Betaing of Thin Plates ais. i0 Bolt Gencing Strength. © |). DL.8 Be Cog Strength Avaya Methods... Balt Snear, Tension & Bending Serengsts Bousaary Cancitions Box Beams Analysis grass Backing CSeificiens ‘Buckling of Fiat Panela with Dissimuar Faces auciding of Flac Sheets ander Combined Loads « ° Dueling st Reeser plates : INDEX ‘Buckling of StUfened Flat ‘Sheets under Longitudinal Compression cat Buckling under Sending Loads C5. 6 Buckling uncer Shear Loads. . C5.5 uekling under Transverse Shear cee Che Garry Over Factor»... ALLS Castigliano's Theorem «| 2. a7.5 Controids - Centor of Gravity. 3.1 Gladding Reduction Factors... 5.5 Column Analogy Method... AL0.1 Column Curves - Non Dimensional. «ee se + Columa Curves - Solution Columa End Restraint... Columa Formulas « Columa Strength. « ° . Columan Strength with Keown” End Restraining Moment. . C2.16 Combined Axial and Trans- ‘verse Loads ~ Generat Aeuan = eee AB AL combined Sending aid) Compression. =... . . C422 Combined Bending and Flecural shears... . C210 Combined Bending ad Tension se CRB Combined Bending and ‘Tension or Compression of Thin Plat Combined Bending & Torsion | ‘Combined Stress Equations « Compatability Equations. Complex Bending - ‘Symmetrical Section. . Compressive Buckling Siress for Flanged Elements Conical Shells ~ Buckling ‘Strengths «+ 2 can Constant Shear Flow Webs | Ald. 10 Constant Shear Flow Webs ~ Single Cell - 2 Flange Beam. A1S.9 Constant Shear Flow Webs ~ Single Cell - 3 Flange Beam. A1S.5, Continuous Structures = Curved Members... ~~ AILST ‘Continuous Structures = Variable Moment of ertia . All. 18 core Shear eee ee ese + « C1226 Gorrection for Cladding: | ¢ 1 C14 orrugated Core Sandwich Paulure Modes coszone Procedsre | reap of Materials | | Creep Pattern ripping Stresses Caieulations cee OUT critical Shear Stress ||| | | C116 Srystallization Theory 2) Guimaatie Oamage Theary. | C122 Curved Bears . “35.3 Curved Sheet Paaeis = ‘Buckling Stress cat Garved Web Systems | 1) | 611.39 Cutouts in Webs or Sita Panella. ss = ee DOT Deflection Limitations in Plate Analvses.-- AIT.S Daflections oy Siastie Weights AT.27 Deections by Moment Areas. AT.20 Deflections for Thermal strains «= oo. ATAT Dallections oy Virewt Work |” A.2 Delta Wing Example Problem. 23.2 Besign for Compression... C42 Design Conditions anc Dasien ‘Weights - AS Design Fiz equiremenis| for atrplane = - - M8 Design Loads DILL age Design for Tension ©2211. CAL Dilferentiat Equation of Deflection Surface...» .% A812 Discontinaities 2 820.15 Distribution of Loads to Sheet Panels. vv ves sss ARLZ uctuiey - DIITIT “aus Dummy Uat Loads Ll age Dynamic Elfect of Air Forces. Ad. 19 -Bilect of Axial Load on ‘Moment Distriution. . . . . ALL 22 lfective Sheet Wiaths ||. C710 Blaatie Buckling Strength af ‘Flat Sheet in Compression. . 5.1 dlastie ~ Inelaste action... BIS Elastic Lateral Support Columns. CaiT Blastie Stabulity of Columa |. A172 ‘lastic Strain Energy cL Elasticity and Thermo: lasticty = One-Dimensional Proolema 326.1 Blastictty and Thermo- elasticity Tao-Diroensional Equations. « ne ASE Electric Are Welding Daz End Bay Bests Lol euts End Momenta for Continuous Frameworks. vss «ALL IO Bquattons of Static Equilibria... es ARE Equiltorium Equations |. | A282 Failure of Columas by Compression. 2s ALBA Failure Modes in Curved Honeycomo Panels... . . C12.20 allure of Structures ©.) BLT Fatigue Analysis - Statistical Distribution =...» 19.4 Extipie and Fail-Safe Design | 19.3 Fatigue of Materials. « Bits Faugue SN Curves «001 Fillers Furning Design Fixed Bod Moments Fixed End Moments Due to Support Deflections . « Eixtty Cooliicients. Flange Design - Flange Desugn Streoses : Flange Discontinuities. Flange toads Flange Strength (Cripeling) - Flat Sheet Web ota Vertical | Susteners Flemural skedr Flow ‘Distribveion vs ss « ALS. 2M Flemura! Shear Flow ~ ‘Symmetrical Beam Section = AL4.S Flemaral Shear Strese. sss ALEStatic Tension Stress Strain Diagram =... . BL Statteally Determinate Coplanar Structures and Loading ee ee ART ‘statically Determtnate ane Tndeterminate Structures ..A2.4 Statically Indeterminate Frames ~ Joist Rotation. . AL2,7 Statically indeterminate Problem sss AB Stepped Column © Strengin | C24 Stifened Cylindrical ‘Stractares ~ Utimate Strengths... tee C88 Stifiness & Carry-over ” Factors for Curved Members All. $0 Stufness Factors s+. sss + ALLA ‘Strain ~ Displacemeat Relations eee ee ee ee Ans ‘strain Energy | DOAnL ‘Strain Energy of Plates Due Bending vee es - A119 Strain Energy in Pare Bending of Plates. ee ALBA Streamline Tabing = Strength | “C4. 12 Strength Checking and Desigs = Problems... .. C45 Strere*" °: Round Tubes -_ ster Combined Loadings .. C4.22 ‘tress Analysis Formulas | | C1115 Stress Analysis of Thin Skin « Multiple Stringer Cantilever Wings ee ee ee es AITO Seress Concentration Factors | C13.10 ‘Stroge Distribution & Angle ‘of Dwist for 2-Cell Thin Wall Closed Section . « 8.7 stress-strain Curve. |.) 1) BLT Stress-strain Relations | |! | Anda Stresses around Panel Culost, 422.1 Stresses in Uprights cuit Stringer Systems in Diagonal Tension. cus Structural Design Philosopay C13 Structural Ficings vo wwe. AZ2 Structural Skin Panel Details | 3.32 Structures with Carved Members... . . ALL 29 Successive Approxiciation ‘Mettoa for Multiple Cell Als.26 Fitting Units... ARS ‘Symmetrica Sections External Shear Loaded « ai, INDEX - Continued Tangent Mogulus... ‘Tangent-Modulus Theory | Tax Loacs Tension Clips | | | ‘Tension-Field Beam Action. Tonsion- Field Beam Formaiaa ‘Theorem of Castiglione » ss = Tygorem of Complemeniary Energy nee ‘Theorem of Least Wore’ | | ‘Theorems of Virtual Wort and Minimum Potential Energy ‘Thermal Detlections by Matrix Methods. = Dimensional Equations. = . Tata Walled Shella = S| ‘Three Cell ~ Multiple Flange Beam - Symmetrical about One ANS vet es ee ets Three Flange ~ Sireie Cali Wiog Torsion - Clreular Sections. ‘Torsion - Btfect of End Restraint ‘Torsion ~ Nos-eireuiar Sections. . cee Torsion Open Sections’ * > >: Torsion of Thin-Walled Shiner 2aving Closed Type jitters ss Torsion Tain Wailed sections | Torsional Moments ~ Beams. Torsional Medulus of Rupture. ‘Torsional Shear Flow in ‘Multiple Cell Beams by Method of Successive Corrections. =» ‘Torsional Shear Sesses in. ‘Multiple-Cell ThineWall {Closed Section - Digtri>ution ‘Torsional Streagta of Round ‘Tubes ents ‘Torsional Stresses ta Maltiple-Cell Thin-Walled Tubes ‘Transmission of Power by ‘Cylindrical Shaft. =. . : ‘Trlarial Stresses =. : ‘Truss Deflection by Method” of Blastle Weights... ‘Truss Structures... . | 1 BLS ale, 13,33 3.2 cunt ema ALS ans Aga Ans 8.39 ans A233 AlS.18 Ano. aB1 A616 8.3 ass AG. 1S 8.5 ASS ait 8.10 AGT 4.17 Redundancy oes... 26. ABT ‘Tubing Design Pacis 2251) Cals ‘Two-Dimensional Proslens, | Az6.5 ‘Two-Cell Multiple Flange Beam - One ds of Symmecy ‘Type of Wing Rite ‘Ultimate strength in Combined ‘Bending & Flemrai Shear... 4.25 Ultimate Strength in Combined Compression, Bending, Flemral Shear & Torsion. . C4.28 Ultimate Strength in Combined ‘Compression, Bending & Torsion sees ca.2e Ultimate Strengt in Combined ‘Tension, Torsion and Internal Pressure p in pat. . 4.26 = AIS IL ‘Aa Uniform strese Condition. | Ci-1 5.15, Aa.2 eee cee es AIOE Saaymmesical Fraes wig Principal Axes 29.18 neymmetzical Structures © Ad. 13 Vv. 'y = Load Factor eee AAT Wagner Equations .. =... C14 Web Bending Shear Stresses C10.5 Web Design oe ee es CLLRS Web splices |. 521121)! cxolt0 Web Strengen, ‘state Webs | | “C10. § ‘Webs with Roun Lightening” Holes... nee = C1OIT Wing Analysis Protlerae °! > “A19.2 Wing Arrangements... <<. AIG] Wing Hitective Section ©.‘ ata. i2 ‘Wing Internal Stresses. . | | 29,44 ‘Wing Shear and Beading Anaiyeia . . ee AIDE Wag Shar and Beidisg waoments wee AB Wing = Shear Lag: 2222 LD aia, 28 Wing Shears ana Moments’ | |” As. 10 Wing Stitisess Macrte. . . | | 23-11 Wing Strength Requirements | Al3. 5 Wing Stress Anaigais Methods 419. 5 Wing ~ Ultimate Strength... A19.12 Work of Structures Group. | |” Ai.2 Y stiffened Sheet Panels... C7.20CHAPTER AL THE WORK OF THE AEROSPACE STRUCTURES ENGINEER ALI Introduction. The first controllable human flight in a heavier than afr gachine was made by Orville right on December 17, 1908, at Kitty Hank, North Carolina. It covered’a distance ot 120 feet and the duration of flight was twenty Seconds. Today, this initial flight sppears very unimpressive, but 1t comes into its true perspective of Importance when we realize that mankind for centuries nas dreamed about doing or tried to do what the wright Brothers accomplished in 1903. ‘The tremendous progress accomplished in the first 50 years of aviation nistory, with most of tt occurring in the last 25 years, 1s almost unbeltevable, Dut without doubt, the progress im the second 60 year period will still be gore unbelievable and fantastic, 4s this 1s sritten in 1064, jet atrlins transportation at 600 MPH 4s well established and several types of milttary atrerazt nave speeds in the 1200 to 2000 MPH range. Preliminary destgns of a Supersonic afritner with Mach 3 speed nave besn completed and the government is on the verge of Sponsoring the development of such a flight vehicle, thus supersonic air trangportation should become comon in the early 1970’3. The rapid progress in missile design aas ushered in the Space age. Already many space venicles ave been flown in search of new knowledge which 1s needed before successful exploration of space such as iandings on several planets can take place. Unfortunately, the rapid Gevetopment of the alsstle and rocket power has given mankind a flight vehicle when combined with the nuclear bomb, the awesome potential to quickly destroy vast Tegions of the earth. Wdntle no person st present knows here or what space exploration will lead to, relative to benefits to mankind, wwe do ion that the next, great aviation expansion basides supersonic airline transportation will be the full develop zent and use of vertical take-off and landing airoraft. Thus persons who will be Living through the second half century of aviation progress will no doubt witness even nore fantastic progress than oceurred in the first 50 years of avtatton history. AL.2 General Organization of an Aircraft Company Engineering Division. ‘the modern comercial airliner, military airplane, missile and space vehicle 1s 4 alghly sctentific machine and the combined know! and experience of hundreds of engineers and scientists working in close cooperation is necessary to insure a successful product. Thus the engineering division of an aerospace company consists of many groups of specialists whose specialised training covers all tlelds of engineering education such as Physics, Chemical and Metallurgical, Mecnantesl, Zlectrical and, of coursa, Aeronautical angineering. It so happens that practically all the aerospace companies publisn extensive pamphlets or brochures explaining the organization of tae enginesrtng division and the dutics and Fesponsibilities of the sany sections and groups and illustrating the tremendous laboratory and test facilities which the aerospace industry possesses. It 1s highly recommended that the Student read and study these tree publications in order to obtain an early general under- Standing on how the modern flight vehicle is conceived, designed and then produced. In general, the engineering derartuent of an aerospace company can be Droken down into six large rather distinct sections, watch in turn are further divided into specialized groups, woich in turn are further divided into sualler working groups of engineers. To tllustrate, the six sections will be listed together with some of the vartous groups. ‘This 1s not a complete List, but it should give an {dea of the broad, engineering set-up that ts necessary. re Pre liminary Design Section. Technical Analysis Section. a @) (3) &) } G) Aerodynanies Group Structures Group veignt and Salance Control Group Foner Plant analysis Group Matertals and Processes Group Controls Analysis Group 11]. Component Design Section. a (2) Structural Design Group (tng, Body and Contrel Surfaces) Systems Design Group (11 aecnanical, Aycraulic, electrical and thermal installations) IV. Laboratory Tests Section, ALLALa THE WORK OF THE AEROSPACE STRUCTURES ENGINEER (2) Wing Tunnel and Fluid Mechanics Test ‘The final results of the work of this Labs. group are formal reports givins c-mp! (2) Structural Test Labs. load design eriteria, with many grachs (3) Propulsion Test Labs. mary tables. The final results (A) Bleetronics Test Labs. plete shear, moment and normal £1 (3) Electro-Mechanieal Test Labs. to a convenient set of V2 axes (5) Weavons and Controls Test Labs. creft untts such as the wing, © (7) Analog end Digital Computer Labs. V. Plignt Test Section, VI. gineering Field Service Section. Since this textbock deals with the subject of structures, tt seems appropriate to discuss in some detail the work of the Structures Group. For the detailed discussion of the other groups, the student should refer to the vartous aircraft company publications. AL.3. The Work of the Structures Group ‘The structures group, relative to number of engineers, ts one of the lergest of the many groups of engineers that mke up Section If, ‘the technical analysis section. The structures group ts primartly responsible for the structural integrity (safoty) of the airplane. Safety may depend on sufficient strengtn or suffictent rigidity. This structural integrity must be accompanied with lightest possible watght, because any excess weight has detri~ mental effect upon the performance of aircraft. For example, ina large, long range missile, one pound of “nnecossary structural welghe may add more than 200 Ibs. to the overall welght of the missile. ‘The structures group ts usually divided {nto sub-groups es follows: (1) (2) 3) a) THE WORK OF THE APPLIED LOADS GROUP Applied Loads Calculation Group Stress Analysis and Strength Group Dynamics analysis Group Special Projects and Research Group Before any part of the structure can be finally proportioned relative to strength or rigidity, the true external loads on the air craft aust be detemmined. Since critical loads cone from asny sources, the Loads Grcup must analyze loads from aertdynamic forces, as well as those forces fron power plants, alferart inertia; control system actutors; launching, landing and recovery gear; armanent, etc. The effects of the asrodyramic forces are initiaily calculated on the assumption that the airplane structurs is a rigid body. afte: the atreraft structure 1s obtained, 1ts true rigidity can be used to obtain dynamic eftects. Results of wind tunnel model tests aro usually necessary in the application of aerodynamic principles ta load and pressure analysis. THE WORK OF STRESS ANALYSIS uD ST aCuP Essentially the primary job of the stross group 18 to help specity or determine the kind of material to use and the thicmoss, size ant crogs-sectional shape of every su: ber or unit on the airplane or aisstle, and also to assist in the design of all Joints and connections for such members. Safety atv: Weight are the parancunt stractural design requirements. ‘The stress group must constantly work closely with the Structural Design Sect’ im order to evolve the best structural over-all arrangement. Such factors a3 poser plants, built in ful tanks, landing gear retracting wells, and other large cut-outs can dictacs = type of wing structure, as for example, a two spar single cell wing, or a multiple star multiple cell wing. To expedite the initial ser: sesign studies, the stress group must supply Lnitiel structural sizes based on approxizate loads, ‘The tinal results of the work by ne stress group are recorded fn elaborate reports which Show how the stresses ners calculated and now the required member sizes wore obtained to carry these stresses effictently. The tinal size of a member may be dictated by one or more factors Such as elastic action, inelastic action, ele~ vated tenperatures, fatigue, ete. To ingure ations, she the accuracy of theoretical cal: stress group must have the assistance of t! structures test laboratory in omer to obtain information on which to base allowable desizn stresses. THE WORK OF THE OYNAMICS ANALYSIS GROUP ‘The Dynantes Analysis Croup has rapidly expanded in recent years relative to number of engineers required because supersonic airplane: misatlas and vertical rising aircraft neve 91 sented many naw and complex problens in the generel field of aynanics. In some alrerart companies the dynamics group {s set up as a Separate group cutside the Stricturas Group, ‘The engineers tn the dynanics group are Pesponsible for the investigation of vibration and sheck, atreratt flutter and the establish— ment of design requirements or changes for its control or correction. Aircraft contain dozens of sechantcal installactons. vitration of any of these installations or sysceas may 32 Sf such charscter as 70 cause faLlsy eperstion or danger of failure and theresore the cynamicANALYSIS AND DESIGN OF FLIGHT VEMICLE STRUCTURES characteristics aust be changed or modifted in order to insure reliable end safe oparation. ‘The major structural units of atreraft such as the wing and fuselage ere not rigid bodies. ‘Thus When & sharp air gust strikes a flexible Wing in high speed [light, we have a aynanic load eftuation and the wing mill vibrate. The dynamicist gust determine whether this vibration 18 Serious relative to Inducad stresses on the wing structure. The dynanies group ts also responsible tor the determination of the stability end performance of missile and flight vehicle guidance and control systems. The dynamles group must work constantly sith the various test laboratories In order to obtain reliable values of certain factors that are necessary in many theoretical calculations. THB WORK OF THE SPEQTAL PROJECTS GROUP In general, all tne various technical Geeta est | ALS groups have a special sub-group which are work— Ing on design problems that m111 be encountered J che near or sistant fusare es aviation pro- ses. Por example, in the “tructures Group, this sub-group atgnt te studying auch problens as: (1) now to calculate the thermal stresses in the wing structure at super-sonic speeds; (2) now to stress analyze a new type of wing structure; (3) mhat type of body structure ts vest for future space travel and what kind of materials will be needed, te. chert 1 tllustrates tn general a typical nake-up of the Structures Section of a large aeroscace company. Chart 2 lists the many’ items which the structures engineer gust be concerned with in Insuring the structural integrity of the flight vehicle. Botn Charts land 2 are trom Chance-Vougnt Structures Design Manual and are reproduced with tnetr permission. an Ter 1 bo a T ~ cnet Structures Section Organization Chance-Vought Corp.Aa THE WORK OF THE AEROSPACE STRUCTURES ENGINEER THE LINKS TO STRUCTURAL INTEGRITY «+++ ARE NO BETTER THAN THE WEAKEST LINK MATERIALS OF ‘CONSTRUCTION asioers ‘STIFFNESS pees ‘CRITERIA eimisions STRESS UTR Sage aL ANALYSIS. coves Sa SBAY nasi anna Ean PANEL FLUTIER-S4iM CONTOURS wenn STRAIM COMPATEBNLITY, cORROL SSE EREEIONS Shin corti Teun ETS saree Macs canon iaaton, eke was OL OM. DTERCENE ‘mena Sats \ -[asooname cower sh etait sous DYNAMIC RESPONSE. Bers ENT wArsis LOADS AND ENVIROMENT uct oxo carain Bone uo ener ‘oka oad laure ovis ‘aroiwe ores Srwanie wernt atoms Sans i asa COMPONENT ALLOWABLES ‘ Sisto ANALYSIS. raceme ema Lo Scr SUIS ‘ano "ds rot Ha un SCS va at essa ht san awuysis ae snare outs iment THERA AySI5 en rs DERaCHON AALS cones srimess sores sueeuNe Chart 2 ‘From Chances Vought Structures Design Manual MATERIALS AND QUALITY CONTROL uct STRESE-STRAIN esau sss ear aca core stmess coRansion santana rasencune speircavon cononmsnes SUE Pn ConACHAPTER A2 EQUILIBRIUM OF FORCE SYSTEMS. TRUSS STRUCTURES A2.1 Introduction. The equations of static equilibrium aust constantly be used by the stress analyst snd structural designer 1 ob taining unknown forces and reactions or unkno“n Invernal stresses. They are necessary whether the structure.or machine be simple or complex. The ability to epply these equations is no Gouot vest developed by solving many problems. This chapter initiates the application of these Important physical laws to the force and stress analysis of structures. It is assumed that @ student has completed the usual college course in ongineering macnanics called statics. A2.2 Equations of Static Equilibrium, To completely define a force, we must know tts magnitude, direction and point of apptica— sion, ‘These fects regarding the force are generelly referred to as the characteristics of the force. Sometimes the ore general term of Line of ation or location is used as e force characteristic in place of point of applica’ Gestgnation. on A force acting in space ts completely defined 1! we know tts components in three Girections and [ts moments about 3 axes, 2 for sxample Fy, Fyy Fg 2nd My My and Me. For Qquiltoritin of a forea system there can te no resultant force and thus the equations of equilitriwn ara obtained by equating the force and nonent components to zero, The equations of statie equilibrium for the vartous types of force systens ill now 32 sumarized. EQUILTSRIUM SQUATTONS PCR GENERAL EF y az = 0 Thus for 2 general space Zorce systen, there are 6 equations of static equilibrium avaliable, Three of these end ao mors can be foree equations. It ts often more conventent so take the moment 2xes, 1, 2 and , as any set ot x, goneat equations about 4 att 2 force equations are written for autually perpendicular axes and need not be the x, y And 2 axes. SQUILISAIUM OF SPA coneurr force system cass through a comen point. ‘The resultant, 12 any, must therefore be a force and not a moment and thus only 5 equations are necessary to completely define the condition that the resultant must be zero. The equations of equilibrium available are therefore:— BF 0 Fy or 0 wa (2.2) Bz = 0 0 A combination of force and nonent equations to make 2 total of not more than 3 can be used. For the aouent equations, axes through the votnt of concurrency cannot be used since all forces of the system pass through this paint. The moment axes need not be the same direction as ‘the directicns used in the force equations out, of course, they could de. ‘SQUILTER: OF SPACE PARALLEL FORCS SYSTEM In a parallel force system the direction of all forces ts know, but the magnitude and Iceation of each 13 unknown, Tmus to determine magnitude, one equation ts required and for location two equations are necessary since the roree 1s not confined to one plane. in general the 3 equations comonly used to make the Taq sultant zero for this type of force system are one force squation and tao moment equations. For exemple, for a space parallel fores system acting in the y direction, the equations oF squilioriun would be: By = 0, Be = 0, 227 -72.5) JQUILIBRIUM OF GENERAL C In this type of force system all forces lie in one plane anc tt texes only 3 equations to determine the magnitude, direction and location or the resultant of such 4 force system. Sitner Zorce or moment equations can be used, except that a seximm of 2 force aquations ean be used. ‘or example, for a force systex acting in the xy plane, whe Zolloniis combination of equtit- britm eqiations could Se used. Hy 20 y= 0 Fy 0 My. BFy #0 or Mp.2 0 or Meas Oor Met 0 24 Beto MFO Meet Dest 0 (the eubsertpts 1, 2 and 3 to atsterent locations for z exes or moment centers.)Az? DAUILIBEIOM OF COPLANAR. Sines 211 forces 1Lo In the sane olane and also pase through 2 common soint, the Tagn:tuce and direction of the resultant of this tyre of force system 15 unknown Dut the location ts ictowm since the roint of concurrency 1s an the Line of action cf the resultant. Thus only tno equations of equilitrium are necessary to Yezine the resultant and sake {t aero. The combin~ ations avatieole are, Hera Her S eet tO BTS] 2s By 20) B20” Bet” Bye =0 (te z axis or monent canter locations must be other than through the point of concurrency) QUILIBRIUH OF CO-PLANAR PURALLEL FORCE SYST! Since the direction of all forces in thts typo of force syston ie Kom and since. ce forces ait lie in one sane plane, {t only takes 2 equations to decine the sagnitide and Location or the ceoulsant of sucha toree systea. tence, there are only © equations of equilisrie avatie able for this type of force system, manely, a force and aoaent equation or two néuont equations. For exaaple, cor forces parallel to Faxis and locatee in the xy plane the equilt= Eeftm equations available would be: ~ Bes Bige = 0 (Te mouent centers 1 and 2 cannot be on the same y axis) By 20 ° My = 0 EQUILIBRIUM OF COLINEAR FORCE SYST A colinear foree system 1s one where all forces act along the same line or in other words, the direction and location of the forces 4g known but their magnitudes are unimowm, thus only magnitude needs to be found to define the resultant of 2 colinear force system. ‘Tnus only one equation of equiltortum ts avaflable, ramely BPSO or M,50 ~--~----2.7 where moment center 1 is not on the line of action of the force systen AZ.2 structural Fitting Umts for Establishing the Force Characteristics of Direction and Pout of Application To completely define a force in space req quires ¢ equations and 3 equations ir the force ts limited to one plane. In seneral e structure 4s loaded by ‘cow forces ana these forces are trensfarred through the structure ‘n some aanner of internal stress distribution and then EQUILIBRIUM OF FORCE SYSTEMS. TRUSS STRUCTURES. equal he various tynes 9: Limited, the structural engi use o: fitting unite which establish th Sirection of an unimown force oF 1 applica’ Gecreasin af unknowns to be deternined. which follow {llustrate the tye: units employed or over ganerai rethocs © taDlishing the Zorc2 cnaracte! qiraction and potnt of application. Ball and Socket Fitting For any space or coplans> and Q acting on the bar, ¢! Such forces must act trroucn bell {2 rotetton of the var 15 a bali and socket joint can se or control the direction and 1 force appliea to a stru: ‘through of fitting. Since the Joint has no rotati resistance, no couples in any plane can 38 applied to tt. ne action of 2 Singte Pin Fitting, for any force such as P and Q acti xy plane, the line of action 9 sut gust pass through the ain center since © fitting unit cannot resist a axts through the pin center. There: forces acting in the xy plane, 0: and Line of action are establisha: Joint as titu mire, Since Single pin fitting can resist moments about axes perpendicular to the pin axis, the ciraction and Line of action of out of plane forces ts fore not established by sizgle a Bt. —<—<—_——- * Tf a ber AB has each ond, then any 2 strgle pin fittings at ree P lying in the xy plens and applied co end B aust nave 3 direction and Line of action colnelaing sith 2 lina Join tne pin canters at and the fittings cannot axis.ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES Double Pin - Universal J Since single pin fitting units can resist, applied moments about axes normal to the pin axis, a double otn joint as {llustrated above fs often used. Tals fitting unit cannot resist oments about y oF 2 axes and thus applied forces such as P and Q aust have a Line of action and direction such 2s to pass through the center of the fitting untt as {llustrated in the figure, The fitting unit can, however, resist a moment about the x axis or in other words, a universal type of fitting unit can resist a torsional noment. Llers In order to permit structures to move at support points, a fitting unit involving the idea of rollers 1s often used, For example, the truss In the figure above 1s supported by a pin fitting at (A) which ts further attached toa fitting portion that prevents any norizontal movement of truss at snd (A), however, the other end (B) 1s supported by 2 nest of rollers which provide ne hortzontal resistance to 2 horizontal covexent of the truss at end (8), The rollers fix the direction of the reaction at (B) as perpend!cular to the roller bed. Since the fitting untt 1s joined to the truss Joint by a pin, the point of application of t Teaction {5 21s0 mnown, henes only one force characteristic, namely magnitude, {8 unknown fora roller-pin type of fitting, for the fitting unit at (A), point of application of the reaction to the truss 1s kncvm because of she pin, but direction and aegnituee are unaom. Lubricated Slot or Double Roller Type ‘tting| uae. Lubricated stot Ans Another general fitting type that 1s used to establish the direction of a force oF reaction is {Ylustrated in the figure at the bottom of the first column. Any reacting force at joint (A) aust be horizontal stnce the support at (A) is 80 designed to provide no vertical resistance. pe P Since a cable or tie rod has negligible bending resistance, the reaction at Joint B on the crane structure from the cable mst be colinear with the cable axis, hence the cable astablishes the force characteristics of direc ‘ion and point of application of the reaction on the truss at point B. A2.4 symbols for Reacting Fitting Unita as Used in ‘Problem Solation. In solving a structure for reactions, aenber stresses, etc., one aust know what force characteristics are unknown and {t 1s common practice to use simple synbels to indicate, wnat fitting support or attaciment units are to be used or are assumed to be used in the final design, Tho following sketch symbols are com monly used for coplanar fores systaas. A ouall circle at the end of a member or on a triangle reprasents a single pin conection and fixes the point ef application of forces acting between this unit and a connecting member or structure. ea © ZS iy ‘yy ‘The above graphical symbols represent 2 reaction in which translation of the attachment point (b] 1s prevented but rotation of the attached structure about (b) can take place. ‘Thug the reaction ts unknown in direction and magnitude but the point of application 1s known, namely through potnt (b), Instead of using Erection as 2n unknown, {¢ 1s more convenient to replace the resultant reaction by tno com- pononts at right angles to each other as indi~ cated in the sketenes,Ana z BILIBRIUM OF FORCE SYSTEMS. TRUSS STRUCTURES. (), waite Boge bepin Rollers Rollers ‘Te above fitting units using rollers fix the direction of the reaction as normal to the roller bed since the fitting unit cannot resist a horizontal force through point (b). Hence the direction and point of application of the reaction are established ant only magnitude 1s unknown. ‘The graphical symbol above 1s used to Fepresent 2 rigid support which 1s attached Figidly to a connecting structure, ‘The Te- action is completely unknom since all 3 force characteristics are unknown, namely, magnitude, direction and point of application. It {3 con venient to replace the reaction R by two force components referred ta some point (>) plus the ‘unknown moment M which the resultant reaction R caused about point (b) as indicated in the above sketen. This discussion applies to coplanar structure with all forces in tne same Plane. For a space structure the reaction Would have 3 furcher unknowns, namely, Rz, My and My. ‘A2.5 Statically Determinate and Statically Indeterminate ‘Structures, A statically determinate structure ts one in whten all external reactions and internal stresses for a given load system can be round by use of the equations of static equilibrium and 2 statically indeterminate structure 1s one in wnich all reactions an¢ internal stresses cannot be found by using only the equations of equilitriun. A statically determinate structure ts one that has Just enough external reactions, or Just enough internal members to make the Structure stable uncer a load system and if one Teaction or member {3 removed, the structure is Teduced to ¢ linkage or a qecitanisn end is, therefore not further capable of resisting the load systen. If the structure has more external reactions or tnternal members than is necessary for stability of the structure under @ given load system it 1s staticelly incever~ minata, and the degree of redundancy capends on the number of unknowns beyond that number watch can be found by the equations o? static equili- prim. A structure can te statically tnaster- inate with respect to external reactions alone or to internal stresses alone or to 50 The additional equations shat are nesded to solve a statically {ndetersinate structure are obtatned oy considering the distortion of the structure. This means that the izo of sll members, the Tateriel tron which members are ade aust be known since distortions must te calculated. in 2 statically ceteninate structure thts information on izes and aatertal ts not required but only the configuration of the structure as a whole, Thus design analysts for statically determinate structure {s straignt forward wnereas a generel trial and error oro~ cedure is required for design analysis of statically indstersinate structurss. ‘A2.8 Examples of statically Determisats and statically Indeterminate Structures. ‘The first step in analyzing a structure ts to determine whether the structure as oresant fs statically determinate. If so, the reactions and internal stresses can be found without ‘mon= tng sizes of meabers or Kind of matertal. Iz not statically determinate, the elastic theory aust be applied to obtain additional equations. ‘The elastic theory 1s treated in considerable detatl in Chapters A? to Al2 inclusive. To help the student become faniiiar with the problen of determining whether 4 structure is statically deteminate, several example problems will be presented. Example Problen 1. Fig. a2 Tn the structure known forses or loads shown in Pig. 2.1, the are the distributed leads of 10 1d. per inch on menver ABD, The reactions a points 4 and 0 are unknown, The reaction at Chas only one unknown characteristic, namély, magnitude because the point of application of Ry tg tarough the pin center ar C and the directten Of Ro must de parallel to line OB because thers 1s a pia at the other end 8 of member CB. At point A the reaction ts unknown in airaction and magnitude but the point of application must be through the pin center at A. ‘Thus there are 2 unknowns at 2 ané one unknown at C or a cotalANALYSIS AND DESIGN OF FLIGHT VEHICLE strucTURES of S. With 3 equattons of equitibrimm avati- able for @ coplanar rorce system the structure is statically determinate. Instead of using en angla aS an unknown at A to find the direction of the reaction, it 1s usually more convenient to replace the reaction by components at right angles to cach other as Hq and Yq in the figure and thus the 3 unicoms for the structure are 3 aagnitudes. Example Problem 2 F P tao) e $e Ye Ha BA Bap Hal h z Fig. A2.3 YA rig.aaa Ne Ny Fig. 2.2 shows a structural crane carrying a known load system P, Due to the sins at reaction points 4 and B the point of application is known ior eaen reaction, however, the aagai~ tude and direction of each’ is unknowm making a ‘total of 4 unknowns with only 3 equations of equilibrium available ror a coplanar force system. At first we might conclude that the structure is statically indeterminate but we dust realize this structure nas an interme) pin at C which means the bending moment at thie point 1s zero since the pin has no reststance So rotatton. If the entire structure is in equilibrium, then eact part aust likewtse be im equilibrium end we can cut out any portion as a 2ree body anc apply the equilibrium equations. Fig. 2.3 shows a trae beay af the frane to lett of pin atc. Taking acments about © and equating to zero gives usa fourth ation to use In deteraining the 4 unknowns, Ha, Yas Vp and fig. The monent equation about ¢ gees Rot include the unknowns Vo and Ha since they have no moment about ¢ because of zero amas. As in exanple problem 1, the reac=ions 2¢ A and 2 have been replaced Sy # and ¥ com Bonents instead of using an angie (direction) as eh unknown characteristic. ‘The structure is statically determinate, Example Problem 3. > a) ty We D Fig. And NRe Fig. 2.4 shows 2 straight aember 1-2 carrytng a ‘mown load system ? and supported by 5 struts ARS attached to reaction points ABCD. Av reaction points A, 3 end D, the reactaon, {8 known in direction and potnt of application Dut the magnitude te unknowm as indicated by the vector at esch support. At point C, the ren getion is unknown In direction decatise 2 struts enter Joint C. Magnitude 1s also unkown but Point of application is imowm since the reaction Gust pass through C. Thus we have S unknoms, namely, Ro, Ray Rp, Yo and He. For a coplanar force system me nave 3 equilitrium equations available and thus the first conelusion might be that we have 2 statically indeterminate structure to (5-3) = 2 degrees requndant. Hone ever, observation of the structure shows tno internal pins at points £ and F which means that the bending moment at these two points is Zero, thus giving us 2 more aquations to use with the % equations of equilibrium. Thus drawing tree vedies of the structure to left of pin E and to right of pin F and equating moments avout each pin to zero we obtain 2 equations watch do not include unknowns other than the 5 unknowns listed above. The structure 1s thers~ fore statically deterainate. Example Probiem 4. He EOS Mc NG te te we Fe ans re ans Pig. 2.5 shows a beam AB which carries a super-structure CED which in turn 1s subJected to the known leads P and Q. The question 1s woether the structure ts statically determinate. The external unknown reactions ror the entire structure are at points A and B, At A due to the roller type of action, magnitude {s the only unknown characteristic of tne reaction since direction and point of application are known. AG B, wagnitude and direction are unknown Sut point of application 1s imcwn, nence wo fave 3 unknowns, namely, Ry, Vp and Hg, anc with 3 equations of equilitriut available we can tind these reactions and therefore the structure is statically determinate with respect to external Feactions. xe now investigate to seo if the internal stresses can be found oy statics after having found the external reactions. Obviously, the internal stresses will be afzectea by ine internal reactions at C and 0, so we draw 4 tree Dody of the super-strusture a3 tllustrated tn Fig. 2.8 and consider the internal zorces that existed at C and D as external reactions. In ‘the actual structure the members are rigidly attached together at point ¢ such as a welded or ALAnt multiple bolt connection, This neans that all three force or reaction charecteristios, nanel magnitude, direction and point of azplicatien are urengim, or in othar words, 3 unknowns exist atc. For convenience me will represent these unknowns by three components as show in Fig. 2.8, namely, He, Vg and Mo, At Joint D in Fig. 2.6, the only lnknown regarding the ree action {$ Rp @ magnitude, since the pin at each end of the sember DE establishes the direction and soint of application of the raaction Rp. Hence we nave 4 unknoms and only 3 equations of equilitrim for the structure in Fig. 2.€, thus the structure 1s statically inteteminate with respect to al) of the internal stresses. ‘The student should observe that internal stresses between points AC, 8D and FE are statically deterinate, and thus the stactcally indeterminate portion is the structural or} angle CEDC. Beample Probl ° > P ig. a2.7 ig a2.8 ha al Hp ali By ity Py SEBS tye MM Ne Pigs. 2.7, 2.8 and 2,9 show the same structure carrying the same iiown load system P but with different support conditions at points 4 and 3, The question is whether each structure ts statically indeterninate and if 50, to wnat degree, that 1s, what number of unknowns Deyond the equations of statics available. Since we have a coplanar force system, only 8 equations at statics are available for equilibrium of the structure as 2 whole. 3 PoP F MA a . In the structure in Fig. 2.7, the reaction at A and also at 8 1s unknown in magnitude and rection but point of application is mown, hence 4 unknowns and mith only 3 equations of statics available, makes the structure statically tndeterminate to the first degree. In Fig. 2.8, the reaction at A ts a rigid one, thus all 3 characteristics of magnitude, iirection and point of application of the re~ action are unknown. At point B, due to pin only 2 unknowns, nensly, magnitude anc direction, thus making @ total of 5 unknowns with oply § equations of statics avatlaple or the structure ts statically indeterminate to the second degree. In the structure of Fig. 2,3, both supports at A and B are rigid thus all'3 force characteristics are unktomm at each support or a total of § unknowns which makes the structure statically indeterminate to the third degree. EQUILIBRIUM OF FORCE SYSTEMS. TRUSS STRUCTURES. Example eroblen 5 et Ya NB Fig. A2.10 Fig. A211 Fig. AR 12 Fig. 2.10 shone 2 2 bay criss supported points A and 8 and carrying a imowm load syst: P,Q. All members of the truss are connscte: af their ends by 4 common pin at each joiat. ‘The reactions at A and 3 ere epplied throu fittings as indicated. ‘The question is whether the structure 1s statically detera: Relative 19 externa] reactions at 4 structure ts statically determinate type of support procuces only one unknowm at A and two unknowns at 3, nacely, Vg, Vg and Hy as shown in Fig. 2.10 and ne nave 3 squations static equllivrium avatlapi: fe now investigate to see 1° we can find ‘che internal member stresses after caving tound tne values of the reactions at A anc B. Suppose we cut out Joint B as indicated by section 1-1 tn Fig. 2.10 and draw 3 tree body es shown in Fig. 2.11. Since the embers of the truss nave Pins at each end, the loads in these meuders gust be axial, thus dlreczion and lize of action is known ana Only magnitude 1s unknown. In Fig. 2.11 Hg and Yy are known but AB, CB, a DB are unknown in gagnitude hence we have S unimowns but only 2 equations of squillorium for 2 coplanar concurrent force systen. If we cut througn the truss in Fis, 2.10 by the section 2-2 and draw a free bedy of the lowar portion as shown in Fig. 2.22, we have 4 unknowns, namely, the axial loads in CA, DA, 08, DB Dut only 3’ equations of equiltbriim evatlable for 8 coplanar force system. Suppose we were able to find the stresses in CA, DA, CB, 98 in some sanner, and we would non proceed to Joint D anc treat it as 2 free vody or cut through the upper panel along, section 4-4 and use the lower perticn as 2 free poay, ‘The same reesoning as used above would show us we have one gore unknown than ths of equilltriun equations available and thus we nave the truss statically inceterzinate to the second degree relacive to internal member stresses. ber Physical; nucture has two nore members than is necessary for the stability the estructura under load, as we coulé leave out one diagonal member {n eachANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES the structure would de still stable and all Rauber axial stresses could de found by the equations of static equilibriua without regard to their size of cross-section or the kind of material. Adding the second diagonal aenber tm each tanel would necessitate ‘mowing the size of all truss members and the kind of material used before member stresBes could be found, 26 the additional equations needed must come from a consideration involving distortion of the truss. assume for example, that one diagonal in the upper panel was left out. We would then be able to find the stresses in the meubers of the upper panel by otetics but the lower panel would still be statically inde terminaze to 1 cegree because of the doub: Giegonal system and thus one additional equation ts necessary and would involve a consideration of truss distortion, (The solution of statically indeterminate trusses 1s covered in Chapter 4@.} 2.7 Example Problem Solutions of Statically Determinate Coplanar Structures and Coplanar Loadings. Although a student has taken a course tn statics before taking a veginning course in aircraft structures, it 1s felt that « Linited problens involving the application of the equations of static aquiltortum ts quite Justified; particularly {¢ the problems are possibly Soneshat more difficult than acest of problems in the usuai deginning course 1n statics. Since one mist use the equations of static équilitriun as part of the aecassary equations in solving statically tndeterstnate structures and since statically indeterminate structurss are covered in rather caplet detat2| tn other chapters of his book, only Limited space will be given to problems {nvolving stetics in this chapter. Example Problem 8 Fig. A214 shows a quon simplified wing structure, consisting of a wing spar supported by Lift and cabane struts wnich tie the wing spar to the fuselage structure, The distributed| air load on the wing spar is unsymmetrical about’ center line of the atrfrae, The wing svar Is made in three units, readily disassembled by using pin fittings at points 0 and 0". All supporting wing Struts nave single pin fitting units a= eaca end. The problem ts to deter~ 20%/in. s0n/i See TT Pre a Sin Fig. A214 eee Ant mine the axtal loads tn the members and the re- agtions on the spar. Solution: The first thing to decide is whether The structure is statically determinate. From the figure it is observed that the wing spar 1s supported by five struts. Due to the oins av each end of ali struts, we have five unknowns, namely, the asgnitude of tha load in each strut. Direction and location of each strut load ts wom because of the pin at each end of the struts. We heve 3 equations of equilivrium for the wing spar as @ single un{t supported by the 5 struts, thus two gore equations are necessary if the S wmkmown strut loads are to be found. It 1s noticed that the wing spar includes 2 ine ternal single pin comections at points 0 and 0 Thais establishes the fact that the moment of all forces located to one side of the pin must bo equal to zero since the single pin fitting can~ not resist 2 moment, Thus we obtain two addi~ ‘tional equations because of the to internal pin fittings and thus we have § equations to find 5 unknowms. Fig. 2.15 shows a tree body of the wing spar to the right of hinge fitting at 0. sons2430 rorse(g0e19)45 By eat ot? ——— I ant Yo E TA Fig. AQIS In order to take moments, the distributed load on the spar has been replaced by the re~ sultant load on each spar portion, namely, the total load on the portion acting through tae centroid of the distributed load system. The strut reaction SA at A aas deen shown in phantom ag it 1s more conventent to deal itn tts com ponents Ya and Xj. The reaction at 0 15 une known in Sagnitude and direction and for convenience we Will deal with {ts components Xo and Yp. ‘The Sense assimed {s indicated on the figure. The sense of a force ts represented graphically by 2n arrow heed on the end of a vector. The correct sense is obtained from the solution of the equations of equiltorium since, a force or moment aust be given a plus or minus Sign in writing the equations. Since the sense of a force or aoment is unknowm, it is aesuted, and if the algebraic solution of the equilibrium equations gives a plus value to the magnitude ‘then the true sense 1s as assumed, and opposite to that ascumed If the solution gives a minus sim, If the unknom forces are axial Loads tn nemhers it 1s common practice to call tensile stress plus and compressive stress ainus, thus Awe assume the sense of an unknown axtal toad ag tension, the solution of the equiltoriumA28 EQUILIBRIUM OF FORCE SYSTEMS. TRUSS STRUCTURES, equations will give a plus value for the magnt- tude of the unknown 1f the true stress 1 tension and a ainus sign will indicate the assimed tension stresses should bs reversed or compression, thus giving 2 conststancy of st To find the unknown Yq we take aoments about point 0 and equate to zero for equiliariim, Big = = 2460 x Gl - 2015 x 100+ Bex, = 0 Hence %, = 204000/82 = 2480 1b. The plus sign means that the sense es assumed in the zigure is correct. ay geometry Xa = 2480 x 117/66 = 4400 1b. and the load in strut ZA equals #2480" = S050 1b, tension or as assumed in the figure. To find Xo we use the squilipriun equation Xo = 4400 = 0, wnence Xo = 4400 1b, IP To find Yo we use, BFy = 0 = 2460 + 1013 - 2480 ~ Yo = 0, whence Yo'= 993 1b. To check our results for equilibrium we will take monents of all ferces about A to see 1f they equal zero. Dy = 2460 x 4) - 1013 x 20 ~ 993 x 62 = 0 check On the spar portion O'A', the reactions ere obviously equal to 40/20 times those Found for portion QA since the external loading is 40 as compared to 20, Hence A'E! ~ 6750, Xo: = 5880, Yor = 1325 Fig. 2.16 shons a tree body of the center spar portion with the reactions at 0 and 0! ae found previously. The unknown loads in the strute have deen assumed tension as shown by. the arrows. zonons0x40 1800080 padi} sor Fig. A216 To find the load in strut BC take moments about Bt Bigh = 1925 x 20 - 2000 x 5 - 1800 x 85 = 885 x 8 + 60 (BC) 30/23.8 = 0 whence, BC = 2720 Ib, with sense as assumed, To find strut load aC! take aments about point C, Mp = 1825 x 65 + 2000 x 30'~ 1800 x 10 ~ 983 x 35 30/83.6 = 0 40 whence, B'C' = 6000 lb. with sense as shown. To find load in member BC use equation = 0 = 1525 + 2000 + 1600 + 963 - 6000 (/33.6) = 2720 (0/35.8) - B'C (80/54) =o whence, B'C = ~ 2635 1b. The minus stan means it acts opposite to that shown in figure or 18 compression instead o7 tension. The reactions on the spar can now be determined and shears, dending moments and axial leads on the spar could be found. The numerical results should de checked for aquili- briwm of the spar as a whole vy Taking aouents of all forces about a different aouent center to see if the result 1s zero. Brample Problem Fig. 2.17 shows a sizplified airplane landing gear unit with all aembers and loads confined to one plane. The brace struts are pinned st each end and the support at ¢ ts of ‘the roller typs, thus no vertical reaction can be produced by the support fitting at point 2. ‘The menber at C can rotate on the roller tut horizontal movement is prevented, A kom load of 10,000 1b. is applied to axle wit at A. The problen is to find the Icad in the brace struts and the reaction at Cy Soluttor Due to the single pin fitting at each of the brace struts, the reactions at 3 end DANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES are colinear with the strut axis, thus direction and point of application are soln for reaction Ry and Rp leaving only the magnitude of each as unknowm. The roller type fitting at ¢ fixes the direction and point of appiteation of the Teaction Ro, leaving magnitude as the only unenowm. Thus there are 3 unos Rg, Ro and Rp and with 3 equations of static equilibrium available, the structure {s statically deterai- pate with respect to external reactions. The Seuse of each of the 3 unknown reactions has deen assumed a8 Indicated by the vector. To find Rp take moments about point B:- Bg = - Looce sin 50° x 36 ~ 10000 ces 30° x 12 = Rp (12/17) 23 = 0 whence, Rp * - 16750 1b, Since the result comes out witt a minus sign, the reaction Rp has a sense opposite to that snow by che vector in Fig. 2.17. Since the reaction ap 18 colinear with the Line DE because of the pin ends, the lead in the brace strut DE 1s 16750 1b. compression. In the above monent equation about B, the reactton Ry was resolved into vertical and porizontal components at point 0, and thus only the vertical component watch equais (12/17) Rp enters Into tae equation Since the korizontal component hes 2 line of action through point 8 and therefore no monent. Ag does not enter in equation as it has zero monent about 8. To find Rg take EFy = 0 RY = 10000 x cos $0? + (- 16750)(12/17) + Ag (24/26.8) = 0 whence, Rg = 3540 Ib. Since sign cones out plus, the sense 1s the same as assumed in the figure, The strut icad BF 1s therafore 3540 1b. tension, since reaction Rg is colinear with line BF. To find Rg take IH = 0 BH_= 10000 sin 50° - 3540 (12/26.8) + (~ 16760) (le/i7) + Ro = 0 whence, Ro = G407 Ib. Result is nlus and therefore agsuted sense was correct. To check the numerical results take mouents about point a for equilibrium. My = 8407 x 36 + 3540 (24/26.3) 12 ~ 3540 {12/26.8) 36 + 16750 (12/17) i2 ~ 18760 (12/17) $6 = 308000 + 38100 = 57100 + 142000 - 426000 = 0 (check) ‘A2.2 Stresses in Coplanar Truss Structures Under Coplanar Loading. In atreraft construction, the truss type Of construction ts quite comon. The nest comon is the tubular steel welded trusees chet make up the fuselage frame, and less freq: A29 the aluminum alloy tubular truss. Trussed type beams composed of closed and open type sections are also frequently used tn wing beam construction. The stresses or loads in the members of @ truss are camonly referrea to as "primary" and "secondary" stresses. The stresses which are found under the following assuaptione are referred to es primary stresses. (1) The members of the truss are straignt, weightless and Lie in one plane. (2) The members of a truss neeting at a Dotnt are considered as Joined together ay a Somon frictionless pin and all member axas in— tersect at the pin center. (3) All external loads are applied to the truss only at the joints and in the plane of the truss. ‘Thus all loads or stresses produced im members ars either axial tension or compres sion without bending or torsion. Those trusses produced tn the truss mem ders due to the non-fulfiliment of the above assumptions are referred to as secondary, stresses. Most steel tubular trusses are welded ‘together at their ends and in other truss types, the members are riveted or bolted together. This restraint at the joints aay cause second= ary stresses in some members greater than the primary stresses. Likewise 1t is comon in actual practical design to apply forces to the truss members between thetr ends by supporting many equipment installetions on these truss embers. However, regardless of the magnitude of these so-called secondary loads, it ts comon practice to first find the primary stresses under the assumption outlined above. GENERAL CRITERIA POR DETERMINING WHETHER ‘TRUSS STRUCTURES ARE STATICALLY DSTERMINATS ‘WITH RESPECT 0 INTERNAL STRESSES. ‘The simplest truss that can de constructed is the triangle which has three members m and three joints j. A more elaborate truss consists of additional triangular frames, so arranged ‘that each triangle adds one Joint and two men~ bers. Hence the number of members to insur: stability under any loading is: ae2]-3------------~- (2,2) A truss having fewer mesbers than required by Eq. (2.8) is ina state of unstable equili— brium and wil) collapse except uncer certain conditions of leading. ‘The loads in the members of @ truss with the rimber of members shown in | equation (2.8) can be Zound with the available equations of statics, since the forces tn the meubers acting at a point intersect at a common point or form a concurrent ‘ores system. For thts type of force system there are two static equilibrium equations avatlable. ‘Thus for J number of joints there are 23$2.10 equations available. However three inenen: squitions are necastary to cetermine the 2x: nal reactions, thus the number oP equations necessary to Solve Zor all the loads in she menbers is 2) ~ 3. Hence tf the munder o7 truss monbers is that given by equation (2.3) the truss 1s statically deterinate relative to the primary loads in the truss members ana the truss ts also stable. If the truss has more aembers than indicated by equation (2.2) the truss 1s considered Fedundant and statically {nceterainace since the member loads cannot be found in all the members by the laws of statios. Such redundant structures If the embers are properly placed are stable and will support loads of any, arrangement. in general there are three rather distinct aethods or procedures in applying the equattons of static equilinriim te finding the primary stresses in truss type structurss. They are often referred to as the method of Joints, nonents, and shears. 2.9 Method of Joints. If the truss as a whole ts tn equiliprin then cach member or Joint in the truss must Likewise ve in equilibrium. The forces in the members et a truss joint intersect in = comon Point, thus the forces on aach Joint fom a concurrent-coplanar force system. The method of joints consists in cutting out or isolating 8 Joint as a tree body and applying the laws of equilibrium for a concurrent foree system. Since only two independent equations are avail~ able for this type of systen only two unknowns Gan exist at any joint. Thus tha procedure 1s to start at the joint where only two unknowns exist and continue progressively throughout the truss joint by Joint. To {llustrate the method consider the cantilever truss of Fig. A2.13. From cbservation there are only two members with internal stresses unmowm at Joint Ly. Fig. A219 shows a free body of joint Ly. The stresses in the members Ly Le and Ly Uy have deen assumed es tension, as Indicated by the arrows pulling away trom tne Joint Ly. Te static equations of equiliorium cor the forces acting on joint L, are 2H and V = 0, BY = - 1000 - LU, (40/60) =0---- ~~ -@) whence, LU, = ~ 1250 lb. Since the sten came out minus the stress 1s opposite to that assimed tn Fig. 42,19 or compression. aK = = 500 - (~ 1250)(0/S0) - Labs = 0 = =f) whence, Les = 260 1b. Since sign comes out plus, sense fs sane as assuned in ¢igure EQUILIBRIUM OF FORCE SYSTEMS. TRUSS STRUCTURES Fig.a2.19 \ oF tension. In equation (5) the loaa of 1250 in LaLa was substituted as a minus value since it was found to act opposite to chat chown in Fig, Ag.19. Possibly a better proceeure noald be to change the sense of the arrow in che trae body diagram for any solved sembers Sefors writs ing further equilitriun equations, ve must proceed to joint La instead Re = 3770 + 20.26 x 114.5 + ($0.28 + 15.33) 1 5 hence Ra = 1295 2b, (The student should always check results oy taking monents about point (1) to see if Zt, = 0) Rear Beam 14. 144/10. 14.5 70.5°—4t The rear beam has the sane span dimensions but the loading ts 14.74 2b/in. Hence beam reactions R, and Ry will be 14,74/80.25 = .4875 ‘times those for front bean. hence Ry = «4675 x S770 = 1a38 Ib, Ry = 4875 x 1295 = 631 Ib, The next step in the solution ts che solving for the arial loads {n all the members. We will use the methed of joints and consider the structure gade up of three trues systems as illustrated at the top of the next column, namely, a front litt truss, a rear list truss and a drag truss. The beans ere comuon to Doth Litt and drag trasses. Table AZ.i gives the V, D and $ projections of the litt truss members ag determined from {information given in Fig. 42.35, The true EQUILIBRIUM OF FORCE sysTEMS. TRUSS STRUCTURES. 4 RB. 3 v 7B FN 7A [ax | Drag Truss SAN NEON 2 "SB 3 Froot Lift ‘Truss A® toe i it ‘Truss member lengths L and 7 gonponent ratios then teliow by simple calculation, sabte 42.1 T [ wor fom) y laps [2 [ale laa [Frese seus] Fa) soe [0 [ue,a4[tia.00],0023) 0 [.e0s! Lins oan [ae © [ite-34|ite-s0|-asaa)_9 _|-asea! Prone Seat | Sy [57.98 | H i4,24 209,70], 4201], bau aa78 Bear Serat_] oy |Sh0@ | 0 /ia.3e]iz0-00|. 488] 0 |-a0a0 We starc the solution of Joints oy starting foint (1). Free body sketcnes are sketche! below, All members are ¢: two-force rembers or taving pins at ‘thus magntvude 1s the cnly unknown character istic of cach member load, The drag truss nen- vers coming in to joint (1) ara replaced sy a Single reaction called 2:. after Da is found, tts influence tn causing loads in drag truss members can then be found nen the rug truss as a whole is treated. In the Jotnt solutcon, the drag truss ms been assumed parallel to crag Girection watch 1s not quire true from Pig. 2.26, but the error on member loacs is nest gible. SOINT 1 (Equations of Squiitbriun) jor wa ra y sp ® ZY 5 S770 x .9986 ~ .0523 FB - IS = = S770 x 0523 - l9e6 FB | BD = .0654 Sp + Dy =O == Solving equatians 1, 2 and 3, we obtain FB T= e515 1b. {compreseton) Sr S833 1b. (tension) Dy 738 1b. (art)ANALYSIS AND DESIGN OF PLIGHT VEWICLE STRUCTURES Joint (3) (Equations of equilibrium) 1838 (drag truso aa_} 1836 “reaction on wo @h p,m) ne ¥-$ plane ‘V-D plane ‘sa, 1858 x .9986 - .0525 RB - .4486 Sq = 0 -(4) 3 9886 RB - S850 8 * 0 (3) 3) Solving equations 4, 5 and 6, we obtain RB =~ 4169 1b. (caupression} SR = 4579 1. {tension} D,=0 Fig, A2.S8 shows the reactions of the Lift struts on the drag truss at Joiata (1) and (3) 28 found above. “ a @ a Fig. 42.38 00 Drag Truss Panel Potnt Loads Due to Air Orag oa It was assumed that the alr load components im the drag direction were § 1lb./in. of wing acting torvard. ‘The distributed load of 6 1b./tn. fs re= placed oy concentrated loads at the panel potnts ae chow in Fig. 42.39. Zach panel point takes one half the distributed load to the adjacent panel point, except for the two autboard panel Doints which are affected by the overhang tip portion. ‘Thus the outboard panel point concentration a Anat points (2) and (4). In the design of the beam and fittings at this point, the eftect of the actual conditions of eccentricity should of course be considered. Combined Loads on Drag Truss Adging the two load systens of Figs. 42.33 and 42.39, the total drag truss loading is obtained as shown in Fig. 42,40, The resulting uember atial stresses are then solved for oy the method of index stresses (art. 42.9), ‘The values are indicated on the truss diagram. It 1s customary to make one of the fittings attach— ing wing to fuselage {ncapable of transferring grag reaction to fuselage, so that the entire drag reaction fron wing panel on fuselage ts definitely confined to one point. In this example point (2) has been assumed as point wnere wag ts resisted. Those drag wires which would be in compression are assimed cut of action. tl 1 “AGT “AL SH ~T0.515 Fig 1908 = Drag Reaction 798 Fig. A240 Ruselage Reactions As a check on the Work as well as to obtain reference loads on fuselage from wing structure, ‘the fuselage reactions will be checked against the externally applied air loads. Table az.2 gives the calculations in table Zora. of 264 1b. 18 determined by taking moments about Table 42.2 (3) of the drag iced outboard of (3) as follows: | -—————— = > . = 70.5 x § x 35.05/S8.5 = 254 13, me TGge3 | me | 9 | a PS 70.5 x 6 x 35.25/58.5 = 254 12 : eee | ume) OF [a | OP 7o simplify the drag truss solution, the . (tension) = CRB - 2280 x 1868 = 0ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES whence CRB = - 1510 1d, (compression) BD =D, - 2860 x .1485 = 9 whence Dy = 835 Ib, drag truss resozton Senter Front Sean sae (1294 - 726) = ss8 (Ref. , Table A2.2) [7 — = 3 Se =) Fo —— 0 Pao 7 R=2898 Beesate Fig, A243 Fig. 42.43 shows the ¥ loads on the center rront beaa and the resulting ¥ component oz the cabane reaction at Joint (7), at Joint 7 2535 ‘vp PianeD IV = 2585 - 721 Op = .S87 cp = 0 28 =~ CFE = .640 Op + S36 Cp = 0 2D - .240 Cp + .397 cy = 0 Solving the thrae equations, we obtain CFB = ~ 2281 (compression) Cr = 26a cp = 1088 Solution for Loads tn Drag Truss Members 42.44 shows al) the loads applica to the center yansl drag truss, The S and D ree getions frm the outer panel at Joints (2) and Panel point 36 Drag Load 336 1 aie, ~ lo | p38 ust t__{isig 10) f usst | % aay vest / a Neer | Es ae} a sweat fe ‘at 908 wad ig. azss reste A219 (4) are taken from Teele 32.2 of problem 1. The drag loa of 336 Ib. at (3) Is due to the rear cabane strut, as is likewise the deem axtel load of - 1510 at (8). The axial beam load of ~ 2282 lb. at (7) {s cue to reaction of front gabane truss. The zerel potat Loads are a3, the given running drag tcad of 6 ib./in, act forward. to. ‘The reaction which holds all these drag truss loads in equiltortum is supplied by the cabane truss at point (7) since the trent and Gfagonal cabane struts intersect to fom a rigid triangle. Thus the drag reaction R aquals one half the total ¢rag loads or 2634 19. Solving the truss for the loading of Fig. 82.44 we obtain the menber axial loads of Pig. A248, 137 368 08 ust & ‘st Xl a Ble ays oy 3 ° 15027|__-11908 -17308_|-15028 Vt D Plant BD = ~ 2534 = .240 0p + 1397 cp SO BV = = .72L Cp - .§97 Sp 20 Solving for Op and Gp, ne obtain = - 2740 1b. (compression) BLO (tenston) 0 adding these loads to those previously calc lated for 117t loads: op = = 2740 + = 105 op = 1058 + S510 = 4358 1b. cg = 2280 1b. uselage Re:EQUILIBRIUM OF FORCE SYSTEMS TRUSS STRUCTURES, Fee | ener ¥ > [2 ia WO tae see Ga as ise | 3 | prosg tare serve | 2000 738 | e280 in @ | Reraite gene | st) cos | © | ono 8 Applied Air Loads V component = 7525 (outer anel) + 65 x 45 10445 (check) 1210 (outer panel) ~ 68 x 6 = ~ 1500 (error 2 1b.) D component = The total st. centerline ot Joad on a vertical plane thru airplane should equal the $ com~ ponent of tne applied loads. ‘The applied sice loads = - 58: 1. (see problan 1). The air lead on center pane} is vertical and thus has zero S component. Fron Table 42.3 tor fuselage reactions nave 23 = 16178. From Fig. 42.45 the load in ‘the front bean at @ of airplane equals - 17308 and 563 for rear bean. The horizontal component of the diagonal drag strut at joints 12 equals 216 x 45/57.6 = 169 1b. Then total S components = 16178 - 17308 + 968 + 169 5 ~ 393 1D. wich checks the side component of the applied sir loads. Example Problem 22. Single Scar Truss Plus Torsional Truss Systen, In snall wings or control surfaces, fabric is often used as the surface covering. ‘Since the fabric cannot provide reliable torstonal Tegistance, internal structure must be of such design as to provide torsional strength. 4 single spar plus a special type of truss system 18 Often used to give a satisfactory structure. Pig. AZ.6 tilustrates such 2 type of structure, necely, @ trussed single spar AEF plus 4 triangular truss system between the spar an the trailing edge 0S. Fig. 42.46 (a, 5, ¢) shows the three projections and dimensions. The 2ir iead on the surface covering of the structure is assumed to be 0.5 1b./in.* intensity at star 2iRe and then varying linearly to aero at the trailing edge (See Fiz. 4) ‘The problem will be to determine the axtal loads in all the aembers of the structure. It w{Ll be assumed that all aeubers are 2 force noabers as 1s usually done in finding the primary leads In trussea structures. 7 ‘Tpanels @ 12" = 84" Fig. 400 souttow: ‘The total atr load on the str the average tntensity ver square inch times the surface area or (0,5)(.5)(26 x 94) = 756 1b. In order to solve a truss system by a meted of Joints the distriputed lead must be replaced by an equivalent load system acting at the joints of the structure. Referring to Fig. (2), tae total air load ons strip 1 inzh wide and 36 tzehes long ts 36(0.5}/2 = 9 1d. and tts vay oF resultant location 1s 12 inches tron Line AB. In Fig. 46a this recultent load of § tb./in. ts smaginad as acting on an imscinary bean Located along the line I-l. Tnis running load apzl aiong shis line is now repiaced sy an equivaiont force system acting at joints OPASEDECA, Tae Pesults 97 this Joint distribution are shom oy the Joint loads in Fiz, aZ.d6. To titustra’ how Hed, lations for loads at foints S80R will be af Fig. a2.4@ shone 2 porticn of che structura to be considered. For @ running leaa of 2 ie./tn., along 1 ine lel, reactions will beANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES ¢ nr err Fig. A248 simple beans resting at zoints 2, 3, 4, 5, ste. fe distance between 2-3 is 8 inches. ‘The total load on this distance 1s @x9=72 1b. One halz or 36 1d. goes vo point (2) and the other half to point (3). Tha o6 1p. at (2) is then replaced by an equivalent force system at © and 8 or (26)/3 = 12 1b. to § and (36) (2/3) = 24 to BZ. The distance between points (3) and (4) ts 8 inches and the load is 8x 9= 72 1d. One hali of this or $6 goes to point (3) and this added to the pravious 36 gives 72 1b. at (8). ‘Tae load of 72 ts then replaced by an equivalent force system at S and D, or (72)/3 = 24 1b. to 8 and (72) (2/3) = 49 to'D. The tinal load at 8 is therefore 24+ 12 = 36 Ib. as chown in Pig. A 2.48. Due to symetry of the triangle CRD, one half of the total load on the distance CD goes to points (4) ana (5) or (24 x 9)/2 = los ib. The distribution to D ts therefore (108) (2/3) = 72 ang (108)/3 = 36 to R. Adding 72 to the previous load of 48 at D gives a total Load at = 120 1b. as shown in Fig. A2.46. ‘The 108 Ib. at point (5) also gives (108)/3 36 to Ror a total of 72 2b, at R, The student Should check the distribution to other joints as shown in Fig. A2.46. To check the equivalence of the derived Joint load system with the original air load System, the magnitude and aoments of each system must be the same. Adding up the total joint loads as shown in Fig. A2.46 gives a total 02 756 Ib. which cnecks the original air load, ‘Tne moment of the total atr lead about an x axis at left ond of structure equals 786 x 42 = 31782 in. 1b. The nonent of the Joint Icad. systam in Fig. 42.48 equals (66 x 12) + (72 x 36) + (72 x 80) + (56 x 2a) + les (24 + 4a) + (120 x72) + (24 x 84) = 31752 fn.1. or a oneck. The moment of the total air load about Line AE equals 758 ¢ 12 = 9072 tn.ib. The ‘gouent of the distributed joint loads equals (6+ 66 + 72+ 72 + 36)36 = 9072 or @ check. Caloulation of Reactions The structure is supported by single pin tunings at points a, Nand 0, with pin axes yarallel to x axis. It iil be assimed that tha citting at i wekes off the spar load tn 2dtrection. Pig. A248 shows the reactions Oy, Ogy ay» Sys lig. 70 find Op take moaents about y axie Glond spar AEFN. Aza Dy = (6 + 66 + 72 + 72 + 06)36 ~ 36 0, 50 waence Oz = 252 1b. acting dow as assumed. To find Oy take moments about 2 axis through point (A). Mz 30+ 36 Oy 50, OY #0 ‘To tind Ay take moments about x axis through point N, The Moment of the air loads was pre= viously calolated 2s - 31752, hence, My = = S17S2 + 9 Ay = 0, whence, Ay = 3826 1b. To find Ny take BFy = 0 BFy = 3528 - Ny = 0, hence Ny = 3528 1b. To find Nz take IF; 50 2g 5 = 252 + 786 ~ Nz = 0, hence Ne = 504 Ib. ‘The reactions are all recorded on Fig. 42.46, Solution of Truss Member Loads For saplictty, the loa¢ systen on the structure will be considered separately a5 two load systems. One system wi1l include only’ ‘those loads acting along the Line AB and the second load system Nill de Tenaining loads which act along line 0S. Since no bending manent can be resisted at joint 0, the external load along spar AE will be reacted at A and N entirely or in other words, the spare alone resists tne loads on line a. Fig. A249 chons 2 dlagram of this spar with its joint external loading. The axial. loads produced by this loading are written on ‘the truss monbers. these member loads.) (The student should check oo ‘ ‘ Lo Oo a9 Ne le y aie Pete] "eg Xe is 7 wae eR Fig. 02.49 TRINDULAR TRUSS SYSTEN ‘Tne load system along the trailing edge 0S causes stresses in both the spar truss and the Giagonal truss systen. The support fitting at point 0 provides a rection in the Z direction ‘but no reacting moment about the x axis. Since the loads on the trailing edge Lie on a y axis throuch 0, it 16 obvious that all these loads flow to point 0. Since the vending strength of the trailing edge aeuber is negligible, theAg32 load of 36 1b, at Jotat § in order to be transferred to point O through the diagonal truss system must follow the yath SDRCCBPAQ. In Ltke uanner the load of 72 at R to reach O must take ‘the path ROJBPAO, ete. Calculation of Leads in Diagonal t fat |yt] we ‘Al Diagonal 45 }12 [36 )30.2| 118] .004] 043 ‘Truss Members Member z | 40, NO 45] 0/36 [s7.5| 120] 0 | 260 onsider Joint The triangular truss SEF cannot assist In transferring any portion of the 36 1b. lead at S because the reaction of this truss at SF would put torsion on 1 no appreciable torsional resistance, Considering Jotat § as a free body and writing ‘the equilibritm equations: whence, 0S = = GS BF, = 36+ 118 08 ~ .118 Gs = 0 Subt. 05 = = GS and solving for oS, gives 63 = 159 ip,(tenston), BS = 159 (compression) Consider Joint o Let Ty and Tz be reactions of dtagonal truss system on spar truss at Joint D. BPq = = 159 x .943 + 1943 DR = 0, hence DR = 159 1b. BF; = + 159 x .218 + 169 x 116-7, = 0 Whence Tz = 0, which aeans the dtagonal truss produces no Z reaction or sear load on spar truss at D. By =~ S14 x 169 ~ 514. x 159 - Ty = 0 whence Ty =~ 100 Ib. If Joint @ ts investigated in the same manner, the results will show that 7, = 0 and Ty © 100. Tue results at joint D shows that the rear diagonal truss systea produces no shear Load EQUILIBRIUM OF FORCE SYSTEMS. TRUSS STRUCTURES. reaction on the = force on the spar cuoes compression truss and tension ‘sop chord bottan chor R ‘The Load to be transferred to truss RGR 45 squal to the 72 1b, at R plus the 36 Ld. et S which comes to joint R frea truss DBC. Hence load in RC = (72 + 36)0.3 x (1/118) 5-457 Ib Whence RJ = 457, 09 = 487 and JQ = Jotnt 9 Load to be transterred to truss 2b. 72 + 36 = 180 1b. Hence load in 4B = (180 x 0.5}(1/.118) ~ 762, whence OL = 762, 8P = 762, LP =~ 762 Joint B Lead = 160 + 68 = 286 Lead in PA = (248 x 0,5)(1/.118) =~ 1040 Whence PN = 1040 Consider Joint (A} as a tree body. BF, = ~ 1060 x .943 + .960 40 20, 40 = 1022 1b. In like manner, considering Joint N, gives NO = = 2022 1, Couple Force Reactions on Sar AS pointed cut previously, the diagonal torsion truss produces a couplo reaction on the spar in the y direction. The megnitude of the force of this ccuple equals the y component of the load in the dlagonal truss members reeting ata spar joint. Let Ty equal this reaction load on tne spar. . At Joint Ty S = (457 + 457).514 = = 287 ID, Likewise at Joint J, Ty = 297 At Jotat 3:- Ty 5 - (762 + 762).814 = - are 479 Likewise st Joint L, Ty At Joint Ty = - (1040 x 324) 226 Likewtse at Joint x, Ty = 326ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES ‘These reactions of the torsicn truss apon the spar truss are shown in Fig, Az.50, The loads in the spar truss nembers due to this loading are written adjacent 79 each truss: member. Adding these member loads to the loads In Pig. 42.48, we obtain the final sper truss momber loads as show in Fig. A2.S2. A229 Ground Line ‘Level Landing with Inclined Reactions ae om 10 uot * sey Zo Fie A250 Level Landing ts Nooe Wino at : Gicr f Grows 3828+ Es ar + a) ke Sls anf a Fig. azst If We add the reactions in Figs. A2.S0 and AZ.51, ne obtain 3528 and $04 which check the reactions obtained in Fig, a2.a6. A213 Landing Gear Structure ‘The airplane ts both 2 lendborne and air borne venicls, and thus a means of operating the airplane on the ground mist be provided wich means wheels and trakes. Furthermore, provision must be made to control the impact forces fnvolved in landing or in taxiing over Fough ground. This requirement requires a speciai energy absorption unit in the landing gear beyond that anergy absorption provided ty the tires. The landing gear thus includes a So-called shock strut commonly referred to as an oleo strut, watch is a menber composed of byo telescoping cylinders. when the strut ts compressed, ofl inside the air tight cylinders is forced through an orifice from one cylinder to the other and the energy due to the landing {mpact is absorbed by she work done in forcing this of] through the orttice. The orifice cen be so designed as 0 srovide practically 2 uniZora resistance over the displacement or travel of the olec strut. An airplane can land safely with the aire plane in varfous attitudes at the tnstant of ground contact. Fig. A2.52 {llustrates the three altitudes of the airplane that ere Specified by the goverment aviation agenctes for destgn of landing gear. In addition to these symmetrical untraked loadings, special loadings, such as a braked condition, landing 9M one wheel condition, side load on wheel, etc. are required. In otter words, a landing gear can be subjected to a considerable mmber of aifterent loadings under the various landing conditions that are encountered in the normal (Ground Line ‘Tail Down Landing Fig. 2.52 Fig. AZ.S3 shows photographs of typical main gear units end Fig. A2.S4 for nose wheel, gear units. The successful design of landing gear for present day aircraft 1s no doubt one of the most atffteult problons which 1s encountered in the structural layout and strength design of atr~ craft. In general, the gear for aerodynamic efficiency must be retracted into the interior of the wing, nacelle or fuselage, thus a re- Lisble, safe retracting and lovering mechanisa system ts necessary. The wheals mist be traked and the nose wheel made steerable. Tue landing gear is subjected to relatively large loads, waose cagnttudes are several tines the gross weight of the airplane and these large loads must be carried into the supporting wing or fuselage structure, Since the welgnt of land~ ing gear may anount to around 6 percent of the wetght of the airplane tt 1s evident that nigh strength/wetgnt ratio ts a saramount design requirement of landing gear, as inefficient structural arrangement and Conservative strese analysis can add uany unnecessary pounds of woight to the airplane and thus decrease the pay or use¢ul load. 2.14 Example Problems of Calculating Reactions and Loads on Membera of Landing Gear Unite, In its simplest fom, a landing gear could constst of 2 single olee strut acting as a cantilever Sean with its tixed end betng the upper end which would be rigidly fastened to the supporting structure. The lower cylinder of the oleo strut carries an axle at its lover Operation of an airplane.Agta EQUILIBRIUM OF FORCE SYSTEMS. TRUSS STRUCTURES. ‘McDonnell Aircraft (Multary Airplane) Jet Aisliner Fig. A2.53 Main Landing Gear Illustrations (One side)Beechcraft Twin Bonanza Piper Tri-Pacer Navy Fé. North American Aviation Co, Douglas DC-1 Air Transport Fig, A2.54 Nose Wheel Gear Installations82.26 end for attaching the wheel and tire. This cantilever deam 1s subjected to bending in cKO Girections, torsion and also axial loads. Since the gear is usually made retractable, it ts difficult to design a single fitting unit at the upper end of the oleo strut that will resist this combination of forces and still permit movement for a simpls retracting mechan ism. Furthermore, 1t would be difficult to provide carry-through supporting wing or fuse~ lage structure for such large concentrated load systems. ‘Thus to decrease the magnitude of the vending moments ané also the bending flexibility of the cantilever strut and also to simplity the retracting problem and the carry-througn structural problem, it 1s customary to add one or two braces to the oleo strut. In general, effort is made to make the landing gear structure statically determinate by using specially designed fittings at member ends or at support points in order to establish the force characteristics of direction and point of application. ‘Two example problex solutions will be presented, one dealing with a gear with a single wheel and the other with a gear involving two wheels. Sample Problem 13 Fig. A2.55 shows the projections of the landing gear configuration on the ¥S and VD planes. Fig. A2.86 1s @ space dimensional diagram, In landing gear analysis it 1s comon to use Y, D and 8 as reference axés instead of ‘the symbols Z, X and ¥. This gear unit is assumed as representing one side of the main gear on a tricycle type of landing gear system. The loading assumed corresponds to 2 condition of nose wheel up or tati dow. (See lower sketch of Fig. 42.52), The design load on the wheel 1s vertical and {ts magnitude for this, problem ts 15000 Ib. ‘The gear unit {3 attached to the supporting structure at points F, H and G, Retraction of the geer is obtained by rotating gear rearard and upward about axis through F and #. The fittings at F and Hare designed to resist no bending moment hence reactions at F and H are unkown in magnitude and direction, Instead of using the reaction and an angla as unknows, the resultant reaction 1s replaced by its Vand D components as chown in Fig. 42.56, The re~ action at ¢ 1s uniciown tn magnitude only since the pin fitting at each end of member GC fixes the direction and line of action of the reaction at G. For convenience in calculations, the reaction G ts replaced by its components Gy and Gp. for a side icad on the lending gear, the reaction in the $ direction 1s taken off at point F by a spectal designed unit. EQUILIBRIUM OF FORCE SYSTEMS. TRUSS STRUCTURES. SOLUTION ‘The supporting reactions upon the 2 points F, H, and G will be calculated os nning step, Tnare re six unknowns, nanely FS, 7, Fo, Hy, Kp and G (See Fig, 42.58), Nite § equations of static equiltsriun availenie for 2 Space force eystem, the reactions oan be found dy statics. Referring to Fig. a2. To find 95 take IS = 0 Fg +050, hence Fg = 0 To find reaction Gy take moments about axis throug2 points F, . Big = S119 x 50 - 24 Gy = 0 \ihence, Gy = 6500 1b. with sense as assured. (tne wneet Load of 18000 ib, tas deen resolved into Y and D components as incicated tn Fiz. 42.65). With Gy know, the reaction @ equals (8800) (21,9/24) = 8610 15, and similarly the component Gp = (6500) (21/24) = 5690 1b Pe fen) Fig. A238 Hid ae J Ra15000 wb.” y= 15000 xc08 150 = 140721 2 ra 8H SLL Ye 0 a OR fi RE ea N yo oN | toss #90 » 4,-s004 b-ses0 sor g 5 oe ear Lo NY. 3 s ~ a ard Fig, 2.56 Fig. 42.97ANALYSIS_AND DESIGN OF PLIGHT VEHICLE STRUCTURES FD te 3820 yp Fs—> op - SSe SF # Epeast ie ‘To find Fy, take moments about 2 D axis shrough point H. IMy(p) 2 18 Gy + 14672 x 8 - 22 FY =O Fig. 2.59 whence Fy = 10063 Lb. with sense as as— sued. To find Hp, take moments about ¥ axis ‘through Fy IMp(y) = - § Op - 22 Hp + Slle x 140 = +6 x 5690 - 22 Hp + Sli9 x 14 20 whence, Hp = 435 1. To find Fp, teke 2D = 0 ID - fp + Hp t Gp - 3g =0 = Fp + 433 + 5600 ~ SlI9 = 0 Wance, Fy = $004 1b. To tina Hy take av = 0 BY. - Py + Oy ~ Hy + 14672 = 0 = 10063 + 6500 - Hy + 14872 = 0 aiios 1b. maence, By g. A2.87 summarizes the reactions as found. The results will be checked for equilibrium of = 1s x 8500 + 14872 x 8 ~ 22 FY = 0 A227 the structure as a whole by taking moments about D and ¥ axes through point A. Ry (p} =~ 10063 x 14 + 6500 x B+ 11109 x 8 = 140882 + 52000 + se8e2 = O(check) Macy) = 5690 x 8 - 453 x 8 - BOOt x 14 48620 - S464 ~ 42086 = 0 (check) ‘The next step in the solution will be the calculation of the forces on the oleo strut unit. Fig. £2.58 shows a free body of the cleo~ Struteatle untt, The trace members BI and Co are two force members due to the pin at each end, and thus magnituce {s the only unknown re~ action characteristic at points Band C. ‘The fitting et point £ between the oleo strut and the top cross member FH is destgned in such a manner ag to restst torsional moments about tha oleo strut exis and to provide D, V and $ force reactions but no moment reactions about D and axas. The unknowns are therefore BI, CG, Bg, By, Ep and Tg or a total of 6 and therefore statically deterainte. The torsional moment ‘Tq 1s represented in Fig. 42.59 by a vector with a double arrow, The vector direction represents the moment axis and the sense of rotation of the moment fs given by the right nang rule, namely, with the thumb of the right and pointing in the same direction as the arrows, the curled fingers give the senge of rotation. To find the resisting torsional moment Tg take moments about V axis through E. Mg¢y) = = 3119 x 8 + Te =O, hence Te = 24952 tn.ib. To find OG take moments about $ axis through B. We(s) = Si19 x $0 = (24/51.8) co xs ~ 24 (21/81.8) CG 3 0 Waence, CO = 8610 Ib, This checks the velue previously obtained when the reaction at @ as found to be 2610. ‘The D and V conponants of Co thus equal, CGp = 9610 (2L/31.) = 5690 Lb. cay = 2610 (24/31.2) = 5500 1b. To find load in brece strut BI, take moments about D axis through point 2. Bea) = - 16672 x 8 + 3 (BI) 22/24.6 + 24 (BI) 12/24.8 = 0 \mance, BI = 8775 1b. and thus, aly = (6775) (22/24.6) = 7940 1. BIg = (8775) (11/24.6) = 3920 1b. To find Be take Is = 0 IS = Gg - 3920 = 0, hence Bs = 2920 bes is2.26 To tind By take 2D = 0 2D = 5690 = BLL9 ~ Bp = 0, hence Ep = 2571 To find By take v= 0 LV = ~ By + 14672 ~ 7240 + 6500 By = 19382 1b, 0, nence Fig. A2.59 snows a tree body of the top member FH, The unknowns are Fy, Fp, Fg) Hy and Hp. The loads or reactions as found trom the analysis of the oleo strut unit are also re. corded on the figure, The equations of equilibriun for this tree body ar: 2S = 0 = - Ge20 + S820 + Fg = 0, or Fg = 0 IMp(p) = 22 Hy - 3920 x 2 - 7840 x 20 - yssse x 6 = 0 woence, iy = 11110 1b, ‘This check value obtained previously, and therefore ts a check fon our Work. Bip(y) = 24952 - 2671 x 5 - 22 Hy 50 wience, Hp = 453 ab. BV = = Fy + sesee + 740 - 0 = 0 whence, Fy = 1006 8 ” = Fp + 2s7l + 453 50 whence, Fy = $004 1b, ‘Thus working through the ree bodies of the cleo strut and the top uember FH, ne come out with same reactions at F and H ag obtained when finding these reactions by equilibrium equation for the entire landing gear. The strength design of the oleo strut untt ‘nd the top member FH could mow be carried out because with all loads and reactions on each member kom, axial, bending and torsional, stresses could now be found, ‘The loads on the brace struts CG and BI are axial, namely, 8610 1b. tension anc 8775 1b. compression respectively, and thus need no further calculation to obtain design stresses. TORQUE LINE ‘The oleo strut consists of two telescoping ‘tubes and sone means aust be provided to trans~ mit torsional moment between the two tubes and still pernit the lower cylinder to move upward inte the oper cylinder, The most common way of providing this terque transfer ts to use a double-cantilever-nut cracker type of stracture.| Fig. A2.60 illustrates how such a torque Length could be applied to the oleo strut in our problea. EQUILIBRIUM OF FORCE SYSTEMS. TRUSS STRUCTURES. oper Z Gylindor iy wey @ Lower Cylinder Fig A2.60 Oe ae Torque oy axe Link ay — \ I OIG ees ber pany The torque to be transferred in cur srob~ lem is 24952 in.Id, ‘The reaction R, between the two units of the torque link at point (2), see Fig. 42.30, thus equals 24962/9 = 2773 1b. ‘The reactions R, at the dese of the Ink at point (3) = 2773 x 8.5/2.75 = 6560 1b. With these reactions known, the strength design of the Link untts and the connections could ve mace. Bample Probien 16 ‘The landing gear as illustrated in Fis. A2.61 is representative of 4 main lending gear waich could be attached to the undar side of a wing and retract forward and upward about line AB into a space provided by the lower portion of the péwer plant nacelle structure. The cleo strut O8 has 2 sliding attachment at 2, wnich prevents any vertical Load to be taken by Jomber AB at 8. However, the 7itting at @ does transfer shear and torque reactions between che oleo strut and member AB. The brace struts 9D, FO and CD are pinned at each end and will de assimed as 2 force members. an airplane level landing condition with unsymetrical wneel loadtug nas been assumed as shown in Fig. Ag.61, ‘SOLUTION ‘The gear is attached to supporting structure at points 4, Band Cc, ‘The reactions at these points will ve calcllated first, treating the entire gear asa free body. Fig. A2.62ANALYSIS AND DESTGN OF Fig. A262 shows @ space clagrem with loads and reactions. Ths reactions at A, B and ¢ have Seen replaced by thelr V and D components. > find reaction Gy take monents about an S axis through points AB. Bigg * - (25000 + 10000) 64 + 24 cy = 0 linenee Cy = 8865 1B. aesumed In Fig. 42.82. uitth sense as ‘The reaction at O mist have a line of action along the line OD since uember CD ts pinned at each end, thus the crag compen ent and the Load In tae strut OD follow as 2 matter of geometry. Op = 66655 (24/28) = S7142 Ib. op = 86666 (36.95/28) = 27900 1b. tension Hence, To find By teke moments about 2 drag axis through point (a). PLIGHT VEHICLE STRUCTURES A229 Diqqp) = - 60000 x 9 ~ 40000 x 20 - 66686 x 13 * 38 By =0 whence, By = 78070 1b. To find ay, take ZV = 0 BY = = 78070 + 60000 + 40000 + ss666 - Yg =0 ‘whence, ay 28896 1b. To tind Bp take moments about V axis through potnt (A). Bta(y) = S712 x 19 - 18000 x 9 - 10000 x 29 ~ 38 Bp =0 wnence, 8p = 17386 1b. To find ap take 2D = 0 2D = ~ S7142 + 15000 + 10000 + 17366 + Ap =o wanes Ap = 14755 1b. ‘To check the results take moments about ¥ and D axes Through point 0. Bo(y) = 5 x 10000 + 14756 x 19 = 17886 x 19 = 0 (check) 2Mo(D) = 20000 x 10 - ses96 x 19 + 78070 x 9 = 0 (check) REACPTONS ON OLEO STRUT 05 Fig. AZ.63 shons a free body of the oleo~ strut 08, “De loads applied to the wheels a ny IP 1 ge r re 4 seeped = % ® i * t we 055 i 100,000 tb, 2 ‘100,000 Ib. Secco'2s, ts0,000, the axle centerlines have been transferred to point (0). ‘Thus the total V load at (0) equals 60000 + 46000 = 106000 and the total D load equals 15000 + 10000 = 28000. The moment of these forces about ¥ and D axes through (0) are Macy) = (15000 = 16000) 10 = $0000 tn.1b. and No(p) = (60000 - 40000) 10 = 2oocoo 1.10. ‘These uonents are indicated in Pig. 2.63 by the vectors ith double arrows. ‘The sense of the moment 1s determined by the right hand trump a442,30 EQUILIBRIUM OF FORCE SYSTEMS. TRUSS STRUCTURES ant flager rite. to t 9s Tho fitting at potnt 5 18 destened to P Fett ventana saseetvabae texte ort sareional Lis zi 3 moment on the ol2o strut, It also can arovide | Shear reactions Hg and Ep tut no bending sola Tesistance about 3 or D axes. i aug vote F Aa Si ies, aT te ons \ | | mast ‘The unknowns are the forces Bg, Ep, DF, DG and the moment Ty. To find Tg take moments about axis OE, Fig. Aa. Blog = - 50000 + Tg = 0, mience Te 50000 tn.2b. = 39218 x 2 - 77461 x 36 ~ 38 By To tind Bg take aorents about D axis 20 through point Be whence, By = 78070 2b, BMp(p) = 200000 ~ 28 Bg = 0, whence 1 Bas to To find Ay take EY 50 2V = e925 + 77461 ~ 76070 - ay = 0 To find force OFy take moments about 2 whence, ay = 86595 15. axis through point G. igcp) = 200000 ~ 100000 x 17 = e6666 x To find Bp take aouents ¥ axis througd A. 17 + 84 DFy = 0 {p} = $0000 + $2143 x 19 - 39 By = 0 wnence, DFy = 77461 1b. whence, 5y = 17386 1b. Then Dfg = 77451 (17/28) rash 7 ‘To tind Ap take ED = 0 D = 17986 = 32143 + Ap = 0, or Ap = and DF = 77451 (32.72/28) 28 : 1D = 0, oF Ap 90503 1b. 14757 13. : Tnese four reactions check the reacztons To find Day take Zv = 0 obtained originally when gear was treated as 4 100000 = 7451 + 66666 - Dey = 0, free body, thus giving e mmericel check on the or Dey = 89235, caleulatians. Then, Dds = 99216 (17/88) = S4iée Ib. With the forces on each cert of the gear we 99215 (32.73/28) = 104190 1d. ‘own, the parts could be designee for strength and rigiaity. The oleo strut would need 2 To find fp take aonents about 8 axts torsion Link as discussed in example problen 15. tnrough point D. end Fig. 42.60. 2ip(g) = ~ 25000 x 35 + 28 By = 0, BP ety. ALIS Problems (1) For the structures rimbered 1 to 10 deter- The results will be checked for static dine whether structure is statically ceter- equilibriua of strut. Take nouents about D inate with respect to external reactions axis through point (0). and Internal stresses. . Bio(p) = 200000 + S464 x 36 - 47085 x 36 — 7143 x 64 = 200000 + Lsassce Bin_Pin ~ 1692828 ~ 487150 = 0 (check) ree i Bois) = S143 x 64 - $7142 x 36 = O(check) oo 8 2 ay w REACTIONS OW TOP MEMBER AB Fig. A2.64 shows @ tree body of member AB ce «) wien the ‘xno applied forces as found tron ‘the previous reactions on the oleo strut. o ‘The unknomns are Ap, Bp, Ay and By. To find By take moments about D’axts trrougn A.(2) Find the horizontal and vertical components of the reactions on the structures {1lus- tpated in Figs. 1I to 15. ofp 098, 204 Jon 2 * . nor Byod ts) apt } med (G) Find the axial loads 1m the members of the trussed structures shown in Figs, 16 to 18. 10 500 Pe «ay ° is mf 0 + an 3 10 - 204.20 Pao Tao 420, (4) Determine the axtal loads in the embers of the structurs in Pig. 18. The aembers are pinned to supports at A, Band c. est 8" 3 eee 4 a . iooe Fig. 19 | 2008 [ts (3) Fig. 20 shows a trt-pod treme for hoisting a propeller for assobly on engine. Find the loads in the frame for a load of 1000 1b. on notst. Fig. 20 (8) Pig, 21 shows the wing structure of an vernally braced uonoplane. Determine the axial loads im all members of the lirt and rag trusses for the following loads. Front beam lift load = 30 Ib./tn. (upward) Rear bean 1tft load = 24 1b./in. (upward) Wing drag load = 8 lb./in. acting att oo sBsegr BAA y tered 351 Pept tty ALP ft Him. T i Ae Tc 5 oo re fi \ te TEL ee IN * aie Pig. 22 shows a braced nonoplane wing. For the given atr loading, find axtal loads tn litt ana drag truss members. The drag reaction on drag truss ts taken off at point A.EQUILIBRIUM OF FORCE SYSTEMS. provides reststa to V, D and § reactions and monents about D and V axes. Fing the reactions at A and B and the load in member CD for given wheel Loading. 10000 TRUSS STRUCTURES, th nembers 3 and BF ere sistance te v, Dand ‘ reactions but Ke 18000” 10000 Fig. 24 only moment resistance about V axis. Find reactions 2¢ 3 snd loads in zembers BF and BO under given wheel Loading. Cessna Aireratt Nose Wheel Installation (Model 182) Douglas DC-8 Jet Airliner Main Landing Gear UnitCHAPTER A3 PROPERTIES OF SECTIONS - CENTROIDS MOMENTS OF INERTIA ETC. Agi Introduction ‘sd terms, center of gravity and moment of iner- tla, are Constantly deing used. Thus, a drief review of these terms ts in order. A3.2 Controtde, Center of Gravity, ‘The cen~ troid of @ line, area, volume, or mass is that point at which the whole line,area, volume, or ass may be conceived to be concentrated and have ‘the saue moment with respect to an axis ae when digeributed In {ts true or natural way. This general relationship can be expressed by the principle of moments, as follows Lings:- = Zi, hence % = 2x = fxaL t areas:~ Ta = Zax, hence & Yolunes:~ 2¥ = Z¥x, nence ¥ Masses: 3M = Zon, hence % =_am = Sect "Mt if e geometrical figure is symmetrical with respect to 2 line or plane, the centroid of the Iigure lies in the given line or plane. This is ‘obvious from the fact that the moments of the parts of the figure om the opposite sides of the Line or plane are numerically equal but of opposite sign. Ifa figure is symmetrical to two Lines or planes, the centroid of the figure Lies at the intersection of the two lines or the two planes, 2nd likewise, if the figure has 3 planes of symmetry, che centroid lies at the intersec~ tion of the 3 planes. 43.3 Moavat of Inertia ‘The term moment of tn~ ertia 1s applied in uechenics to a mmber of mathematical expressions which represents second goments of areas, volumes and masses, such a Syren, frtav, freee eve. Ad 4 Moweat of Inertia of an Area, As applied coan area, the term moment of inertia nas no physical significance, but represents a quantity entering inte a large number of engineering problems or caloulations. However, it may be Sonsidered as 2 factor which indicates the In~ rluence of tue area {tself in determining the total rotating moment of uniformly varying for— ces applied over an area. Let Fig. AS.1 be any plane area referred to Saree coordinate axes, ox, oy and 92; ox and oy being the plane of area. ‘Let GA represent an elenentary area, with coordinates x, y, and r as shown, In engineering calculattons,| sea tele, ty Sst2h, Te where Ty, 1, ares about Ss48 and Iz are nenonts of inertia of the ine axes 3x, yy and 22 respectively. Aa,5 polar Moment of taertig «In Pig. AS.2, the moment of inertia Iz =/r ‘dA about the Z axis is Teferred to as the polar soaent of inertia and Gan be defined az the xonent of inertia of an area with respect toa poine in its surface. Since r4=x*+ y* (Fig. A3.1) Iz =S{y* + x+) da = Ty + ly oF} the polar monent of inertia 1a equal tothe dum of the coments of tnertia with respect to any to axes In the plane of the area, az right angles to each other and passing thri the point of intersection of the po Tar exis with the plane. A3.6 Radius of Gyration The radius of gyration of a solid is the distance from the inertia axis to that point in the solid at which, if its ene tire mass could be concentrated, its aoment of Inertia would remain the same. thus, fred =@*H, where @ is the radius of gyration 1, then £ =@at or @2\ Vix since, Jraat = the case of an area, T a By analogy, in tChore AacT_ Parallel Axis Theres In Pig. 43-2 let Ty ‘be tne moment of Inertia of the area referred to the centroidal axis y-y, and let the moment of Inertia about axis yxy: be required. vay. ts parallel to yy. Consider the elenentary area dA with distance x +4 from yy. then, ty, =f(e + x)" Sara + cara + a San asa43.2 CENTROIDS, CENTER OF GRAVITY, MOMENTS OF INERTIA Fig. A3.2 1 | I I ty y The first term, /x7dA, represents the nonent of inertia of the body about its centroidal axis y-y and Wi11 be given tne symbol 7. The second term 1s zero decause/'xdA ts tero since yy 1s the controtdal axis of the body. The last tern, 44/ dA = Ad? or, area of dody tines the square of the distance Detween axes yy and HaYae ‘Thus_in general, retest This expression staves that the amount of inertia of an area with respect to any axis tn the plane of the area {s equal to the aouent of thertia of the area with respect to a peraliel centrotdal axis plus the product of the area and| the square of the distance between che two axes. Parallel gxts Theorem For Yasses. If instead of area the mass of the body is considered, the parallel axis can be written: 1=T+Ma*, wnere M{ refers to the mass of the body. AS.7a Masa Mosents of Inertia ‘The product of ‘the mass of 2 particle and the square of its distance from a line or plane is referred to as, the moment of inertia of the aass of the parti- ele with respect to the line or plane, Hence, I= aMrs, If the sumation can de expressed by 2 definite integral, the expression may be written T=/r* a Moments of Inertis of Airplanes, In both ‘lying’ and landing conditions the airplane aay be subjected to angular accelerations, To determine the magnitude of the accelerations as well as the distribution and magnitude of the aass iner- ta resisting forces, the aoment of inertia of the airplane about the three coordinate axes 1s generally required in making a stress analysis of @ particular airplane. ‘The mass monents of inertia of the airplane: about the coordinate X, Y and Z axes through the center of gravity of the airplane can be expressed ag follows: Ty = Daya + aaa * ZAly, ly = Dana + Daze + 2aly Rectangle area = ba Fog TABLE 2 Section Properties of Areas 4(2b + d.) a Ta, = 49 (024400, +, 4) “Bp tot area = t(D + 6) Raptect Fs a% ear Bes ay bee) Fede 2 {Hew Sty" -aty-294] ‘ y= Leb x)8 + axe (are) | tana =2 ter, where ty = abe tpi ate 4 Sint + fy 0008 = sin 20 v ae LaeP Ty C088 + Ly stn? B~ Lay sin 20 Ig = 2mm? + gw + BatyANALYSIS AND DESIGN OF ‘TABLE 1 - Continued FLIGRT VEHICLE STRUCTURES aaa area =n (aids aad4] Area = t(d + 2a) F = ade Ty = ba* = ala ~ 20)? — a Ty = a(d +a}? ~2a%¢ - Gabe van 29 = (at t*) (b*~ dt) x Ty Pot For I,_, and Ty, $88 oqua~ ‘ions as given’for angle. Be = 8 (0%, = a4") Serasecircte Se a ait, a 4 tao Sta a 5 (TORE, ne me ae mat 3 wea nT nee Ream cine et : 7. sgn nr 2 TABLE 2 Properties of Solise Tas een Dw (ego Fe wuscrgn Wh RPL (r= radius) (Potal we.) za Ys sear Ty = 1098 re Ox = 264r m[ee + ysl he oliow Cize. Gr eS - ‘C rrouatie A Vol. =n L(ry*- r4 Rhea tMryt+ rye Tyna = MQ? + ye LE /BVE For thin hollow cire. cyl. Semi-circular Ring area =a (R*- Pr) zg ( in Re? Tx=.1098(R*=r+). ye Ror Ig(approx)=,St(r +2) * when x! MAP eH) approx. F=2r T emt Bor) radius T,-, 7 HE? Rect.Prism 3 Vol. = abl, MW Z & eo, Tay = M (at + pe . AY Ses mene oe eee 2s los 2S : F Soltd Sphere Veh. = (4 nF9/5) Spore Tapout axis * ary: Thin doltow sphere “91+ * 3% (a=?) 71 = radius of aphere 1 ‘about dia. ~aoe GENTROIDS, CENTER OF GRAVITY, MOMENTS OF INERTIA = cont (ores a) _ TABLE 2 - conttaued tg Fate = ESE. Gawls TRG ay « mass por une volume mae ER & of beay. ass 4 ® x Tx tantaat( + (Se */4)) centroids of Traptzoidei areas Iyy=lan*Re* (4R*+ at) y wee L [r > a ‘TABLE S toa section Properties of Lines (e ts sual] in comperison to radtus) Distanse = Bistace 7 Circular are Area > 2m rt Tha saree rey par 2 Mt Ox = .707F Opolar Aznrt 6366 x Ganersicuar Are area = Be : + at : 5 = 5006 5 pT? ays arse Oral ay aor Teng = 149 590 COROMLAR ARS Area = are a tn Radiane ESL oy sa Fs rev sina) gin 20 z tn 2a 7 Moe = AF = ret (1-cos a} ad = FE to - 1528) esse eH] a8 8aseeRs: sayANALYSIS AND DESIGN OF FLIGHT VERICLE STRUCTURES 43.3 ‘rote ¢ a a Sipe gee T Fropelier we) es 3h 8 E feelge treme | He | ue “S| aps] 8 3 Ribeltee’ tie | foo 282 | ttese| i fing Sfowe'"” | too [38s | *hes00| aa 5 seth Ee ‘so [jee | T¥te0| 78 : BSS | ies) at $ Sig | iio) is : yoo His | a0 | 30 joo jies | 3508) “Ye 3 40136 | Hoss] 7 en “are where Ix, fy, and ig are generally referred to | gxample Problem 2, Determine the moment of tner- as the Tolling, pitching and yawing moments of |tia about tne norizontal centroidal axis for the inertia of the dirptane. w= weight of the items tn the airplane xX, ¥ and z equal the distances from the axes thru the center of gravity of the sirplene and the Wefgnts w. The last tora in each equa tion 1s the sumation of the moments of inertia of the various items about their own X, Y and Z centrotdal axes. If w ts expressed in pounds and ces in inches, the uoment of inertia is express~ ed in untts of pound-tnches squared, which can be converted into slug feet squared by multiply~ ing by 1/S2.16 x 144. Bxample Problea 1. Determine the gross weight Sunes or Gravicy of the airplane Seow In S16. 43.3. The airplane wetgnt has been oroken down into"ths 10 itens or weight groups, with their individual c.g. locations denoted ty the symbol Solution. The airplane center of gravity will be located with respect to tno rectangular axes. In| this example, 2 vertical axts thru the center- Ling of the propeller will be selected as a ref erence axis for horizontal distances, and the thrust line as a reference axis for vertical dis-| the dtstan~ tances. The general expressions to be solved are! = Bue = distance to atrplane c.g, trom ‘Gr ref. axis 8-3 ¥ = Dy = distance to atrplane c.g. from IN oraf. axis X-X Table 4 gives the necessary calculations, whanae = 427180 = 183.3" art of f propeller ‘sis 5480 = -1.74" (below thrust Line) Fig. A2.2 area show in Fig. Ad4 Solution, e first find the moment of inertia about a Horizontal reference axis. In this solution, this arbitrary axis hes been taken 25 axis x'x’ thru the base as shown. Having this moment of inertia, a transfer to the centroidal axis can be made. Table 5 gives the detailed calculations for the moment of inertia about axis x'x'. For simplicity, the cross-section has been divided into the five parts, namely, A, B,C,D, and 5. Iox 18 moment of inertia about centrotéal X axis Of the particular part being considered. Distance from axis x’x' to centroidal nortzontal axis = y = Zav = 17.97 = 2.91" TA by peraliel axte thooren, wo tranator the nocent of Inertia from aris sx" to control axis x. , Tee tyrgt ] aes, co wes | - | 37 | as.rsy] 2,082] 2,908, 50 saiass eeaalies |= | 35 | 2a,sa] sea] 2,064,n7 695,388 zeo[s00 | =| 53 | a.ee0] .288/ rig.s09 2. 880 Sutte paaen sosel aoe-3|i08 [28s] $3 | .2;2ea| 5:70n) 2.1355 503 myses Sutee Faset Sousl 35:3 {199 |ige| 43 | lacey! apiite) 1240!s40! 1 gisieee Gutee ganet Fase| “eeig(i32 [igs] $4 | lacegal *Sc24H) 1038953 ae eet iitetoas SUS fis a |i) 230 30855 eralser | = |,99-7] stegee] geagzlta.zm2.ana 93h. SEUSS | 2 [abs°7] THSSS) 3:588)45: 250589 28h S53 siecle |= | or | 55,284) 25,436) 9,706,486) |a,ceo,284 2,870,000) 4,476,286 s9.3| 99 2.39 258,300 sss] 136,400 Bast 358 = | Berkeg Sah) Be n9}_79 zal - 19,499) 91.520 3398 exol 803 B38 waco, coms ol 0 = | 49.000 Fabia ee) 3) a48 2) H8883| so] a,203| : a Fusco} 384) ERS he Boon $3 - s,ago| € 3,568;800 suse. 3 20,3 Be sBxe 2 Fou SESE, je BE eee xsl 0,702 353,08 759,369 Bei09 eel Hei iro 23. ee (Table 64 Prom X.A.C.A. Teck. Rate #575)ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES ‘The principal moments of tnertia are gtven oy following equation. Iyp7 ty cos? 3+ Igsin’d - Iyz sin 2. (see art. A311) Typ= ly Ig = ly sin? S+1, cost + 1, sin2 do Substituting 5061 x (0.9998)* + $086 x (0.0800)* - 161 x 10509 = 3056 Tap * 2062 x (0,0500)* + $096 x (0.9996 + 1eLx 10599 = 9102 tp Typ = 6680 43.7 _Problens gies [* z Fo, Tt x T r Ie | i | | at pot ae ———on Fig. A3.6 Fig. 3.7 (1) Determine the monent of inertia about ‘the horizontal centrotdal exis for the beam Section shorm in Fig. A3.6. (2) For the section as shown in Pig. 43.7 calculate the moment of inertia about the cen— ‘roidal 2 and X axes. ity \ , - eo } Tae 5 5 2 * gon ole ara ft ah Fig. 3.9 Say reas (3) Determine the moment of inertia about ‘the horizontal centroidal axis for the section shown in Pig. AS. (4) In the beam cross-section of Pig. 43.9 agsuze that the four corner members are the only effective saterial. Calculate the centrotcal gonents of inertia about the vertical and nort~ zontal axes. A3.8 Product of Inoreta Tn various engineering problems, particu larly those Involving the calculation of the nonente of inertia of unsymmetrical sections, ‘the expression / xy dA 1s used. This sxpresston ig referrad to as the product of inertia of the area with respect to the rectangular axes x and y. The term, product of tnertia of an area, ane wil be given the symbol Tay, hence er ‘the unit, like thet of morent of inertia, ts expressed 4s inches or feet to the 4ch poner. Since x and y may be either positive or negative, the tora Igy may be zero or either positive or negative. Product of Inertia of a Solid. The product of inertia of a solid is the sum of the products obtained by multiplying the weight of each sualL portion in whicn it say be sssumed to be divided by the product of its distances trom two of the ‘three cooreinate planes through a given point. ‘Thus with respect to planes X end T yy * J xy OH Tyg = x2 lye ove a 3.9. Product of inertia for Azes of Symmetry. If an area ts symetrical about two ree— tangular axes, the product of inertia about these axes is zero, Thts follows from the fact that symmetrical axes are centrotdal x and y axes. If an area 1s symotrical about only one of two rectangular axes, the product of inertia, (feyds, 18 zero because for each product xydk for an element on one side of the axis of symetry, there 1s an equal product of opposite sign for ‘the corresponding element dA on the opposite stde of the axis, thus making the expression /yda equal to zero. 89,20. Parallel Axis Theoren ‘the theorem states that, "the product of tnertia of an area with respect to any pair of co-planar rectangular axes 13 equal to the prad- uct of (nertia of the area with respect to a pair of parallel centroldal exes plus the product of ‘the area and the distances of the centrota of the total area from the given pair.of axes". Or, ax- pressed a3 an equation, ++ -@) ‘This equation is readily dertvadie by re~ forring to Pig, Ad.10. 1¥ and Xf ere centroidal axes fora given area. YY and XX are parallel axes passing tarough point 0. The product of inertia about axes YY and xr 18 Igy = Je eG ED ae = fla GF oR SE Sy as Tf xen The last to tntegrais are each equal to zero, since /yaa and /xdk refer to centroidal axes. Hence, Ixy = /xyda + 37 dA, which can be written in the form of equation (2).43.20 | | yf 3.11 wonents of Inertia with Reapect of Taclined bes Fig. 8.10, Unsymetrical Dean sections are very com~ mon in atrcraft structura, because the airfoil shape 15 generally unsymmetrical. Thus, the general procedure with such sections 1s to first fing the moment of inertia about sone set of rectangular axes and then transfer to other in clined axes. Tus, in Fig. A3.11 the zoment of inertia of the area with respect to axis X,%, is| Tg, =/¥,2 th = /ly cos B~x sin g)*ea = cosa a / yaar sin®d /x*dn- 21nd cos 8 f xysk = Iz cos" B+ Ty sia" 6 ~ 2 Ixy sing cos 6 (3) end Likewise in a similar manner, the following equation can te derivad: ty, “Tgsin" + ty cos*8 + 2 Igy sin 8 cos a-- +4) vy CENTROIDS, CENTER OF GRAVITY, MOMENTS OF INERTIA or the sum of the moments of inertia of an with respect to all pairs of rectangular axes, Thru a common point of intersection, ts constant. A312 Location of Axes for which Product of Inertia i Zero. In Pig. aS11 yy fey, oa=/(xcosd+y sind) (y cos d- OY ysl ada (cos*g=sint 8 / xyca+ cos asin of (ewe) ca why oo8 29+ 1 (xs ty) sin 2 Therefore, Iy,y, ts nero when tan 20 = 2lyy Tyaig 43.13. Principal axes. In problems involving wnsymetrical bending, ‘the moment of an area is frequently used with re spect tc 4 certain axts called the principal axis. A principal axis of an area ts an axis about which the moment of snartia of the area ts either greater or less then for any other xis passing thru the centroid of the area, axes for which the product of inertia ts zero are principal axes. Since the product of inertia 1s zero about symetrical axes, 10 follows thet symetricel axes are principal axes. ‘Tae angle between 2 set of rectangular centroidal axes and the principal axes ts given by equation (6). Example Problem 4 Determine the moment of tnertia of the angle as show in Fig. 43.12 about the principal axes passing through the centroid, Solution: Reference axes X and Y are assured as shown in Pig. AB.12 and the moment of inertia ts first calculated about these axes, Table 6 gives the calculations. The angle ts divided into the two portions (1) and (2). Pig ana aaa 1 me fuer | wt Ded oe | | Tia ts pues [rows [ann wse [oa [aos yan ma | aoe [sierra raat ae ore ar 2a oe a = ror toes {tome iss err}ANALYSIS AND DESIGN OF, Log and 1 = noment of inertia of each portion about their own X and Y centrota~ al axes, Location of centrateal axes:~ 3 = zay = .o719 = 757" TA SS ax = SASS = 392" KES ‘qranster aonent of inertia and product of tner— tia tron reference X and Y axes to parallel neroldal axes:~ 1 CAI? = 985.075 « TET? = 40 1281 ~ 878 x .SBE* = 4187 sIS2 ~ 1278 x .767 x 302 = Ix Zynty oak? Ty sty -AR Tay = hay - “fs0 calculate angle between centraidal x and f exes and princtpal axes through centrotdal as follows: van 2 9=2 fey = 1.06 Ty- ix 20246 - 40" d=25? ~ 20" falculate somente of inertia about centroidel principal axes as Tgp = Ty cos? 6+ Ty sin® 0-2iyy stagcosd = 44x STG" + 187 « S5S*-2(-.150) x 19968 x .918 = 604 11) Type iy sta’ O+Fy cos? 8+ 2igy sind cosd 44x 5965 + 187 x .5TS* + 2(-.160) x follows :~ & , ro, ‘ ae rile # Tet jitter alt ye Bxample Problem Fig. A5.15 Shows a typical distributed flange ~ 2 cell ~ wing beaa section, The upper and lover surface 1s stiffened by Zand buld angle sections. Determine the coment of inertia of the section about the principal axes. Solution: ‘The properties of the cross-section depend upon the effective material wnich can develop resisting axial stresses. ‘The question of effective material is taken up in later chapter. ‘Table 9 shows the calculations for the moment of inertia about the assumed rectangular referent axes 1 and YY (see Fig. 43.13). The cross- LIGHT VEHICLE STRUCTURES sect as 1 the 1 the ¢ consi: the s 28 thicknesses Aaa on has been broken down into 18 stringers sted in column 1, For the top surface, 2 width of 20 thicknesses of the .CS2 skin 1s as~ sumed to act with he stringers and a sldth of ne .04 sin (sae Col. 3). on ower surface, the skin hal? way to adjacent stringers 1s assumed acting with each strtaser, or the entire skin ts effactive. Column 4 gives ombined area of each stringer uit and Is dered as concentrated at the centroid of stringer and effective skin. Ail distances, x and y, colums 5 end 8, have been scaled from a large drawing. Fig. Ada sisteaices*t axes, Location of centrofdsl axes with respect to ref. = 166.5 ins iy 7 1548.56 - 30) 70x I5.727= 431.7 ins ty=as, 58 - (5.70 .396 x 18.74) =36.41 Ins 2 Tyy=2 (235.41) = - 28686 Eph, Bi-ties 2 O=i6%-52.5', O= 8-16.25" Typ tx cos d+ Ty sin? 6-2 Ixy sind cosa = 185.46 x 9886" + 431.7 x .figs* -2 2 ye1.2 ine Ese6.dix 19898 x (~.1488]AgI2 CENTROIDS, CENTER OF GRAVITY, MOMENTS OF INERTIA ix sine B+ Ty cost d+2 Try in AB. 14 Seetion Properties of Typical Atrerait Structural eos + eve cease Jectione, 186 45 x. LEEEZ® FASL.7 x 2895) Table AS.i0 throush A3.15 and zive me section properties of a few si Shapes comen to aircrazt, Use of these tal MALL be sade in later chapters of this Sook.ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES Aus arr Table Aa. 12 bat oe Nee Properties of Extruded Aluminum Sear USS Sess yous aa. GSS ‘cow Handbosk) ee Dea re SET. ati eee en Pe ep ‘sva] sae] va | 211 | 0.004] 0.293] o. 202 J .cors] 0, x17 aval vie] iva | “aap | 0°04] 0°20| 0299 | “cora| o! Lez sya] aaa | a | “ise | 0/08] a°ata| 224 | ooee| 0: tat Sa] ie | ia | csty | o:o0a| o:ain| oa? | ooea] 0: tat yas oan 9.198 va oat pared va 2.8 Oi we 3B Oise waa oca8 om va oar on vis H om ane eH om 338 co 230 Sas “3 0 38 3S ote [ou fo ans fej ante [ost | olde 3/32 ss | ale [ose | 0.35 ant 3 | oler [050 | 036 is sr jose [om | ose a8 2 | orse [ons | 0.36 va BL fost f98 | 0.10 vs at jose un [oa va eo forse [ie | o39 wa solos: bir [oie nF Tale =a este P= LeeAgu CENTROIDS, CENTER OF GRAVITY. MOMENTS OF INERTIA ‘Table A218 SECTION PROPERTIES OF TYPICAL AIRCRAFT EXTRUDED SECTIONS Sseces| Dinensio | n2eq | properties about 1-r! propersies about 12 ee a ey ee { ri (Lh wae) tie las [eo | oso] ite [ose | woosse sivas ois tes | | TP [rae | .080 80] 080 | | oat gees a0 aos | 7 | nna Yaa | Wa Tae [ie [eas] a2 | toes] 0100 | ata ane one | a8) ne | [_9ns| we [3718 [000 ].080] _o/a2 .t0ar .olaa [mia _.9e9 | oats | nia | -as0 {9ns| va [sne|.oso|.o7s| 2/22 | toe! .o120 | 320 | 248 | 00202 | 6s | .209| [3a [ia [ae|ors| ors] aie | iste) ons [267,408 oosso | 00 | 1a [icsae [ore [57a [arse] evea | aa oso ai | ate Toone [at] a Loe ee pe Sie oss aS ORT oe oe We | we) saa [aia| ae | avs | osetia) eats oom | oe ere) waa [eae] asians amo ,ank a0, -onae | ta | a! wave) ure] [aaa |svan] ayaa | 35am, osa6 1 aa0 as -o0een | 206 | 210 i [iaame[i | -‘[aAe|sas] aaa [seas | asso | ar? a0 ons [aos] ate peia [fa fe | ae ost sas | aso | ome | a ae T Wis | a ee ee | OTE | SET Wari ine ane is eats ne sae aaa Rese we [oat [vs |iae]—_a/ae (400 | .00de0 tee And 0088 [as | ala [rsa aa [san lis | svae] 3764 ane, 00ses | 1a ase | ae te [2 [tate [ave Savon arse) 2/2 | -one ons | ze” 201, ov [son 0 pe [hia eae [ue joe) asa sas) ours [an 300 eat | aoe pa ew IA [sae ae oa owe ow a | 2eAe La sas [sian isan’ 5/2080 | 2078 | so) .988 2826 | 590 hie! naa ewe) [ie [vas] ae aot ose | ones oe] a0 a | Rwe le T we — 17a .304 i882 | 6a —~6S% | 0848 | 902 5 750) M [han [@ 1 [aa [ive |e Teo ats [maa 757 | 0800 | amt sa8| 21a [Ree WA [sae | 97s [te 1.00 | wees oso ae] a | we? inva) saa 57aa.a0s aso tse | te (2.00 7080, 080 wena | 40a —-300 | -o1es | -m4m | «6828 tas [_aarae [aire [aris [000 [080 | 088 a | woa7s | 486.562, 0205 | 405 _| « nas) | 3/8_[mue late | d30 [17s | Wie "7a | oes | sas Twist | m8 | 800) 30 | 34s lia [ase [Ae |e ase visez | 792 1,00, or0t | ase | 725) mam [2 a fine [aaae aie _o7aa | wae | 773 1,00, ona |r| to) x 1s | a/a2 | 9/52 9/32 120 785 _1,00_| .0768_| 388 | 78 | E “as; | 0086 |. 208 | | wa | ve | ieae lie) aie [tts | .o1ss oes | .ooase | 07 | -s00 | 3 | ee [a [ 1s | .080 | 3/98 [aoase } .ae7__.500 [ooze | ase [532 | ae | eae (tua) _[.ovs| ors, 2/32 |.a70| .oa04 | 47s, 628 | -o0ete| 237” | 525] a | ae hia |e | t00) aaa sa | -oss0 | sos _.o78 | ote | .s09 | .008 t i I i wees eve vise | __a/a2 902 | ano soa, 700 | oak | 360 | tar aA iP vom) we jes | ied | 157 Loo fone [.367| 7] ea iF aaa] [aa ates | 7 1.00] i wee [368] 700 |‘TaKNvo 40 JuoraH = A313 ove sso ane “Snare, seca wd weld ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES ‘uHaNt 40 INEHON = wee) SNOLLOUS TANNYHO CANOE JO SALLIGAONA L'eY AIH.ASS CENTROIDS, A.15 Problems Fig. A218 (1) Por the section of Fiz. the moment of inertia about =: pal axes Xp and Zp. (2) Ciculate the aonent of inertia of the section in Fiz. 43.15 about the principal axes. ch of the arts 1 Fig. 3.16 Fig. ASAT (3) Pig. AS.16 tllustrates a box type bean section with etx longitudinal stringers. De- termine the monent of Inertia of the beam sec~ tion about the principal axes for the follon= tng assumptions:— (a) Assume the beam {s bending upward putting the top portion in compression and the lower portion in tens{on, Therefore, neglect sheet on the top side sthee it has very little resistance to compressive stresses. The sheet on the bottom side is effective since it ts in tension, For simplicity neglect the vertical weds in the calculations. (b) Reverse the conditions in (a) thus placing top side in tension and lower side in compression. (4) For the three stringar single cell box beam section in Fig. 48.17, calculate the moments of {nertia about the prineipal axes. As sume all weo or all material ineffective. Fig. A028 AB.L4 determing CENTER OF GRAVITY, MOMENTS OF INERTIA Fig. A019 (6) Fig. 43.19 shows a wine beam sect with 2 cut-out on the loner surface. the aoments of Inertia apout th assuming the eight stringers are the only 2: tive material. (7) Ftg, 43,20 shows 2 3 flanze bean, The 7 flange members on the uocer face of beans have an ares of .3 sq. In. eacn and those on the bottom skin 0.2 sq. in. eacn, The bottom skin is .0S inches in thickness. Compute the moments of inertia about th principle axes assuming that the flange members and the bottom skin comprise the ezfestive aaterial. \—curout for door Fig. A221 (3) Pig. 43.21 shows the crass-section of a small fuselage. The dasied line represents a cut-out in the structure due 1 oT. AB sume each of the 18 stringers seve an area of 0.2 sq. in, Consider fuselage skin Ineft20" Calculate the moment o¢ inertia of the effective Section about the principal axes.CHAPTER A4 GENERAL LOADS ON AIRCRAFT Ad.1 Introduction. Before the structural design of an airplane can bo gade, the external loads acting on the airplane in flight, landing and take-off con— ditions mist be known. The complete determin ation of the air Loads on an airplane requires a ‘thorough theoretical Imowladge of aerodynamics, since modern aircraft fly in sub-sonic, trans= sonic and super-sonte speed ranges. Aurther- more, there is a wide range of wing conftgur~ ations, such as the straignt tapered wing, the swept wing and the delta wing, and many of these wings often include leading and trailing edge devices tor promoting better 11ft or control characteristics. ‘The presence of power plant nacelle units, external fuel tanks, otc. are units that effect the airflow around’ the wing and thas effect the magnitude and distribution of the air forces on the wing. Likewise, ‘the fuselage or airplane body itself influences ‘the airflow over-the wing. The theoretical calculation of the airloads on the airplane is too large 4 subject to be covered in a Structures book and it {s customary in college aeronautical curricula to provide a separate course for this, subject. Tn gost airplane companies the loads on ‘the airplane are determined by a group of en- Gineors assigned to the Structures analysis Section and this group is often refarred to as the Aircraft Load Calculation group. While the work of this group 1s primarily based on the uso of aerodynamics, {t is that phase of aaro~ dynamics which is conserved with determining the magnitude end distribution of the air loads on the airplane so that the airplane structure can be properly designed to support these air forces safely and efficiently, The engineering department of an airplane company has a distinct or separate aerodynamics section, but in general their responsibility 1s the use of the subject of aerodynamics to insure or guarantee the per formance, stability and control of the airplane. A basic general over-all imowledge of the loads on aircraft 15 desirable tn the study of aireraft structural theory, and hence this chapter attempts to give this informtion. in a later chapter dealing with wing design, this subject will be further expanded, AM Limit or Applied Loads. Design Loads, Because an airplane 1s designed to carry out a definite Job, there result many types of aircraft relative to size, conziguration and performance. for exanple, ¢ comercial transport Like the Douglas O08 is dasimed to do a Job of transporting 8 certain mumber of passengers safely, sffictently and comfortably over various distances between airports. On the other hand the Air Force Fignter type of air~ craft has a Job of shooting dow eneny aireratt or protecting slower friendly aircraft. To do this job efficiently requires a far different configuration as comparad to the DC-2 transport. Furthermore the Fighter type airplane aust be maneuvered far more sharply to do its required fob as compared to the J0-8 in doing its re~ quired joo. In general the aagnitude of the air forces on an airplane depend on the velocity of the airplane and the rate at whicn this velocity ts changed in magnitude and direction (acceleration), The menitude of the flight acceleration factor may be governed by the capacity of the mnman body to withstand these acceleration inertia forces without injury which {s the situation in a fighter type of airplane. On the other hand the maneuvering accelerations for the DC-8 are not dictated by what the human body can withstand, but are determined by what 19 necessary to safely transport passengers from one airport So another. Designing the airplane structure for loads greater than the airplanes suffers in the par- formance of tts raquired Jod, obviously will add considerable weight to the airplane and decrease its performance or over-all effictency relative to the Job it 1s designed to do. To particularly insure safety in the air ‘transportation, along with uniformity and ef— fictency of desten, the government aeronautical agencies (civil and atlitary) nave definite requirements for the vartous types of aircraft relative to the menitude of loads to be used tn the structural design of aircraft. In referring in general to these specified aircraft loads two terms are used a8 follows: Limit or Applied Loads. MENTHO terms lini? Gnd applied refer to tho same loads with the civil agenctes (C.A.4.) using the term limit and the military agenctes using the term applied. Limtt loads are the maxtmum loads antict- pated-on the airplane during its lifetime of service. ‘Tho airplane structure shall be capable of supporting the limit loads without suffering detrimental permanent deformations. At all loads up to the limit icads the deformtion of the structure shall be such as not to interfere mith the Saze operation of the airplane. Ultimate or Design Loads. GETisie Ure Tenis SPS used im general to mean Aa4.2 GENERAL LOADS ON AIRCRAFT the same thing. Ultinate or Design Loads are ogiel to the Lintt ‘cade sultisiied by a factor of safety (F.S.) or Design Loads = Limtt or Applied Loads times F.S. In general the over-all factor of safety ts 3.5, The government requirenents also spect?: that these design loads be carried by the structure without re. Although afreraft are undergo greater loacs than the specified Litt loads, @ certain amount of reserve strength against complete structural failure of a unit ts Recessary in the design of practically any na~ chine or structure. This {s due to nany factors such as:- (1) The approximations involved in aercdynante theory and also structural stress analysis theory; (2) Variation tn physical Properties of materials; (3) Vartation in fab- Tioation and inspection standards. Possibly the most important reason for the factors of safety for airplanes 1s due to the fact that practically every airplane 1s limited to the uaximum velocity it can be flown and the maxtmum acceleration {t can be subjected to in flight or lending. Since thesa ara under the control of the pflot it is possible in energ- ency conditions that the limit loads may be slightly exceeded but with a reserve factor of safety against fatlure this excesding of Limit load should not prove serious from an airplane safety standpoint, although it mich cause permanent structural deformations thet might require repair or replacements of small units or portions of the structure, ‘Load3 due to airplane gusts, are aro{trery in that the gust velocity 1s assumed. Al- though this gust velocity is based on years of experience in measuring and recording gust forces in flignt all over the world, tt 1s quite possible that during the lifetime of an alr- Plane, turoulent conditions near stera areas or over mountains or water areas aight produce air gust velocities slightly greater than that Specified tn the load requirements, taus the factor of safety insures safety against failure 1f this situation would arise. Rot supposed to ‘The broad general category of external Joads on conventional aircrart can te broken down into such classifications as follows: Dus to Airplane Maneuvers. (under the control of the pilot), Due to Air Gusts, (not under contro? of pilot). (1) Air Loads Landing on Land. ski type). Landing on Water. Arresting, (Landing on Air- crart Carriers), (wheel or (2) Landing Loads (4) Take off Loads auxtitary enor shrust «nits. Hoisting Airplane. ‘Towing Atrclane. Beaching of ull type Fuselage Pressurtains. (5) Special Loads anplai (s) Netent and Inertia Loads. In resolving external loads for 5} analysis purposes, 1t 1s convenient vo seve a Set of reference axes. The reference axes XZ passing through the center of gravity of the airplane as tllustrated in Fig. Ag. normally used in stress anslysis wore as well 2s for aaradymanic celouletions. Fer the reference axes are orten re: Fig. M40. AG4 Weight and Inertia Forces. ‘The term weight {s thet constant force, proportional to 188 qass, woich tends ta draw every Physical bedy toward the center of the earth. an airplane fn steady flight (uniform veloetty) is acted upon by 2 system of forces in squilid- Tium, namely, the weight of the airplane, che air forces on the complete airplane, and the poner plant forees. The pilot can change this balanced steady fiight condition by changing the ongine power or by operating the surface controls to change the direction of the airplane velocity. ‘These unbalanced forces thus cause the 2irplane to accelerate or de-acceierave, Inertia Forces For Tend Bodies Ir the unbalanced forces acting on a rigid body cause only a change in the magnitude of the velocity of the bedy, dut not its direction, the notion ts called translation, and from basi¢ Physics, the accelerating scrce F = Ma, wnere 4 tS the ass of the body or w/g. In Fiz. 44.1. the unbalanced force system causes the rigid dedy to accelerate to tne right. Fig. A4.2 shows the effect of this unbalanced fores ta sroducingANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES F = unbalanced Etlective Force ‘external force ‘e2mas Ma ma Bat Motion 3 ; = t = w = Assume Friction tw zr maa Fig. Ad. Fig. A6.2 a force on each mss particle of ma, maa, ete, ‘thus the total affective force is Ina = Ma. If these effactive forces are reversed they are referred to as inertia forces. The external forces and the inertia forces therefore form a force systen in equilibrium. Frog basic Physics, wo have the following relationships for a motion of pure translation Lf the acceleration is constant:~ ave sat------- @ Ss = vot + hat! @) vioewt=ms -------------- (3) where, 8 = distance moved in time t. initial velocity final velocity after time t. ve ve Inertia Forces on Rotating Rtgid Bodies. ‘A common airplane manouver ts a motion along a curved path ina plane parallel to the XZ plane of the airplane, and generally referred to as the pitching plane. A pull up from steady flight or 4 pull out from a dive causes an air~ Plane to follow a curved path. Fig. Ad.3 shows an airplane following e curved path. If at point A the velocity ts increasing along its path, the airplane 1s being subjected to two Aocalerations, nazely, a,, tangential to the curve at point A and aquél in magnitude to Center of Curvature 0 ME ite Mv4/E Fig. A4.3 ast a, = Fa, and a, = Fo an acceleration normal to the flight path at A and directed toward the center of retation (o}. From Newton’s Law the effective forces due to these accelerations are:= Ryo Mfat = tits 22 eee a Fpetfe ---------+-------- where @ = angular velocity at the point A. a = angular acceleration at potnt A. F = radius of curvature of flight rach at point A. ‘The imertia forces are equal ané opposite to these eftactive forces as indicated in Fig. 44.3. These inertia forces can then be con sidered as part of the total force system on the airplane which {3 tn equilibrium, Tf the vatoctty of the airplane along the path is constant, then a, = zero and thus the inertia force Fy inertia tore F,. If the angular acceleration ts constant, the following relationships nold. 9, leaving only the normal ora @ = ust + dott ot = 9" = 208 where @ = anglo of rotation in time t. @ = initial angular velocity in rad/sec. @ = angular velocity after time t. Tn Fig, 44.3 the moment T. of the inertia forces abgut the center of rotation (o) equals We(F)2 wea, The tera MF 41s the mass moment of inertia of the airplane about point {0}. Since an airplane has considerable pitching Goment of inertia about its own center of gravity axis, it should be included. parallel axis ‘Tus by the (2) g 7 Roment of inertia of airplane about ¥ axis through c.g. of airplane, Inertia Forces for Pitching Rotation of Airplane ‘about Y Ais Through c.g. airplane. In flight, an air gust may strike the horizontal tail producing a tai force which has a aonent about the airplane c.g. In some lending conditions the ground or water forces do not pase through the airplane c.g., thus producing 2 moment about the airplane c.g. These moments cause the airplane to rotate avout the T axis through the c.g. ‘Therafore for this effect alone the center of rotation in Fig. a4.3 ts net at (o} but atfew the c.g. of airplane, or F = 0. Thus F, and F, equal zero and thus the only inertia force for ‘the pure rotation is I, a, (a coupe) and ‘thus the moment of this inertia couple about the t As explained before if the inertia forces are included with all other applied forces on the airplane, then the airplane ts in static equilieriug and the problem is handled by the static equations for squilivrim. Ad.5 Air Forces on Wing ‘The wing of an airplane carries the major portion of the air forces. in level steady flight the vertical upward force of the air on the wing, practically equals the weteht of the airplane. The term atrfo{l 1s used when re~ ferring to the shape of the cross-section of a wing. Pigs. A¢.4 and 44.5 illustrate the air pressure intensity diagram due to an air— Angle of Attack = 89 Fig. AGS stream flowing around an airfotl shape for both @ positive and negative angle of attack. The ghape and intensity of this diagram is influenced by many factors, such as the snape of the airfoti itself, as the thicimess to chord Tatio, the camber of the top and bottem surfaces ete. A normal wing is attached to a fuselage and tt may support external poner plants, wing tip tanks ete. Furthermore the normal wing 1s usually tapered in planform and thickness and may possess leading and trailing slots and flaps to produce Aizn Litt or control effects. The airflow around the wing 19 affected by such factors as listed above and thus wind tunnel tests are usually necessary to obtain a true picture of the air forces on a wing relative to their chordwise spanwise distribution. Regultant Air Force, Center of Preseure It {8 conventent when dealing with the balancing or equilibrium of the airplane 2s 2 whole, to deal with the resultant of the total GENERAL LOADS ON AIRCRAFT sir forecs on the wing. For example, consider the two air pressure intensity diagrams in Figs. a4.6 and A4.7, These distributed force systems can be replaced by their resultant (R), which of course must be kncwn in magnitude, direction and location. The location 1s specified by a ‘erm called the center of pressure waich is She point where the resultant R intersects the a. foil chord line. As the angle of attack ts changed the resultant air force changes in magnitude, direction and center of pressure Location. Fig. A468 Fig, AGT Litt and Drag Components of Resultant Air Force. Instead of dealing with the resultant tore R, it 4s convenient for both aerodynamic and stress analysts considerations to replace th resultant by {ts two coaponents perpendicular and parallel to the airstream. Fig. 44.3 11+ lustrates this resolution into 1{7t ang drag components. y Fig. Ad. Fig. Ad. 9 Agrodymantc Canter (a.c.}. Since an atrplane flies at miny different angles of attack, At means that the center of pressure changes f¢r the many flight destan conditions. It so nap~ pens, that there ts one point on the airfoil ‘that the moment due to the Litt and Drag forces is constant for any angle of attack. This point {s called the aerodynamic center (a.c.) and its approximate Location ts at the 26 percent of chord measured fron the leading edge. Thus the resultant A can be replaced by 2 lift and drag force at the aerodynamic center plus a wing Roment %q.c, a8 {llustrated in Fig. as.3. AM.& Forces on Atrplaiie in Flight. Fig. A4.10 tllustrates tn general the min forces on the airplane in an accelerated fltgnt condition,ANALYSIS AND DESIGN OF Fig. A410 r L = total wing Lift plus fuselage Lit. D = total airplane drag. Ma = aorent of L and D with reference to wing ‘a.c. (aercdynante center) W = meignt of airplane. inertia force normal to flight path. tmertia foree parallel to flight path. rotation thertia zoment. = tet] load normal to flight path. For a horizontal constant velocity fltzht condition, the inertia forces 1,, Ip, and 1, would be zero. For an accelerated flight con~ dizion involving translation but not angular acceleration about its own ¢.g. axis, the inertia moment 1, would be gero, bub fy and Ip would have values. Equations of Equiltorium For Steady Flight. From Fig. a4.l0 we can write 0, D+ sin@Tcos B= 0 LW cos Gf sin p+ E=0 = Mg - Ia - 0b + Tc cos B+ Eo Bquations of Equilibrium in Accelerated Flight Woo, 0+ ¥sin@-=T cos B-I,=0 ary BY, #0) + Na - la =o, L-Ncos@+TsinB-1,-E50 - db + To cos p+ Ee tis {Foress = Plus 1s up and toward tail, Monent - Clockwise 1s positive. Stgns used: Distances trom c.g. to force = Plus 1s up and toward tail. A. Load Factors. ‘The term load factor normally given the symbol (a) can be defined as the numerical mul- The bar through letter 7 has no significance. ing without bar, Same mean- LIGHT VEHICLE STRUCTURES A taplying factor by which the forces on the air— plane in steady fltght ara multiplied to obtain a static system of forces equivalent to the 4y- namic force system acting during the accelera~ ‘ion of the airplane. Fig. Ad.11 {llustrates z Le total ft (Wing & Tail) TD Fig. ASL 4 steady horizontal flight. L repra~ total airplane lift (wing plus tail). L =H, Now assune the airplane ts ac~ upward along the Z axis. Fig. Ad.12 additional inertia force Wa,/e acting forces in sents the ‘Therefore colerated shows the dowarard, or opposite to the direction of Gecsterstton. ‘the total airplane 1ift L for the z 1 ae ra Fig. Ad.12 { Ted Cea unaccelerated condition in Fig. Ad.11 mst be multiplied by a lead factor n, to produce static equilibrim tn the & direction. 4 Tas, aL -W#- Lasso Stnee L Hence 3, ‘4n airplane cen of course be accelerated along the L axis as well as the S axis, Thus in Fig. Ad.13 the magnitude of the angine thrust T is greater than the airplane Drag 9, which ‘a a ee Fig. Aa 12 ‘T te greater than D fe @ causes the airplane to accelerate forward. It is convenient to express the Inertia force in the X direction in ters of the load factor 2, and the46 weight W of the airplene, nence a4 tee 4 ays fe, (eee rig. ana.15) BF, = 0, whence T-Daat = 0 =D once a, = 2 Therefore the loads on she atrplene can be ats~ cussed In teraz of lead factors. The applied or Limit load factors are the maximum loa 4 Ghat mignt occur curing the servic> of the par- ctoular airplane. Thes2 loads as itscussed in Art. Ad,2 must be taken by the airplane structure without appreciable permanent deformation. The destgn load factors are equel to the Limit load factors multiplied by the factor of safety, and these desim loads aust be carried by the structure without rupture or collapse, or in other ords, conplete failure. AGB Design Flight Requirements for Airplane. ‘Tae Civil and Military Aeronautics Author- ities issue requirements which specity che design conditions for the various classiftcation of airplanes. Generally speaking, any airplane flight altivuce can ba defined by stating the existing values of load factors (acosleration) and the airspeed (or more properly the dynamic pressure). The accelerations on duced from two causes, namely, maneuvers and air gusts. The accelerations due to maneuvers are Subject to the control of the pilot who can manipulate the controls so 4s not to exceed = certain acceleration. In highly maneuverable military airplanes, an acceleroneter ts in~ cluded in the cockpit instruments as a guide to Limit the acceleration factor, For comerctal airplanes the maneuver factors are sade high enough to safely take care of any maneuvers that would be required tn the necessary flight opera— ‘ions of the particular type of airplane. These Limiting maneuver factors are basad on yaars of operating experience and have given satisfactory results fren a safety stendpoint without pen- alizing the airplane trom a weignt design con~ sideration. ‘The accelerations due to the etrplans striking an air sust are not under the contrat of the pilot since 1t depends on the dirscticn and velocity of the atr gust. From mich ac- cumlated data obtained by installing accelero- meters in comercial and ailltary atrerart and flying them in all types of weather and loca CLons, {t nas deen found that a gust velocity of SO tt. per second appears suttictant. The speed or velocity of the airplane ‘he loads om tha airplane. The higher the velocity the nigher the aerodynamte wing moment. Furthermore the gust accelerations increase with airplane velocity, thus tt airplane are pro- GENERAL LOADS ON AIRCRAFT a definite zaximan elec: For com mercial airplanes the velocity is limitad oo 2, reasonable = take care of Ide speed Naich 1s sufficient t operas: easonazle ILt ‘A49 Gust Loaa Factors, ‘aon & sharp scze in a direction normal to the thrust 1 axis), a sudden changs taxes place in angle of attack with no speed. Tae norsal force be ascumed to vary attack. Thus in F: seat the normal necessary to meinta:n leval flight velocity ¥ and potnt (2) the value Sa in the angle of attack without chang: ‘The total increase in the airplane lead in the Z sn can therefore ce expressed ay the v Fig. b Fig. a ‘The load factor increment Jue to the gust KU can then be expressed 2, = (Mim) (ots) KUvS, Bi a v BSW U = gust veloc! K's gust correction factor depending on wing loading (Curves tor K are provided dy Civil Aeronauttes authorities), = indicated air speed in atles per hour. 5 = wing area in sq. ft. M = gross weight of airplane. in tt./sec. ‘NACA Technical Note 2964 (June 1959), proposes that the alleviation factor & should be replaced by a gust ‘actor, Bg = 0.88 Mg/G.2 + Mg). ia this exprescion Mg ia th alfplane mass ratio or mass parameter, 2 W/Gpces, 12 is customary to iimit the particular airplane to which ¢ 15 the mean geometric chord in feet andg the ac~ celeration due to gravity.ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES If vis change in C, aken as 30 ¢t./sec. and 2 as the with respect to angle of attack A tn absolute units per degres, equation (4) ree duces to the followiag Sony nes @) Therefore the gust load factor n when atrplane is flying in horizontal altitude equals 2+ SW lle nai, Se --- ©) and when atrplene HW lee (9) WS ‘AS.10 Ilustration of Main Flight Conditions. Velocity-Load Factor Diagram, As indicated before the main design flight conditions for an airplane can be given by Stating the limiting values of the acceleration and speed ang in addition the maximm value of the applied gust velocity. as an {llustration, the design Leading requirenents for a certain airplane could be stated a3 follows: "The proposed airplane shell be designed for applied Positive and negative accelerations of + 5.0 and -3.5g respectively at all speeds trom that corresponding to CL, up to 1.4 tines the saximm level flignt speed. Furthertore, the airplane shall withstand any applied loads due to 2 30 ft./sec. gust acting in any direction up to the restricted spead of 1.4 times the maximm level flignt speed. 4 desim factor of y of 1.5 shall be used on these applied Loads”, In graphical form these design require tents can be represented vy plotting load tac~ tor and velocity to obtain a dlagraz whtch 1s generally referred to as the Yeloc!ty-accelera— ‘ion diagram. The results of the 2bove speci- fleation would be similar to that of Fig, Adsl4. Thus, the lines AB and CD represent ths restricted positive ana negative taneuver Load factors which are limited to speeds inside 1 SD which ts taken as 1.4 tines the maxtmum level flight speed in thts {llustration. These tricted mansuver lines are terminated at points A and © by their intersection with Axim C, values of the airplone. At speeds detneen A and B, the pilot must be cai to axceed the maneuver accelerations, since in general, {5 would be possible for him to man— tpulate the controls to exceed these valuas. Ab Speeds below A and C, there need be no cre of the pilot a3 far as loads on the airplane re conberned since a sanouver precuetng C, ne ALT would give an acceleration less than ited values given by Lines 48 and cD. The positive and negative gust accelera- ttons due to a GO'Pt./sec, gust normal to fliznt path are shown on Fig. adlld. In this example diagram, a positive cust is not critical within the restricted velocity of the airplane since the gust Lines intersect the line 80 below the Line'sB. For a negative gust, the cust load factor becones critical at velocities between F and D with 2 maximm acceleration as given by point £. For airplanes wnich have a relatively low required maneuver factor the cust accelerations way be critical for both positive ana negative accelerations. Examination of the just squatton indicates that the most lightly leadea condition (smallest gross weight) produces the nighest. gust load factor, thus thvolving only partial Pay load, fuel, etc. On the diagram, che points A and 3 corre. spond in general to'what 1s referred to as high angle of attack (H,A,A,) and low angle of attack (Lea.k.) respectively, and points C and 0 the inverted (H.A.A.) and’ (L.A.A.# conditions spectively. ‘Generally speaking, tf the airplane is de~ Simed for the air loads produced ty the velocity and acceleration conditions at points A, 3, 5, F, end C, 1% should be safe frog a structural strength standpotnt 1f flown within the spectfted Limits regarding velocity ana acceleration. Basically, the flight condition requires ments of the Civil Aeronautics authority, amy, and Navy are based on consideration of specified velocities and accelerations and 2 consideration of gusts, thus 2 student understanding che basic discussion above should have no difficulty understanding the desiga requirements of these three government agencies, For stress analysis purposas, all speeds are expressed 2s indicated alr spseds. Tne "mdicated" air speed 1s defined as the speed. which would be indicated by a perfect atr-speed Indicator, that 13, one that would indicate true air speed at Sea level under standard at~ aospheric conditions. The relation between the actual air Speed V, and the indicated air speed Vy ts given by the equation lim wnere ¥, = dicated atrspeed ¥, = actual atrspeed , = Standard air density at sea Level f= density of air in which V, is attained " fsAue = Acceleration in Terms of GENERAL LOADS ON AIRCRAFT the Loads on the wing should be checked ror cases where the ‘to the wing and are located forward ding edge. Tn cases where the lanaing gear is atta ‘to wing or when the fuel and engines are th and on the wing, the leads produced ot wing structure In a landing condition nay be critical for some portions of the wing struct) inboard of landing gear and engine attachmen points, A4.12 Example Problems lnvolving Accelerated Motion of Rigid Airplane. AS previously explained, it is general practice to place the airplane under accelerated conditions ef motion into a condition of static equilisrim by adding the inertia forces to applied force system acting on che airplane. It is usually assumed that the airplane is 2 rigtd body. Several oxample problems will be 2re~ sented to illustrate this general procedure. Example Proplen 2 eee AM11 Special FUignt Design Conditions. There are many other flight conditions watch aay de critical for certain portions of the wing or fiselage seructure. ost airplanes are equipped with flaps, to decrease tie 2and— tng speed and such laps are lowered at speeds at least cwice that of the minum landing speed. Since the flapped airfoil has different values for the sagnitude and location of the airfot characteristics, the wing structure mist be checked for all possible flap conditions within tha specified requiresent relative to mxinun speed at wnich the flaps may be oper- ated. Generally speaking, tha flap conditions wilt etfect only the wing portion inboard of the flap and it 18 usually only eritical for the Ter beam web or shear wali and for the top and bottan walls of the torsion box. This 13 due to the fact that the deflection flap noves the center of preeaure considerably aft thus pro~ ducing aore shear load on the rear shear wall as well as torsional aoment on the conventional cantilever box metal deus. The airplane met iikewise be investigated for atleron conditions. Operation of the ailer~ ans produce a dizterent air lead on each side Sf the airplane wing wnien produces an angular Tolling acceleration of the airplane. Turther= nore, the deflected ailerons change the mas~ Rizale and location of the airfoll character~ fsties, thus calculations mist be carried out to determine wiether the loads in the aileron con~ ditions are more critical than chose for the normal tlignt conditions. For anguiar acceleration resulting trom pitening moments due to eir gusts on the tail, Fig. Ad,15 tllustrates an airplane landing on a Navy alferatt carrier end being arrested Sy 2 cable pull T on the airplane arresting neck. If the airplane wetznt is 12,000 Ibs. and the airplane {5 given a constant acceleration of 3.5¢ (212.7 tt/sec*), find the hook pull 7, the wheel reaction %, and the distance (d) between the line of action of the hook pull and the airplane c.g. If the landing velocity is 60 M.P.K. wnat is the stopping distance. Ws 12000 tb, R Fig. A418 Sotuticn: = on contact of the airplane with the arresting cable, the airplane 1s decelerated to the Fight relative to Fig. 44.5, The motion 1s pure translation horizontally. The inertia force is cs wa. (12000 . $= (Be) s.5¢ = se000 1. ‘Te inertia force acts opposite to the direction of acceleration, hence to the left as showm in Fig. Ag.15. ‘Tha unknown forces T and R can now be solved for by using the static equations of equilibrium. IF, = 42000 + T cos 10° = © hence, = 42700 Ib. 2, =12000 + R - 42700 x sin 19% = 0ANALYSIS AND DESIGN OF FLIGHT VENICLE STRUCTURES ences R = 19420 1b. To find the distance (a) take moments about c.g. of airplane, Bigg, = 1820 x 24 - 42700 4 hence, a= 10.9 tn Landing velocity Vs = 60 M.P.H. = 88 ft/sec. v - Vo" = zas Subt: - a8? = 2(-112.7) 8 hence stopping distance s = 34.4 ct, Example Problem 2 ‘An airplane equipped with float 1s cata- pulted into the air trom a Navy Cruiser as t1- lustrated in Fig, 44.16. The catapulting force P gives the airplane a constant horizontal acceleration of 34(96.6 ft/sec"), The gross weignt of airplane 9000 1b, and the catapult wack ts 35 ft. long. Find the catapulting force P and the reactions Ri and Rs from the catapult car. ‘The engine thrust {1s S00 1b. nas is airplane velocity at end of track run? At henca, Ra = 26800 tb. (up) BB, = 29600 ~ $000 + Ri = 0 hence, a, = = 20800 1b. (acting down) ‘The velocity at ond of catapult track can be found from the following equation Woo ve" = as Woeo= 2x 6.6 x35 ‘V = 62 ft/sec, = 56 M.P.H. ‘ample Problem 3 Asie that the transport airplane as 11~ lustrated in Pig. 44.17 has Just touched down in landing and that a braking force of 35000 Ib. on the rear wieels ts boing applied to bring the airplane to rest. The landing horizontal velocity is 85 MAPLE. (125 ft/sec). Neglecting air forces on the airplane and assuming the propeller forces are zero, what are the ground reactions Ry and Ry. wnat 1s the landing run distance with ‘the constant braking forcer W = 100, 000 1b, Solution: ~ ‘The forces Will be determined just after ‘the beginning of the catapult run, where the car velocity 1s smll, and thus the litt on the airplane wing end the airplane drag can be neglected. Horizontal inertia terce acting toward the atr~ plane tail equals, a = (222) 3,05 = 27000 ab. From statics: - BF, = -900 - P + 27000 = 0, henea P = 26100 Ib. To find A, take moments about point A, BY, = 9000 x $6 + 27000 x 78 - G00 x 83 - ER, Fig, AIT ‘The airplane ts being decelerated hortzon— ‘tally hence the inertia force through the air~ plane c.g. acts toward the front of the afrplane. Since the braking force 1s given we can solve for the deceleration factor by the equiliorim equation, EP, = $5000 - May = 0 hence, tm. = 35000 “ a, = 38000 hence a, + (OO) s2,2 = 11.27 ft/sec To tina landing rum (2), ve o- Bs es2as = 2 (-11.27) hence, 5 = 595 ft,M10 To find Re take moments about point (A) BH, = 100,000 x 21 - 35000 x 9 + 38 Re = 0 Ra = 47000 1b. (2 wheets) IF, = 47000 ~ 100,000 + A: = 0 Ra = $8000 1b. Syample Problem 4 The airplane in Fig. Ad,18 watshs 14,000 1b. It ts flying hortzontally at a velocity of 500 M.P.R. (82 ft/sec) when the pilot pulls tt upward into a curved path with @ radius of curva ture of 2500 ft. Agsime the ongine thrust and airplane drag equal, opposite and colinear with each other (not show on Fig. 44.16). Find: = @ () {ce} Acceleration of airplane in 2 di= rection Wing Litt (L) and Tati (7) forces Airplane Load factor. Engine Thrust te Fig. A418 Solution: ~ ve accoteraticn a, == BE or 214,5/82.2 = 8.575 (uprard). Tie inertia force normal fo the flight path and acting dom equals = 214.5 tt/sec* ‘a, (79922) 6.675 = 3700 1b. Placing this force on the airplane through the c.g, promotes static equiliorim, hence to find tail load T takes amsnts about wing aerocy- namic center (¢.p.) m5 hance = = (14000 + $3700) 8+ glo T= 0 ‘T= 4100 1b. (down) To find Wing Lift (1) use EF, = ~ 4100 - 14000 ~ $700 +L = 0 L = 111800 Ib. GENERAL LOADS ON AIRCRAFT Aizplane Lose Factor = AitBlene Lite = Wls00 = 4100 T4006 ’ ‘Drample Problem § ‘Assume the airplane as used in example problem 4 {s in the same attitude as used in that example problem. Now the airplane {s further manguvered vy the pilot suddenly push- ing the control stick forward so as to atve the airplane 2 pitching acceleration of 4 rad/seo” (a) Pind the mertia forces and the tatl load T, assuming the lift force on the wing does not change. (>) Find the forces on the Jet engine waten welgns 1500 Ib, and wnose c.g. Location 1s shom in Fig. 44.19. Assume moment of inertia I, (pitching) of the strplane squais 300,000 1b. sec*, tn. 99700 tp. Fig. AS 19 Solution: ~ Pig. A4.19 shows a trea bedy of che aLr~ plane with the lift and inertia forces as found in Problen 4. ‘The additional inertia force due te the angular acceleration a = 4 rad/sec™, equals, Ia * 300000 x 4 = 1,200,000 im. 1b. watch acts clockwise or counter to the di of angular acceleration. The airplane ts now in static equilibriur and to find the tatl load T take moents about airplane c.g. yet ion, at, ts g, = 19200,000 - 112600 x @ ~ 216 7 = 0 T = 1400 1D, To find Ma, take, BF, = 111600 - 14000 + 1490 - Ma, 0 Ma, = 99200 1b.ANALYSIS AnD besten of nonce, B00) 5 oon g select ape (GB) enna resect. The c.g. of the engine ts 50 inches aft of airplane c.g. as shom in Fig. Ad.19. The foree on the engine will be {ts om weight of 1500 1b., and the inertia forces due to a, and a the Tmertia force due to a, equals, m= (28 2) 701g = 30880 tb. Inertia force due to angular acceleration @ equals, 2 tO . Mra = gO x 50 x 4 = 778 1b. (down) Thea the resultant force on the engine equals 1800 + 10630 + 778 = 12908 1b. (down) Note if the engine had been forward of the air plane ¢.g., the Inertia force of 778 1b. would act upard instead of downward. In caloulating the inertia forces on a certain airplane item due to angular acceler= ation, the equation F = Mra assumes that the particuler Ltem nad negligible mass moment of inertia about {ts own centroidal ¥ axis. In ‘the case of a large iten this centroidal mass moment of inertia aay be appreciable and should de included in the Z, of airplane. ‘Then to find the inertia force for such an itea the equation F = Mra should be modified to be Fe a)/r where Teg. T= distance or arm from airplane c.g, to Cg of item. = mass moment of inertia of item about airplane c.g. equals I, + M* where I, 18 aass mement of inertia of Item sbout its own centrofdal ¥ axis. inertia force in lbs, normal to radius r. Buample Problem 6 SER Ey trae shove a large tranepore atrplane whose gross waight 1s 100,000 1b. The Airplane pitching mss moment of tnercia 1, = 40,000,000 1b. seo. in. ‘The airplane 1s making a level landing with nose wneel slightly oft ground. The re~ action on the rear wheels 1s 319,000 1b. in- elined at such an angle to give 4 drag com ponent of 100,000 1d. and a verticel component of 800,000 1. log. FLIGHT VEHICLE STRUCTURES Ag Pine: {a} The inertia forces on the air~ plane. (>) The resultant Load on the pilot whose weight 1s 180 1b. and wncse Location ts shown in Fig. Ad.20. Soiution: - ‘The wing Lift will be neglected tn this example probien. ‘Ths inertia forces on the airplane are forces Ma, and Ma, and the couple I.) a. To tind Ma, take, BF, = 100,000 - Ma, = 0 Ya, = 100,000 1b. 40,000 , (209,000 ieoteas) § 7 38> To find Ma, take, BF, = 300,000 - 100,000 - Ya, = 0 Ma, = 200,000 1b, hence 200,000 , (200,000), 2,7 Ps (Beideg es ae + , take moments To find the inertia couple Tyg. about airplane c.g., “ mM, tg.g. 7 7 100,000 x 120 = $00,000 x 64 Teng, yg = 977200000 2b. ton a = 322200,000 _ hence angular acceleration a = Sba+he5 = 0.98 rad/sec*.A432 Calculations of resultant load on pilot: - 2.08 a, 2 ag ef cues att hee ows tans be = reot Pe A Fig. A&.21 shows the airplane c.g. accelerations The forces on the pilot consist of the pilots wetght of 160 1b, and the various inertia forces as indicated in the figure. va, = 22) ag = 200" seg « Ce) 208 = seo The inertia force due to the angular ac~ celeration o acts normal to the radius arm detween the airplane c.g. and the pilot. For convenience this normal tores will be replaced by its s and x components, Ge = gH rae oss si aebbggs ore x 0.05 = 51 2. Total force tn x direction on pilot equals 280 = 17 = iss 2. Total force tn # direction = 360 + 180 - 181 = 379 1b. Hence Resultant force R, equals Vo + IF = 410" MA.13 Bifect of Airplane Not Being 2 Rigid Body. ‘The example problems of Art. A¢.12 assume thet the airplane is a rigid body (suffers no structural deformtion). On the tasie of this sssumption the applied loads on the airplane tn either flignt or landing conditions are placed tn equilforiwm with the inertia forces which occur due to the acceleration of the airplane. It is obvious that an airplane structure like any other structure {s not Tigid body, particularly a cantilever wing watch undergees rather large bending deflections in Doth flight and landing conditions. Figure AS.21 shows a composite photograph taken of a ‘test wing for the Sosing B47 airplane. The naximm uptard and coward deflections shown GENERAL LOADS ON AIRCRAFT are for design loads, which tn general are 1.5 ‘times the applied toads. It would not te correct to say thet the wing deflections under the a= plied loads for these two Hizn angle of attack Conditions would be 2/8 the deflections show tn the photograph since under the design Loads a considerable portion of the wing would de stressed fond the elastic limit of the matertal or {nto the plastic range where the stiftness modulus 1s Fig. A421 considerably less than the modulus of olesti- city, hence the deflections under the applied Loads would be somewnat less than 2/3 those shown in the photograpn. This photograph thus indtestes very strikingly that a ming struc is far from being a rigid bod: Static loads are loads which are gradually applied and cause no apprectable shock or Vi- bration of structure. On high speed aircraft, air gusts, flight maneuvers and landing actions are applied quite rapidly and thus can be classed as dynamic loads. Therefore wnen these dynamic loads strike a flexible (non rigid) airplane cantilever wing, a rather large Wing deflection ts produced ana the wing tends to vibrate. This vibration therefore causes additional accelerations of the mass wits of the wing which means additional inertia forces on the “ing. Furthermore if the time rate of application of the external applied forces approaches the naturel bending frequencies of the wing, the vibration exeited can produce large additional wing stresses,ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES Up until World War I practically all etr- planes wera assimed as rigid Dodies for struc~ tural design purposes. During the war failure of aircraft occured under load conditions watch the conventional design procedure based on rigid body analysis, in¢icated satisfactory or sate stresses. The fallures were no doubt due to dynamic overstress because the airplane {5 not a rigid body. Furthermore, atrplane design progress nas resulted in thin wings and relatively large wing spans, and in many cases these wings carry concentrated masses, suct 28, power plants, bombs, wing tip fuel tanks etc,, Thus the flexibility ofwings have increased which means the natural bending frequencies have decreased. ‘Tals fact together with the fact thet airplane speeds have greatly increased and thus cause air gust loads to be applied more rapidly, or the loading 1s becoming more dynamic in char— acter and thus the overall load effect on the Wing structure {s appreciabla and cannot be neglected {n the strength design of the wing. General Oynamte Effect of Air Forces on fing Loads ‘Me orttical airloads on an airplane are caused by minauvering the airplane by the pilot or in striking a transverse air gust. 4 transport airplane does not have to be designed tor Sharp maneuvers producing high airplane accel~ erations in its job of transporting passengers, thus the time of applying the maneuver loads 1s considerably more than a fighter type airplane pulling up sharply from nigh speeds. Fig. AG.22 shows the result of a pull-up maneuver on the Douglas D.C. airplane at 160 MAP.H. relative to load factor versus time of application of lead, as indicated the peak load of load factor 3,25 was obtained at the nd of ane second of tine, Fig. A422, t Sb ae 2 Pull-vp of DC-2 Airplane at 180 mph. ‘The author estimates the natural frequency of the 0.2.3 wing to be around 10 to 15 cycles per second, thus a loading tine of 1 second against 2 tine of 1/10 or 1/15 for half a wing deflec~ ‘tion cycle indicates that dynamic overstress should not be apprectable. in general, it can be said that dynamic over-stress under maneuvering loads on transport airplanes is not great as from other conditions such as air gusts or landing. Dynamic Setact ‘of air Gusts. ‘The higher the air gust velocity and the higher the sirplane velocity, the less the A833 for applying the load on the wing when striking the air gust. NAGA Technical Note 2424 reports the flight test results on a twin-engine Martin transport airplane. Strain gages were placed at various points on the wing structure, and strains were Tead, for various gust conditions for which the normal airplane accelerations were also recorded. ‘Then Slow pull-up maneuvers were run to give sim{lar airplane normal accelerations. The wing had @ natural frequency of 3.3 cps and the air~ plane spsed was 250 M.P.H. Two of the con clusions given in this report are: - (1) The bending strains per unit normal acceleration under air gusts were approximately 20 percent Aigner than those of slow pull-ups for all mea~ suring positicns and flight conditions of the tests, and (2) The dynamic component of the wing dending strains appeared to be due primarily to excitation of the fundamental wing bending node, ‘Taese results thus indicate that afr gusts apply a air load more rapidly to a wing than a maneuver load giving the same airplane normal acceleration for a comercial transport type of airplane, and thus the dynamic strain effect on ‘the wing’ {s more pronounced for gust conditions. Figs. 44.28, 24 and 25 show results of dynamic effect of air gusts on a large wing as de~ termined by Bisplingnoft*. ‘The results in these figures show that dynamic effects tend to con~ siderably tnerease wing forces on same portions of the wing and decrease 1t on other portions, Fig, A423, [Comparative shear “Disertbution ——Dymanic Analysis Rigid Airplane Analysis 10 Fig, Ad 24 Comparative Bending Moment Distribution ° 6 8 10 . ie 14.28 4 THT), comatice Bate THE 2 “iatte tore a CPE cua tt, Be 5 cexerai é =| Wing Span = 189 ¢t. 1 Gross Wt. = 184000 Ib. : | Selabe el ath ae o 2 4 6 8 10 Fraction of Semi-Span ‘Report on an lavestigation on Stresses in Aircraft Struc~ tures under Dynamle Loading. M.1.T. Publication.It has also deen found that landing Loads applied through the conventional landing gear or by Mater prassure on 2 flying dost are applied rapid encugn to be classed ¢s dynamic loads and such loads applied to wings of large span pro~ duce dynamte stresses which cannot ce neglected in the safe design of such structures. A.14 General Conclusions on Influence of Dynamic Loading on Structural Design of Airplane, The advent of the turbo-Jet snd the rocket type engines has opened up a Tange of possible airplane airspeeds hardly dreamed of only 4 few years ago, and already trans-sonic and super~ ‘Sonic speed airplanes are a coumcn development. From an aerodynamic standpoint such speeds nave dictated @ thin airfoil section which nas thus prouoted @ high density wing, Thus for atr— planes with appreciable wing spans Like Milt- tary Dombers and near future Jet comercial transports, which usually carry large concen- ‘trated masses on the wing such as angines, fuel tanks etc., the assumption that the airplena {s a rigid vedy ts not sufficiently accurate enough Decalise the dynamic stresses are appreciable. ‘The calculation of the dynamic loeding on ‘the wing requires that the mass and stiziness distribution of the wing structure be Imam. Since these factors are not known ‘han the structural design of a wing is started, the general procedure in design would be to first Daze the design on the assumption that the wing is a vigté body plus correction factors based on past design experience or available research information to approximately take cars of the tnfluence of the elastic wing on the airplane aerodynamic characteristics and the bulld up dynamic inertia forces. with the wing thus in itially designed by this procedure, {t then can de checked by a complete dynamic analysis and modifjed as the results dictate and then re calculated for the modified elastic xing. Tnts procedure is now practical because of the avail- ability of aigh speed computers. ‘4.18 PROBLEMS, (1). The airplane in Pig, a¢.26 1s being Jauiched trom the deck of an airerart carrier by the cable pull T watch gives the airplane e for~ ward acceleration of 3.26e. The gross weight of the airplane is 15,000'25. (a) Pind the tension load 7 in the launching! cable, and the wheel reactions Ry and Rae (>) If the flying speed ts 75 MAP.) what launching distance 18 required and the launching time t? (2). Aseume the airplane of Fig. 24.26 1s landing at 75 M.P.H. on a runway and brakes ars applied to the rear wheels 2qual to .4 of the vertical rear wheel reaction. What is the nort~ GENERAL LOADS ON AIRCRAFT. Ww = 15000 1b. A426 zontal deceleration and the stopping distance for the airplane? (2). the flying patrel boat tn Fig. a4.27 uakes @ water lending with the resultant bottes water pressure of 250,000 1b. as show tn the Tigure. Assume lift and taf} loads as shown. Toe pitching moment of inertia of the airplane is 10 afllion 1, sec,? in. Determine the atr~ plane pitching acceleration. wnat is the total load on the crew member who wetgns 200 1b, and is located tna seat at the rear end of the mull? Fig. A427 ro9y (4). The Jet-plane in Pig. A4.28 1s diving at @ speed of 600 M.P.H, when pilot starts 2 es pull-out. Wetgnt of airplane is 16,000 1b. Assume that engine thrust and total airplane drag are equal, opposite and colinear, (a] Find radius of flight path at start of pull-out. (b) Find inertia force in Z dirsetion. (c) Pind tft L and tafl load 7. Fig. A4.28. CHAPTER AS BEAMS - SHEAR AND MOMENTS AS.1 Introaucti In general, = structurs] aember that sup~ ports Loads perpendicular to 1ts longitudional axis 1s referred to as a beam. The structurs of aircraft provides excellent examples of beam Units, such as the wing and fuselage. Tery Seldom do bending forces act alone on a major airoragt structural unlt, but are accompanied by axial and torsional forces. However, the bend~ ing forces end the resulting beam stresses due to bending of the beam are usually of primary importance in the design of the beam structure. AS.2 Staulcally Determinate and Statleally Indeterminate Beams. A beam can be considered as subjected to imown applied loads and unknown supporting reactions, [f the distribution of the applied ‘mown loads to the supporting reactions can be determined trom the conditions of static equil~ torium alone, nemaly, the summation of forces and moments aqual zero, then the beam ts con- Sidered as 2 statically determinate beam. Hon ever, Lf the distribution of the known applied Loads to the supporting beam reactions is in~ fluenced by the behavior of the beam material during the loading, then the supporting reections| cannot be found by the statical squilforium equations alone, and the dean ts classified as a statically indeterminate beam. To solve such a been, other conditions of tact based on the dean deformations aust be used tn combination with the static equilibrium equations. 5.3 Shear and Bending Moment, A given beam {5 subjected to a certain ap~ plied inom loading, ‘The beam reactions to hold the beam in state equilibrium are then calculated by the necessary equations of static equix Librium, namely: ~ BY = 0, or the algeprate summation of all verti~ cal forces equal zero. TH = 0, or the algebraic sumation of all hert~ zontal frees equal zero. BM = 0, or the algebraic sumation of all the moments equal zero. With the entire beam tn static equilibrium, te follows that every portion of she beam aust Likewise be tn static equilibrium. Now consider ‘the Deam in Fig. A5.1. Tne knonn applied load of P= 100 1b, ts neld in equilitrium by the two reactions of 25 and 75 Ibs, as shown and are calculated trom simple statics. (Beam weight is neglected tn this problem). Now consider tne deam as cub at section ana and consider the Pe 100. 20" o" ¥ at 4 ot } 8 hs, t 100 = P bsp 04 at I "I Fig. AS.2 2 ig. as.3 am cry * 8 100 «1 tl Rg = hs Fis, A5.4 right side portion as a tree bedy tn equilibrium as shown in Fig. AS.2. For static equiltorium, 3Y, IH and IM must equal zero for all forces and motents acting on this beam portion. Consider- ing BV = 0 in Fig. 45.3: — W278 - 1005-25. ---~-- (1) ‘thus, under the forces show, the force system is unbalanced in the V direction, and therefore an internal resisting force 7; equal to 25 1D. must have existed on section aa to produce equilibrium of forces in the V direction. 48.3 shows the resisting shear fores, Vy = 25 1b, watch aust exist for equiliorium, Considering mM = 0 tn Fig. A5.3, take Roments about scme point 0 on section aa, Fig. Be = = 75 x 15 + 100 x5 = ~ $25 in.1b2) or an unbalanced moment of - &25 tends to ro~ tate the portion of tha besm about section aa. A counteracting resisting noment M = 625 must, exist on section a-a to provide equilttrim. ‘1g. A5.4 sows the free Dody with the Vy and My acting. Now ZH aust equal cero, The external reas as well as the internel resisting shear ¥, have no horizontal components. Therefore, the tnternal forces producing the resisting, mozent My must be such as to nave no horizontal AS.48.2 BEAMS -- SHE, unbalanced force, which means that the resisting moment My in the form of a couple, as shown in Fig. AS.S, or M, = Cd or Td and T must equal C to make BH = 0. ‘The tendency of the Loads and reactions acting on a beam to shear or move one portion of a deam up or down relative to the adjacent por— tion of the beam 1s called the Bxterm) Vertical Shear, or comonly referred to as the beam Vert= ical Shear and {s represented by the tern 7. From equation (I), the Vertical Shear at any section of a beam Can be defined as the al~ gebraie sum of all the forces and reactions acting tone Side of the sedtion at anion the Suess Wesivea. “Ty tis portion of the tear to the left of the section tends to aove up relative to the right portion, the sig of the Vertical Shear is taken as positive shear and Rogative if the tendency is opposite. Or in other words, if the algebrate sum of the forces 4s up on the left or down on the right side, then the Vertical Shear 13 positive, and nega= tive for down cn the left and up on the right, From equation (2), the Bending Nonent at any section of a beam can be defined as the al= gebraic sum of the mments of ell the forced acting to either side of the section about the Section, if this bending moment tends to pro= Giice compression (shortening) of the upper fibers and tension (stretching) of the lower tfbers! of the beam, the bending moment 13 classed as a positive bending monent, ang negative for the reverse condition. A9.4 Shear. and Moment Diagrams, In afreratt design, 2 large proportion of ‘the beams are tapered in depth and section, and also carry 4 variable distributed load. Thus, to design or check the various sectiond of such beams, it is necessary to have a complete pic~ ‘ture as to the value of tha vertical shear and bending moment at all sections along the beam. If these values aro plotted as ordinates from a base line, the resulting curves are roforred to as Shear and Monent diagrams. A few example Shear and Mozent diagrams will be plotted, to refresh the students imowlsdge regarding these diagrams. Beample Problen 1 ‘Draw a shear and bending moment diagram for ‘the beam shown in Fig, A5.8, Neglect the weignt of the dean. In general, the first step is to determine the reactions, To find Rg, take moments about point A. IM =~ 4 x 500 + 1000 x 5 + 500 x 15 - log = 0 hence Rg = 690 1D. TY = ~ 500 + Ry - 1000 ~ S00 + 680 = 0 hence Ry = 1110 Ib, AND MOMENTS gle 600 Fig, A5.6 calculations tor Shear Dia: - anes SUAPT GE the ISTE ond of the been. Considering a section just to the right of the 500 Ib, load, or section 1-1, and considering ‘the portion to the left of the section, the Vertical Shear at 1-1 = IV =~ 500 (negative, down on left.) +810, +810 tb +300 tb, 200 Ib. -s0015, S001, 880 = 80 tb. Fig. AS.7 (Shear Diagram) ~2000 53. Ib. pig. 45,9 (Bending Moment Diagram) Next, consider section 2-2, Just to lett of reaction Ra. ae 500, or same 26 at section 1-1. Next, consider section 3-3, Just to right of Rae BY = - 500 + 1110 left side of section). Next, consider section 44, Just to lett of 1000’ toad. BY = - 500 + 1110 section 3-3). Section 5-5, to right of 1000 lead: 2Y = ='800 + 11t0 ~ 1000 = - 380 (awn on left). - Check this shear at section 5-5 by using the portion of the beam to the right of S-5 ssa free boay. BY == 300 + 690 = S60, watch checks (sign of shear ts minus, because ZV ts up cn Tight). Section 6-8, use the portion to right asa free boty: BY = ~ 300 + 680 = 390 (ninus shear), Section 7-7: ca 620 (postive, up on 610 (sane a8 at 300 (positive shear, down on rignt)ANALYSIS AND DESIGN OF FUIGHT VEHICLE STRUCTURES Section 8-8: BV = ~ S00 (positive shear}. Pig. AS.7 shows the plotted values on diagram. Caloulation of the Moment Diagram, Start at section 1-1, and consider the forces to the left onl: B= ~ 600K 0=0 Since sections 2-2 and 3-3 are only a dtfferen- tial distance apart, assume a section Just above Ay and Consider the forces on the left side only: IM = - $00 x 4 =~ 2000 tn. 1D. (Negative moment, because of tensfon in the top fibers). Consider tke section under the 1000 in. 1b, Ioad¢| EM to left =~ $00 x 9 + 1110 x 5 = 1080 In, Ab. (positive monent, compressing the top fibers}. Chack by considering the forces to the rtgnt: WM right = 300 x 8 - 690 x $ = - 1050 in.2D,| Next, consider a section over Rg: mM right = 300 x 3 = 900 in, 1b, (Negative moment, tension tn top fibers). Ab Section 8-8: Bt rtent Fig. A5.8 shows From the above results {t cay be noticed ‘that hen the bending moment is obtained trom the forces that lie to the left of any section, ‘the bending moment 1s positive when {t ts clocke wise. If obtained from the forces to the right, ib 18 positive, when counter-clockwise. The student should sketch In the approximate shape of the deflected structure and determine the signs from whether tension or compression exists in the upper and lower fibers. le Provlen 2 BIGIELISIIStS UNA drow tho shear and zonent diagrams for the beam and loading as showm tn Fig. 5.96 Pinet, determine the reactions, Ry and Rg: ~ IM, + 36 x 10x 18 + 120 x 9 = SéRg =O. hence Ag = 210 1b. BY = - 120-88 x10 +210 +2, 20, ence Ra = 270 1b. 300 x0 = 0 the plotted values, Sheer Diegram: ~ “The verticai snear just to the right of the reaction at A 1s squal to 270 up, or_positiv ‘This t3 plotted es Line aD in fig. AS.10. The vertical shear at section C just to the left of the load and considering the forces to the lert of the section = 270 ~ 9 x 10 = 180 1b. up, oF positive. The vertical shear for any section between 4 and Cat a distance x trom A ts 45.3 Yq = 270 = 10K, and hence, the shear dew creases at aconstant rate of 10 1b./in, trom 270 at A to 160 at c. ‘The vertical shear at section D, Just to the right of load ts, Yp = Were * 270 - 10x 9 - 120 = 60 up, or positive. gies rf wai /in, ap ER dy a mateo Pie. 88.9 Moment Diagram The vertical shear between points D and 8, when x Is the distance of any section between D and B trom A: Yon = 270 - 120 - 10K ~ -- At point B, x = $5: hence ¥g = 270 = 120 - 10x36 =~ 210 1b., which checks the reaction Rg. Since the Vertical Shear decreases at a rate of 10 Ib/in, from D to B, it will be 6" tram D toa point where the shear is zero, since the shear at D ts 60 Ib. ‘Tnis point could also be Located by equa ting equation (1) to zero and solving for x as follows: 0 = 270 = 120 = tox, or x = 3SP 158 = 15" ran AL Ie the shear diagram has passed through zero under the concentrated lead, then the method of equating the shear equation to zero and solving for x couid not be used, thus in general, {t is best to draw a shear diagram to find waéa shear is cero. Fig. A5.10 showe the plotted shear diagram. Moment Diagram: - “SRE Section A just to the right of reaction Rg the Dending moment, considering the forces to the left, 1s zero, since the arm of Ry ts zero. ‘The bending moment at any Section between A and 0, at a distance x from the left reaction Ry, 135AS BEAMS -. ‘The equation for the bending moment between D and B (x greater than 3) 1s Me ERK =P (e9) HE 2-2 ----- Lox? 270 x ~ 120 (x-9) - HOE 3000 + 150 x = Gx" = = = = @ At section 0, x = 9", substitute in equation (4) Yq = 1080 + 150 x 9 - Sx 9 = 2025 InvIb. {posttive, compression tn top fibers). At the point of zero shear, x = 15", M = 1980 + 150 x 18 ~ § x 184 = 2205 ‘Thus, by substituting tn equatien (2) and (4) the mément dlagram as plotted in Fig. AS.2L is obtained, AS.S Section of Mazimum Bending Moment. ‘Tea general expression for the bending mo~ ment on the beam of example problem 2 is from equation (3): Ye Rx = P (zee) = Now, the value of x that will make maximum or ainimm is the value that will make the first derivative of My with respect to x equal to zero, or Het RA> Pome ~~ ++ +--+ --- 65) Therefore, the value of x that will make My @ maxim oF minimm may be foun from the equation| Ry -P- we 20 observation of this equation indicates ‘that the term Ra ~ P - wx is the shear for the Saction at 4 distance x from the left reaction. Therecore, where the shear 1s zero, the bending monent_is maxima. Thus, the Shear diagram watch Shows where the shear is zero is a con— Wentent nediim for locating the points of maximm dending moment, A5.6 Relation Between Shear and Bending Moment. Equation (5) can aiso be written t= y, since the right hand portion of equation (5) 1s equal to the shear. Hence, @t = vex (6) Which moans that the difference at petsen ‘the bending aonents at two sections that are a Gistance dx apart, 15 equal to the area Véx under the shear curve between the two sections, Thus, for two sections x, and Xay SHEAR_AND MOMENTS Me Xe vax Mh x Tus, the sree of the sheer ciesran benmeen an faa foints equals che shange in Sanding aeReaT Sater these oe ese the shear Glagram in example problem 2 (Fig. ‘A5.10). The change in bending moment cetween tho leis Fonction’, ane tho ieee ts equal os the area or tho sheds diagan totnoen these two Sots. os HO 520 5 3 5 aces in.tb, since the tending szont at the left support 18 zero, this change cerofors cquala tho Hus aazest af 2 seccien Enter te 10d 2. Adding to Cite te aren of the angie" totneen poise Basa the solne snear, or 2x 6 = 160, we obtain 2205 1 small tri- of zero be ac the maxim aozent, This can be checked Sy ‘taking the area of the shear dlasran Setmean ‘the point of zero shear and point 3 = 239 x 21 = 2208 tn. 1b, Example Problen 3. Gg Tliustrates 2 landing sear oleo strut ADO braced by etruts 3) and Cz. A lené= ing ground load of 16000 ib. ts applied through ‘the wheel axle 2s shown, Let it be required to find the axial load in 412 members and the shear and bending manent diacram for the olso strut. Yop 18.9 bsTH vy ie eae ‘- Ea 10" i 1" 5 Fig. AS.12 . Ping at Pots B, Erbe woverda | [yg 6000 0, | Resistance at Point * | SOLUTION: - To tind Vg take moments about point 5, By = = 15000 ¥ 0.5 x 42 + 15000 x 0.866 x'6.77 © B0.77 ve = 0. hence, ¥, = 12880 1b, ‘The axial load in member CE therefore equale,ANALYSIS AND DESIGN OF FLIGHT VEHICLE sTRUCTURES 1180/08 30° oF 19380 1d. J, Hy = 118580x15/26 = 6660 To find Hy take monants about point 0, ‘Mp = 11650 x 15 - 6660 x 10 ~ 1s000 x 0.5 x 22 + lous = 0. hence, Hq = 18340 1b. To find Vg take IFy = 0, BF = 15000 cos 30* + 11850 ~ ¥3 = 0. hence, Vg = 24550 1b. ‘The axial load in member 8D therefore equals 24550/eos 30° = 26360 1b. (compression]. The reaction Hp therefore squals 28360 x sin 30 = 14180 tb,” Fig. AS.1S shows the olec strut as a tree body with the reactions at A, D and Sas calcu~ lated. Fig. A5.1$ also shoxs the axial load, vertical shear and bending moment diagrams. ‘The bending moments due to epplied loads without regard to bending’deformation of the beam are usually referred to as the primary bending moments, If a aember carries axial loads additional bending moments will be produced due to the axial loads times the lateral deflection of the beam, and these bei Ling Z0— ments are usually referred to as secondary bend~ ing noments. (Arts. A23-30 covers ‘he caloula~ tion of secondary momenta). Fig. AB. SOLUTION: ~ Calculations of reactions at A and 8: — To find Vg take moments about point A, Bh = > 500 x 7 ~ 500 x 6 + 2000 x 20 + 1000 x Sih 45° x 10 + 2000 cos 45° x 2-22 Vg = 0. hence, Yp = 989.3 1b, (up). To find Yq take BY = 0, DY = 999.3 ~ 1000 ~ 1000 sin 45% - 500 + ¥, #0. hence, Vy = 1207.8 1b. (up). Jo find Hp take MH = 0, Bi = = 500 + 1000 cos 45° ~'Hy = 0, hence Hy = 207.1. With the exception of the 1000 1b, load at 45°, all loads are applied te brackets which in ‘um are fastened to che beam. Therefore the Rext step {8 to find the reaction of the loaded brackets at the beam centerline support points. The load at E and the reaction Hy at 8 will be alse referred to beam centerline. Fig. 45.15 (@,b,c,d) show the cantilever brackets as free bodies. ‘The reactions at the base of these cantilevers will be determined. These reactions reversed will then be the applied Loads to the beam at points C, D, F and , sso se \ston (9 sap 8 — rt a 7500 245s 14180 130001, oy Compression 139 sear Diagram Se Oe SLE 7 Soe LLY YY is. Fig. AS. 13 #20000 =199440 in Ib, Beamplo Problem 4. Fig. AD.i# Shows 2 beam loaded with both transverse and longitudional loads, This beam loading 1s typical of interfer beams in the air plane fuselage which support all kinds of fixed equipment. The reactions for the beam are at points A and 8. Required: - Shear and bending moment diagrens. Fig. A513 For bracket at C, to find Ho take TH = 0, or obviously Hy = $00 1b, In like manner use BY = 0 to find Vg = 500 Ib, To find My take moments about point C. Me = - S00 x 2 + 500 x 8 ~ Me = 0, hence Me = 3000 fn. 1b, The student should check the reactions at the base of the cantilever brackets at D and F (See Fig. b,c).45.8 BEAMS -- SHEAR AND MOMENTS ‘The load of 1000 at 45¢ and spplied at point BY will be referred t> potnt £ the centerline of ‘Dean. Fig. 4 shons the reaction at E due to the load’at 2". The reaction at B should also be referred to the beam centerline. Fis. A8.16 shows the beam with the applied leads at points C Og’ Panda’. Figs. 45.17, 18 and 19 show She axtel loed, vertical shear and dending noment diagrans under the beam loading of Pig. 45.16. 300 ona ‘4000 a0. Fig. A5.16 sim m. 2m. pxal ine Da nese, [BR wha ‘ ar ster ia vig, 5.18 [ -099.3 207.1 Pig. A5.19 ‘The shear diagram is determined tn the sane manner as explained before, The applied external couples do not enter into the vertical shear calculations. The bending moment diagram can be calculated by taking the algebraic sw of all couples and moments of all forces lying to the one side of a particular section. If it is de sired tq use the area of the shear diagram to obtain the bending aments, {t is necessary to add the couple moments to the sear areas to ob~ sain the true bending moment, For example, the Dending moment just to the lett of point 2 will be equal in magnitude to the area of the shear diagram between C and B plus the sum of all applied couple moments between ¢ and 5 but not ine cluding that at £. To illustrate the calculations are: ~ (- S00 x 5) + (707.8 x 10) = 4876 in. 1b, (trom area of shear diagram). {3000 = 4000) = - 1000 in, 1D. (trom sim of couple moments). ‘Thus bending ment at & 3578 tn, Ib. left The Dending moment at Epyght Wil equal that at Bert plus the couple moment at 2 or 3578 + 707 = 4285 in. ib. ‘The student should rsalize that when couple onents are applied to a beam it is possible to = 4578 ~ 1000 = Bending Moment Dia. have maximum peak goments without the Vertical Shear passing through zero. To tllustr: this fact, consider the beam of Fig. 45.20, namely, & Simple supported beam with an exe ternally applied couple aoment of 10 in. 1 magnitude at point ¢ the center point of the beam. The shear and bending mouent dtagrans are as indicated and 4 maximm bending soment occurs at C but the shear dtazram does not pass through zero. Moro" ‘Rat ae Re 19 Shear Dia. Bending Fig. 5. ‘Moment Dia. 8 A couple 1s two equal and opposite forces not in the sane straight line. Lat it de as- ‘sumed that the 10 in. 1d. couple is made up of forces equal to 100 1b, each and an arm beoween ‘hen of 0,1 tnch as illustrated in Fig. 8.21. 400, 100 T 4a. i ~y Shear Dia. Fig. A5.21 9 The shear diagram ts as shown in Ftg. A5.21 and Tow passes through zero under each of the couple forces. Thus ir we assume the couple aoment has @ dx arm the shear to the right of © is one 1b. and then changes to some unknown negative value and then back to one 1b. positive as the distance dx 1s covered tn going to the left. Thus ‘the shear goes to zero twice in the region of point C. AS.T Moment Diagrams as Made up of Parts. Tn calculating the deflection of statically deterainate deans (See Chapter a7) and solving statically indeterninate structures (See Chapter AS), the area under the dending aoment curve is, Tequired, thus it is often convenient to treat each load and reaction es a separate acting force and draw the moment diagram for each force. ‘The true bending aoment at a particular point will then equal the algebraic sumation of the ordinates of all the various moment curves at this particular point or adding the various separate monsnt diagrams wil give the true dending moment diagram. Figs. A5.22 and AS.23 illustrate the drawing of the bending moment diagran in parts. In these examples, we start from the left end and procead to the Tent end and draw the moment curve for each force as though the bean was a cantilever wits the fixedANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES wp 0 eh Petoe oe Len! | { we10 Ib/in. et Re Rys100b. Ryet00 Reg Ratd pus 2Ba——“Fs00 120 2 ove wo, 620 D8 0 Thon z puete Pe 500 DHE ey 300 a ——— 5 Ts Final Moment Dia, Final Moment Curve Fig. a8.22 Fig. A529 support at the right end. The final bending moment curve for the true given beax then equals ‘the sim of these separate diagrams as {Llustra— ted in the figures, STATIC MOMENE CURVES IN SOLVING STATICALL: INDETERMINATE STRUCTURES The usual procedure in solving a statically Indeterminate structure is to first make the structure statically determinate by removing the necessary redundant or unknown reactions and ‘then calculating the deflection of this assumed Statically determinate structure as one step in the overall solution of the problen (See Chapter AS). In the solution of such structures tt ts Likewise convenient to treat the bending aonent diagram as made up of parts. To illustrate, Fig. AS.24 shows a loaded rectangular frame fixed at points A and 8. The reactions at both pee Pa Pastoe a) be press, Fe - 3 io 89-50 Py =10# 4 t = é lio a” 3 ag 4 3 2\ 3 a tg Fa AS.24 |, 5 0g Fide 60-30 39 a0 Fig. A525 ig as.26 r 10a points A and B are unknown in magnitude, direction and location, or each reaction hes 3 un- monn elenents or a total of § unimowns for the two reactions. With § static equilibria equations avatlebie, che structure is statically indeterainate to the third degree. Fig. 45.25 ‘llustrates one manner in watch the structure can be gade statically determinate, oy freeing the ond 4 to geke 2 bent cantilever beam fixed As. at 8 and thus leaving only 3 uninowm elements Of the reaction at 3. Fig. A5.25 shows the bending monent curves for each load acting separately on this cantilever frame. Pig, A5.26 shows the true bending coment as the sumation of the various moment curves of Fiz. AS.25. As another solution of this fixed ended frame, one could assume the statically deter~ minate modification as a crane pinned at A and pinned with rollers at B as {llustrated in Pig. 45.27, This ascimed stricturs is statically deterainate because there Peale are only S unknown elements, namely the magnitude and direction of the reaction at A and the magnitude of the re- Palo action at 8. For conven- tence the reaction at a 1s la ls Pin, 20, resolved tnto to magnitudes: Ha as Hand V components, The n10 “Fgea0 reactions Vs, i, and Vg can a ‘then be round by statics ana Fig. 5.27 the results are shown on Fig. A5.27. Fig. 45.28 shows the bending moment diagram on this frane due to ach Load or reaction acting separately, starting at A and going clockwise to B. Fig. 45.29 shows the true bending moment diagram as the summation of the separate diagrams. 5 -50—Due to P, OS 20 pe 8 Bi -100 sole 0 -60 240 4 240 -100 -60 -50 i? Z als a 2 2l/ 2 | eh gis g si/a |g alle a a 4 76 Fig. AS. 28 + 38 al Sse Tit oeig ce ve (Tension on inside Gon des te An 8.8 Forces at a Section in Terms of Forces at a Previous Station. STRAIGHT BEANS ‘ireraft structures present many beans wnich carry a varying distributed load, Mini- jum structural weight 1s of paramount importance 4n afroratt structural design thus it 1s dew sirable to have the complete bending moment Giagraa for the structure so that each portion of the structure can be proportioned sffic~ tently. To decrease the amount of numerical work saquired tn obtaining the complete shear3.8 and bending moment dlagrans it usually saves time to exress the shear and moment at a given station {m terms of the shear and manent at a previous staticn plus the affect of any loads lying between these two stations. To {llustrate, Fig, AS.80 shows 4 cantilever beam carrying a considerable number of transverse loads F of dif-| ferent magnitudes. Fig. 45.31 shows a cree body be Pitsere t Bia, cehy m0) idm. ae oe Be as.20 ig abo of the beam portion vetween stations 1 and 2. The Vertical Shear V, at station 1 equals the sumation of the forces to the left of station 1 and Mf, the bending moment at station 1 equals the algebraic su of the moments of all forces lying to left of station 1 about station 1. Mow considering station 2: ~ The Vertical, Shear Va = Vi + Fine, oF stated in words, the Shear V, equals the Shear at the previous sta~ tion 1 plus the algebraic sum of all forces F lying detween stations 1 and 2, Agein constder- ing Fig. AS.G1, the bending nonent M, at station 2 can be aritten, My = th * Vid + Fines OF stated in words, the bending moment Ny at sta- ‘ion 2 1s equal to the bending moment My at a previous station 1, plus the Shear Y at the previous station.1 times the arm d, the distance between stations 1 and 2 plus the moments of all forces lying between stations 1 and 2 about station 2. 45.9 Equations for Curved Beams. Many structural beams carry both Longitud~ tonal and transverse loads and also the beams may be made of straight elenents to form 2 frane or all beam elements my be curved to form a curved frame or ring. For axanple the airplane fuselage ring is 2 curved beam subjected to foress of varying mgnitude and direction along its boundary due to the action of he fuselage skin forces cn the frame, Since she complete bending moment diagram 15 usually desirable, it 1s desirable to miniaize the amount of numerical work in obtaining the complete shear and bending monent values. Fig. A5.S2 shows a curved beam loaded with a mumber of different vertical leads F and horizontal loads Q. Fig. AS.53 shows the beam portion 1-2 cut out'as a treo body. Hy represents the resultant horizontal foree at station 1 and equils the algebraic swmation of all the Q forces to the left of station 1, V, represents the resultant vertical force at station 1 and equals the sum of all F forces to left of station 1, and M equals the bending moment about point (1) on station 1 due to the moments of 211 forces lying to the left of point 1, BEAMS -- SHEAR AND MOMENTS ‘Then from Fig. 46,33 we cen write for the recultant forces and moment at point (2) at station 2: Va = Va + Pace acre =a My EM + Vad = tha + Fy Having the resultant forces and moments for 2 given point on a given station, t ts usually necessary {n finding beam stresses to resolve ‘the forces into components normal and paralle? to the beam cross-section and also transfer their location to 2 point on the neutral axis of the beam cross-section. For example Fig. 45.4 shave the resultant Fig. AS. 38 ve s MosMy Fig. AS.36 Fig. 45.24 forces and moment st point 1 of a dean cross section. They can be resolved into 2 normal force Nand @ shear for $ plus 4 moment M, as shown in Fig. 45.35 where, N2zcosa+vsing $= Vos a-Hsing later on when the beam section 1s being de~ Signed it may be found that the neutral axis lies at point 0 instead of point 1. Fig. 45.36 shows the forces and moments referred to point 0, with M, being equal to M, ~ Ne.ANALYSIS AND DESIGN OF FLIGRT VEHICLE STRUCTURES ‘45.10 Torsional Moments, ‘The Loads which cause only bending of a boam are located so that thetr Line of actton passes through the flexural axts of the beam. Quite often, the loading on a beam does not act through the flexural axis of the beam and thus the beam undergoes both bending and twisting. The mouents Waicn cause the twisting action are usually referred to as torsional moments. The airplane wing is an excellent example of a beax structure that is subjected to combined bending and torsion. Since the center of pressure of the airfoil forces changes with angle of attack, and since there are many flight conditions it 1s impossible to eliminate torsional moments under all conditions of fight and landing. For the fuselage, the Vertical tall surfaces 1s nora~ ally located above the fuselage and thus a load on this tail unit causes combined bending and twisting of the fuselage. Fig. A5.34 illustrates a cantilever tube being subjected to 2 load P acting at point A on a fitting attached to the tube end. ‘The flex- 2 Pr P Fig. A8.35 Fig. A694 ural axis coincides with the tube centerline, or axis 1-1. Pig. A5.5 shows the load P being moved to the point (0) on the tube axis 1-1, honever the original force P had a moment about 40) equal to Pr, thus the goment Pr must be added to the load P acting at (0) if the force system at point (0) 1s to be equivalent to the original fores P at point A. The force P acting ‘through (0) causes bending without twist and the moment Pr causes twisting only. For the resolution of moments into various resultant planes of action, the student should refer to any textbook on statics. 5.11 Shears and Moments on Wing. Arts. Ad,S and A4.6 of Chapter A¥ discusses the cirloade on the wing and the oquilforiin of the airplane as a nnole in flight. As explain ed, it is customary to replace the distributed a1 forces on an airfoil by two resultant torees, naxely, 1ift and crag forces acting ‘arough the aeredynanic center of the airfoil plus 4 wing aoment. The airflow around 4 wing is not witform in the spanwise direction, snus the airfotl force coefficients CL, Cp and Cx vary spanwiee along the wing. Fig. 45.86 sions a typical spanwise variation of tha Cand Cp roree coefficients in terms of untférm spali~ wise variation G and Op. any particular type of atrplane ts aestened to carry Gut a certain Job or duty and to 40 thet job requires a certain mextmm airplane Tig. AS30 SPANWISE DISTRIBUTION LIFT AND DAG COEFFICIENTS tn terme of worm eetriatica) velocity with the maneuvering Lintted to certain maximm accelerations. Thess limiting accelerations are usually spectfied with reference to the X ¥ 2 axes of the airplane. Since the directions of the litt and drag forces change with angle of attack it 1s simpler and conventent in stross analysis to resolve all forces with rer eroneo to the X YZ axes which remain fixed in direction relative to the airplane. As a time saving element in wing stress analysis, {t ts customary to make unit load an— alysis for wing shears and moments. The wing shears and moments for any design condition then follows as a matter of stmple proportion and addition. For example {t ts customary: ~ (2) To assume a total arbitrary unit load act~ ing on the ing in the Z direction through the serodynamic section of the airfotl section and distributed spanwise according to that of the Cy or litt coefficient. (2) A simtlar total load as in (1) but acting tn the X direction. (3) To assume a total unit wing load acting in the 2 direction through the aerodynamic center and distributed spanwise according to that of the Cp or drag coefficient. (4) Same as (3) but acting in the X direction, (5) To assume a unit total wing moment and distributed spanwise according to that of the Cyy_, OF moment coefficient. ‘The above unit load condtttons are tor con~ ditions of acceleration tn translation of theAst0 airplane as 2 rigid body, Unit load analyses are elso made for angular accelerations of the airplane which can also occur in flight an¢ landing maneuvers. ‘The sudject of the calculation of loads on the airplane is far too large to cover in 2 structures book. This subject {s usually covered in a separate course in most aeronautical curricula after a student has had initial, courses in aerodynamics and structures. To 11- lustrate the type of problem that is encountered in the calculation of the applied loads on the airplane, simplified problems concerning the wing and fuselage Will be given. AS.12 Example Problem of Calculating Wing Shears and ‘Moments for One Unit Load Condition. Fig. AS.37 shows the half wing planform of a cantilever wing. Fig. 45.38 shows a wing section at station 0. ‘The reference Y axis nas been taken es the 40 yercent chord Line watch, happens to be a straight line in this particular wing layout. agseg 1 | | Area = 17760 5q.10. 22 22222888 100 40% of chord line is straight Root Gecilon 240" Fig. 5.38 ‘The total wing area 1s 17760 34. in. For convenience a total unit distributed load of 17760 lbs. wili de assimed acting on the halt wing and acting upward in the 2 direction and through the 2irtofl aerodymante center. The spanwise distriputton of this load will be according to the (C,) Lift coefficient spanwise distribution. For simplicity tn this example At WLLL be assumed constant. Table A5.1 shows the calculations in table form for determining tha (V,) the wing shear in the Z direction, the bending aoment Me or no~ went about the X axis and My the momemt about ‘the ¥ axis for 2 mmber of Stations between the wing tip station 240 and the centerline station o Colum 1 of the table shows the mmber of stations selected. Colum 2 shons the Cig. BEAMS -- SHEAR AND MOMENTS ratio or the spanwise variation of the 1itt coedficient ¢, ia terms of a unizorm distribu tion G,. In this example we have taken this ratio Hs unity since wa nave no wind tunnel or aerodynamic calculations for this wing relative to the spanise distribution of the 1ift force coefficient. [nan actual protien involving an atrplane a curve such as that given tn Fig. 48.36 would be avatlable and the values to place tn Column S of Table 48.1 would De read from such a curve. Column (2} gives the wing chord leagth at each station, Column (4) gives the wing running ioad per inch of span at each station point, Since a total untt load of 17760 1b. vas assumed acting on the half wing and since the wing area {s 17760 sq. tn., the runing load per inch at any station equals the wing chord length at that station, In order to find shears and aoments at the various station points, the distributed load ts now broken down into concentrated loads which are oqual to the distributed iced on a strip and this concentrated strip load 1s taken as acting through the center of gravity of thts distributed strip load, Colums 5, 6, and 7 show the calculations for determining the (arg) strip loads. Colum shows the lo= cation of the Pz load which is at the centrota of a tropizctdal distributed loed whose end values are given in Colum (4). In determin tng these centroid iocations it 1s conventent to use Table AS.4 of Chapter AS. ‘The values of the shear Vz and the no ‘nent My at each station are calculated by the aethod explained in Art. 48,8, Colimms 9, 10, Ui and 12 of Table AS.1 give the caleulations. For example, the value of Me = 9864 in Col- um (12) for station 220 aquais 2436, the My moment at the previous station in Coluan (12) plus 4906 in Column (10) which {s the shear at the previous station (230) tines the cist~ ance 10 inches plus the aoment 2540 in Column (8) due to the strip load between stations 280 and 220, which gives 2 total of $684 the value tn Colum (12). ‘The strip loads ’4P; act through the aeredynantc center (a.c.) of each airfoil strip. Column (13) ene (14) give the x amas whieh 15 the distance trom the 2.c, to the reference Y axis. (See Fig. 48.28). Colum 15 gives the My nowent for each Strip load and Colum 16 the My moment at the various stations wnich equals the sumation of the strip moments as one progresses from station 240 to zero. Fig, 45.38 shone the results at station (0} as taken’ tron Table 45.1. ‘| VariT760 1875630 Crpava Pig. 45.29ANALYSIS AND DESIGN OF FLIGHT VEHICLE sTRUCTURES asin TABLE AB CALCULATION OF WING SHEAR V, AND WING MOMENTS Me AND My DUE TO TOTAL UNIT DISTRIBUTED KALF WING LoaD oF 1fté0 Los, ACTING UPWARD IN 2 DIRECTION AND APPLIED AT AERODYNAMIC CENTERS OF WING SECTIONS, (See Figs. A537, 30 for Wing Layout) “Sattone ny 4 Arm wo comvold Salton + y= Distance Fro Wing ot Sortion fy + Distance swe Bip Lad A ¥s to Cal 12 previous s Col. (cot. 7) C21. 6) Gal Jo | ana le | Ms Ts Ere Tae r nae Tae | aaa OE aaa i TF z Toe Tae eee Taste eee i Tesco [See | TE TET Taso aanage ag [aE Fe aa: 170 Cheens Toul Limit Load Aseumed oo Hail linen the time comes to design the structura: make-up of a cross-section to withstand these applied shears and moments, the structural de~ sigmer my wish to refer the forces to another Yaxis ag for example one that passes through the shear center of the given section. This trenster of a force system with reference to another set of axes presents no difficulty. SHBARS AND MOMENTS CN AIRPLANE B00Y AS.13 Introduction, ‘The body of an atrplane acts essentially as & beam and 1h some conditions of flight or Tana~ ing as 2 beam column witch may be also subjected to twisting or torstonal forces. Thus to design an airplane beay requires a complete picture of The shearing, bending, twisting and axtal forces which may 3e encountered in flignt or landing. Tn the load analysis for wings, the direct air forces are the aajor forces. For the body load analysis the direct air pressures ara sacondary, the aajor forces being of a concentrated nature in the form of Loads or reactions from units attached to the body, as the power plant, wing, landing gear, tail, ote. In addition, since the vody usualiy'serves a3 the load carrying medium, important forces are produced on the body tn resisting the inertia forces of the welznt of the intertor equipment, installations, pay load etc, as tn the case of the wing, a large part of the load analysis can be made without auch consideration as to the structural analysis of the bedy. The load analysis of an airplane body tnvolves a large anount of calculation, and treatment in this chapter must be of a and 1s presented chiefly for purpose of shawing the student in general how the problem of load analysis for an air= plane body 18 approached.ASD AS.14 Design Conditions and Desiga Weights. The airplane body aust be designed to withstand ali leads trom specified flight conditions for both maneuver and gust conditions. Since accelerations due to alr gusts vary tnversely ag the airplane welght, 1t customary to analyze or cheek the dody for a itznt load condition ror flight conaitions. ia general, the dasig weignts are specified by the government agencies. For landing conditions, however, the normal gross weight 15 used since it would be more critical than a lightly loaded condition. ‘The general design condttions whien are usually Investigated in the design of the body are ag follows: Flight Condttions: HeAsAs (Hitgn angle of attack) LAL (Low angle of attack) Tibia. (Inverted low angle of attack) TIHAlAl (Inverted high angle of attack! ‘Tha above conditions generally assume only translational acceleration. In addition, ‘t ts sometimes specified that the forces due to 2 certain angular acceleration of the airplane about the airplane c.g. aust De considered. ‘The body ts usually required to witastand spectal tail loads ooth symmetrical and unsya— metrical which may be produced by air gusts, engine forces, ste, Also, the body should te checked for forces que to unsyumatrical air loads on the wing. landing Conditions: Im general, the body 1s investigated for the following landing conditions. The detafled ree quirements for each condition are given in the goverment specifications for both military and comercial airplanes. Landplanes: Level landing, Lavel landing with side load. Three point landing. ‘Three point landing with ‘groumd loop. Nose over or turn over condition. arresting. (Usually for only Navy Carrier based alr- planes}. Seaplanes or Beats: ep landing with and without angular acceleration. Bow landing. Stern landing. ‘Dwo wave landing. Beeching conditions, Catapulting conditions (Navy airplanes]. Spectal Conditions or Forces: SEN NS ToWinE OF aitplane. Body supercharging. BEAMS -- SHEAR AND MOMENTS: AS.I8 Body Weight and Balance Distribution. 1g rasisting inertia ferces due to the deca weignt of the dody and its contents plays an important pert in the load analysis for the airplane body, ‘hen the initial aerodynante and general layout ané arrangement of the airplane is mde, tt is necessary that 2 complete Weight and telance estinate of the airalane be nade, This estimate ts usually m¢e Dy aineer from the wetgnt control section of engineering cepartuent who as liad experience tn estinatin« the weight and distribution of airplane units. This estimate whien 1s brew sented in report form gives the weizhts ang (e.g.) locations of all aajor atrziane units or instaliations as well as for many of the minor units which make up these major airplane assemblies or installations. This wetgnt and balance report forms the tasis for the dead weight inertia load analysis which forms an important part in the load analysis of the eir— plane body. The uss of this weiznt and balance estimate will de {Llustrated in the exenple problem to follow later. A518 Load Analysis. Unit Anaiysis. Due to the many design conditions such as those Listed in art. 45.14, the general procedure in the lead analysis of an airplane body is to vase {t on a sertes of unit amiyses. ‘The loads for any particular design condition then follows as 2 certain combination of the unit Fesults with the proper multiplying factors. A Simplitied exemple problem follons which illustrates this unit method of approach. AS.17 Example Problem Diuatrating the Calewlation of Shears and Moments oa Fuselage © e to Unit Load Conditions. Fig. 5.40 and 45.41 shows a layout of the airplane body tobe used in this example pr~b- lem. It happens to be the body of an actus. airplane and the wing used in the previous ex= ample problem was the wing that went with the airplane. ‘Reference Z ania Front View Fig. A541 207-0ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES Table 45.2 gives the Netght and Salence astimte for the total airplane, This table is usually formulated by the Weight and Balance Section of the engineering department and tt is necessary to have this {nformtion before the 5.33 (WEIGHT AND BALANCE OF AIRPLANE LESS WING GROUP "AND INSTALLATIONS I AND ON WING 7 ssid arm from Garvie, = la whe airplane load analysis can be wade. a on: | IG ance trom 5 Rise” fora ore] we vom | oe rc _ z 0 same ™ Arm | Mom. 1. TABLE AS.2 Be P| Arm | Mom. | Arm | Mom AIRPLANE WEIGHT AND BALANCE Yer (@) anne aeaure en Cet Une Reference } (a 99) SEE } fk Bb arms seared from ae Bretec penne Rel wale | || tee) sae ie | io. ‘oS a a} 8 3 i) st i at | ake t ie | “ite id ae | tie ; ES : of | ake ; ne 3 | 8 |e [sete omnty «| waite | | wetgnt aistribution should study the inboard yo mat Ed | «|i! | Frorite drawing of the afrplane waten shons the | Bj Remm | wat |-aal) | general arrangement of all the installations and LPR Pate «| 000 Lame |" [feae| | cquiment. furthermore, he should study che Tasaon 6-8. cnn cvoral? structural arrangement as co tts posst— ‘ieee . . ble influence on fuselage nefeht distribution. The wtole process involves considerable comen Sense if a g00d approximation to tho weight die soLutrene tribution ts to be obtained. Fortunately the WGIGHT AND BALAN OF g0DY TTBS. BIGHT DISTRIBUTION. Table A5.5 gives the weight and balance calculations for all {tems attached to fuselage or carried in Te fuselage, except the wing and tens attached to the wing'as the front landing gear and the fuel. In order to obtain 2 close approxigation to the true shears and aonents on the fuselage due to the dead weisht Inertia loads, it is neces~ sary te distribute the weights of the vartous [tens as given tn Table 45.5. Fig. AS.4@ shows side view of the airplane with the center of gravity locations of tha wetght {tems of Table 45.3 indicated ty the (+) stans, In the various design conditions, the direction of the wetaht tnertia toreos changes, thus it is conventent and customary to resolve the inertia forces inte X and % components, Thus, in Pig. 48.43, the Weights as given tn Table 45.3 are assumed act= ing in the Z direction through their (c.g.) Lo- cations, The loads as shom would not give a ‘true picture a8 to the shears and aoments along the fuselage, thus these loads should ve distributed ina manner which should simulate ti actual welzht distribution. In most wetzht and balance reports, the weight itens are oroken dow into considerable more detail than that shown in Teble AS.S, which makes the weigat ats— ‘srtbution more avident. The person making the large dead weight loads, such as the power plant, tail, etc, are definitely located, taus suall erroré in the distribution of the minor distributed weignts does not change the over= all shears and roments an appreciable anount. In order to obtain reasonable accuracy, tae fuselage or bedy is divided into a series of stations or sections, In Fig. A5.42, the sections selected are designed as stations which represent the distance from the 2 reference axis. The general problan is to distribute the concentrated icads as shown in Fig. A5.49 into an equivalent eystem acting at the various fuselage station points. Obviously, 1f @ weient tten trom Table AS.3,, represents a concentrated load such as a pilot, student, radia, stc., the weight can be dis~ tributed to adjacent ‘station points inversely as the distance of the weight (c.g.) from these adjacent stations. However, for a weignt ‘tem suen as the fuselage structure (Item 2 of Table 46.3) whose c.g, Location causes it to fall be- ‘ween stations 40 and 120 of Pig. a5.43, it would obviously be wrong co distribute this waignt only to the two adjacent stations since The wetent of 36= 1s for the antire fuselace. This weight {tem of 3807 shoul‘ thus de distributed to all station points. The controlling Tequirenent on this distribution is thet the mocient of the distributed system about cha ret- erence axes must equal the moment of the original wetsht about the same axes. Fig. aSAdBEAMS -- SHEAR AND MOMENTS WEIGHT DISTRIBUTION 70 FUSELAGE STATIONS 170 200 230 280 280315 oi: 50 @0-« 20.170 200 200 250 900 31S Fig. AS.42, Location of weight items of Table AS.3. 470 200 290 260 Fig. A5.49. Weight items of Table AS. 3 acting in Z direction. 470 200 230 280 poe oes oi 90 89 120-470 200 23 260 290 315, . Final weight distribution to station points. i I lara See we oe Hig -GoeGa0r C09 Sau 639 cs a | i Peloy i | Fig. 48.49. Final weight distribution in X direction referred EAPALT SiS oper ceases,ANALYSIS AND DESIGN OF FLIGHT VEHICLE sTRUCTURSS shows how the dead wetght of SSO was distribu ted to the various station points considering ‘the weights to be acting in the Z direction. Table 45.4 shows the results of this sta- ‘ton point weight distribution for the wetght items of Table A5.3. The values in the horizontal rows opposite each weight item shows the distribution to the various fuselage stations. ‘The sumation of the weights in each vertical colum at each station point as given in the third horizontal row from the botten of the table gives the final station point metent. ‘These weignts are shown tn Fig. A5.45 for Weights acting in the Z direction. The moment of each total-station load about the Z exis is given in the second horizental row tron the Dottom of Table A5.4. The summation of the monents in this row aust equal the total wx nonents of Table 45.3 or 219700", This check 45 shown in the last vertical eolimn of Table AS. The distributed system must also be distributed tn the 2 or vertical direction in such a ASS mamer as to have the same resultant c.g. location as the original welaht system when (3 illustrated in Fig. 45.46, Fig. A5.47 illus ‘trates how the fuselage weight distributed system as shown in Fig. AS.44 1s distributed tm the vertical direction at the various station points so that the moment of this system about tha X axis 1s equal to that of the original fuselage wetgnt of 35c#. For convenience, these distributed fuselage weignts can be transferred to the X axis plus a moment as shown in Fig. 3.48, Table A5.4 shows the vertical distribution of the varfous ttens at the various station points. The bottom horizontal row gives the Moment about the X axes of the loads at each station point, which equals the individual loads times their Z distances. The summation of the values im this hortzontal row must equal ‘the total wa moment of Table 45.3. This check 1s shown at the Dotan of the last vertical column. Fig. A5.49 shows the results as given in Table AS.4 for the weight distribution In the X directton. PANEL, ROOT WEIGHT DISTAIRUTION Taososoee on a eage Tiga ease [ee] 9 5.18 Unit Analysis for Fuselage Shears and Moments. Since there are many flight and landing conditions, considerable time can be saved if a unit analysts is made for the fuselage shears, axial and bending forces. The design values in general then follow as 2 summation of the values in the unit analysis times @ proper multiplica- tion factor. Tae loads on the fuselage in general con- Sists of tail icads, engine loads, wing re~ actions, landing gear reactions (f attached to fuselage and inertia forces due to the airplane eeceleration which may be due to both trangla- ‘tonal and angular acceleration of the airplane. For sinplicity, these loads can be resolved int components parailel to the Z and X axes» To illustrate the unit analysis procedurs, @ unit analysis for our example problem will be carried out for the following unit cond Sa oat eee) weet a es aT) se ae (2) Unit acceleration or load factor in Z direction and acting up. (2) Untt acceleration or load tacter in x @irection and acting rorvard. (3) Unit tail load normal to X axis acting down, Unit ansiyses are also usually carried out for engine thrust and engine torque, side load on tail and angular acceleration, but to keep the example caloulations from becaning too lengthy only the above 3 unit conditions wil de carried out in detail. The others will be discussed in detail in later paragraphs. Solution for Unit Load Factor in Z Direction, "ig. A5.50 Shows the dead wetgnt loads35.18 acting in the Z direction as taken trom Table AS.4 or Pig. A5.a5. The wing {= attached to the fuselage at stations 73 and 116 as shom cn Fig. 45.50, The f{ttings at these points are sssumed ag designed to cause all che drag or reaction in the X alrection to be taken eff entirely at tne BEAMS -- SHEAR AND MOMENTS TABLE 5.5 FUSELAGE SHEARS AND MOMENTS FOR ONE LOAD FACTOR IN Z DIRECTION ‘(Deas Weight Acting Down) Front iteting on stacton 7S. TpP2 to 1. ‘no place the fuseiazs in equiltoetis,, the aay eri Bm ming Peaction will be calculate axa, | Loagor |V = shear/ax = Dist Now w | fos.) | stations: Bh #0, Ry + OF 0, hence Fy =O + ~ 3 Meration 0 7 219700 - 118 8p - 73 Ae tay | [See]. 2 os (Note: 216700 from Table 45.3) 2 me 220 By += 2555 + Ap + Bt - eee 3) eu | | o| 10 : ] ss ao'| go) saan | - 2500 Tea) i | Rp = 1780 1., Ry = 775 1D. 9 + mo‘) 2] | VT sia0 | smo Table AS.§ gives the calculations for the x —! fuselage shears and bending Ronents at tha ver 7) ) me | tous station points. woc}em | oe | tee | aan ps0 zo aro ao aso 260 wo sts [aa] gg] | 21000 | -so020 ' rot 1 I , | “ ! | 1 i 4 i | ior 1 1 1 1 I 1 +| oO sez | | rot | Mod FA tf] ag] a8 [ste | sae | av 4 I 6 | I i ae ee Te Sg et 87) Loot) 499 | 7 6912 | -6o800 1 | |e j wt) nd) 8 = soe | 2420 Fig. A559 2 SAT staus sos) ae -2eeen | 827 2° Soutioa for Uhit Load Factor is X Direction. a) yaar 1 Fig. A5.S1 shows the panel point dead weight distribution tor loads acting in the x direction and aft, as taken from Table a5.4 or Fig. AS.40, To place the fuselage in equili— brim the wing reactions at points (a) and (3) will be caleulsted, BP 5 2665 ~ Ry FO, hence Ry = 2685 1d. (forward) Take soments about poine (a) By * 2555 x 17 + 5920 - 43 Rg = 0, hence Ry = 1147.8 (up) (8820 equals the sum of the couples trem Table 48.4. aP, = 1147.8 - Rp = 0, hones Rp = 1247.2 2b. (dom) Table 45.6 gives the calculations for the shears moments and axial loads for the loading of Fig. 45.51, (3) Up on left and down on right side of a positive shear. (6) Tension in upper fuselage portion is negative bending moment. a) # relere to aft side of station. + relers to forward side of station. © MOP Mat previous station in col. 8 plus 4M in col. 5 ction is 13072 200 280 250, 290 818 1 1 og 8 fe ad aa Ban BS OF te 907 Geog Sale ve G10 Gar CA RAEI on aan e Tigei97.8 Reelin. Fig. A8.51 STATS. STA.116ANALYSIS AND DESIGN OF FLIGHT VESICLE STRUCTURES AST t TAM A8.8 | TABLE 5.7 [__rveag cans, mons 8 aa gun rm FUSELAGE SHEARS & woME‘TS FoR UstE Saat eS toad Tone HON ri oat st Doe) ! « | 1 2 a 4 3 8 - oad ot tae an| qe Imeal Taal || ow SSS | gat | | Ds sa Soca YY Se [2n| «| oo 1 ° 4 as) fs] 3 og | s “fo ° wo th o ° as Toe Te ¢ msl ad | we a) ws To] 00 Tine wo o 100 7 ATO) | 1750 . ° +75( 5 = ve] 00 = of er.s| get | ater | ° 23007 0 100 + 3000 | 7 4750 a ame 0 ao EL soe [0 Tae 7 or tite) ner op a seen too: o 300 = 3000 | = 1750 eT assyuner ha S| eaeas: 0 aT) 0 oro Ln ape Pee 7 A of) Li 2} oo 100 [7002 | s1ors0 a te 6 al v| 8 | 100 ~15750 2 3 oo. 0 100 I + S000 | “is780 ‘ 2) ee —taiso Go 1 3 C9 ree a | ne 1) as.6 | -378.6 + 2 | e150 (Col, 3) Mos Mai previous station lz Col. # pies Mt of col. € plum 36 AME of toh. . i (RR ERAA Mees meseeoeeme | $ xa | 2B : Sstuen for Unt Herts Tall Land Acting Down Doe Pe mal? t] k.s | 75 4 ‘The fuselage shears and moments will be a computed for a unit tall load of 100 1b. om the sot] 8 3 ° fatt Seting sn the 2 etrection, witn balancing ! os Fancvions ct the Wing ebeachaent points, The joe pe Q center of pressure on the horizontal tail is at a “8 o 2 Stition a7higs Pigs Abc2 shows the fase2ags loeding. To find wing reactions at (A) and (B); (Col. 6) we Marrone station in Col. 6 plus Diy = 100 x (277.5 - 73) ~ 43 Ry = 0, hence Ry = 475.58 (ur) BF, = = 100 + 475.8 - Rp = 0, hence Rp = 375.58 (down) Table AS.7 gives the detatled calculations: for the shears and moments at the various ste- Using the results m Tables 48.5, 45.6, and ‘tion soints. he applied shears and moments tor a given 11 $0 81203700 290-260 200 325 ‘adition ZoLLOW as a matter of procor= 1 | thon and adattton. To illustrate, the applied yelues for ome flight condition will be elven. Tt will be assumed that the aerodynamte calculations for this airplane for the (H.A.A.} figh angle of attack conditton gave the follom tng results, which the student ili have to accept without imowleae of how they vere optained. CALCULATION OF APPLIZD FUSELAGE SHEARS, MOMENTS aND AXIAL LOADS FOR A SPECIFIC PLIGHT CONDITION. Rpr3t5.5 Fig, A852 STA.TS STALE16 Applied load factor in Z direction = - 6.0 down Applied lead factor in X direction = 1.853 art. Applied tafl load = 120 1b. up. Thus with the load factors in the Z and directions and the tail load imonn, Table 45.8 can de filted in as illustrated, In similar uamner the values for other flight conditions can be found, the only difference beme a new set of multiplying factars since the applied loads would be aitterent. [ee | APPLIED FUSELAGE SHEARS MOMENTS & AXGAL, LOADS FOR AS.19 Example of Puselage Shears and Moments for Landing Conditions. Fig, A5.83 illustrates the airplane in level landing condition, Tho ground reaction ts assumed to pass the center of landing gear waeel and c.g. of airplane. The fuselage sears, no~ aonts and axial’ loads are required when tnd vertical ultimte load factor ts 7, (Gross wetgnt = 43008). SOLUTION: The vertical or 2 component of the ground Teaction R is specified as 7 lead factors which equals 7 x 4500 = 301008. One half or this is acting on each wheel, The horizontal or X conzonent of R is 30100 ‘tan 25° = 425 x Solo0 = 12600$ and acting aft. ‘The horizontal load factor on airplane equals 22800/400 = 2.68. BEAMS -. SHEAR AND MOMENTS ‘The resistance so these X and 7 compen of the ground reaction & 18 provided oy the ertie forces of the airplane in the K 2nd Z dic rections. Tables a5.5 anc aS.8 show the 258 sheers, acments and axial Loads for inercta loads due to one load factor in the 2 anc £ directions respectively, ‘Taus to obtain the fuselage forces for this given landing cont: it is only necessary to miltipt these two tatles oy the croper factor and the results. ‘Thus Puselage forces cue co vertical factor of 7 would equal 7 tines the 7alu: columns (3} of Table a5.5 to obtain snaar an ‘times colum 6 to obtain bending aonent. Likewise the torces due to the 2.93 lead factor in X direction would equai (-2.96) tines the values tn columns (4), (5) and (8) of Table a5.é to obtain axtal leads, shears and bending moments respectively. ‘The final or true forces woulc be the algebrate sum of these recults, Landing with angular Acceleration na level landing condition, tt is seme- ‘imes specified that the horizontal component of the ground reaction must be a certain proportion of the vertical component, which causes ‘the Line of action of the ground reaction 2 tn Fig. 85.83 to not pass through the c.g. of the airplane, which creates an external Sitchin moment of the airplane. This sonent ts usually balanced by the inertia rorees cue to the angular acceleration produced by the unbalanced moment about the c.g. The shears and monents on the fuselage cue to this externel Bement could be found as explained in art. 35.20, AS.20 Inertia Loads Due to Angular Acceleration, In some of the flying conditions, it ts souetines specified that the airplane must ce subjected to an angular acceleration as well as ‘translational acceleration. This angular celeration or the airplane produces inertia forces watch must De calculated if the airplane is to be treated as a body in staticANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES equilitriua., In some cases, a tail load due to a gust on the tail ts specified wnich produces a monent about the airplane c.g. which produces anguler acceleration. cf the airplane, In cer= tain landing conditions, the growd forces do not pass through the airplane c.g. thus pro~ Guotng a moment about the c.g. ahicn for stress analysis purposes 1s valanced by inertia forces. Moment of Inertia of Airplane The calculation of the moment of inertia of en airplane about the center of gravity axes was axplained on page AG.S of Chapter AS. A detailed example Solution was given in detail in Table SA of Chapter AS, The general quations for the moments cf (nertia of the airplane about ‘the reference axes aret Ty Shwe +See + 2b ly Ig 2 buy tlw. +2 ty Int wy tlw +2 ate ‘The last term in eack of the above squa— tions represents the moment of inertia of each weient item about Its om centroidal axes par~ allel to the reference axes. ‘AS.2L Solution for Inertta Loads Due to Unit 100,000 In. Lbs. Pitching Moment. Zo illustrate the general procedure of de~ termining the balancing inertia loads when the airplane :9 subjected to an unbalanced moment about the ¢.g., an analysis will be made for a anit 100,000 in, ib. moment. Table AS.9 gives the necessary caleulations. From ktnettes: Pitching angular acceleration a (rad/sec.*) where = unbalanced external pitening aoment about c.g. of airplane. pitehing moment of tertia of plane about airplane c.g. = Zer* ‘The tangential inertia force F for a mass w/g due to an angular acceleration « equals, w Yy petra, putas z Ty F hence ty Feqywr, were r ts the distance cron the weight w to tne airplane c.g. It Le conventent to treat the inertia force Fas resolved into two components F,, and Fz. hence, 45.19 Re @ Fe (2) From Table 45.9, ly = 16097600 My vas assimed as 100,000 hence Fy = 290000, Py = GOOOOR, w xe = 0062 W te . By = .00621 W Ze where Zp and Xq are the z and x distances of weight W to the airplane c.g. Columns No. 9 and 10 of Table A5.9 gives: the values of these inertia components. Fig. 5.94 shows these inertia loads applied to the fuselage. The reactions at wing attachaent points should be computed and then a taple of fuselege shears, aonents and axial loads should be made up. This unit table could then be used for all conditions involving angular acceler ation of the airplane. Tt should be realized that the inertia resisting loads in Table AS.9 are only approxi~ mately, since the moment of inertia neglects the centroidel manent of inertia of the big items, such 23 the power plant, wing, etc. The example ts only for the purposeyof illustrating ‘the general procedure of determining the inertia resisting loads due to angular acceleration. ‘The same general procedure can be followed in considering unbalanced external mouents about ‘the Z and X axes, comonly referred to as yan— tng and rolling moments. a3 | x cae SBBER BEE epeeeeieceeBEAMS -- SHEA! amo 790 260 290248 Fig. A5.54 AS.22 Problems (1) Draw the shear, dending moment end axial load diagrans for loaded structures Ia Figs. SS to 60. w/e 2 wee aa! $20 F104 os) “F204 ey x eas 2% sno” ss = ie wo asa (sta ow" sok 10 © a ____Labie 490 oD _e a rio 4 yf — 12" —t— 9 ~— t is r 500 Lb. sep mow. 58) 1094 a Si 100 HL 100 apr Leb og UB epi Lory Hom a" 109/13. 100 se ae bn s 100-4} 100 (Eon T 38 2 é rT " x ab on HOA 80, 80, (11) Draw bending aoment diagram tor structures and loading in Fig. 45.61, ave. 4008 i iG 1001008 t 00 fa) 16" 310%" Lo fatten, e tb) be ted (RAND MOMENTS 1. 2 shous Assume 4 loa on the 3 30 ib./sq.tt. unite exor: and bending moment on wing and 7: at 25, 100, 250 and 260 tn Fig. 48.62 ~ so" Ay ret. axis LOSE TT TTT TIT, Relative soar 0 e “Soil. wise Distribution Fig. 45.85 shows plan form of 2 cai ‘The total distributed air load normal to survece 1s 10000 1b, ‘The relative spanwise cistriaut: {a shown. Take center of sressure at 24 percent Of chord tro leading edge. Divide wing into 10 inch widtn strips and calculate Vz, My and My, an¢ plot curves for same. tlever wing. IV. Fig. 45,64 shows an oxtarnelly braced monoplane wing. Take an av Load of 90 Ib./sq.ft. norm center of pressure at 27 percent of the chord from leading edre of ing and calou~ late an¢ draw the front and rear beam pri~ gary shears and bending acnent diacrass, woth, Fig. AS. 64 runt strut Rear Sra Fig. AS. 61ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES Asa BENDING MOMENTS - BEAM - COLUMN ACTION AS.29 Introduction A dean-colum {s a member subjected to transverse loads or end moments plus axial leads, ‘Tae transverse loading, or and moments, produces bending moments wich, in turn, produce lateral bending deflection of the zember. The axial Loads produce secondary bending aoxents due to the axial loed times this lateral deflection. Coapressive axial loads tend to increase the primary traneverse Dending moments, where as tensile axial loads tend to decrease them. ‘Beam-column members are quite comon in airplane structures. For exexple, the beams of externally braced wing and tail surfaces are typical examples, the air loads producing trans verse bean loads’ and tha struts introducing axtal beam loads. In landing gears, one member 1s usually subjected to large bending and axtal joads. In Cubular fuselage trusses, Jateral Joads due to installations supported on mexbers between truss Joints produce bean-colum action. In general, beam column members in airplane structures are comparatively long and slender compared to those in buildings and bridges; ‘thus, the secondary bending moments due to the axial loads are frequently of considerable proportion and need to be considered in the design of the members. ‘This chapter deals briefly on the theory of single spen bean-column asubers. A summary of equations and design tables is included to~ gether with examples of their use. The {nforaa- tion in this chapter 1s used frequently in other chapters where practical analysis an destm of peam-coluun members is considered. For a con pleted and comprehensive treatment of beam-colunn theory and derivation of equations, see Wiles and Newell-"Airplane Structures", AS.24 General Action of a Member Subjected t9 Combined Axial and Transverse Loads. Gub-Pigure 2 of Ftg. 45.65 shove 2 nember subjected t6 transverse loads ¥ end axial com pressive loads P. The transverse loads W produce a primary bending distripution on the meuber as shown in Fig. b. This Dending will produce @ transverse deflection curve as illustre! ed in Fig. c. The end loads P now produce an additional secondary vending zoment due to the end load P tines the deflection 6 , or the bending moment diagran of Fig. d. This first sec~ ondary moment distribution produces the addi~ tional lateral deflection curve of Fig. @ and end load P will egain oroduce further bend ing monents due to this deflection. If the ax- deflections will gradually converge and the meber mill reach a state of equilibrium, These secondary bending mozents could be found by suc~ ceseive steps by the various deflection princtples given in Chapter A7. However, for prismatic ‘beams this convergency can be exprossed as @ mathematical sertes and thus save much time over ‘the above successive step method. For members of variable moment of inertia, the Secondary moments wil usually have to be folmd by successive steps. If the end loads P are tension, they will tend to decrease the primary nouents; thus, in general, the case of axial compression 1s more Important in practical design, since Duckling and instability enter into the problem. AS.25 Equations for a Compressive Axtally Loaded ‘Strut with Uniformly Distributed Side Load, Fig. AS.68 shows a prismatic beam of length L subjected fo a concentric canpressive load P and a uniformly transverse distributed losd ¥, with the bean supperted laterally at each end, and with end restraining moments M and Me. It is assumed that the general conditions for the beam theory hold, namely; that-plane sections Tenain plane after bending; that stress is proportional to strain in both tension and compres- ston. ‘At any point a distance x from the beam end, ‘the monent expression 1s, (ert) whe mt T =e mom, + laath) , We ME py From applied mechanics, we know thet M*3I SY gerotore, cxtterentteting eat- tion {A5.1) twice with respect to x gives aM, P Ween eee eee ee Las eta Me {aS.2) we meats pop : Mg ~ My con | 7M My ae cad EL sy 5 # Max? Mi, > aE pam we ==, 4 + Gas Sie tose TO Sa Yomaste Swern, . tooo oi? | on? Max? W982 L) ree MP ES Taiform Side Lead Pius Bhd Moments Maax* Phe wi? sa {TET Me say TE) a | tect j z Die Mags oi? Sogeguaed Sica ad. No __[ 3en, = Wain ene st Moneys ° ° Max * (617407 re o— set 7 PS foaming |- mama] o e144 ee 7 Trigngulas Loading. Wo (NOTE A) End Moments + 0? o "To obtain Maximum Moment, compute ee TTT wh ‘moment at 3 or 4 points in span. Draw a smooth St 7 ‘curve thru plotted results. Saar Triangular Loading. No End wi? WF! 2x | (See Note A) eee Couple Leading (Choekmse) | xca,smeces | eo 2 | (See Note AY co we — 4 : ® Kare P | x>a,-mcosa | meosa) 0 Eee] > - sons - ‘= or W is positive when upward. Mis positive when it tenas to cause compression on the upper ‘Gbers of the beam at the section being considered. Reference: ACIC #493; Niles, Airplane Design; Newell and Niles ‘Airplane structures For Table of many other loadings, see NACA T.M. 985.NATERAL SNES, COSNES, 480 TANORITE ANGLES I RADIANS fy psy || ketsTABLE AS. 0 NATURAL SINES, COSINRS, AMD TANGENTS OF ANGLES IX RADIANS ssa “3 9.14112 T-1SKE je 12129 TIT PULSE! 08150, 0.07155 0.06155 0, 05156 5, 0: wrx 0.00160 =0, 0OBaT 0, 0184T, =0, 02840 =0, 05838. =0, 06836 =0. 07833 =0, 08823, 0, 05825, =0, 10820 A a L, L, L sin 2/, | com by, con by =0. 98999 0, 00136 099135, DOLE, 0, 00166 0, 60087 008: 00086 0, GO93T 5007 WO o. 00027 =a. =0, 99060 5, 01060 =0, 99830, = E =0, 3968; =0, 99608, 59516 4, tax ty, oD 0, DIDI, 9, sate 9, 01008 ‘a, or oor 0, 0100 00, 02841,AS. 28 BEAM these values in the general expression for Mas given at the top of the Table. In @ beam-colunn member, the tending mom ents do not vary directly as the load is increas} ed. Thus, the student should realize that merg— ing of safety based on direct proportion of mom~ ents to loads are incorrect and lie on the un- safe side. It 18 recomended that four stgniftcant figures be used in computations, making use of ‘the so-called precise equations, since the results in many cases involve small differences between large numbers. AS.20 Example Problems Beample Problem gi Fig. A5.c7 illustrates @ typical upper, outer panel wing beem of a biplane. Let tt be required to determine the maximum negative dending monent between points (1) and (2), gene erally referred to as the maximum span moment. To obtain the true bending moments on the Dee, the axial beam load 2s well as the end moments at (1) and (2) are necessary since they influence the deflection of the bean. Solution:~ To obtain the aorizontal component Ty of the Lift strut load, we take moments about the {1800#eReaetion from strat to Lower Wing . Fig. 45.87 2200/10. aeao"e 4204 show Fig. A5.68 hinge point at the left end. Bh, = = 2000 x 50 ~ $40 x 116 = 1500 x 100 + 70.75 = 0 hence Ty = 48208 ‘The axial compressive load induced by the Lift strut at point (2) then equals ~ 4420¢. ‘Taking ZH = © for the load systen of Fig. A967 gives P= - T) = 44208. ‘The end moment on the dean at (1) equals the end load tines the eccentricity of the hinge from the neutral exts COLUMNS of the bean, or 4420 x .75 = SS15"A positive because it produces compression in tha top fibers. ‘The monent at (2) due to the cantilever overhang, equals (20+ 30) G6 x 16 = S640"F, FIZ. 45.58 (eto) shows the beam portion between points (1) and (2) as a free body. From Art, A¢.25, we nave the following precise equations for a beam carrying a transverse uniform distributed load with end compressive loads. tan ¥ =n. = 2, cos $ io at D, sink and Nee Ba te J Byaluating teras for substitution in these equa~ tions, we cbtain, My > Sas"# = scante = 4420# compression 10 in* given and assumed constant ‘tnroughout the span. 3 7 VESEE ERS . asi = cs.c3 wg" 20 x 2961 = 58820 Di = Mi - J? = 3315 - S8éz0 = - 55505 Dy = My - wg" = 8640 - 88820 = ~ So180 L . 30, § = Bo - 1 . prom Table 45.12 sin E = 96290 ane coe Y =~ 2598 Substituting in equation (A) Fed, - & tan = d= 0, cos § D sin 5 = - SOL80 - (-55505 x -:26981) _ ~65156 ay - a seaiat * 22192 3 = tan” 1zise = .seses Hence, x = .88383 x $4.25 = 48%, which equals tne Gistance from the left end of the bean to the point of maximm span moment. Mogg ETN) 008 Fs cons rem tape 5 AS.IL Hence = - 53505, 0 = 2 “oa * ~ UpShde 7 89620 = - 20,700" ‘To obtain an {dea as to the magnitude of the secondary bending aoment, that 1s, the moment due to the axiel load times the lateral bean deflection, the primary bending aoment at a point 48*ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES from the Left end wlll be computed, Mya = 3315 + 45x 20 x 26~ 940 x 48 = ~16765"F ‘Thus the secondary Sending renent oquals = 2a700 15786 = 9988" woh 19. large per~ Gentage of the prisary sment. The transverse Gofiection of the beam av the point of mmx. span Sonont shen aquele ~ 9906 = 2.28 Inches upiard, - GO bending Moment at any Point Along Span seth Sasa eer poTHt TO" from point (2) de required. In this case, x = 100 - 10=90 aeniian stn ecco H] ems" iRet, 35. 48.7) x. 90 x Fs eqigy «1.8506, 91m F = .90608 cos. 5 = = .08867 ven XB-1,2192= wlue for x at the point of 4 maximum bending moment Hence, =~ 55505 [(1.2192.x .96608) +-.06867 ] + 58820 =~ 3664"# Example Problem #2 Fig. Ac.cO shows a simplified landing gear structure carrying a vertical Load of 12000# on. the axle. Member ABC 13 continuous thru B end pinned atc. Lett be required to determine tho bending moxent at the midpoint of member 2c and its lateral deflection due to the 12000# vertical design load. Fig. A569 Line of action of DB goes through E solution:~ Solving for reactions at C by statics, ne evtain the axial load tn 3C = -20000. The Jending aoment at 8 due to S* eccentricity of the ges] lead = 3 x 12000 = os00"#. Pig. ag.79 shows a free body of portion 8¢ of aeader ABC. From Table AS.I in f+ cy cos Set Me sin + Ca cos F A520 substituting values of C, and Cy and f(w) from Table A5.I 1m the above equations: (Mast cos L/J) sin x/} sin t/T 2.1/2 ~ 009 Steel Tube 4M cos ¥/3 38,000 52,0008 Ls 41.762 Fig. AS.70 But, My = 0 tn our problem, hence, M = Masia x/; Scare Sl OIE yess. 1 =f Bae Veer = 2.028 uy = Beh = ser = ~~ atm Ld = 20682 x = L/2 = 20.881 a/} = 30:88 «e005 - - = erm x/4= 72827 Substituting In the above equations for Mf ~~ = 36000 x .72327 . x = SOs = Bees" # Tats compares with a primary soment of 3600/2 = 180009. “The deflection at the aid~ potat of 80 = 26066 = 19009 = .408 in. —~qo000 ‘Que saximan aoment is given by tho equation: = ate 21 Mag" FET ont {occurs at x= (see 3 ‘Table 45.1) AS.29 Stresses Above Proportional Limit Stre: Material. ot ‘The equations as presented in this chapter assume that © is constent or in other words the stresses are within the elastic range. In air= craft structural design the applied or limit loads must be taken without suffering permanent Gezormation, hence £ {s constant under such loads. Hewever the atreraft structure aust take the design loads which equal the Limit loads times @ factor of sazety (usually 1.5) without fetlure. In many cases structural failure will occur under stresses in the plastic range where the gatertal stiffness 1s less and not constant. A good approximation for an effective modulus B' 1s obtained as fellows: (2) Sompute P/k for the given number. (2) with this value of Fo enter the baste column curve diagram for the given material (for end fixity ¢ = 1) and tind value of L'/o corresponding to the stress F..45.20 (3) Using these values of L'/p and Fey compute 2 BIS) coon (2) Basic colum curves for various materials are given in ancther chapter of this book. Ie AS.20 Problems Fig. AS.72 Fig. AS.72 (1) Fig. 45.71 shows a 1-1/2 ~ .065 stesi tube subjected to both end and lateral loads. Determine the maximus bending moment on the tube, Compare the result with the bending moment due to the side toad only, B= 29 x 10" pel. I of tube = .075 in.¢ Compute lateral deflection at point of aaximm bending sonent. (2) Te beam column menber in Pig. 45.72 1s) nade of 2487 aluminum alloy. Calculate and plot 4 curve of the bending monents on the mender. Also pict bending moment due to lateral loads only. E= 10.3 x lo*psi. 1 8.0 in (3) Determine the maximum bending morent for the wood wing beam and loading of Fiz. I of beam section #17 in.t B= 1.3 x Lae he | a re ttt Fig. AS.74 BEAM -- COLUMNS (4) Determine the vending moment at the centerline of the beam-columns show in Fig. 5.74, Assume 64,000,000 1b. In. sq. aoe Err PE a a Fig, A575 “{8) For the beam-column in Fig. 45.75 calculate the bending moment of the centerline of the member. Assime £ = 1,300,300 pst. and T= 10 int rin. pee ET fj 208i. : = $= —s0008 fo sy oe 4 toon sdo¥ 000+ Fig. 48.78 (8) For the beam-eclumn loading in Pig. -76, calculate bending moment 2t center point of beam. Take 2 ~ 1,200,000 psi and T= 10 ins AS.31 Beam-Columns in Continuous Structures, ‘The secondary moments tn a particular eaber due to deamacolumneaction also effect or influence the deflections in adjacent members of @ continuous structure, Tais rather tnvoived problem can be handled quite simply and rapidly by the moment distribution method as explained and Lllustrated in Arts, Al1.12 to 15 of Chapter 421CHAPTER AG TORSION. - STRESSES AND DEFLECTIONS AG.1 Introduction, Problexs involving torsion are comion in afroraft structures. The metal covered airplane wing and fuselage are basically thin-walled tubular structures and are subjected to large torsional moments in certain flight and landing conditions. The vartous mechanical control systems in an airplane often contain unite of various cross-sectional shapes which are sub— jected to torsional forces under operating con~ ditions, hence 2 iowledge of torsionel stresses and distortions of aenbers is necessary in aircraft structural destsn. A6.2. Torsion of Members with Circular Cross Sections. ‘The following conditions are assumed in the derivation of the equations for torsional, stresses and distortions: ~ (1) The member {s a circular, solid or hollow round cylinder. (2) Sections renain circular after application of torque. (3) Diameters remain stratght after twisting of section. (4) Material 1s homogeneous, tsotropic and elastic. (5) The applied Loads 1te in a plane or planes perpendicular to the axis of the shaft or cylinder \p Fig. Ab? Pig. AG.1 shows a stratgnt cylindrical bar subjected to to equal but opposite torsional couples, The ber twists and zach saction ts subjectad to a shearing strass. Assuming the Lert and as stationary relative to the rest of the var 2 Line AB on the surface will move to 4! under these shearing stresses and this rotation ab any section will be proportional to the distance from the fixed support. It ts sumed that any radial line undergoes angular displacement only, or CB remains straight when moving to 0B". ‘The unit shearing strain in a distance & equals, 2 Ee BT Lot @ equal modulus of rigidity of the material and let t equal the unlt shearing stress at the extrege fiber on the crosssec- tion. eo, THES In Fig, 46.2 let tp equal the unit shearing stress on a circular strip dA at a distance p tron 0. Then oe sRted mp rors pes oeg j ‘The aoment of the shearing stress on the circular strip dA about O the axis of the bar is equal to, 7044 a =z, pda = OSM ternal torsional resisting moment is, and thus the total in 2 { Sotoaa. goreaa Por equilitrim, the intemal resisting onent equals the external torsional moment 7, and since G8/L is a constant, ne can write, rr a6 ge Pe Ming = where J = polar moment of inertia of the shart cross Section and equals twice the moment of in— ertia about a dlaneter. tron equation (1) 32 x, int. wax = tlle uence, P22 -------------- 3) es we---- |) also from equation (2), Solving for the twist @, 2B -------------- s) (@ 18 measured in radians).8.2 AG.3. Transmission of Power by a Cylindrical Shaft. The work done by @ twisting scuple noving through an angular disslacenent 13 {othe product of the aagnitues of cae 2our and the angulw diszlacenent in radians. angular displacenent Ig one revelution, the done equals 2nT. I: 7 te axrressed in inch pounds and N ts the angular velocity in revels tions per miauts, then the horsenower 1 autted by a rotating shaft may be writs SP. wnere 396000 represents inch pounds of work of one horserawer fer cne minute, Equation (6! aay be write a= Ess x 86000 , 63025 H.P. a wo TTT o Problem 1. Fig. A6.5 shows a conventional control stick-torque tube operating unit. for a side load of 180 1bs. on stick grip, determine the shearing stress on aileron torque tube and the angle of twist between points A and 8. SOLUTION: Torsional stick force of 50 x 26 = 3900 in. 1d. to this torque ts provided by the | 24ST Al Alloy Control stick“ G = 3,800,000 pst | sev Blgy.Controt Wires, |aiteon Horn Iayeron operating Tapeh fo oa se searing) Fig. a6.3 atleron opsrating system attached to afleron horn and the norm pull aquals 3800/11 = 356 1b. ‘The polar moment of tnertia of a if - 0.088 round tube equals 0.1568 int, Maxtmm Shearing stress = t = Tr/d (300 x 0.75}/0.1368 = 21400 psi. ‘The angular twist of the tube between points A and B equals sth. 3000x298 L © ay > Se00,000 x 0.1a68 * 0-2) Padiens or 12 degrees, TORSION ‘on sontron, (1-1/4 + .0a3 in ize} suppor hinge brackets ing stress 1 aileron fe as indicated in Fig. also compute the ans ‘sMist of tube b Tween horn section snc end of atleron. Pressure Tverage london sustacen 40470 1 = oe Tubta Alloy) eS las oo ue SOLUTY The airload on the surzace sends to rotate the aileran around the terque tube, Sut nove ment is prevanted or created Sy a control rod attached to the torque tube over the cent supporting bracket. ‘The total load ona tneh wide = 40(15 x 1/ Let w equal intensity of loading per tneh of afleron span at tha leading edge point of the aileron surface, (see pressure ciacram In Fig. A6.4). Than & + (0.5 w)le = 416 hence w= 0,863 10. ‘Tha total load P, forward of the center~ line of torque tue = 0.453 x 3 = 1.369 1b. P, the load cn aileron portion aft 57 ninge £0,483 x 0.5 x 12 = 2.778 iD. 4 ee na) moment per running inch of torque tube: = ~ 1.59 x 1.5 + 2.778 x 4 = 9.0 fn. 1b. Hence, the maximum torque, which occurs at the center of the alleron, equals $.0 x 2¢ = 261 in, 1b. Te torst ar T 251 x 0.825 Os07S Tax.) = 2450 pst. (J = 9.08678 tnt.) Since the tube section 1s constant fe torqueANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES vartes ¢irectly 2s the distance cron the end of the aflerons, the angle of twist 9 can be cea puted oy using the average torque as acting on Sntire lengtn of the tube to one side of hora or 2 distant L = 29", nance Th 281 x 29 5: ot. Sees ° 0.86. OF * EE SECOONO x C.UeETE degrees AS. Torsion of Members with Non-Circular Cross Sections. ‘he formulas dertved in Art. A6.2 cannot be used for non-cireular shapes since the assump~ sions mace do mot hold. In a ciroular shaft Subjected to pure torsion, the sheeting stress distribution ts as indicated in Fig. As.5, namely, The paximuz shearing stress 1s Located ab the most remote fiber fron the centerline axis of the ber and ts perpendicular to the radians to the stressed point, At 2 given diste anee from the axis of rotation the shear stress mx Fig. AS Fig. A6.6 Fig. AGT tg constant in both directions as {Llustrated in Figs 6.5, Niich qeane that ends of segments of the bar ag it tists renain parellel to sach other or In other words the bar sections do not warp out of their plane when the bar twists. If the conditions of Fig. A6.5 are applied to the rectangular bar of Fig. 46.6, the mast stressed fibers will de at the corners and the Stress will be directed 2s show. The stress nould then have a component normal to the surface as well as along che surface and this ts ot true, The theory of elasticity shows that ‘the saximum shear stress occurs at the center= line of the long sides as illustrated tn AS.6 and that the stress at the comers {s cero. ‘Thus wnen 2 rectangular bar twists, tne shear stresses are not constant at the sane distances irom the axis of rotation and thus the ends of Segments cut througn the bar would not remain parallel to each other when the bar tists or in Scher words, warping of the section cut of its plane takes place. Fig. A8.7 illustrates this action in a taisted rectangular bar. The ends of the bar are warped or suffer distortion normal to the original unstressad plane of var ends. Further discussion and a sumary of 2qua~ tions tor determining the shear stresses and 8.3 twists of non-circular cross-sections 1s given in art. 46.8. ABS: Elastic Membrane Aoalogy the shige of a“darped cross-section of 2 non-ctrowlar eross-section in torsion is heeded In the analysis by the theory of otas~ fietty, ang as a result only a few shapes such ae rectangles, 2bLipsss, triangles, etc., uve deen colved Dy the theofetica: approach.” Hon over, a close approximation can be made. ex- perigentally for almost any shape of cross Section by the use of the aenbrane anelosy. ft wa pointed out by Prandtl that the equation of forston of a bar and the equation for the deflection of a xembrene subjected to unitom prestura have the saxe fora. Taus if an clastic membrane {s stravched over an open the watch tas tha sane shape 2s the cross~ section of the ber veing considered and then If tho aeubrane is deflected by subjecting 1t to 8 Siight dizcerence of pressure on the two sides, She resulting deflected shape of tho sexbrane provides certain quantitiss which can be nes fured axperizenvaily and then used in the theoretical equations. However, posesbly tke sain advantage of the uenbrane theory is, that {t provides a nethed of visualizing to 2 considerable desres of accuracy how the stress Gonaitions vary over 4 couplicated cress-soction of a bar in torsion. The menbrene arlogy provides the foLlow- ing relattoneatpe detween the deflected xem brane and the twisted bar. (2) Lines of equal deflection on the membrane (contour Lines) correspond to shearing stress lines of the twisted bar. (2) Tae tangent to a contour line at any point on the mambrane surtace gives the direction of the resultant shear stress at the corre= sponding point on the cross-section of the var being twisted. (3) Te maximum Sofe of the deflected membrane at any point, with respect to the edge Support. plane is equal in aignitude to the Shear stress at the corresponding point on the cross-section of the onisted bar. (4) The applied torsion on the twisted bar 1s proportional to twice the volume included Detween the deflected membrane and a plane through the supporting edges. To Sllustrate, consider 2 bar with rectangular cross-section 4s indicated tn Fig. 36.8, Over an opening of the same shape we stretch a thin membrane and deftect it normal to the erese-seetion by 2 small uniform pres~ sure, 3qual deflection contour lines for this deflected tembrane will take the shape as 11- lustrated In Fig. 46.9. These contour lines wedon correspond to direction of shearing stress In the twisted bar are nearly circular near the center region of the bar, but tendABA Tor to take the shape of the bar boundary as the boundary is apprcached. Fig. A6.9a shows a sec~ tton through the contour Linas or the ). ap aut = 547 5 Fig. A7.90 ‘This is the Theorem of Complenentary inerey, Now tor’ the Tinoarty slastis body a vary important theorex follows since (Fig, 7.30) qu sau" 30 that 2 ave au 6 Fig. AT.9¢ wu @ 79 --+---+-------2 2-8. = (a7) Tn words, "The rate of change of respect to Toad ia squat £6 defiost isn EQ, (27) and the above quotation are state= nents of Castiglieno's Thooren, For 4 ‘ody uider tie Simultaneous action of several loads the theorem is written 30 a6 C0 apply indtviduaity to sacn lose and its assost- strain energy with the asséctated ated’ aetiection, muss a Be wees eee eee trey The partial derivative sign in aq, (17a) indicates that the increment in strain anergy is due to 8 small change in the particular load Py all other loeds held constant. ' Note that by "load" and *deflection® nay be meant: _Assoctated cn tection ——_—____ “The proof af the theorem for the case of multiple loads is Generally formulated more rigorously, appeal to & simple diagram such an Fig. A7-d¢ being lesa etfective. See, for example, "Theory of Elasticity" by &. Timoshenko, DEFLECTIONS OF sTauUCTURES Force (1bs.) ‘Translation (inches) Moment (in. ibs.) Rotation (radians } Torque (in. los.) Rotation (radians) Pressure (1ds/1 lume (in*) Shear Flow (1bs/1n} area (tn*] any generalizations of the meanings of “force” and *dezlection" are possible only so Jong as the units sre such that taelr product yields the units of strain energy (in. ios]. Once again for emphasis {t 1s repeated that, While the complementary rature of eqs. (14) and (17) are clearly evident, tne use of eq. (27) (castisiiano’s Tneoren) ts restricted to linearty elastic structures. A brie? example will serve for illustration of the possible pitfalls, ‘The strain energy stored in an initially straight untierm colum under an axial load P when deflected tnto a hal? sine wave is = PASLt Us Er > ¥evo sa 22; Me py Consider Flemurat Eneszy Only i one eae > 9 af(gra ts: pe Page x Fig. ato TET Where 0 ts the end shortening cue to bowing, Because the detlections grow rapidly as P approaches the critical (buckling) load the Broblen {8 non-linear. ‘The detatis of tho ~ calculation of U are given with Fig. A7.10, Now according to the Theoren of Virtual Work (eq. 14) au Bee out wu 6 ° Wer Therefore or 2 WE Pete (Suler formula for unirorm coluan). correct reeult. Application of Castigitano’s Theorea. ea. (17), leads to the erroneous result, aw ges e aU | 2PoLs @ * Er e¢ AF. A17.8, Chapter A-17 for detailed derivation of this equation.ANALYSIS AND DESIGN OF F. EL p= SF (incorrect) Moral: D0 not use Castigiiano’s theorem for non-linear problems. Fortunately the above restriction upon cha use of Castigiiano’s theorem 1s not a very severe one, the majority of avery day structural problems being linear. Castiglisno’s theoren is quite useful in performing deflection calcu lations and 2 veriety of applications will be made tn the following sections, ATS Calculations of Structural Detlections by Use of Gastigliano's Theorem. AS the examples of Art. A7.3 have {llus~ trated, tne strain energy of a structure can be expressed as a function of the external loadings provided the internal load or stress distribue tion 1s calculable. Having the strain energy so expressed the deflections at points of load application can be determined with the aid of eq. (17), Castigitano’s Theorem. In the examples to follow the deflections of a vartoty of statically determinate struc— tures are computed, Methods of nandling redundant structures are considered in subsequent articles. Example Problem 8 Find the vertical deflection at the point of load application of the crane of Fig. A7.11. Cross sectional areas are given on each meuber. The stranded cables have effective nodulti of 13.5 x 10* pat. © = 29,000,000 for other members. c Fig. AT 1 Solution: ‘Tae strain ensray considered here was that due to axial lording In each of the four nem bers. The load distribution vas obtained trom statics and the energy calculation was made in tabular form az follows: IGHT VEHICLE STRUCTURES ‘Then 2U = 43.25 x 107 PA LB PT au o $9.26 x 10" Pt Note that tn this probian the only use mde of the calculus was tn the differentiation. a simple sum was used to form U, as per eq. (2). Example Problem 9 DoFive Bredt’s formula tor the rate of twist of a hollow, closed, thin-walled tubé an ghg§ Snot. cmps as, ‘The strain energy ts stored in shear according to 4 aiff vey vas ff te . Tais is the only enerey stored, secondary ef- facts neglected. Over a given cross section @ 18 a constant and 1s given by Bredt’s equation a=. , where 4 te the enclosed area of the tube (Ret. chap. 4.5). Solution: Since the twist per unit length was desired ‘the strain energy per unit length only was written. Thus, assuming y In the axial direction, no integration was mde with respect to y. The integration in the remaining direction was to be carried out arcund the perimeter of the tube and so the index was changed from "x" to the more ap= propriate "s". Hence (compare with Sxample Prob. 7) veg SUB dnb ‘me symbo2 § means the integration 1s carried out around the tube perimeter. Therefore a, 2 pes os Ge aaht - USB OF FICTITIOUS LOADS In the following example the desired ceflection is at the free end of the bar waere no load ig applied. A fictitious load will be added tor purposes of the calculation. The rate of change of strain enargy with respect to this ttetittous lead will be found after wnich the load will be { TABLE af.2 Set equa: to zero, ‘hia tectniqve givas the de~ 5 sired result in a much as the deflection ts i MEMBER | Sips | bpp | AEX10* SL x10¢|| equal to the rate of change of strain energy | tes | a with respect to the ioad and such a rate exists { 30? | «0 | 136 espe |} even though the load iteelt be zero. “I 90 P| 90.0 [ue 26.48 P| AC. Ese FL 83.0 1:8 [i835 B¥7 | cxemple Problem 10 (egress | Ceapute Ene axial notion of the free end of [E2418 2] | the tapered bar of Fig. 47.12.Ana etotteg) 4 stv. az, [Ratt LIES constant, Lo = 40in, BB Lice a 7% is Fig. AT.12 Solution: After addition of th ‘Lous end load R the axial load from statics vas found to be S(x) = R+g (L =x) Hence, since loadings other than tenstis are of a secondary natura Before ovaluating the Integrals it was observed that the steps to follow, in which U was to be differentiated with respect to P and the subsequent setting of R = o would drop out doth the first and last terms. Hence only the secend term was evaluated. 2" Be vex+ 2B mes Thon DIFFERENTIATION UNDER THE ITSGRAL SION 4n important labor savings may be had in the calculation of deflections by Castigiiano’s thecren. In the strain onergy integrals arising in ‘this class of problems, the lead P,, with respect to wnich the deflection 1s to be found, acts as an Independent parameter in the Integral. Pro— vided certain requirements tor continuity of the functions are met - and they invariably are in ‘these problems - the differentiation with respect to P may be carried out before the inte gration is made. The resulting integrals gen- DEFLECTIONS OF STRUCTURES erally are easier to uate. Example Problem 12 “Tid the Geflection st point 8 of Fle. 47.33, El constant ay web pee of the beam Pig. ATS Solution! A fictitious load P, wes added at point 8 and the bending moment diasran was drawn in two perts. . PL y fe eres Mp, eee Fig. AT. 138 OcxeL/3 Mp 2 Ei osyen/3, 3 as ma ocaenya + ‘Then neglecting the energy of shear as being suali* uy vege (er +d ad a "Sememy ot -a| (2p ara] a. Differentiation uncer the integral sign with respect to P, gave ae os 2 ade al [ter + eg ax uy wa a| oy -af Epic on] GE aw ‘For beams of usual high Length - to - depth ratios the shear strain energy is small compared to the encrgy af Dexure. Neglecting the shear energy is equivalent to neglecting the shear deflection contribution. (see p. 7.13) Ln _ i - Base] Bien — ay) + Be = sy) et dla - avi + Ba an ]epn syANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES The fictitious load P., havingyserv: pose, was set equal to zero before the work. 2.2 (yore (au - 3y)(L + By) ey 1 ‘Se, - 3a)" ao 1 ‘Ts ST Exemple Prob: Fig, AT.1e shows a cantilever round rod of diameter D formed in a quarter circle and acted upon by a torque Ts. Find the vertical zov ment of the free end. Fig, AT 14 Solution: Fig. AT-Ada shows tne vector resolution of ths applied corque Ty on been elezents. 1. (6) Tages @ and the soaert MA(@) #7. sin’@. Ape plication of 4 flotitious + vertical load‘ (down) at T the point of desired de— Hection gave she loadings Shown in Fig. 47.143. Fig. At. t4a ans ‘Tae total loadings were Metittar(ToPR)ein 6 pun view TeteeTas Tacos @ + FR ‘4 (2 “cas 9} AX af \y Dus _ Fig, AT. 1 (7 ulsy (Te = PR)" Sint @ Rd @ a J7tzecos 9+ PR(L - cos 61] RA Qiote use of "Rdg" for the length of ¢ air— ferentisi beam element instead of "dx"). Off, ferentiating under the integral sien ay (te = PRR? (sip ay, a smu (" ‘sit 6c a (Te, + recos 9 + mas ~ 208 @) a cos 9) ee Putting 2, the fletittous load equal to zero and integrating gave 4 au, TR? (4 err “ap * “a” Gr -3) Since J = 21 ana o = "/2.6 the deflection wes negative (UP). |A1.7 Calculation of Structural Deflections by the Method of Dummy-Unit Loads (Method of Virtual Loads). ‘The strict application of the celculus to Castigiiano’s theorem as in art, 47.5, leads to a number of cumbersome tectniques ill-culted to the solution of large complex structures. A nore Flexible azproaeh, readily adapted to improved "book Keeping" techniques ts the Methed of Dumy- UaLt Loads developed independently dy J. o. Thet the equations for 1 attested to by the great vartety of tains the equations by 2 reinterpretation of appeal co the concepts of strain snersy. mechanics. course, yiela the same resul: Mexwell (1g64) and 0. 2. Mohr (1874). Nathod of Dumy-Unit Loads may be derived in 2 number of ways ts names applied to shis method tn the itterature ® below are tao derivations of the equations steming frou ¢: he symbols of The other derivation uses ste 2rinciples of rigid dody Based a6 they are upon a comon set ot Presented arent viewpoints, One derivation ob- fastigLianc’s theorem ~ essentially an insistent agemptions, 211 the methods must, of I - From castigliano’s Theorem Beginning with strain energy, aq. (38) paral. expression for If - From the neiple of Virtual Work ‘hen a system 02 forces whose resultant is zero (a system in equilibrium) ts displa variously called the Maxweil-Mohr Method, Method of Virtual Velocities, Method of Virtual Work, Method of Auxiliary Loads, Dummy Loads, Method of Work, eteana Teen ved (MK a} (itse yk [Mee 3) GE 2) Sr 72) a digterentiate undor the integra) stan with Fespect 20°, co 20taia as Sar 7E consider the symbols as ot at WF, Oe eR a, [ake . mat Sach of these 1s tha "rate of change of so wand-so with respect to Py" or “how much so and-so changes when Py changes @ unit amount" OR QALY, "the so-and-so loading for a unit Lead Ptr ‘Thus, to compute these partial dertvative terms one need only compute the internal load= ings due to @ unit load (the virtua load) ap- pifed at the point of desired deflection. For example, the term at{,_ could be computed in either of the two ways Shown in Fig. A7.l6a. RATE METHOD war reryop 4 4 ; Te. 1. Fig. AT. 152 Me Px m= dumy loading am an ap, * o* (- a) tkentco, 4p, where Py 12 a toed (coat or tieeteious) appiiia Se jomre wot Page Seide| tb given oy the Leadings cor the wit Lad ape aS Bowe Ih practice she use of the ante toed ts woot Gouvonieass‘Uetng toe antation 238 Fed or a a a, re ee decenes oof Sgt [SP [RE + jBe-ppadew .-------- as) “01.0 FseRe Fig. AT. 150 "a" Loads due to a unit (virtual) load. DEFLECTIONS OF STRUCTURES, Bcon'g, small amount withcut disturbing the equiz balance, the work done 13 zero ~ obviously, since zero resultant force noving through @ distance does zero ork. Consider now the set of equilibrium forces applied to the truss of Fiz. A7-18(b). The set may de divided into two Darts: the “external system", consisting of the unit load applied 2c the peint wnose deflection ts degired anc the two reactions fixing the line of reference, and the "internal system" consisting of the axial loads acting on the cruss aenbers to produce equilitrim, Trese latter are denoted by the symbol "u". This set oz forcas 15 2 small enough so a8 not to.affect the actual be havior of the structure during subsequent application of a real set of major loads. This unit load set 1s oresent solely cor aatnenat ical reasons and is called e "virtual loading” or "dummy leading*. ASSume AoW that the structure undergoes 4 deformation due to application of a set of reel Joss, the virtual leads "going along for the Fide’. Sach newer of the structure suffers 2 deformation denoted by 4° , Tne virtual loed~ ing system, Deing {n squitibgiua (zero resultant) by hypothesis, does work ("virtual work") equal to zero. Or, considering the subdivision of the virtual loading system, tae work done by the external virtual load must aqui that absorved by the internal virtual loads. The work dct ‘the external virtual forces 1s equal tc one pound tines the deflection at joint 0, the reactions Ra and Re not moving, That (3 Externa? Virtual Work = 1" x 6, ‘The internal virtual work 1s the sum over the structure of the products of thd aember virvual loads u by the member distortions 4. That ts, Internal Virtual work = 2 ud . by a SU/s5 for each member and one has. eey St tha argnene given above my oe exence virtual bending moments (m), torsion loads (t}, shear loads (v), anc shear fiows (3) doing werk SET ae (ce sens fae, sag torques (T), shear loads {¥J, and shear Zlows (q). Tne general expression becomes aa “jypgse 9, , Se © Note that the deforsations are not restricted to those due 49 elastic strains only. They may be the result of clastic or lalnatie string, temperature strains of mission ment corrections. ‘rains ‘eatenBPR ANALYSIS AND DESIGN OP FLIGHT VEHICLE STRUCTURES In applying 2q. (18) the labor of 2 deflection calculation divides convententiy tate several steps: 1, Calculation of the real (actual) load distribution (8,17, ete.) 11. Calculation of the unit (virtual) 2oad distribution (um,t, ste.) due to a unit (virtual) load applied at the point of desired deflection and reacted at the reference potat(s) | ttf, Calculation of flexibilities, 1a sey ur ote iv, Sumetion and/or Integration. Here again note the general nature of the terms "Load" and "deflection’. (See 9. 47.6) ‘Tae following examples apply the method of dumay-untt loads to the determination of ab— solute and relative ceflections, both rotation and translation. fxample Problem 15 ~~ The pin-jolited truss of Fig. A716 is acted upon by the external loading system shown. ‘The member loads are given on the figure. Mem der properties are given in Table A7.3. Find the vertical movement of joint G and the hortzontal movement of joint #, : 10908 900% 10008 Fig, ATI6 Solution: only the energy of axial loadings in the aeubers was considered. Unit (virtual) loads were applied successivaly at Joints G and # as, show in Figs. A?.16a and A7.ieb. All § and u loadings were entered tn Table A?.d snd the careusation sor 9 = f SESS as carr out by Sulyaz tems tor the members sul, forming the sun of of the truss, f.8. G=E An tie ray" loads Fig. AT. 16a, le ee |e | a [a aso|-a rt “a 7a [| ae aaah Ta cafe] na eTepPle 2 aE [0] v0.18 of am |e ao to) 3a [aw fe a fo] aie cnepe feo = oy fe] sere nee} e fe] © 7 ca [ep ame cote a] @ z be lw] nove zoe) efoto . Le mea ote. answers 0,286" eR 4, oR the Joint moves to the LEFT since the unit load was dram £0 the RIGHT in Fig. 47.169). = -.0587" (the negative sign means Sxample Problem 14 ‘OF the tPiss of Fig. 47.16 find the fol~ toning relative displacements of joints: ¢) the movenent of Joint ¢ in the direction of a diagonal line joining ¢ and F. 4) the movenent of joint G relative to line Joining points ? and Relative deflections are determined by applying unit (virtual) lads at the points where the deflections are desired and Dy support- tng Such unit leed systems at the reference points of the motion. Thus, for solution to part (c) a unit load system was applied as Shown in Pig. A7,16¢ and for the Solution ofAra part (4) the system of unit Loads of Fig. A7.16d has used, Table A7.4 conpletes the solution, the real leads and member’ flexibilities L. (4) seine the sane a2 for exaaple provian 15. DEFLECTIONS OF STRUCTURES Rotations, both absolute ana deternined by azplying unit to the menver or portich of struct: rotation is desired, The unit couple ed by reactions placed cn the line of for the rotation, Thus Pigs. i7.18e ena 27. show the unit (virtual) leadings fer zerts (2) and (f) respectively. Table a7.5 completes tne Galculation, the real Loads and member Plexi: tt1es 7, oo (U/gq) Pens the sane as tor axanple provien soe 1/40e_oms 028.0255 Fig. Atte Fig, aT.6 4 Y jo Ties OS = OE CNY 80 TABLE Ane "ee" loads Fig, AT. Abe Fig. a7. 366 le a ‘Sea x 109|Saek x 169 sea Sp, fee jee | ae eae = snl af -e |e a . = aao|-e | am] aa] ae : Ey azo] 2 ve TABLE a oF ca.m80] 9 @ 76 rato] @ ene wl Sab ete seen iB] He [samc sacnne] Sa 0x e@ ope 2 ae aE ae) : rere te) a 2.00 Lo ao me | sme = Lae 2s 2 -B oD xi 38 a oF a ae or 5.591 7 o so id a ro $.501 73 o 2m 2 ia oe 3 3 poem re || Cor > 7 a | ee a Toe ba | oat |3m0| 05) 0 | 8 | 9 ar | oa [ae] e |e |e @ ca] inna fam) spe) ° Tuerefore the aovenent of Joint ¢ tomrés | Poo 2 joint F was 6 = .06587 inches and the motion of | a I Joint G relative toa line between F anc H was 2 6 = ~.0194 inches, the negative sign indicating an uplard movement. ‘eample Problem 15 For the =riss of Fig, A716 deterstne 9) the absolute rotation of member DG 1) the rotation of member BO relative member °¢. Therefore the absolute rotation of member DG was @yq 7 00475 radians and the rotation of BO relative to CG Was @,, _ oq = .0O0SS radians.ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES ATI3 Beemple Problem 15 points 2, D, 2, the attachment points of an SOREMME tne vertical deflection of point ¢ for| aileron or a flap, The wing beam deflection the cantilever Seam of Fig. A7.17 carrying 2 bends the aileron or flap structure by applying ThesGevated icas Pat its ene. aleo find slope | a lead at D thru atleron supporting Prackat, To or elastic curve at 0. faow this fore the deflection of the wing dean 30) Selavive to Line CB aust be nom. Solution: a Soutien: er Origin at - ite origin at 8 x N=30x+ 35515 x ta Pe (Fig. 47.27) : For virtual loading (Fig. A7.278) os m= 0, for x0 Go" ae 2p 10h aot msl (x - 0}, for xb Fig, AT.18 i mace Made = - Px(-x + Didx = (Px* - Pox}dx met ny L L P 2 2 BP ee _oxt = os menee 39 = {, (xt = velox al x. L ra ati 4! ‘ zl But , b* m= 0, x< 20 * 31 ~ = z a = -.5 (x- 20) =-.5x + 10, whanx = 2000 SO m= -.5 (x-20) +1 (x 50) = .5x-40, when If > = gero, then 6, PLY/SE x = 50 to 80 *ocee) * [SE . c are en co ye i a Feary + (2 +2 4 ¢ . i; 1S xt (-.5x + l0lee + 1s xt Borin ie leo B Jeo 3 (5x = 40) oe poo ari ait _afese, wor!) 2 fzsxt _ soos]? “EL € s 4 s &= 3 [nave + 5.25 + 300 - sco + sor sistant Tosaing a0 F'8. A7.17B 72.9 ~ 100 = 11.72 + 28.0] a= 0, xb, B= 1, 00 6 set dence Maax = fex)(-L) ax = Px ox a meratore detlection of point D relative to Line joining 22 ts down because result cones out nega~ (us = bt) Sive and therefore opposite to directicn of tual load. Example Problem 12 Pm Tr L xax = ~~ For the unizo aly loaded centilever beam 2. 47,18, find the deflection of point D lied im diag. ive to the Line joining points © anc & on eo eer | the alastic curve of the Dean, This {8 repre- 7 i Rati Bagge | RAYE sentati gi proviem in saronsutics, . . couple in th far Wing beam and ow o Fig. A719 Fig. A7.20ALi DEFLECTIONS OF STRUCTURES Pind the horizontal deflection of point c for the frane and loading of Fig. A7.19. A150 angular deflection of C with respect to Line OD, Solution: Fig. > shows the statie moment curve for the given loading and Fig. ¢ the moment diagran for the virtual loading of a unit horizontal load applied et C and resisted at 0, 2 (ME; eM lt aes coma) * [FER = SE 2 FoRee ‘wx? nh = may: To find angular deflection at c apply a unit imaginary couple at C with reactions at C and 0. Fig. 47.20 shows the virtual a diagran. oe 2 /.- -) Ee af ‘4 «-a(e-e] - ‘Linear Deilection of Beama Due to Shear by Virtual Work, Generai2y speaking, shear deflections in beams are suall compared to those due to bending except for comparatively short beams and therefore are usually neglected in deflection caleulations. A close approximation {s sometines nade by using @ nodulus of elasticity slightly less than that for bending and using the bending deflection equations. The expression for shear deflection of a bean is derived trom the sama reasoning 2s in previous derivations. The virtual work equa~ ‘ton for the hypothetical unit load system for ¢ shear detrusion dy (Fig. 47.21) considering only dx elastic is 1x S-vdy wnere v is shear on section dus to unit hypothetical load at point 0, and dy is the shear detrusion of the clement @& due to any given load system or any other cause. Fig. A721 “ 1 instead, See p. AT.4 dy = Vac and $= 5h, more 4 = cross sectional ares of bean at section and 25 = acdulus of ptgteteys and assuming that the shearing stress f ts uniform over the crass-secticn. Therefore 1 x 6 = WE Es for the shear slips of all elenents o: equals ‘Then the total deflection bean the —~ta) L 8 total [ ee , where Y is the shear at any saction due to given loads. v = shear at any section cue to unt) hypothetical load at the point where the ceflec— tion {s vented and acting in the desired direction of the deflection. ‘The reactions to the ayp- thetical unit load fix the Line of ence for the deflection, 4A 1s the cross-sectional area and 35 the modulus of rigicity. Equation (a) 1s sitzhtly im error as the shearing stress is not untfora over the cross-section, e.g. deing rarabolic for a rectangular section.” Konever, = shearing strass gives close results. For a uniforg load of w per unit center deflection on 2 sidply supported beam is: - tL ® seencar = 2 |B WS ee PL is A o AEs wL* aE Por bending deflection for a stuply supported beam uniformly loaded the center deflection is Swit . sant Hence wi" Ey . ie » using Ey =. Seer, P= radius of gyration. For t veuns ang chamois ris approxizatety } 4 and for rectangular sections t= ay lay geANALYSIS AND DESIGN OF PLIGH ant ote Oy = HO + 1.4) Px 10 a . a SE Too ooo ms enteferae 247 (1600) 10 tn = 265 in. hye Ga, y t A Detietons Dee to Thermal Strains. Y LU AS noted in the "virtual work dertvation" | 4 . pire of the cumy-unit load deflection equations, the real internal strains of the structure may be due to any cause including thermal effects. Hence, 2rovided the temperature distribution and theraal properties of 2 structure are imowm, che gumy-unit load method provides a ready means for computing thermal deflections. Buample Problen 22 Find the exis) movement 2t the free end of @ uniform bar due to heat application to the fixed end, resulting in the steacy state tem~ perature distribution shown in Fig. 47.28. As~ sume mater{al properties are not functions of ‘temperature. ‘7 = temperature above bient tempera Tob ToveTo Gotan thre = - k= an empirical con- une stant depencins —t upon thermal prop~ prttes ang rate of Fig, ATI heat adgitimn. Sotution: ‘The thermal coeffictent of expansion of the rod material vas ¢, Hence a rod slenant of length éx axpertenced a thermi deformation Q=sa-+T+ dx, application of a unit load at the bar end give 4-1, Therefore 238 undergoas rapid “eating af th: upper flange to a cenperature T, uniform span= sige, above thet of the lower flange. Deter- mine’ the resulting displacement of the free end. Solution: (The axial deformation of a differential elenent of the upper flange (subseript U) was assumed given by dy = aT dx where @ was the material thermal coefficient of expansion. ‘The lower flange, having received no heating undernent no expansion. Inasmuch as a thermal expansion is uniform in all lrections no shear strain can oecur on a material element, Hesce no shear strain aceurs in tho wed, The apparent anomaly here = that wed elements appear to undergo shear cstoraatons ¥=3E5 (rg, AY. i06)- exlaind at fallow, Pe eopefanre varies lei oes he Seam ‘alee peafots mica tea ber as cree Seth itnnews shy nary an ate mane igi graye spect sear stration. No So eee crc ee Sohne Lee ‘Sak the ecettion ot a uate (virtual) toed ta the free snty ie WiréaL loadings 02> ‘Stnet sn tuo Flanged were! Lex By FE 74 then che deflection equation nas os fous faaclE vata-fo ams oat Byenple Problem 24 The first Step in computing the thermal stresses in 2 closed rine (S tines Indeterminate) Involves cutting the Ting to make {t statically determinate and finding the relative movement of the bwo cut faces. Fig, 47,80a Shows a unifora ctroular ring whose iasize surface {s neatgd to 9 tampsrature P apove the outside surface. The temperature is censtant around the circumference and 1s assumedAns DEFLECTIONS to vary linearly over the depth of the cross section, Find the relative aovenent of the cut surfaces show In Fig. 47.30. 1 att Mh a 2 ® . Fig. AT.30 Solution: An element of the beam of length Red te shom tn Fig, 47.30¢. Due to the linear t perature variation an engutar change 40 2a? 4p occurred tn the slenent. The change in length of the aidline (centroid) of the section was 4 = 83% ag. unit redundant loads were applied at the cut surface as show in Fig. 47.30) giving the following unit loadings around tne ring. From unit recundant couple (x} m= 1 (m positive if tt tends to open ‘the ring). axial load) = 0 (positive of tensile) He From unit redundant axial (horizontal) load (¥) ay = -R (1 ~ ces 6) uy = 608 > From unit redundant shear (vertical) load (3) me -Rom$ uy zo sin Te deflection squation by the dumy-unit lead method is os furacfan. ao Ten 2n Bog a on BE seo fos = [FRB ob = %ye [= oR {6 A(i-ces 4) BE ap x - SURAT (negative indicating movenent to ‘the right} STRUCTURES 8 [Fs 4) Bg E ein 6) SE ab Remarks: In the three elementary examples given above no stresses ware developed Inasmuch 2s the idealization yielded statically determinate structures which, with no loads applied, can have no stresses. Indeterminate structures ara created tn Chapter 4.3, AI.9 Matrix Methods in Deflection Calewlations. Introduction. There {3 auch to recomend the use of matrix methods? for the handling of the quantis: of data arising in the solutions of stress and deflection calculations of complex structure ‘The data is presented in a form suitable for use jn the routine calculetory procedures of high Speed digital computers; a flexibility of opera— tion 1s present which permits the solution of sd~ ditional related probleus by 4 simple expansion of the progran; The notation itsel suggests new and improved methods both of proach ane work division. ‘The methods and notations employed here and later are agsentially those presented by Wehle and Lansing? tn adapting the Method of Dumy~ Unit Loads to matrix fotation, Other appropriate references are listed in the bibliogranay. BASIS OF METHOD Assume the structure to be analyzed has begn idealized into a truss-lixe assenbly of rods, bars, tubes and panels (sheets) upon which are acting the external loads applied as concentrated loads fy or Py, each With a different numerical a @) e) Fig. A7.31. Idealisation into an assembly of bars and panels. ‘For the reader not famillar with the elementary arithmetic rules of matrix operations employed nere, a abort appen~ Gis mas been ested. SL. B, Woble Jr. and Warner Landing, A Method for Re- ing the Azalyeis of Complex Redundant Serato tS a (outine Procedure, Journ. of Aero. Sciences, 19, OctoberANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES subscript. Thus the system of Fig. A7.Sla is idealized into that of Fig. 47.313. With the above idealization an improved Scheme may be employed to systenatize the con- putgtion of deflection calculations. Tha following steps sumarize the procedura which {5 discussed in detail in succeeding sections, 1. A set of internal generalized forces, denoted by a1 or q3 (2, J are airfarent mumeri- cal subscripts), 13 used to deseribe the internal stress distribution. The q’s may represent axtal leads, moments, sheers, etc. In con Junction with a set o¢ member tlexibiizty coor- Tctents, a,4, the q's are emplaved to cesrass fhe stdin endrey U. ats sivesythe displacenent of point f per untt torea at pofat f+ It, Bquiliorium conditions are used to relate the Internal generalized forces qt, Qj to ‘the external applied loads, Pq or Py. With this relationsnip the strain energy expression obtained in I, above {s then transformed to sive U as a function of the Pra, TIE. Castighiano’s Theorem Is used to compute deflections. CHOICE OF GEVERALIZED FoRcES Consider for example the problem of writing, the strain energy (of flexure) of the stepped santilever team of Fig. A7.S2a, assuming external loads are to be applied’as transverse point loads at A ané 8. The set of internal Generalized forces of Fig. a7.32 will com pletely determine the bending moment distribution in the beam elements and hence the strain onergy. Set (b) then {s 2 satisfactory choice of generalized forces. It should be pointed out that set (b) ts not a unique set. “Other satisractory onoices (not an oxheustive display) are shown in Figs. 7.320, 4 and 2, The Pinal selection may be wade for convenience or personal taste. Note that only es many generalized forces are used per element as are required to ceter- aine the Significant loadings in that slenent. Ee a mp oS io = Tt ha) a on ate =" to as i) a Meax Oc aw + — M=qs+qay Oy placing unit loads successively at external loading points 1, 2, 3 and 4 (Pig. a7.360). he results were collected in matrix form as inches ‘The results here give the deflections of all four points. Since only the deflections of points 3 and 4 mere destred the first two rows a i ney be areppeg out. Te sumo rest could have been achieved by leaving out the coe oso ig] (ae meee o fe }) ‘The matrix form of squation above ts useful in organizing the computation of deflections for a nuuber of different loading conditions. Thus, Should there be several diferent sets of external loads P,, corresponding to various load= ing conditions, each set ts placed tn column form giving the loads as the rectangular matrix [x]; k different maerical subsey‘ipts for the load conditions. The matrix producttions at each point (m} for fonditions (it). Skanple Sroblea 27 “SNpetlections at pcints 1, 2, 3 and 4 of the truss of 71g. A7.25 are desired for the follow ing loading conditions: AL23 Le L aus = Gan = AB + Gan = AB a oo Lu L a= a= BO + Sey = BC he os Le Sen = BC , Et gengttton No. |p, | Ps | Py | Py (see Fie. Coltected tn matrix torn (noting that t [este [ss | 8b [ 850 47.386) Bs 2 [ =izoc |-200 [-2i00 [#1760 e 3 [zee [taro [-is00 [1360 TeTs>s[s]s ley Solution: Teapepet ep eye yey z]ofsefo| ej e\ala| o Tha uetrix product formed per eq. (260) was jo} else} 0 | ol ala} o set up 2s slo [ oo |ssco| ae} 0 | | 0 = sjo\ajo| sso) 0 | 0 | 0 0 369 287 380] | 2500 -1200 1800 elojalo, o | |seo| a | oe 2 - So - n00 14% nlolefet © fol} o wel o = — _ , [829 927 282 769] [2000 - 200 470 ee ee ee ae Sax] * | 287 282 257 282|| 800 -2100 -2200 seo 7a9 282 769] | 450 -1750 -1100 Zeanple Provlen 2e For the iefiging gear unit of Sxeaple Prob Jen 20, Fig. A7.23 find the matrix of influence coertictents relating ceflections due to Litt and drag loses acting 20 point A and torque out the axle a8. Solution: The structure was divided into and che set of internal generalized forces ap plied es shom tn fig. a7.a7a. (Torques and oments are shou vectortally by R.i, Tues). stresses wera neglected in 0-8. elements Bae ~ |p at ae ae x Nar SAG AB as Fig, A137 Fig. A737 The following member flextbitity coeff ictents here deternined The anit toas atstrsbattons [G,,] were obta:nes by applying unit external appited loacs, numbered and directed as in Fig. 47.37. El: Ab this point problem as solved, tsa routine opera! EE. axanpie Pro! Seam of example sroblem 21 is to be solved by the matrix methods vresented nerein. the engineer zay consider th for the renatning computation tonsA.38 Influence are to be coefficients for points F, d and H found, so2ution: Fig. 47.38 shows the chotee and numbering of generalized forces. a fag t Sag Set ef metry i } ——— ff ay i—— Fig. A738 No forces were shown applied to the lower flange elexents as these were imonn to Ue equal to those of the upper flange due to symmetry. Gutries were made for a,, in matrix form as be~ ny low. were quadrupled as Mntries for agg and ag, these occur in tuo identical embers each on top| and bottan, intros f6F Wee) ee aie Se wore doubled. (See Art. A7.10 for coefficient formulae.) Weel rT 2) ss ; ss Bae: : ss ts Untt load values were obtained as in Pig, A7.27, considered to be external loading number "iva", Similar dtagrans sere drawn for unit loads at points ‘one™ (H) and "three" (Flice rig A 1-25) Rete ts afi ° | 3 2} .06e7 ° s}aea | o | o [és] sl] 0 [10 ° s| .ce4o} 0600) o 6) 2a lic | oo 7) 0 0 fio | .o17| ose | .cses 2 | 3.00 | 2.00 | 1.00 DEFLECTIONS, OF STRUCTURES ‘The matrix trip: product, PoE completes the calculation, Scriond of statically indeterminate seructuras often say de computed suscessfutly by the aethoda of this chapter provides shee sens auxiliary means is eupleyed 9 obtain an approxisation 20 che true invernal tore dis~ tribution. The exact internal force aistrisue Sion is not necessarily required in making do- flection caloulactons inasmuch as such a calcu lation anounts to an integration over ta structure - an operation mnicn tends to average out aay errors. Taus one my use the angineering Sheory of bending (8.2.5.), experinontal data, Freviows experience, sve, to obtain reascnaele Setinates of the Internal fore distr ‘bution unit loadings. In the follontng problem the aatrix of in~ fluence cossficients 12 deterainsd for a Siar cell, three-day box deen (S tines indeteraizate) by sling the 2.7.8. Pig. 47.360 stone en idealized doubly symetric single cell cencilever Sox dean sayin three days, Determine the setrix of influence ceetficients for the six point nat indicated. Constant = 32 in ® Fig. A739 Solution: Pig. A7.39> ts an exploded view of the beam showing the placement and numbering of the in= ternal generalized forces, Note that only tae upper side of the beam was numberad, the lower Side being identical by symetry. Menber flexibility coefficients were com puted by the formulas of art. 47.10 and entered in matrix form as below, Note teat all entries for which there were corresponding Leads on the lower surface of the beam were doubled. 5y this aeans the total strain energy of the beamANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES SN, was accounted ror, Note also that entries for Gee, Gans Gee BNE Crosse Were Te-doudled as each of these q’s act on two (Identical) men ders. Fig. AT. 296 MCLG GG [elelelelela | ahs i Pp al | clay ; E M4 Pn Note: VOID SPacus moreats os. Unit load distributions were sbtained for guccessive applications of unit loads to points one through six (Fig. A7.d0a}, Ths internal, forces predicted by the 2.7.8, for a load Through the shear center (center of beam, due to symmetry) were superzosed on the uniform shear sta fs le ous | o.0s 9.05 | -0.05 cs | ous ws | 1s Ls | Las sms} crs] 0267 | -.0287 0179} os 2of2 fa |e azote oj: dk Anat tion (< +B) cue 20 tke torus cevetored ta transtecriae tke 2006 to on sice, ‘The matrix triple product Ce) &I EE] AT.12 Detlections and Angular Change of the Elastic Curve of Simple Beams by the "Method of Elastie Weight. (ohr's Method}. In the caleulation of structural deflec- ‘ions there occur many steps involving simple integral properties of elementary functions. ‘The Method of Elastic Weights (and the area Moment Method to follow in Art. A7.14) owes tte popularity in large measure to the fact that it enables the analyst to write dom many of these integral properties almost by inspection, rely- ing as it does upon the analyst’s fanilierity wath the properties of simple geometric figures. For finding the deflection of a point on a simply supported beam relative to a line joining She supports, the Methea of Elastic watehes states: Tue deflection at point A on the glastie curve of 2 Simple beam is equal to the bending mount st A due to the 7 diagram acting ag 3 distributed bean oad. Spelled out in ste t= tas % ciagran 19 cram just as * occurs due to the applied bean icad ii = This diagram is visuaiized as being ‘he loading on second bean (the conjugati beam) supported at the points of reference for hs deflection destred iii - The bending monent in this conjugate deam is found at the station where the deflec- ‘ion of the original beam was desired. This bending moment 18 equal to the desired deflection, To prove the theoren, consider che dumy- unit load (virtual work) Squation Biceaccoee BR, Tuts expression oquates the external virtual work done by 2 unit load, applied ab a point dew fecting an axount 5, tothe internal virtual work on a beam elenent axperteneing an angular change ¢ 6 = "SX. the sum (integra) of such expressions throughout @ beam gives tha totalaq. 12), we now 2 exsrasston, using > Gane a2 th simple beam loaded by uation, 15 © momont expression cor + Max or: In Pig. AT-40, the loading of (a) produces the real moments of (>). Consider the Zeflec~ tions of points 5 and C due to tna angular change HEX chenge an “elastic wetgnt” tna beam element at At (Fig. 27. Py te (4) mp, moment diagram for unit Toad atzing downward at point B 3 du (e) mg,. unit toad at point © Max. Ler 3 Max o Mee aa ar Fig. AT.40 For a unit load at point b, Pig. d shows the m disgram. The value of m at the midpoint of dx (point a} = 1/8. Hence oy = SE Es ee Sr 8" 53 For deflection of point c, draw m diagram for 4 unit load at C (see Fig. 9). Value of z on ele pont ax = b Hence ‘For sumpitcity the points A, Band C were piacea at the one- quarter span points, The reader may satisty himeelf with the general character of the proof by substitating Xa, Xp ard Xe {or the potat locations and then fallowing through 2he afgument once again. DEFLECTIONS OF STRUCTURES 1g. f consider vex as a load on ply supported at potats ‘Tasse values of points > and ¢ are identical to che deflection: ac Dando Dy the virtual work equations. Th noment diagran 2 for a unit load at > and ¢ (Pigs. ¢ and e} {8 mmerically prectaely the as the infls: ‘Therefore ae de determined by considering ti ‘Lections of 2 siapie beam can ‘Mf curve as an = ‘The dending zonent at loading equals the de~ imaginary bean loading. any point due to this a flection of the teas under the given loaas. Likewise tt ts eastly 2roved that the angus lar change at any section o? a simply supsoccen beam is equal to the shear 2t tnt secuion cue to fhe E Glagran acting as 2 seme toad. A712 Example Probleas Example Pr: a. Getlection and slope of points a and > for and loading shown in Fig, A7.41. Tae lower shows the moment diegraa for load Paci center of 2 simple beam. FI Constant Fig. ATL PL? PL a i 18 Deflection at point 2 equals sending zon ent due to M diagram as a load divided by ST. (See lower Fig, of Fig. A7.ai) 2 2 = (2 bw oe (FE a 2 2 =( bw (5-8 ‘The angular change of any point equals the shear due to M/SI diagram as 2 load. wf e)a . se - at(B-8 le are(Slope ts hortzont~ al or no tron orig rection af deem axis.) Determine tha Loaded untfornly as shown in Dending nogent exxression tor @ simple team ATA. The a uniform los: tt = wiz = wae ar parabolic as 2S shown in 81g, APS2a, The deflection ae mid potnt equals gram as a load. Sending moment due to M dia me eo Tr zr Fig. Ata M Diagram Le Awe = area z are: Fig. AT.423 Scenter a aS slope ac ae Slope at supports a reaction ample Problem 33. S Fig. 7.48 shows the plan view of one-half of a cantilever wing. The aileron 1s supported on brackets at points D, & and F with self-elian-| ing bearings. The Drackets are attached to the Wing rear beam at points A, Band C. when the ing bends under the 2ir load the aileron aust Likewise bend since {t 1s connected to wing at In the destn of the 2ilersn beam for cases of wihg flaps this deflection produces critical bending moments. As Suing trat the running load distribited to tre rear beam as the ming bends 2s a unit is a shown) in the Pig., tind the deflection of point 3 with respect to straight line Joining 2oints A ard c, anion wi21 de the deflection of E with respect fo) Line joining D and F if bracket deflection 13, neglected. The moment of inertia of the rear beam between & and C varies as Indicated in ne Table A7.6 An.20 |¢, Airplane Plan View 1/2 Wing 0 ie toad cee PP] Zand on Rear Dea ¥ ase Fig. ATS Solution:- Due to the beam variable monent of inertia the bean length between A and C will be divided {nto 10 equal strips of 10 inches each. ‘The bending moment Mat the midpoint of each will be calculated, The elastic welzht for each strip will equal Ns, where ds = 10" and I the moment of inertia at aidpoint of the strip, These elastie loads are then considered at loads on an fm- aginary beam of length AC and simply supgorted at And C, The bending moment on this imaginary deam at roint B will equal the deflection of B with respect to Line Joining ac. ‘The bending moment at c= 15 x 30x 15 + 10 x 15 x 10 = a250"# The shear load at ¢ = (18 +25) x 30= 600% 18-28: Bending moment expression between points C and A equals, M = 8250 + 600x + 12.5x2, where x = 0 to 100. ‘Table A7.6 gives the detatled calculations for the strip elastic loads. The I values assumed are typical values for a alumimm alloy dean carrying the given load. The modulus of elasticity £ = 10 x 108 ts constant and thus can de omitted untt] the final calculations. The figure below the table shows the elastic loads on the imaginary bean. 3 3 4 1 lo SIT DWE SITIOS ‘299080A.30 Bending aonent at point 8 due to above elastic Loading = 7,100,600 .". deflection at B relative to line AC = 7,100,000 = .71 ine 510,000,006 Sample Problem 34 ig. A7.4Ga shows a section of a cantilever wing Sea plane. The wing beams are attached to the tall at points Aand 3. Due to wing Yoads the wing will deflect vertically relative to attactment points ag. Thus installations such a8 piplng, controls, etc., gust de so located as not to interfere with’ the wing deflections between A and 8. For {llustrative purposes @ simplified loading has been assumed as shown in the figure. ET has been assumed as constant whereas the practical case would in~ volve variable I, For the given icading deter~ mine the deflection of point ¢ with respect to ‘the support potnts A and 8, Also determine the vertical deflection of the tip points D and &. Bt 3 8 arr beets 4 (ii ET Constant Moment Diagram Fig. AT. 43a] ! Big. 7.430] 1 Fig, AT.4e | g = Fig. AT. 40d] solution:~ diagram for the given wing loading. deflection of C normal to line joining AB we treat the moment diagram as 4 load on @ imaginary’ deam of length AB exd stuply supported at 4 and B Fig. 7.43 shows the bending monent To find tne (See Sig. A7-4c,) The deflection of ¢ 1s equal numerically to the bending moment on this ficti~ cious beam. Hence B10, = 25920 x 40 = 25920 x 20 = 518000 or 8, = S10 To find the tip deflection, ne place the elastic loads (area of moment diagram) on an imaginary Deam simply supported at the tip D and 5 (See Fig. A7.4Gd). ‘The bending moment on this imaginary Dean at points 4 or 3 will equal mumerically| the deflection of these pointa with respect to the tip points D and B and since points A and 8 actually do not move this declection will be the movement of the tip points with respect to the deam support points. Bending Ronent at A = 190420 x 700 - 40000 x} 433 - 127500 x 124 = Loze00000. .". Seip = 102,200, 060 =r DEFLECTIONS OF STRUCTURES AT.M4 Deflection of Beams by Moment Area Method* For certain types of beam problems the meth 04 of moment areas has advantages and this metn- od is frequently used ta routine aralysis. Angular Changs Prinetple, Fig. A744 shows 2 cantilever dean. Let it de required to deterine the angular change of the elastic line be~ ‘teen the points Aand B dus to any siven loading. From the equation of virtual work, we have Fig. AT. 44 Max Max. where 2 is the moment at any ire off mug teres wie Sarre Seto Se ssa ne bE Therefore ag fC Mex ‘3 ET career let re 8 en samaANALYSIS AND DESIGN OF PLIGHT VEHICLE STRUCTURES Referring to Fig. A7.4, this exoresston sents the area of tI re eM disgrem between points Sand A, Thus the first prineiple:= "The ange in glope of sie slsstis ine of = seas ipieey ay Se pole a an Efe nme Sqal isthe arsi or the 4 Cideras vemteen = Siiares etneen vhese two points." Deflection Principle in Fig. 47.4 deterine the detection of point 3 hora: io tangent of elastic surve at A. In fig, Aves this deciection would be vertica! Since tangent to elastic iine at A 13 horizontal, = ate aa snere 3 is the aoaent st any section a distance yitroa 8 cus to @ unit Aypotnetical vertical oeg acting av 3. Hence © point becieen B and A. Prom virtual work expression 65 2 fence AH mats exoreseion represents tite lee aonent of oe M diagram about a verticel thru 8. Thus the deflection principle of the monent area aethod can ve stated as follows:~ ‘he deflection of a joint Aon the elastic Line Of 2 beam in dendl ot tetas aero te elise Tasers point 5 1 equal numerically to the statical go- dest Of the M afea betneen points TAT and 7B" bs about point A". HLlustrative Problems ‘Example Provlen 35. Determine the slope and vertical deflection at the free end a of the Bheleover bean anew ta Figs ayeie. SY te cone sean: : js ___, rig ata® ~ eM biagran “QE 2p —— solution: The noment diagran for stven toad ts Suinguiar as tnam tk Figs Arete, since one beam is fixed at A, the elastic line at a ts horizontal or slope 1s zero, Therefore true slope at 5 equals angular clange between 4 and 2 mnich equals area of ament diagram between A 2 (pL. We) B= ‘he vertical deflection at 8 is squal to the lst onent of the Regent dlagrem about point B di vided by 81, since tangent to elastic curve at A 4s horizontal axe to Zixed support. Alsi Hence az 2 Je “ata Example Problem 36 Fig. A7.46 {Illustrates the sane simplified wing and Loading 2s used in example problem 34. Find the deflection of point C norzal to line Joining the support points A and B, Also find the deflection of the tip points D and Z relative to support points A and B. eu ~ SI 2 3 3 soiution:: ‘Due to symmetry of loading, the tangent to the deflected elastic line at the center line of airplene is horizontal. Therefore, we will find the deflection of points A or B anay trom the norizontal tangent of the deflected dean at point ¢ nhich 1s equivalent to vertical defiection of C with respect to line AB. ‘Thus to find vertical deflection of A with respect to Rerizontal tangent at C take moments of the M diagram as a load between points a and Bs ¢ about point A. finenee (area) 9 2 (bores : tidy (tangent at c) = & (S25S) co x20 2 (s1e100) = ceflection of ¢ nornai to AB. po find the vertical deflection of the tip point Dowith respect to line A, first find deflection of D with respect to norigontal tangent at C and subtract deflection of 2 with respect to tangent atc. 8 : zk 7 sal (respect to tangent atc) = gy (40000x 267 + 127500 x S76 + 25920 x 720) (see Fig. a7.46 tor gran). Subtracting spect to ¢ as found (202700, 000) areas and aras of 1/31 ata~ ‘tho deflection of A with re- above ve obtain 25. (respect to line AB) x 518400) = 7 {102,180,000} & cace,r00000 -Ale AT,15 oan Fixed Sad Noueats by Method of Ares Yoxents From the tea principles of area mozents given tn Art. A718, 1t ie evident that the deflection and sloze af the elastic curva dasend on the amount of bending moment area and its lo~ cation or tts center of cravity. Fig. A7.47 shows 2 bean fixed at tne ence and carrying 4 single load P as shown, The dend: ing monent shown in (c) can be considered 2s Bade up of two parts, namely that for a load P acting on a simply supported team which zives the triangular diagram ith value Pa (L-a)/L for the moment at the load point, and secondly a trapezoidal moment diazran ot negative sign with values of My and Mg and of such mgnituce as to make the slope of the beam elastic curve zero or norigontal at ths support points A and B, since the beam ts considered fixed at A ana 8 ‘The eng moments Ma and Mp aro statically indeterninate, however, wita the use of the tho moment area principles they are easily aetermin- ed. In Fig. > the slope of elastic curve at A and 8 is zero or horizontal, thus the change in slope between A and 8 {s zero. Sy the 1st > 2 erconnan_pf — 1. 2 Ane a A cient Fig. AT.AT 2 = i = principle of area moments, this means that the algebraic sum of the moment areas between A and B equal zero, Hence in Fig. ¢ 1) Le e-a) bag ae cE In Pig. b the deflection of B anay from a teng- ent to elastic curve at A ts zero, and also deflection of & anay from tangent to elastic curve at B ts zero, ‘Thus by moment area principle, the aoment of the moment diagrams of Fig. ¢ about points A, or 8 te zero. ‘Taking moments about soint A:- By = Pa 8(L-n).25+Pa (Loa) *x(a + Loa oMgl x ta > Pala) ZerPabea) Mak Se oS Bpixe L=0 3 Solving equations A and 8 for My and My DEFLECTIONS Of STRUCTURES =~ Bab? and Mp = = Pra*/L* where b= (L~a) Tr To find the fixed end moments tor a team #Ith variable moment of inertia use the t/I diasrans in place of the nomen diagrens. FIg. Av.a8 shone 2 cix-ended seen two consentrated loads. Pind the fixed-end moments Mq and Mg. 2008 100 i I tip @) o Fig, AT. 4a Solution:- Fig. D shows the static moment dtegran assuming the bean simply supsorted at A and 3. For simplicity ta finding arees and taking moments of the ndnent areas the moment dlagram nas been divided into the 4 simple shapes 2s shown, Tae centroid of each portion is shown together with the area watch ts shown as a concentrated load at the centroids, Pig. © shows the moment disgrans due to un- wmonn aoments My and Mg. ‘The area of these tri~ angles is shown as a concentrated load at the centrotas. Since the change in slope of the alast: curve between A end B ts zero, the area o: moment disgrans must equal cers, nence S265 + 14040 + 2360 + SEES + 1EMy + 15M =O. or LStty + 18h + 2850-0 The deflection of point A anay crom tar astic curve at 8 1s zero, cherefore the aoment of the moment diagrams about point A equals zero. Hence, 5265 x 6 + 15 x 14040 + 17 x 2260 + 24 x 365 + ISCHA + 308 Mg =0 oF 150 Mq+ G00 Me + ansg00-O- - - ~(2) Solving equations (1} and (2), se obtain Mg == G16 in, Lo. My = - 1074 in. lbs. With the ond moments xnown, the ual. Thus for deflec- vion of Joint 0, apply @ unit vertical loed act~ ing down at joint 0. The stress m in bar be due to this unit load = 2. 2b = 4P . Tharecore aor & Sof tbye . 4P - The deflection at other H " lower chord joints could be found in a similar aenner by placing a unit load at these Joints. Diagran (2) shows the resulting deflection curve, ‘This clagram 1s plainly the influence line for Stress in dar be multiplied by Abye. Diagram (3) shows an Imaginary beam loaded with an elastic load glee acting along a vertical line thru joint 0, the moment center for ob-, faining the stress in'bar be. The beam reac~ tlons for this elastic Loading are also given. Diagram (4) shows the Daan bending moment diagram due to the elastic load at point 0. Ib ts noticed that this mouent diagran 1s identical to the deflection diagram for the truss as snovm in| + dtagran (2). ‘The elastic weight of 8 member 1s therefore equal te the zenber deformation divided ty the arr to its moment center. If this elastic load is applied to an inaginary beam correspond- tng to the truss lover chord, the bending moment on this imaginary beam will equal te the crue truss deflection. Diagram 8, § and 7 of Pig. A7.49 gives a similar study and the results for ¢ aL Langthen- {ng of member OK. The stress coment center for This diagonal zeaber Lies at point O, wnich Lies outside the truss. ‘The elastic weight aL at FL point O' can be replaced by an equivalent systen| av points 0 and < on the tmaginary beam as shown im Diagran (6). These alastic Loads produce a bending sonent dtagram (Diagram 7) identical to deflection diagram of diagram (5). ‘Table 47.7 gives @ sumary af the equations! for the elastic veignts of truss chord and wed nenbers together with their location and sign. o —e4 ‘S Panels @p ‘ sar te a linn ara weg tte ara ae Tenor oe Elastic Load on Imaginary Beam fr Abbe 1 (3) EE Siemans tain ae ber ck at{asto) 6 ae Fao ae See a ae 6 Fig. A749 ‘Table AT.7 Equations for Elastic Weights Elastic-Weight for Chord Members (See Member ab) Lower Chord wel Upper Chord is Member Amal Deformation __ | aL, ” “Perpendicular Arm to Moment Canter" ¢ ‘The moment center 0 ofa chordmember fs the intersection of the oter two menbers cut by the section used in determining the load In that aemb= jer dy the method of noments ‘The sign of the elastic weight w for a chord member {s plus 1f 1t tends to produce dcwnward deflection of its point of application. Thus fora, Simple truss compression in top chord or tension in botton chord produces downward or positive Ab 7 AES: elastic weigntana DEFLECTIONS of sTRUCTURES Twaneane 10003690 0001080 Table at. WEB DIAGONAL s sit (continues) (See Member 30) 30) é S / OF ay %S a La20 0) esto TF ss10 Pro © ae fe fos “2000 “2000 “2000 “se00 aga0" » 20" —+ Fig, 47.50 this step are given in the figure, the stresses deing written adjacent to each menter. The next step oF steps is to compute the sender elastic eights, their location and thefr sense or gic~ Paget P= For a truss diagonal member the elastic lNeignts P 4 Q have opposite signs and are assumed to be directed tomard each other or jaway according as the member is in conpras- sion or tension, In fig. 2, P ts greater fthan Q and P ts located at the end of the di- jagonal nearest the moment center 0, Jownward elastic wetgnts are plus. TABLE A7.7 (CONTINUED) ‘Truss Verticals (See Member 20) ‘The elastic welgnt P acts at foot of vertical and downward if vertical 13 in tension. IQ acta opposite to P at far end of chord member cut by index section 1-1 used in finding stress in ab by method of sections. Getion,” Tables 47-2 ane A7.9 gives choos, cal iseions. So smears] te] ee + eg sage ot Eeaenios | jee at] tho | 28 Seeg caf [Ele Paaes Bee Se Cab aUeHel i Se area t n n Ey PERSEEEE| mer Fig. A7.51 shows the elastic wetgnts obtained from Tables 6 and 9 applied to an imaginary beam Whose span equals that of the given truss. ‘These elastic weights are the algebraic sum of the elastic wetghts acting at each truss Joint. A717 Solution of Example Problens. ‘The method of elastic wotghts as applied to truss deflection can be best explained by the solution of several simpla typical trusses. Example Problem 38 B- A7.50 shows a simply supported truss symetrically loaded. Since the axial deforma ‘ions in all the members aust ve found, the first step 1s to find the loads tn all the zenb- ers due to the given leading. ‘The results of ei yks BR: iter eee HH. @ speces at 1s" —— ee The deflection at any joint equals the bending nonent on the tmaginary beam of Fig. A7.S1. Defi. at A = (.005412 + .o185)15 = .oo7eé2.x 1s -1o3* Defi. at D = 1109+ (07262 - .000507)15 = .109 + 008755 x 15 = .209" Defi, at B = 1209+ (,008788 - .002547)15 = .208 + 7008408 x 15 = .275*ANAL ISIS _AND DESIGN OF FI Defl. at c = .275 + (.004408 - .0027)15 = .3OL" ‘The Slope of the elastic curve at the truss Join points equals the vertical shear at these points for the beam of Fig. A751. Example Problem 39 Find the vertical deflection of the Joints of the Pratt truss as shown in Fig. A7.62. The member deformationsal for each member due to the given loading are written adjacent to cach member, Table A7.10 gives tha calculation of menber elastic wetgnts. Fig. a7.58 shows the imaginary dean loaded with the elastic weights from Table 47.10. The deflections are equal mmerically to ‘the bending moments on this beam. .OL8SS x 25 = .465" 1465 + .0S3 (AL in Bar Bb) = S18" 301855 x 50 - .00887 x 25 - .833" +835 + .OGL (AL of Ce) = .364 TOLESS x 75 - .00387 x 50 - .00623 x 25 = 1.03" - v An. 38 deam loaded with the elastic weights from Table A711. Table A712 gives the calculation for the Joint deflections. Fig. A754 sapte a1 Hastie Wight of Gora ener “ues ae]: | eedk | com = 2 & z S St $ FY i i z i = 3 per ip =a [sot Tenet sane St Fig. AT.52 = ' 3 é zg ? Joint e ce 3 2 § 3 a e E 3 sox 5 Fig. ASS 2 ante 17.18 op Fig. A1.59 . Example Problem 40 Find the vertical joint deflections tor the unsymmetricelly loaded truss of Fig. 7.54. The aL deforuations for all members are given on the Figure, Table 47.11 gives the calculation of the elastic weights, their signs and points of application. Fig, A7,SS shows the imaginary wournt = 12.5 7 Sess" de seer Point ‘Bxample Problem é1 Fig. A7.06 shows a simply supported truss with canttlever overhang on each end.’ This sin plified truss {s representative of a cantilever Wing beam the fuselage autachment points being~ @ and simply supported at these points. A138 DEFLECTIONS 0 at eand e", ‘The AL deformition tn each truss member due to the given external loading !s giv- enon the figure. The complete truss elastic loading will be determined. With the elastic Joading known the truss daflections from various reference Lines are readily determined. reste tas ieatie teignta of chord snabere wor | a |e opi ae ! lad 3 r 8 : # $— 2 3 8 : 2 : haste voles of tab arabe weer [a poe [epi] ve | oe Gipol iat ssi & : $ $ 2 3] & i! ‘Table A7.13 gives the calculations for the magnitude of the menber elastic weights. The signe and also the Joint locations for locating the elastic icads are also given. Combining alge~ praically the elastic wetents for eack jotnt fram] Table A7.13 the beam elastic loading as show in Fig. A7.87 ts obtained. Let it first be required to determine the vertical deflection of Joint f relative to the ‘russ support points at e and o'. ‘To determine the deflections of the truss debwecn the supports oa 4 9! tt 19 only neces sary [0 consider the elastic weignt loading be. Sween these points. Fig. A7.S8 sons the portion) of the imaginary beam of Fig. A7.57 between these| points. The deflection at f relative to line ee! is equal to the bending moment at f tor the portion of the imaginary beam between Joints ¢ and Hence deflection at f= ~ .ocsiz x 30+ 00282 x 15 = .0686" (upward since minus 1s up), = STRUCTURES lever over-hang portion of the truss relative to support points 2 and e!. Since the cantilever portion ts not fixed at estraint ts deternined by the truss and e, tots fact must be taken tnto account in loading the cantilever portion. The Teactions on the beam of Fig. A7.S8 represent the slope at e due to the elastic loading Detween eande’, This elastic reaction in acting tn the reverse direction ts therefore applied 2s 2 toed to the imaginary beam between ¢ and 2 as shown in Fig. A7.59. 16 g oon Re 00012 R= .00s12 Fig. AT.58 In finding deflections this overhang elas tically loaded portion 1s considered as fixed at end free at e. The bending moment at any point on this dean equals the magnitude of the vertical deflection at that point. ‘Thus to find the detyection of the truss end (Joint 2) we find the bending moment at point a of the imaginary beam of Fig. A7.59. Hence deflection at a = Me (calling counterclockwise positive). = (.01922 - .00312) 80 + .coaks x 70 + .CO333 x 60 + .00239 x 80 + .00as8 x 40 + 100284 x 30+ 00543 x 20 - .0OL4S x 10 = 2.15" upward. due to symmetry of truss and loading of the truss we iciow the slope of the alastie curve a the center line of the truss is horizontal or zero. Thus to find the deflection of any potat with reference to Joint f we can make use of the deflection principle of the moment area method, ‘Taus in Fig. A7.60 the vertical deflection of any point for exauple joint a, relative to Joint f equals the moment of ali elastic leads between 2 and f about a. g Free ky R= ,00312 (Erom Span ee") 0248 0008 ‘op 90200 [00460 of 00098 ol 00088 >| 00189 | ovate Fixed Fig, A759 Big = 2.077" (student should make calcula~ ttons)." Previously the deflection of f with respect to e wes found to be -.0586". Thus de- Flection of 8 with respect to point e = 2.077 + 0566 = 2.105" which checks value found above. Let it be required to tind the deflection of Joint © relative to a line connecting Joints > and d. Find the vertical deflection of the canti- sg RESa8 a2 28 ee pba ibaa Te ere ban le Fig. 7.60ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES For this problem we need only to consider the elastic loads between points > and d as, loads on a Simple bem supported at b and d (See Fig. A7-61) The dafiection at c with re~ Spect toa line bd of the deflected truss equals the bending moment 2t point ¢ for the loaded beam of Fig. 47.61. hence Oe * .COHMES x 20 = .00889 x 10 =,07 inches aostts yecoteer Pg. At. - Aras reepiens 22°" oop 24 c E,- arr 4 . o Sch S ad BLL 0: Pa ates ig. ar. 84 (1) Pind vertical end horizontal defiec~ ‘ion of joint 8 for the structure in Fig. a7.62. Area of AB = 0.2 sq. in. and BC = 0.3. B= 10,000,006 pst. (2) For the truss in Fig. A?.63, calculate the vertical deflection of joint C. ‘Use AE for each nember equal to 2 x 107. (3) For the russ of Fig. 47.5% determine the horizontal deflection of Joint a. area of each truss member = 1 sq. in., B= 10,000,000 pst. (4) Determine the vertical deflection of Joint § of the truss in Fig. a7.54 {5} Determine the deflection of joint 9 normal to a line Joining Joint CE of the truss in Fig. a7.64. 100" 1" 'p | = 30" By ey re oe wl Fig. AT.65 Fig, 47. 66 (8) calculate the vertical displacement of joint ¢ for the truss in Fig, A7.65 due to the load at joint 8, Members a, D, ¢ and hi have areas of 20 sq. tn. each. Members ¢, ¢, f, & and { have areas of 2 sq. in. each. & = 30,000,000 pst. ana (2) For tne truss in Fig, a7.65 calculate the deflection of Joint ¢ along the direction ca. = 30,000,000 pst. senor cma ate J\s ams « BS : » 3 ast Fig. A788 the truss in Fig, A7.67, tind the horizontal displacement of Joints c area of all members carrying tension each and those carrying compression as S sq. in. each. § = 30,000,000 pst. (3) For the truss in Fig, A7.68, determine the horizontal displacement of points C and B. = 28,000,000 pst. (8) For vertical and and D. Take es 2 sq. in. 1090 1090 1000 s00# (aon Fig, A7.69 Fig, A710 (10) For the truss in Fig. A7.69, find the vertical deflection of joint D. Depth of truss 10". Width of each panel ts 190", The area of each truss member 1s indicated by the mumber on each bar tn the figure. © = 30,000,000 pat. Al- s0 calculate the angular rotation of bar DE. (11) For the truss in Fig. 47.70, calculate the vertical and horizontal displacement of Joints A and 3. Assume the cross-sectional sree for members in tension as 1 sq. in. each and those tm compression as 2 sq. in. E = 10,300,000 pst. (12) For the truss in Pig. 47.70 calculate the engular rotation of member AB under the given truss loading. 1000 $000 000 rT Pie fio py oe Sid Fig. ATT Fig, A172 (23) For the beam in Fig. A?.71 deteraine the deflection st points A and 8 using method of elastic eights. Also determine the slopes of the ‘Taka B = 2,000,000 elastic curve at these points. pat and I = 1296 tnt (14) For the beam in Fig. 47.72 tind the de~ ‘Als0 the slope of the ‘Assume EI aquais to flection at points A and B, elastic curve at point c. 5,000,000 Lb tn. sa.1 r so + sl _ coo pn 10" 30" oy (15) Fig. A7.73 {2lustrates the atrloads on a flap beams ABCDE. The flap beams is supported at B and D and a horn load of S0# is applied at 20008 Fig. A781 Fig. AT. 92 (23) In Pig. A7.81 find the vertical aove~ nent and the angular rotation of point A. HI = 12,000,000. ‘Take c. The beam is made from 4 1"~,049 aluminum (24) Determine the vertical deflection of alloy round tube. I = .01659 in*; E = 10,300, point A for the structure in Pig. A7.82. EI = pei. Conpute the deflection at points ¢ and £ | 14,000,000. And the slope of the elastic curve at point B. 1oo# 100200 woo 10100 eepepepsd eo tr 3 rig 7.83 f > B A = Pig. A114 Fig. A118 (38) For the been of Fig. A7.74 doteraine v the deflections at points Cant D in teras of SI tray walch 18 constant. Also deteraine slopes of the \ plastic curve at these saue points. v (17) For the contiiever bean of Fig. 47.76 | wort determine the deflections and slopes of the s adh elastic curve at points A and B. Take BI as A 'P=500% Constant. xpress results in toras of SI. Front View side View X00# 100 100 1008 = 400# (25) The cantilever deam of Fig. A7.83 1s Ma Snore parae oY ig" wg | loaded normal to the plane of the paper ty the E to loads of 100f sack as show. Find the de cD iB “he ‘Mal flection of point A normal to the plane of the Fig. AT. 76 ‘Fig. 41,77 paper by. the method of virtual work. The rec- (28) For the loaded beam in Fig. 47.76 determine the value of the fixed end aonents Ma and Mg. BI is constant. Also find the deflection at points ¢ and D in terms of EI. (19) In Fig. 47.77 determine the magnitude of the fixed end moment Ma and the simple support Rp. > p 4 | 1 ' ad a i N f po Mga Fig att Fig.at.80 (20) In Fig. 47.78 EI ts constant tnroughout Calculate the vertical deflection and the angu~ lar rotation of point a. (21) For the curved beam tn Fig. 47.79 tine the vertical deflection and the angular rote tion of point A. Take EI as constant, (22) For tho loaded curved beam of Pig. 47.80, dotermins the vertical deflection and the angular rotation of the point a. Take EI as constant. vangular moment of tnersia for the tube 13 0.0277 int. § = 29,000,000. (26) ‘To cantilever landing gear strut ia Fig. A7.94 1s subjected to the load of SOO tn the drag direction at point A and also a torsion~ al moment of 2000 in. Ib. at A as shown. Do ‘ermine the displacement of point A in the drag direction, The tube size for portion CB 1s 2" 083 and for portion RA, 2"-.065 round tube, Material is steel with £ = 29,000,000 pst. (27) Usiag the matrix equation 20 = [4 | [ees] for} ecnuee the strain eeray tate truss of Fig. A7.63 (Problem 2). The member Hanioflety Gooriutunt fara tomes ae uniform axial load ts L/aB (see Fig. a7.35a). Miat'Gete tn (28) Using matrix equation (23) compute wns strain ana nano bens ot Rig. 7, be made so aso evaiv sonpastion of she Season cloxosisay sueczietone oy sue eauattone Ope aver ase d= S688 Oe {29) Re-solve the problem of example prooisn 26 fora scape? cantiiover sede moce I doubles at point "e* and doudies again at "S*. (Heaviest section at built-in end.ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES (30) For the truss of Fig. a7 $8 determine the influence coefficient aatrix relating vertical deflections due to loads P,, P, and P, applied as shom. Menber areas are shown ‘on the figure h. Ao io 2 he ans snower. re ee | aer ono net 38.0 Tha]-} |aer soz: cago sa.0 33.00 38.00 38.00 33.0, (Si) For the truss of problem (30) determine vintch of the following tuo loading con~ ditions produces the greatest deflection of point 4, (All loads tn pounds}, Condition Yo. (92) Determine the matrix of influence coefficients relating drag load (positive eft), braking torque (positive nose up) and moment in the ¥-S plane (positive rtgnt wing down) as applied to the free end of the gear strat assembly of Problem 26, answer, s00 80.7 [ag]+ wt |so7 eer 7” ° o 6400. (83) Find the deflection of the load applied tc the cantilever panel of fig. 47 6. (Asaime the web does not buckle). Use zatrix notation. ans, 6 = 7.34 x 10-* inches. ar39 + 85 108 pst ‘AE = 7.52108 ibe) SAE = 3108 ~ ibe. 008 Fig. AT.86 (34) Find the tnfluence coefficients re~ lating deflections at points 1 and 2 of the sinply supported beam of Fig. a7 87. Use matrix methods. er r AE=3x10° ibs AR=8 2100 1b too 53,950 tet o of? Fig, AT.87 7.202, 7.102 12.34 Nowe to It will be highly instructive to re-gork Pollen SU an 04 tsing the alternate choice of generalized the stringers {rom those ied in your Lirst solution, See p. AT. 22 for alternate generailzea forces on a stringer. References for Chapters A7, AB. ‘TEXTBOOKS ON STRUCTURAL THEORY "advanced Mechanics of Matertais", F. Seely and J. 0, Smith, 294 Ed., Jonn Wiley, New York. “advanced Strength of Materials", J. P. Den Hartog, MeGram-dil1, N. ¥. theory of Slasticity", S. Timosnenko, MeGran— an, ¥. ¥.An 40 TEXTBOOKS Wi PLICATIONS "Elenentary Matrices", R. A. Grazer, Ne J. Duncan and A. R. Collar, Cambridge University Pros. “Introduetion to the Study of Aircraft vibration and Flutter", R. Scanian and 2. Rosenbaum, Mac Millan, New York. ‘TOOHNICAL PAPERS Bengeoter, SU.» “The Partitioning of tat 4p Structural andiySis", Joufa. of ADpl. Vou. 15, 1968. DEFLECTIONS OF STRUCTURES Wenle, L. ant Lansing, w., 4 Method tor Redue: the Atalysts of Jouplex Aadingent Struc Phousiny Hosea down oe Se. Langeters, B., Analysis of Matrix Transformation 91fh Sent Monocoque ) SEE, Lansetors, 2. 23 Structures, Journ. of foro, Sci Falkenheiner, H. Syetomatt: Slastic Sti Journ. of sers, Sc Argyris, 3. and Structura) analy Get. isse, ot. seg. Many problems involving calculation of deflections are encountered in the structural design of a large modern airplane suca as the Douglas DC-CHAPTER A8 STATICALLY INDETERMINATE STRUCTURES ALERED F. SCHMITT ‘A8.00 Introduction. A statically indeterminate (redundant) probiea {s one in which the equations of static equilibrium are not sufficient to deteratne the internai stress distribution, Additional relationships between displacements must be written to permit a solution. Tne "Theory of Slastictty" shows that a2 structures are statically indeterminate whet analyzed in minute detail. The engineer now= ever, 1s often able to make a number of as~ sumptions and coarse approxtmaticns which render the problem detarzinate. In addition, auxtliary aids are available such as the Snginesring ‘meory of Bending (XE) and tho constant-shear- flow rules of thumb (q = 1/2h) (see Chaps. A-5, 4-6 and A-13 through 4-15). While these latter are certainly not laws of "statics", the engineer enploys then often enough so that prob— lens in wnich they are used to obtain stress distributions are often thought of as being “eeteratnate’. It ts frequently the case in aircraft structural analysis that, in view of the rex Quirenents for efficient design, one cannot ob tain a determinate problem without sacrtficing necessary accuracy. The Theory of Elasticity assures the existence of a sufficient mumper of auxiliary conditions to permtt a solution in such cases. This chapter employs extensions of the methods of Chapter A~? to effect the solution of typical reauncant problens. Special aethoas of handling particular structural configurations are shown tn later chapters. ‘A8.0 The Principle of Superposition. ‘The general princtple of superposition states that the resultant effect of a group of loadings or causes acting simultaneously 1s equal to the elgabreic sum of the effects acting separately. Tne principle is restricted to the condition that the resuitant effect of the several loadings or causes varies as a linear function. Thus, the principle does not apply when the member material 1s stressed above the proportional limit or when the member stresses are dependent upon member deflections or deformations, as, for example, the Deam-colum, a ember carrying bending and axial leads at the sene time. AB.1 The Statically Indeterminate Problem. Several characteristics (and interpreta ‘tions thereof) of the statically indeterminate problem may be pointed out. These characteris~ tics are individually useful in forming the bases for methods of solution. ‘There are more members tn the structure than are required to support the applied loads. If n members ay be removed (cut) while Leaving a stable structure the original structure is said to be “n-times redundant”. COROLLARY In an netimes redundant structure the aag— nitude of the forces in n members may be assigned arbitrarily while establishing strasses in equilibrium with the applied loads. Thus, in Fig. 48.1 (a singly redundant structure), the internal force distribution of (a) is in equilibrium with the external loads for any and all values of I, the force in member BD. Fig. AGL ‘Singly redundant stress distribution, (a) consisting of a stress 4g static equilibrium with the applied loads, (b), with one aero-regultant streas distribution, (c), superposed. Only the system (b) ts actually required to equilibrate the external leads (corresponding to X= 0). Note that the systen (c) has zero exe ternal resultant. [El or al the possible stress (force) ais- tributions satisfying static equilibria the ong correct solution Is that one which results in kinenatically possible strains (displacenents), i.e, retains continuity of the struc~ ture. ‘Tnus, for example, there are an infinite number of bending moment distributions satis— tying static equilibrium in Fig. A8.1 (d) since My can assume any value. Of these, only one. Will result in the zero deflection of the rignt hand beam tip necessary to maintain structural continuity itn the support at that point. aaa38.2 pi’ u - p> Me/2 + PL/A Fig. A8.id Singly redundant beam with reot bending moment My undetermined by statics, COROLLARY If n member loads have been assigned arbitrarily while establishing equilibrium with ‘the external loads, relative movements of the elements will result, violating continuity et a points. n zeroeresultant stress (forea) dts— tributions wey then be superposed to reduce the relative motions to zero. The resulting stress distribution ts the corract one. A8.2 The Theorem of Least Work, A theorem extremely useful in the solution of redundant problens may be obtained from Castighiano’s Theorez. Consider first the problem of recundant reactions such as in a deam over three supports (Fig. 38.2). One of the reactions cannot be obtained by statics. L_, Hoc rE ET = constant 500 + Ry 500728 Re Fig. a8.2 A singly redundant beam with one reaction given an arbitrary value (R,). if the unknown reaction (say that on the Zar Fight) is given a symbol, Ry, then the remaining Feactions sna the bending moments may be de~ termined trea statics. The strain sneray U may then be written 25 a function of fy, tiesy Wat (Ry). Next torm au aR 7 Rx This 1s the deflection ey Ry due to Ry. But this must also be zero, sings the suport 1s rigid, Hence wu By 0 Trt ttt tener eee ay . (1) 4s true for 911 redundant reactions occuring at fixed supports, Bacause it corre sponds to the sathenatical concttion for the mininum of 4 function, eq. (1) is said to state STATICALLY INDETERMINATE STRUCTURES the Theorem of Least Work. In words; "2! of change of strain eheray with resnect Tixed redundant resetion ts zero”, AB.2.1 Determination of Redundant Reactions by Least Work, Sample Problem a By Way of illustration, the problex posed by Fig. 48.2 ms carried to completion. The Dending moment was given by (x, y, @ measured trom the left ends of the three beam dtvis:ons) Ms (500 + oed/e a) (500 + Rx) L = + (Ry-500) y ree, Mahe (b= 8) ome, Then Ye afm 2 (7% . 13) bf hosoree Ly A . [ fection + (be + soos] ay L + delat - ite Difrerentiating under the integral sisn, Pe AT.) (see Le bye, abl (00+ aeiL aiANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES ena ed Completion of the computation gave 1500 : Rx = - HBS is, = ~ 95.8 ibs., the negative sign indicating that Rx wes dom. sample Problem 2 “~Datermins the redundant fixed end moments for the bean of Fig. 48.2(a), ¥ EI * constant M, ees Mp Lit» —] ig, A8.28 A-doubly redundant beam with two reactions given so aroitrary values. Solution: ‘The redundant end moments ware designated ae My and My for the left and right beam ends respectively and were taken positive as show. The uouent equations tor the tao beam portions (e from left and, y trom right) were Mt 2PL = I, wey RTE ocean x a sm wen se ty owen a ren fh, eats Digterenttating under the Integral sign (see remurks on. 47-3) BAG -&)® Jae aa. L + oP = ¥ Beeoe mye ROPE] x 3 [Ce ee ESE De Evaluating the integrals and solving simultaneously gave 8.2.2 Redundant Stresses by Least Work. ‘The Theoren of Least Work my be applied to the problem of determining redundant meber forces within a statically indeterminate structure. Thus, than n-times redundant structure 1f the redundant member forces are assigned symbols I, Y, 8, > ~~ etc., the values watch these rorces must assume for continuity of the structure are ‘Suen that the ¢isplacesents associated with these forces (the discontinuities) aust be zero. e8, by an argunent parallel to that used for wundant reactions, one writes,--@) In words, "the rate of change of strain enerzy vith respect to the reduncant forces is zero”. qs. (2), Like aq. (1), are statemonts of the ‘Tnacran Of Least York.’ They provide a aquations for tae A-tines redundant structure. The similtaneous solution of these equations ytelds the desired solution of the problem. ‘Seanple Problem ¢ cantilever beam and cable system of Fig. A6.3(2) te singly redundant. Find the ember loadings by use of the Least work Theorem. A sat x Hees au IF 87-00 Yroooe fa) (be) Fig. A8.9 A singly redundant structure with one member force given an arbitrary value (©). Solution: ‘To tensile load in the cable was treated as the redundant load and was given the symbol X (Fig. 6,3(d)). The strain energies con Sidered were those of flexure in rortions AC, CD and BC an¢ that of tension in che cable AB Gergies due to axiel forces in the beam port- sons ware considered negligible. ‘The bending moment in SC (origin at 3} was we «(000 = yx) x In AC, (ortgin at A): 50. Mao * geek Y In cD: Mog 7 50,000 ‘The strain energy was therefore o(z) ve E(E £(#), Obviously there was no need to consider &) dn GD as its load and hence could not enter the ferentiating under the integral sim energy ting gave STATICALLY INDETERMINATE STRUCTURES, dtd not devend ups problem. a esr < s, wo (om = 22 2B, CH) |e Se ~ 50 YY _X 2 “(a) ae fee (Ba) ge |) sean, gw x, sox Bg Mie Be ate ett Bec Agg = 0-085 in fp = 8-0 Int Typ = 10.0 tnt X = 613 lbs. ‘ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES supportea and leaded as show’ in Fiz. 8.3/0). AS a first 2zprcximation the aortzontal floor tie ts te D2 assumed rigid axially. Find the dending moment distribution in the ring. Solution: ‘The axial load tn the floor was taken to be the redundant (since the floor sas assumed rigid, tats could nave been thought of as a re— dundant floor reaction from fixed supports}. The loading is shom in Fig. a8.3(4). ‘The bending moment distribution was M=XRSing@-FR(1- cose) — Ocecco® KAS O- m/e s0°25c90" ‘The axial loadings were S=Pose+ksing oxacso? S*Xsine 60% acg0" The strain eneray (for omy half the structure) was * re [Ere neal ae 90° . . ode | [kota off se 60° FO Sivan eoergy im the rigid floor (AE —» o#). 8.5 Digzerentiatin: under the integral sign 60° ule : x “Es sunt 628 RE["C ccs 9) sina coe + EE [ge tece a |g! 60° Rm 90° 60° Bvaluating, go o-3x(E-k) ‘Therefore Example Problen 2 "The porta: Trane of Fig. 46.5(8) ts Shree times redundant. Set up the simultaneous equations in the re= dundant forces. The relative Dending stift- nesses of the secuents are given on the figure. Solution: ‘The redundant forces selected wera the bending ‘noment,“the transverse shear force and the axial torca,'all at point A. The four figures A6.3(¢) through 48,5(1) show the bending moment diagrans of the structure due ‘to applied leads and due to redundant forcesrc acting individually (it being easter to cone pute the Loadings In this fashion]. The com plete loading was obtained sy superposition. Fig. asx 30,000 ig, a8. 3g Sov (1+ sin 0) SOT (1 + cos 6) a Eber s v 7 Toor Pie Ath Ye ne ana ‘The composite bending moments as functions of 8, 0 ands! were hp 2+ Ve Mgq = - 50,000 sin @+ M+ SOT + Sov sine ~ SOT (1 ~ ces 9) Myy = 1000 s' +H - Ye! + SOV -1ooT ‘Then since . ww aul au weg ant B= P= Feo, one has, * au om + vs) os (EA as STATICALLY INDETERMINATE STRUCTURES -ffeee eee ee pee at eo mt 220 on) ox g os |Gemeswees Bees ¢ . [fstonessessemame at Agter evaluation of the integrals the equacicns obtaines were s149at + 9,682v ~ 7.4597 = 1.212 x 10° 9.68 + 763.17 - 424,07 = 268.3 x 10° “745M = 484.00 + 514.90 = 111.15 x 10° ‘A832 Reaundant Problems by The Methods of Dummy- Unit beads. while the Theorem of Least Work may be made the basis of redundant problem enalysis, its direct application by the calculus, as tn Art. AG-2.1 and A8~2.2, 13 often impractical. For the majority of problems the work is facilitated 1r carried out by the tectniques of the Metiod of Dummy—Unit Loads. ‘The following derivation 1s for a doubly redundant truss structure, The extension to = more general n-times redundant structure, in wnich other loadings in addition to axtal (flexure, torsici and shear) are present, is indicated later. Consiser the doubly redundant truss of Fig. AG.$(a}. It may be made statically determinate by “cutting” two members such as the diagonals indicated. application of the external leads to this deterainate ("cut") etructure gives 2 load distribution, "S", computed by satisfying static equiliorium.’ at this time discontinuities appear at the cuts "x" and "y" dus to the strains developed. Pe RE ER ig loads ty loads Fig. ABAD ig. A8.4e Fig. A8.4a To compute these and subsequent displacements the Method of Oummy-Unit Loads my bo used (art. A7~7). For this purpose virtual loads are placed alternately at tne x and y cuts as in Figs. A8.a(b) and (¢). From the cumuy-unit load equationsANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES Suk Oe = aE z Sayl om Uy and uy ere the unit-redundant stress dts- tributions as indicated in Figs. Aédd, c. The subscript "o" indicates these relative displace~ ents occur in the determinate ("cut") structure with the "original* stress distribution. It ts now desired to close up the dis~ continuities by application of redundant loads Xand Y to the x and y cuts, respectively, as in Fig. A&-4(4}. Load X causes @ stress distribution Xu, and, Likewise, ¥ causes 2 distripution Yuy. The relative displacement at cut x due to redundant load X 1s given vy (0,, 18 Tead "displacement at x due to x"). WA Fig. 48.4 and ae cut, y by (read 6, yodue to x"), oe a Met ex yt me aE aE as “displacement at Similarly the load ¥ causes displacements at the outs y and x given respectively by sate ye Wy" and fly ah yg Wy Oxy zy 2 Now the net relative displacement at each under the simultaneous action of the three stress systems S, Ku, and Yu, ts eos Suh. yg tah Seo * Soe * Say 82 SES EE vr ee and ABT For continuity these net relative aisplacoments must be zero, Equating the above expressions gach to Zere, and rearranging, gives the simal- taneous equations uth, yz Belge LL y Such xe AE AE 2 AE Eqs. (4) are two similtaneous equations in the ‘two unknowns X and ¥. Upon solution for X and ¥ the true stress distribution may be computed as Sqpyp 8 + + = = For a structure wnich 1s only singly re~ dundant, es. (4) and (5) are appliad by Setting Y= o giving yt. uyth ® yk iy*L xe bee ys We (8) ucth Sux tie TE or, simply, 34 2 ee aE ee (4a) aoe aE and, Soup 7S * KU tee A8.4 Example Problems - Trusses With Single Redundancy. xample Problem #1 ES 8 ike Fig. A8.5 Fig. A8.6 Fig. A8.5 shows a single day pin comected truss, ‘The truss 1s statically determinate with respect to axternal reactions, but statically indeterminate with respect to internal member loads, since at any joint there are 3 uniowns with dnly tv equations of statics avatlable for a concurrent force systen. The truss ts therefore redundant to the first degree. The general, procedure for solution is to cake the truss statically determinate by cutting one of the emoers;on Fig. 28.5, member be has Deen selected as the redundant member, and {t {s cut as show. ‘The aeber stresses S for the truss ofAg Fig. A8.6 are then determined, the results detng| recorded on the menbers and also entered in Table 46.1. In Fig. A8.7, 2 unit Le tensile dummy load has been applied at the cut section of the redundant member be, and the loads in all) ‘the members due to this unit load are calculated ‘The results are recorded on the figure and also in Table A8.1 under the head of u stresses. The solution for the requndant load X in the redundant aember be ts given at the bottom of Table 48.1, The true load in any member equals the S$ stress plus X times its u stress. Tape 18.1 Fig. A8.8 shows 2 singly redundant Sumember frane. Find the member loadings. Member areas are show on tne figure. Fig. Ag.8 Solution: Member 0C was selected as the redundant and was cut in figuring the S-loads, as in Fig. A8.9. Fig. A8,10 shows the u-load calculation. ‘The table céapletes the caloulation. wu [True Load Men} tj als} s Bij age Xu 40 [1ai.a]o.2] 0 | Laz] 0 | i080] 98.8 BO | 100. 0} 0. 2} 1000] -1. 366] -8.83x105) 933] 625.6 co |z00.0f04| 0 | x00] 0 | s00| ama E [683x105] 2492] nee 276.319, STATICALLY INDETERMINATE STRUCTURES a4 cA 8 % Vic00 Fig. 48.9 Fig. 8.10 S toads Sloacs, Exemple Problem 41-4; Deflection caioulation tn Sane tees Calculations of the deflections under load of a redundant structure are made by application af the methods of Chapter A-T. Since, however, there are certain pitfalls as regards symbols and also some impartant special techniques, the following examples are given at this time, The extension of the method to more complon structures is immediate and no further work on deflections of redundant structures (9 given sn this chapter (excepting in the case of the matrix methods of Arts. AB10 et. seq. ). Find the hortzontel movenent of potat "4" of Deample Problem #1 under the ection of the load applied thers. Solution: The equation used tot deflection is Sq. (18) of Chapter AW7. for application te truss deflections it ts Now for a deflection calculation the symbols "3" and "uY must be carefully retntersretad fron thetr Seantngs in the reduncent stress calculation. For a de¢lection caleulation the symbols of eq. (A), above, mean: *S-l0ads" sre the true loads of the redundant structure due to application of the real sxterne) loading; Yucloads" are the loads due to a dumy-untt (vtrtual) load applied at the extemal soint were the deflection is desired and im the di rection of the desired deflection. ‘Tus, the "S loads" for use in Sq. (A) are the true stresses (the Solution) of Example Problem #1. The Tu-loads” represent scdittonal tntorma— ‘tion which would, in general, aopoar to necessi- vate another redundant stres? calculation. as will be seen, such 1s Cortunately not the case. In the present sroblem the cumy-unit load 1s applied identically as ts the 1000¢ real lead and hence the usloads are simply equal to the "Saleacs" properly scaled dom, The following table completes the calculation,ANALYSIS AND DESIGN OF FLIGHT VEMICLE sTRUCTURE 28.9 Men] t ] a | ® | = | SA ue wen. | | ats |» x | [a | | 08 | 4600 1 AO M4 2 | 338.5 v 5 is wa | a |i | -6os | 008 | 10,980 fe a wo | io | .2| 66 | a | -aien @ 1% Ja | 18 | a9 | 4600 co | mo | a| mil o a = {0 [a | o| 39 | 4000 . eb 424) 2 359 +559 6, 620 o2arei) aa | ae] is | oss | 058 | 20,000 22,700 2 | 2,200 = + Identical with the "true stress" of Table aa. . s+ stuply 1/loo0tn of the "Sstoads* stce the Gumy-unit load 1s applied exactly 38 ts the 1000¢ real Load. Bxanple Problem +2-A, Find the horizontal deflection of point 0 of Sample Provism #2 under application of the vartical 1000# load shown in Fig. 48.8. Solution: To compute the deflection use Sul erie Again the symbols "S" and "u" are to be ren {nterpreted for 2 deflection calculation as explained above in Gxample Problem #1-A. The ‘"S-Loads" are now the "true loads" computed in Example Problem #2, above. The "u-Leads" are Loads due to plactng a cumy-unit load acting horizontally on the structure at point 0. Since this load acts on a redundant structure 12 would appear that another redundant stress However, tats 1s not calculation ts required. necessary. caloulstien any set seatie aquilisrti: may 3e ‘This theorem says that to get the "u-Loads’ tor this deflection calculation #e may "cut" any one of the three tembers and get satisfactory set of u-loads oy simple statics! Before proving “he theorem we complete the calculation tn tabular form as shom. The ‘usloede* were obtained oy cutting némber OC end applying 2 unit load horizontally at 0. * true loads trom Exanple Problex #2. By way of demonstration another set of usloads, called u', were found for this same problem, this time by cutting member OA. The corresponding calculations follow: su 5 w * 395.5 0 ° 625.6 | -878 | 1. a0an10® aia. | -1.a56. | 1.58905 2 | o.azsxio® ‘The results are tdentical (allowing for round-off errors). Proof of Theorem To prove the theorem above we return to the virtual work principle and the argument from watch the dumy-untt loads deflection equation, Bq. (18) of Chapter A-7, was derived (refer to D. A7.10). it will be remembered that the de~ flection was shown to be equal to the work done by the internal virtual loads (u-loads) moving through the distortions (4) due to the real loads, t.e., = 2ua, The internal virtual Loads are these loads due to a unit load acting at the point of desired deflection. Now for the statically indeverminate structure these internal virtual loads (u-loads) are, in general, indeterminate since the dumy- unit load ts applied at an external point of the structure. Honever, we recall that, 1 = any stress distribution in static equilibriim with the "applied load" (ror the moaent now we are thinking of the dumy-unit load ag the "applied load*) dizfers from the correct (true) distribution only by a stress distribution having zero external resultant (>. a8.) Al =a zero-resultant stress distribution moving through a set of displacements does zero work. ased these potnts are; cc * Yeo WHEE Usyorg 18 2 Unload distribution cbtained from statics ina simple “cut” structure under the action of the externally applied aumay-unit lead and u25 18 the zero-resultant u-load system which must be superposed to sive the true u-load distribution 4X tag 20 It follows, therefore, that any set of usloads in static equilibclim with the externally applied dummy-unit load will do the same amount of virtual work when the structure undergoes {ts distortion as would "true” set of usloads computed by an indeterminate stress calculation. That is, OF Dax ups = Bax Usrore * 1-2 Xy20) BA x Usmore * 24% tray * BAX Uscanic ‘Trusses with double reduncancy are handled directly vy Eqs. (4). By nay of illustration, the structure éf Fig. A8.4, trom which Eqs. (4) were derived, will be solved for a loading P, = 20008 and Py = 10009. Chotces of redundents were made identical with those of Fig. aS.da. Figs, A911, 48.22 and A€,15 show the S, u, and u) load calculations respectively. te wo oe v2 ZV" P = . Pig. 8.14 Sioa STATICALLY INDETERMINATE STRUCTURES ° LN fn ‘ Fg. At.tt Ps. at.a8 eae fart ‘We note here the rule ty which the degree of redundancy of a planar pinjointed truss can be determined. For a truss of m members with p joints, the trues is n times redundant where a=m-(2p- 0), ‘Fora Spatial truss (3 dimensional truss) nem-(3p -6). in the present oroblem n = 21 ‘The calculation ts form in Table 48.2. The given in the tapie, ried out tn tabuler ember dimensions are True id= | cf ae = acy a | = ib 1 = a0) a ae) Substituting trom the table into Bqs. gives (common factor of £ divided out) Berea — eater a Hes Su $4.1 K + 56.6 Y = ~ 562,000 36.6 X + 452.6 ¥ = ~ 478,000 @ Solving, X =~ 15508 ys- ssw Finally (see Table 48.2) True Load = § + Xuy + Tuy.ANALYSIS AW. ‘B.S Trusses With Double Redundancy, cont'd. Example Problen 3 Fig. AB, 14 Shows a structure composed of four co-planar members supporting a 2000# load. With only two equations of statics available ror the concurrent foree systen the structure, rel~ ative to loads in the members, is redunasnt to the second degree. Solution: vE ag Solving the two equations for X and Y, one cb= tains X = 521¢ and Y= 416#. The true load in any member = S + Ku, + Yu, which gave the values tn the last colum of the table. 8.6 Trusses With Multiple Redundancy. By induction, eqs. (4) may de extented tor application to trusses which are three or more times redundant. ‘Thus for a triple redundancy, egy te Suh Fig. A8.15 shows the assumed statically aime 72g determinate structures the two members CE and DE were taken as the Tedundants and were cut at cyst gut points x and y as shown. The member stresses +as 2-2 = bis) for this structure and loading are recorded on Aaa aE the members. Figs, 46.16 and 46.17 give the UW, and U, member stresses due to unit (1#) ast ee tensile loads applied at the cut faces x and y. #5 aE Table 48.5 gives the complete calculations for Bae (a) ead (ede The dead tn the sadunmang” | and agter solving tor x, ¥, 8, manber CE was designated X and that in DE as Y. stenatec an a as True Stresses = S$ + Ku + tu, + gus --- - - (7) pa marae y BRAT lh AB.T Redundant Structures With Members Subjected to \ . Loaaings in Adetion to Axial Forces, Eqs. (6) are extended readily to cover 4 Y. problems’ tn watch flexural, torsional, ana shear 20004 ‘200% loadings occur, Thus, for @ three times redundant Fig. 48.14 Fig. A815 structure a a Le Rigg * Ty * Bays = > Oy aa Q ue Wat Ye + aa == 8. q “ ie 7 yy * yg = Op be (8) sy Houde vy loads : Kage + gy * Egg =~ 5, Fig. A816 Fig, A817 ea ee 50 ugh uguyl Subs wnere xe hy Lt Substituting values from table 2AK6 X + 2350 ¥ = 2,253,000. uaa yy Uy" Suyi xe sy, wh. St substituting: 2850 X + 8089 Y = 2,488,000, mu | uyMt funupulerve Lone ge | a tere tae fees [nwo [eee] 0s von) | a 2 [sas] eo] ae fave pow [ao] andAba STATICALLY INDETERMINATE STRUCTURES Solution: tinix (Tex Se Too shear flow in the chest pensle vee © chosen as redundant, Because o2 y the problen was only singly racundant. Fig. 43.16 axty Shows the uy and Gy loadings tue to che redind~ = ant shear flow X= 2. The real loading in the determinate structura consisted of a oenstent load P im the central stringer alone. The myae | ty equation solved was (ref. eqs. (3)). eT a x ai, & ay , ro ote. where and where $= = constant, tn contral sti REAL ($= 9 tn side stringers 8, M, 7, q are the rea loads in the ae. | LOADS terminate’ structure; —— ano Yes Oy ty, A, are the unit (uirtwel) load crmmuny [77% 20 880 stringers ings due to 2 unit load at cut x Tone” 4 * 2(e-L) tn contrat stringere Bye yy Sys U, are for a untt Lose at out a7 19 yh toe when evaluated, (note that tha double inte ; grals simply réduce to 2 constant cimesytne The redundant force(s} need not be an axial | Panel area) force out may be a moment, torque etc. After solution for the redundants, True axial Forces = S+Xut +3 arta r eu ‘True Bending Moments = ------- 09) MeXa+ a+ an, ete, Example Problem 4 1@ SyimetFic sheet stringer panel of Fig. 48,18 is to be analyzed for distribution of lead P between stringers, As a first approximation, assme constant shear flow in the sheet panels. All stringers ave the seme ares. oe t mye - id X=L01 Sy TT «Cie #) +e) x-Z ps ect ‘Therefore the trua stresses were Bagge tn contrat oorincer ar P, noon * P-2Lk = P or + Bape tm site stringers ia Sranple Problen § . THE probled of Fig. 49,20 is doubly ro- duicent ag shown. Deteraine the Danding noaent distribution, otk members have aqual secticral properties. es "Clamped" Lal La ao tp Fig. A819ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES Soluttor Tha bonding moments 2t sotat ¢ in member BD and at potnt 8 in member 3B were selected as rodundants ytelding (when cut) the pin Jainted determinate structure of Fig. A8.20. The virtual loadings wera ge shown In Figs. 43.21 and 48,22, sa Tey eae Mote that in Fig. 42.22 the untt redundant Loading was applied as 4 self-equilibrating set of unit couples. ‘The real and virtual load~ ings were a follows: (member portion 8D, having no virtual loadings, vas omitted. ‘It could not enter the calculations) Z veuber] 8 [Bl [mx] | we [eel oben fo Lae y, EZ BC ~ep | Py|- VL ea. yeas which were solved were (Ref, “(Ge-{ se). (Ce [3 . (ee . [s mye G82 |) Gee L | =) After evaluation of the integrals and qultt- ‘these become plying through by L* L x(a+ae) es won (@2gBE- ag.32 (GE S)(¢ - 2am) «21 GE For @ specific case (t mas assumed that az Br AB s 100 giving 3N18 X + .1826 Y= .09011 PL +1826 K+ 0121 T= .3616 AL X= .0845 PL Ys 458 PL Then as usual True Stresses =S + Xu, + Yu, True Monents = M+ tn, + Ya, AB.8 foitial Stresses. In a redundant structure initial stresses are developed 1f, upon assembly, certain mex~ bers qust be forced into place because of lack of fit, In some situations intentional misvits are employed to obtain more favorable stress @istributions under load (*prestressing"). If, in Pig. A8.4(a), the redundant member wite the "x cut" was initially oversize (too tong) an amount 6, (an oversize, corresponding to a distortion in the positive X direction, ts @ positive 6,,), the modified condition tor continuity at the x cut would be (compare with ‘the equations just preceding eqs. (4)). 80 * Sxt * See * Say Sintlarly if the ¥ redundant member were too Long, Byq + 8yy 7 Og * yy #0 ‘Then using the previous notations, the appropriate equations for the redundant forces arz Ef fee, | aul uth suk x er$te- =) ae AB ae Tae "8 loads" of 2g. (10) are present because of applied external loads. ‘These may or my not be 2ere depending upon the problem. = (20)AB.1e Example problem “Trin example problem 3 aember C= was 0.0L inches too short before assembly, deternine the stress distribution after assembly and load application. Solution: Data obtained from the previous problem was substituted into eqs. (10) along with Og > Oy, 0 OL" (negative because "too short") to give Baas X + 2550 Y = 2,253 x 10° + OLE 2350 X + Zos9 ¥ = 2.486 x 1o* with © = 29 x 10* the redundant forces were X = 985 Ibs. T= 97 Ibs. ‘Then, as usual, True stresses = 8 + tu, + Tu. Example Problen 7 ‘Assume that in the structure of exemple problem 5 an angular misalignment occurred be~ ‘een members AB and CBD at Joint 3 sucn that the end of mamber AB had to be rotated 2.7° = clockwise to fit upon assembly. Determine the moments developed without external leads applied. Solution: The tnittel imperfection wes 6, = ~ 2.7 /gy 3 = - .0471 radians The sign was determined by noting that the original misalignment was in the negative direction of tha redundant couple T, ‘The equations used from the previous problem were (noting the equations there had been miltiplied by L* x BI/L¥ = BI/L). 3716 X* 61526 T= 0 aseex+ suits .omB solving, X= .0258 BI/L ¥ = .0630 BI/L True initial stresses and aoments were deter— mined as usual. STATICALLY INDETERMINATE STRUCTURES If, a6 1s sometines the case, the number of nisaligments exceeds the mumber of redundancies, or if the misalignnent does not cofncice with the redundant cut chosen but occurs elsowhere, ene may use the virtual work principle to con pute the effect of these misalignzents on the redundant cuts proper, Thus, referring to the ‘virtual work" éerivation of the Dumy-Untt load equations, (Chap. a7) one has FE a pe ee eee ee 0) Set where 6,, is the initial misalignment in the determinate structure at the X redundancy-cut due to initial imperfections (equivalent to initial strains) 4, throughout the structure, u, as before, te the unit loading due to a virtual lead at out xq, (11) and similar expressions for the Y, &, ete. cuts my be in- Serted in eqs. (10): Bxample Problea 8 Referring back to example problems num ders 3 and 6, assume that uember BE 1s .025" too long, Determine the initiel stresses if the other members are of proper length and no external lead 1s applied. Solution: t ‘To employ the sams equations as those of example prehien 3, the initial imperfections occurring at the same x and y cuts used there were computed, in this case due to the initial elongation of BE. Thus Op FZ Udy = (-1,564)(.025") = -. 0897 n 47 BUY, = (-1.728)(.086) = -,0492" &y ‘Then, use of those previously computed coef: efents in eqs, (10) save, 2As6 XK + 2350 Y= .03S1 2350 X + 3039 ¥ = .04g2 with & = 29 x 10* pet X = 263 Ibs. ¥ = 209 Ibs. Finally, ‘True Initial Stresses = S + xu, + Tu, AG.9 Thermal stresses, Stresses induced in redunant structures by ‘thermal strains may be computed by application of methods presented above. The problem may De approached from the point of view of computing theANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES relative motions at the outs of the determinate structure caused by the thermal strains ana then restoring continuity by applying redundant mem— der forces to the cuts. Specifically, consider a doubly redundant truss such as that of Fig. a6.4(a}, Arter making cuts x" and *y* to render the structure determinate, the application of the temperatura distribution 1s visualized. Relative ¢isplace- nents occur at the cuts, dencted bY On and Oa, ‘Thase displacements may be computed by the Oumy-Unit Load method as shown in Art. A7.8 of Chap. A?. After this calculation 1s acconp— lished, the problem proceeds as for initial streing, art. 48.8. Thus, the continuity con= Gition at the cut gives (compare with the equa— tions immediately preceding eqs. (4) and as~ sue for simplicity that the external loads are absent, making 0, =O, = 0) 4 et * One + Say #0 sy +o tos Dyn t Oye + By = 0 Ina truss, the thermal strains produce relative displacements at the cuts given by the “virtual work® derivation of the Dumy-Unit Load equations (ref, Chap. A?, Arts. A7.7 and 47.8) as Oy 8S ga Tex -- (2) Syy 3S Uy a Pee Where a 15 the material thermal coefficient of expansion, T is thé temperatura above the anbient tenperature and u, and u, are the unit load distributions due to virtual loads at the and y cuts, respecttvely. The sums in eqs. (12) are written as integrals rather than fiatte| suas to allow for possible variation in a and 7, along the members as well as trom aember to Gsnber. ‘Then the final equations for thermal stresses in a doubly redundant truss becoze atl Pugh x Ar fat [uate ae a 4, ~- 3) agit gu, xfer Qe lore 4c az v Equations (15) my, of course, de extended for application to structures other than (pusses. The expressions appropriate to other Toadings nave deen developed previously (eqs. (8) et seq. in this chapter and other equations in Art. A7.3). ALIS eample Problea ¢ Send upFight of the truss of Fig. 48.23, 1s heated to the temperature distribution snow. Determine the stresses and reactions developed. A TyaT es ‘s Al oto t germite sliding Fig. aga Siding ag, a2 Solution: ‘The structure was nade determinate by cuts X and yas im Fig. A8.24. The untt loadings are showm in Fig. 48.25. i Bt Fig. A8.25 ‘The thermal coeffictent a was assumed constant. The calculation was set up in tabu- Jar form, (ler =o ser wes ee Substituting into eqs. (13) 8.008 X = 90 ¥ = .90X+ 180 gaTx 10° ° Solving, K 22.0107 x 108 Y2l2lat x 10% ‘True stresses are given in Table 46.4. Example Problem 10 ~The upper furface of the built-in bean of Fig. A8.28 15 heated to @ uniform temperature T.8.16 ‘Through the depth of the beam the temperature varies linearly to normal (7 = 0) at the lower STATICALLY INDETERMINATE STRUCTURES Then the equations corresponding to eqs, (13) wore wedtten (see also eqs. (3)). JS2)(34-[82) sees ee ra tend ecu aenissen = Seping sink swtnat te asaes | (|S reneesing = iS : rt ! ie pt a y —_—_——_) T eT ensass Via Latin BOW meas ‘The problem was only singly redundant because of symmetry and was nade determinate by cutting the end bending restraints. Applii tion of unit couples (Fig. 4€.26b) gave m= 1 = const. Then (see Erample Problem 24, Art, A7.8) the thermal deflection at the fout® was L Ly va = [ae | 1 ‘The redundant moment equation was (by analogy to ea. 13) Therefore Les XB t- tan X= + Tosr = The redundant moment compresses the upper fibers a8 Was Co de expected. Sxample Problem 11 Coaplete the problem begun in Sample Problem 24 art, A7.8, viz, that of computing the thermal stresses in a'closed ring whose inner surtece is uniformly heated to a temperature T above the outside. Solution: ‘The ring was made determinate by cutting at the top as in Fig. A7,30(b). The unit loadings and thermal deflections were determined in the Teferenced example. The results of aeflection calculations made greviously were Ou = On Re? bh yp = -2n R2at ¥ Evaluated, the equations were bee oxsons(heSt)a Nose shat tron che tery of tase squation a oad TELS Tacha: Sitesi dita.” Seating tne Sitee wo oobatlons aur x & Yeo A non-zero value of Y would sroduce 2 very ing bending moment which cannot be because cr symetry. Hence this result too, is rational. AB-10 Redundant Probiem Stress Calcwations by Matrix Methods. In the following section the indeterminate structural groblea 1s formilated in masrix notation, the reader is assumed to be familiar with the natrix applications of art. 47.9 and the slonents of satrix notation and artinetic {see Appendix). The stress distribution of the structure ts specified by 4 set of internal generalized sarees, Gy, ay." (ret. Art. a7.9). Unlike tne case OF the determinate structure, thase q,, ¢, cannot be ‘In the case of indeterminate structures, wherein some of the support reactions may also be redundant, these reactions lao are denoted by q's. (see Example Brobiem 132).ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES related (mediately to the external leads vy the equations of statics. Thus a certain subset of the a,, 4, are redundants and are denoted by Gyr Gye (Fy 8 different numerical subscripts). shen, finally, the redundant forces, Gy ays are computed (by satisfying continuity} the true| values of all the q,, 4, say be found by statics. SmBoLs Gps 4j — ~ Anterma generalizea sorces acting on structurel elenents and reactions at support points. Gyr Gy = Peduident generelized corces and redundant reactions PP, = applied external Loads z, = che value of a, (a)) in the de- terminats strufture/onder appli- eation of a unit external load Ppt 1 @, =D Bpp (*8jq) ~ the value of a, (Qy) tn she den terminate structure due to application of a unit redundant force ett @ed Gy (20,,) ~ the true value of @, (a4) in the redundant structure due to application of lead P= 1 (PR) = 1) + the true value of g, (@,) for a unit value of applied load PRel (P= 1). a, ~ neuber flexibiuity coefficient: deflection at point {for a unit roree, q, = 1 (see Arts. A7.9, 10}+ oon = influence cosffictent for the determinate structure: displacement ab external loading point 2 for a wit applied load, P= Le (a= ay) } 7 iMfluence coeffictent for the de~ terminate structurs: displacement at redundant cut r (s) tor a unit applied lead, P= 1 (P= 1}. (oof, = yp) a, = influence cgeffictent for the de- * rninate {"cut") structure: displacement at reduniant out r Zor a unit redundant foree q, = 1. lang = Oey) . rs * Sor Aon a8.7 SWMBOLS ~ continued he temperature (above normal) at points t, J. Syqe yy > She ember thermal distortions as- Soctated with a) qj. TT - In the notation of Arts. 48.3 et seq, the Final true values of the stresses were expressed as (eq. (5) art. 48.3) Sup 78+ + Wy yee (5) In the notation to be eaployed here, this equation is restated as as) {4} [ea] (%}* Gel fe }o- 2 +--+ Kere: [Fin] (-{8sn]) #9 te metrsx of wmst-asse stress distributions in the determinate ("cut") structure found by the application of unit (vtrtual) loads at the external loading points. Tee product [fq] {Pq} shen st¥e8 tae reel 2eaes in the determinate structure, corresponding to the "8" loads of eq. (5). (ez ]¢[ts2]) is the matrix of unit-load stress distributions found by application of unit (virtual) forces at the redundant cuts in the determinate ("cut") structure, Hence this 18 the matrix of uy, uy, ste. leads. the fg.) of Soae, corzpant t0 Xs fy ote. Mote that the [e,q] and (,,] trices are load distributions computed and arranged in much tna oane fashion as was [0,] of art. A7.9, The small letter "5" ts used vo indicate load dis- ‘tribution in the "ut" structure. By nay of tliustration, che final result for Gxample Problen 3, Art. 48.5 is expressed below. FIRST, in the tora of oq. (5} true Stresses = 8 +X (a) + i) © + S21 (,808) + 416 (1.154) BE = 2000 + S2b (-1.56¢) + 416 (-1.728) 0B = O + Sel (2,00) + 426 (9) 8 = 0 + S21 (0) + 416 (2.00) + Note that within each of the sets of subscript symbols (i, 1), &, 3), (im, 8) the symbols may be used interchangeably.A818 SECOND, in the form of eq. (14): a ° +806 1.187] [se as L 1.554 -1.729 ae f 7 00% 2%) 100 ° 416 a 0 © 2.00. Note that in this case [6,_)] consisted of only ong column, inasmuch as there mas only a single external load. In Art. A7.9 the strain energy was written *=La JE {5} - wtsere [0,5] {5 the matrix of member flex:pitity coefficients (art. 47.10). If now eq. (14) and 108 transpose are used to substitute Into (15) ‘the expression becomes {note the use of (1,3), (z,3), and (n,n) interchangeably) *(L. a)* Ls} Bil) *Bal*(Es} Ral): Multiplying out Ls |) Bea] Boal P}* * Ls JG El Bal{%}- Ls Jo El Bel {} ‘The reader say satisty himself that the “cross product" term in the aiddle of the above result is correct by observing that, because of the emetry of [a,j], Lj es] Ess) Beal {=} UJ) El Eel {s} ‘Tas various matrix triple products occurring| above are assigned the following symbols, each aaving the Interpretation given (compare with eq. (24) of art. 47.3) STATICALLY INDETERMINATE STRUCTURES of external-point influence s0effictents tn Nout" structure: deflection at point 1 per load at point n. the unit iF tpl faa eh fal ey nt joads: p unit lead at point a, displacements at aisplacenent at cut as) of snfluence coeffictents relating relative is~ placements at the "outs" to reduidant ioacs at the "cuts": displacasant 2t cut r par unit re dundant fores at cut 8. With the above notation one may write (a | Eal{%}* ? Lt Bale} Now according to the Teeoren of Least Work Wag, = 0 for continuity. ‘Then, ditterentias eq. (19): x ro=(Bal{Pabt Gs] {tsp ---- (20) This last result my be verttied by writing eq. (19) out in expanded form, differentiating and then recombining in matrix form. Rearranzec, eq. {20) gives Dre] {%} > Beal {Ps} Eq. (21) ts a sat of similtaneous equations for the redundant internal forces @., a. It ay be compared witn og. (6) of art. 48.6, to hich to serresponss. iq. (22) say oe solved Utrecely fron tue form share aispiaved oF 13 solution say bo obtaines oy seaputing [B,, the inverse of the matrix of coef: {%}*- Es Bal {es} ‘The matrix product - tents, giving values of the syabot of the redundant forces Zor unit values external loads. Fag] 5° tae ‘This my be given the Ga RIGDG -y7225 = the matrix {S}-Baltt}ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES If now ea, (22) is substituted into eq. (1g) one gets (with exchang: of r for s and m for n) {sp Ga] 73> Ge] Good eal {5} -(Gal- ES BEE) {4} - (Gals al Eel {3} value of the external loads. It is given the symbol Eas Gal BAGAIEa --- © “HEALS so that {}> Bal fFe} >> Bqs. (25), (24) constitute the major re~ sult, fnasmuch’as they present the aeans for computing the internal force distribution tn a redundant structure. Example Problem 13 SREDoe GnbEy Tetundant dean of Fig. 48.27 (a) is to be analyzed for the bending moment distribution. The deam 18 loaded by couples over ‘the supports as shown. (24) Et constant Po ——— Re a <= =P Oo) a ” “eC a Fig. A827 Solutien: ‘The choice of internal generalized forces is show in Fig. a€.27 (2). The appropriate nenber flexibility coettictents wei a in matrix fora as (raf. art. A7,10 for coefficient expressions). ae.19 bys Wz 8 0 te to 0 bial: o 0 M/s be la The moments qe and dy were taken as the redund~ ants. With these set equal to zero, the internal force distributions due to application of untt values of P, and P, were determined, giving “yo ao a) * oly 0 6. With the applied loads set equal to zero, unit values of the radundants were applied ylelding (Ps) (2) () ~ A o 1 ° Es] “loin Ma ° 2 Note that redundant load q_ was applied as @ self-squflibrating internal couple, acting on both beam nalrs. ‘The following matrix products were formed: Es) GEIS ables Tis ve 0 0 fn 0 Blo on illig 2 0 ofr 0 0 0 MMs well iy tn o 0 we atio 2 Ed Ea) GOED Ba 0 0 Ms wello mm58.20 ‘The inverse of appendix) [op,] #85 found (ree Ge Next, the unit redundant load distribution was found (eq. 22). ea - Ged: 408 286 Finally, the true unit stress distribution was computed. Gal * Ga] * Ee] Es] T.286 as 29] F%, 07] [-.2es -.926 ,joof | 2 0 | Lass -.296 Blog wy eel Lo of ey: 2 1] tae | ee =} [Ee seen [oar S|. | 058 4 [14st | -.286t (The tabular form of presentation of the matrix G,,, above, 1s used here only te indicate Clearly the functioning of the subscript nota- tional scteme. In general, {t should be unnee- essary to call out the subgeripts in this fashion excepting for the larger matrices, tor the handling of which, the tabular fora may prove helpful.) Example Problem 186 i@ Fedundant beam problem of Fig. 4.27 ts to be re-solved using the redundant reactions 25 the unknowns. Solution: ‘The support reactions under the loads P and Pa were given ‘the symbols qs and Ge, reszec~ tively, positive up. These forces did aot anter into the strain energy expression so that, al- STATICALLY INDETERMINATE STRUCTURES Though the matrix of member flexibility coe fflolents was expanded vo a 6x 5, the oostficients for Gy ard q, were aero. Thus, Lys be 0 9 00 bg 1 6 0 60 El o 0 bs ly a 0 0 9 be 16 2 09 9 ao 6 9 0 9 o8 With the redundants q, and q. set equal zero, successive applicatices of unit external couples P, and P, gave the stress distribution Gs] - and, with P, and P, zero, successive applica~ tions of unit redundant forces ae and de gave 8 Le a4 fr]-|a 1 oe ° teeny suttoiying out ger agp. (7) axe 0) - Ea-& Ef we fe a Ge] +8 FS 2 tie inverse mas foun: Gey [2 3 Ess] - Ga) - Ge] BJ Ea]ANALYSIS AND DESIGN OF PLIGHT VEHICLE STRUCTURES i CY] acs 1.286 174 428. (Compare with solution to Example Problem 13). le Prob The sontiaucas truss of Fig. 48.28 is tale redundant. It 1s desired to analyze it for stress distributions under @ variety of loading conditions consisting of concentrated Vertical Loacs applied at the four external points indteatad. ‘ro Try 8 panels at 20" = 160" Pig. AB. 28 Selution: ‘The internal generalized forces (4, ,4,) eaployed were the axial loads in the vertous necbers. These vers nusbered from one to tnirty-one as shown on the figure, “The member flexibility onefrictents in this case were of the corm a, = '/Ad (Ret. Fig. 7.352). The cooffictents are written as a colum matrix be~ low, (They vere anployed as the diagonal elenents of a square matrix in the matrix malti- plications, ut are written here as @ column to Conserve space.) Menber leads q, and q, vere selected as redundants. with 4, and q, set squal to zero (Tout"), unit loads ‘ners applied successively at external loading points one through four, the four stress distributions thus found being arranged in four colums giving the matrix [Eig] (Petom. ay vey of timustration, che loading figure usd to obtain the second column of [yg] $= shown ta Fs, 48.28, Next, unit forces sere applied successively| at the redundant cuts "three"and "tive" as shown tn Pigs. 48.302 and 48.305. Thesd Teads were arranged in to columns to give the satrtx [8,7] ABZ toh ts Loading for column {gi2} of the matrix [Bim] Fig. 48.29 {Go} tnd TIRE ig. A890 fei) nang aps ee RESESESESSRSSEEB RBA bbbberenrrrebe IB BER 8iT NOTE: VOIDS INDICATE ZEROES sutislying ois ewe eaedea-: [2] Bd- BUSI ED 72.0 -16.0 0 ° o 0 -15.0 8.9 ‘The inverse of appendix) Ga- was found (ret. 3] Gal-ste (My Next, the values of the redundant forces, for wit values of the applied loads, were found per eq. (22). Pd--EAGS STATICALLY INDETERMINATE STRUCTURES TOs9 311.0088 .00a 20085 .0065 121 «C59. The calculation was completed as per eq. (23) to sive (@,,] , the veluss of the xenter forces for unit applied external icads. Ze 30 3 Example Problem SEENT ie-aT stows tho two bays of 2 st tubular tail fuselage truss which 1s loaded by tail air loads to be resolved into three con— centrated leads applied as show. Tne fuselage bulkhead at the attach-points station (A~f-F=<) is heavy enough so that it may be assumed to be Tigi¢ in its own plane. Hence, the truss may be analyzed as 1f cantilevered’ from A-E-P-K 25 shown. All members are steel tubes, their lengths end areas being tabulated below, Solution: ‘The generalized forces ware taken to be ‘he member axial loads, these being numbered as im the table below, Member flexibility coertic~ tents, L/a, (2 set equal to unity for conven tence} were also tabulated.ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES 8.23 Fig. A8.31 ese | MBER | Lovord | anne La = a oF efsielesta 108.8 ne 3 Tm stress 41 Y ree tines redundant, ons of statics may ce written Here, nowever, stress details 1n AEM are ts de sacrificed; six equa~ 38 quetions; 24 aenber unknom. 2, &% enc 24 vere cut, Text step seg to compute the unit strigutions [7,5] 884 [aye] «RAY than apply successive unit loads and forces at points m1, 2, S andr = 22, 23, 24 and then ry each loading through the structure, an other procedure, often better adapted co large complex structures, was employed. In the metned usec, chs equations of static equilibrium were urttten for each of che seven Joints." Sumation of forces in three directions on each foint gave 21 equations in 24 q’s (the unlmoms) and the three applied loads P,, P, and Py. Some typical equations obtained were: From 2 F on Joint ¢, PASE dag = 21968 G4 + ESSE, = 18H Ge = Pe wr aso eet ae = cess From 2 P on Joint 8, 081759 Qi ~ 26K de = ae OBL? aay = S227 dae ae e-Bay zo sano a+ fae + MS gue = 8823 ee And so forth, for the other joints. Note that tn'each case the equations wer arranged with the applied loads (P_) and che Fedundant 9°S (qazy Ges, Gea) erouped on the right hand side of the equal sign. This ar- Fengement was observed for all 21 equations, after which the equations were placed in natrix Boreacci) td = Ly eee BE a 22,2,3 S = 22,23,24 (ote that there were 24 equations here, the additional three equations being the ident ities Gas = Ges as = Gas Qa Fee) On the right hong side of the above astrix aquation the matrices are shown "partitioned". ‘The first three colums of [Dara tne coatric~ fants of PL and che last chad are the For structures other than trusses the equilibrium equations are written for the various structural elements, equilibeiam, bf jolnts alone being inappropriate,Ag 26 costtictents of ene a,. The satrix equation es solved tor the ay by finding one snveres of [0,5] (see aprencte, fh Esiedl{e [a is]sEo"]Bai, Thus, the unit stress distributions tn the determinate structure were found by a procedure having as its main advantage tha reduction of ‘the sork to a routine mathematical operation. In the conduct of this mors appropriate stani- ardized techniques may be enployed.* ‘The result in this case was were ‘The member flexibility coerrietents “/s wore arranged as the diagonal elements of the matrix [aj]. Then, multiplying out accorcing to eqs, (17) and (28), 477.9 1137.0 ~6430. a]+ | 29:0 - 151.7 2263.0 1644 181.8 2e7.g 2970 so 1145 es}: | nso i2l -1035 148-1035 eee, = 1 See references in appendix, STATICALLY INDETERMINATE sTAUCTURES inverse was found (see appendix ) ana (22a) a3 2239 7] -238 8 Finally, the complete unit stress ¢istrizucon was obtained as Ges] « [ia] * Eee] Bea] 1 aug | Lars 2OOOLT | 2. BOr TOL 33 The dowiy symmetric four flangs idealized Dox beam of Fiz. 46,32 is to be analyzed for strssses due to load application at the six points indicated, Flange areas taper linearly from root to Baal {Fs} Buaple Problem 26 ~—“Deternine the matrix of influence coertic~ jents for the reaundant truss of example problem Li, Solution: From the products ware previous work che following aatrtx formed 8.27 Fry 15.0 ° a0 88 wo 08 2 38.0 -15.0 0G iso 168 Ga] -GIEI&d +3 7572 888.0805 1388 1.855 .0a7s 0825 0975 1.665 028.0505 988 028 0805 888 472 IGIE-4 Finally, the sun of the above so matrices gve aa 9 = ose = 088 oe ee Gea]=2 |- cose - ose ass -as.0 = 08 + 0s 15.8 1a fxample Problem 17 Determine the matrix of influence coerfictents for the box beam of exemple problem 15. Solution: An alternate procedure to that shown by eq. (25) was followed. The influence coeffictent matrix was formed es in Chapter A-7, art. 47.3, dy the product, Ba]: Gd Ed Eel This product was formed resaiiy, trasmueh as [sn] "#2 avaitabie cron exanpie problem 15, Tre result was a7 1522. 7 304-2] 35-7] 7 AG.12 Precision and Accuracy in Redundant stress Calculations. Matters of prectston are dependent upon the number of signif {cant figures obtained and re~ tained In dealing with the geometry of the structure and in the care with which arithmetic operations are performed. In the discussion to follow tt 15 assumed that all due caution 13 exercised with regard to the precision of the work. Matters of accuracy'have to do with the mum- ver of significant tisures finally obtained in the answer 28 influenced by the sammer of formu8.28 lation of tke oroblem. The accuracy of the re SuLt may be arfected Sy 4 nunber of factors, To of the mest important of which are discussed here. ‘Two factors ini 15 accuracy often are considered together under the heading "sholce of reaundents". They are: number of significant cigures waich may be retained In the solution tor the inverse of the redundant force coefficient ma- crix, [ag] + (ea. 21). 0 ~ tho magnitude of the reduncant corces Felative to the size of the determinate vorces. These tuo factors are concernad respectiva- dy sith the Lett hand side and the riznt nand side of eq. 21, viz. eli }> > Bal {a} they are discussed tn detail below. - scoomecr oF nvmste o° (3.5) 5 THE CONDITION OF a HAOREK. Te characteristic of tne antrtx [6,5] watch determines the accuracy with watch its ine verse can be computed 1s {ts conaii The condition of the matrix ts an Indication of the Sagnitude of elements off the main dlagcne] (upper left to lower rignt) relative to those Qh. The smalier are the ralative sizes of Slements off the min diagonal, the better ts the condition of the matrix, A wel2 conditioned matrix is gore accurately tnvei in poorly ‘oned dng. TWO extTens cases ars aGw diven for Tilustration: a) the diagonal matrix. Ite off-diagonal alements are @3r9, 90 that it 13 ideally son aitionad, Tous, the inverse of ° a7 9 goo ts easily and accurately ootained as Ve 0 0 o ly 6 o 8 D) a macrtx 212 of wnose elements are equal am each row. 11 the elements or? the main Glagonal are equal to those on. Ths determin- ant of such a matrix ce zeroaia hones Tee STATICALLY INDETERMINATE STRUCTURES inverse cannot be found ( condition 1s terrible. 2 € 3 3 Jr the matrix [o,,] the off-diagonal ole ents are messuras of goubling of one redundant forse Mich dmonrer. ‘trengtn of this cross—sourling Ls ¢e: ent upon the choice of redundants made: "oubting" the structure to make it staticelly determinate. FOR EXAMPLE, the doubly redundant bean of Fin, 48.35(a) may be made statically determinate by "cutting" any two constraints. rus iti- @ = Tq fe =e @ Fig. 48.35 Fig. 48.35(3) shons the chotoe of zoneral. ized forces. Only two (q, end qa) ar required to describe the strein energy, but che central Support reactions were also ziven eyazols as {¢ was desired to consider them in the discussion, Then 2% We 0 0 . |’ % 0 0 Ue) o 6 3 8 oo 0 8 FIRST, suppose the beam was made determine: Selecting the support reactions qy and a. as Fedundants. The "cut structure’ in this case nay de visualized as the beam of Fiz. A8.25(2) whose central supports have bean removed. Application of unit redundant roreas qy 2 oy rans Ys ]s v|-e2[-de Gal: PaigeANALYSIS AND DESIGN OF FLIGHT VERICLE STRUCTURES Multiplying out, --GOGIGa Baal ] rs ‘The condition of this matrix ts poor. Paysically, a unit load at point "Ss" causes aimost as auch deflection at point "4" as at "S" itself, Tne crosg-soupling is large. SECOND, suppose the moments 4, nd qe had deen chosen as redundants. The "cut structure” in thie case deine visualized as tn Fig. A8.35(¢). Application of unit redundant forces Land qa = 1 gave Were Tape ott Gi]: |S] aie sit Multiplying out, GI -RIGdGd-& [TJ] ry s=1,2 ‘Tae condition of thts redundant matrix is obviously vetter than chat obtained with tne first choice of recundants, There is less cross-coupling between the ‘redundant forces. ‘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 impossible to invert 2 large, ill-conditioned matrix with the limited number of significant figures available {rom the Initial cata, ‘The following statements and rules-ol-thumo may be Useful in the treatment of highly redundant problems, (1) Et ts always possible to tind a set of seguncenes Zor-aRlvh The Cross 0UpTing Ts 503 may be reduced to a diagonal mtrix (1dea.ly conditioned]. 3 some epecis sTemes, orthogonal Surugeures, Sucn as Fines 8.29 Tedundants may be chosen simply. In most struc— tures, however, the additional labor involved in Seeking an orthogonal set of redundants is not warranted. (2) Im choosing 2 set of redundants watch win yield 3 welt-oanttotoned Panursane SOE, ‘Tes best to gaa such Touts” as will ladve a Statically determinate structure Fetainine 23 fany of tho characteristics of the orisinel Strugture a5 p08e wus, One Gay consider Ghee the Stricture of Fig. a8.35(¢) retains more of the features of the original continuous-bean structure than does that of Fig. 8.35(¢). (S) The degree to watch one recundant in~ siuences afstasr [ovtets oF cresiecoupttes] saa be-visulized by observing Row mush thats Inet vidual initelcad diagrans Toverlao", CESID storate ECR TSE eink, rerer once again to the above {llustrative example. The cross-coupling of q, with q, my be expected to be large ir their unit-load dlegrams are drawn as Delow in Pig. 48,36(a). ‘This deduction follows easily tf it Is reeslled that the dummy-untt losd equation for such 2 cross-coupling term ts or tne tora | mam, ax » obviously largs for as, Strictly, the comparison 1s with the terms [ae enents of the matric [2,6] | S520 mich sora tne nectagona2 “L_ AM | e) Fig. 48.38 Study of the untt~load diagrans for the jundant qas Qe (Fig. 45.350) reveals that tke cross-coupling should se small sere Mats x since an tntecral can co [ vee [BREE ts the canter span only, obvicusly considerably sation then {#8 or [SLE ae, a visual (nepetton of Fig. A.3e ravens thet a, 4, 15 @ better choice of Gs Qe recundants than 1s + see ea. (8) Art. 48,7, BSAB. 30 STATICALLY INDETE: (4) Combinations of redundants say be om Ployed to yield nev miltredmeant Stress dis- SEibutlons which do not “overlap” 25 axtensival: spae thoes 3) oe Tee ne aily shosen- For sxanple, suppose that in the previous 1llustretive problen, the choice Qs, 24 for r= dundants naa been sade originally, Jeacing to the unit~load diasrans of Fig. A8.37(a). Ine spection of the Clagrans leads one to anticipate a strong degree of crocs—coupling ana henes & nen set of Teduncants 1s sought. Rather than return to the structure to choose new "cuts", combinations of the a, and x, diagrams are fooked for anich wil ave léss “overlap” and hence less cross-coupling. It 1s observed by inspection that two new stress distributions Walch fave the cesirad property may be formed from the 3, 24 diagrans by proper combination. Thus, if ane halt the By Giagran 15 subtracted trom the ay igeran to Ting one Tei stress distribution and one halt tue a, diagran ts subtracted from the m, die to form the ofher new stress dletribution, the results are as shom in Fig. A.37D. There ts obviously less "overlap" of the diegrans for these now combinations. ay 4 my -$m 1 sat tim, 3 * 4. my Lan note he (a) (ey Fig, 48.37 New unlt-redundant-force stress distributions (b) obtained by combining previous distributions (a). In this way two new unknowms are introduces by Linear combination. In aatrtx notation, the old stress distributions (r] are transformed to a now set [3,,)] by forsing Gd-Ea ea o Ho ete ome (ATE STRUCTURES such a way that the product (6,.] [8.9] d2e8 guactly what 1s desired, viz., "take one tines the qy column ataus one hal? the q, column to give the firet column of the new distribution ", Ten * e] axe ulus one half ef the qs column plus one second column of nes she now aistrsbution [5,5] "- 12 G4 colum to give the multiplying out in the above exaze TEL o] b 0 -RL & Bele) ok & -Foon Mow form the matrix of redundant coefficients (Fe) * for the Rew unimowns (subscripts 9, 93 [810] ) Ce) IEE The condition of this matrix te greatly taproved over that obtained for as, ay alone (previously computed), viz., Gea] =} “est E Hl (, 825, 4) (That the [o,_) antrtx obtained here nappens 2] to be statlar to mat obtained for a, Ge tna previous example, is coincidental.) Once 'a erandcormacton sas beon perforned, vescung fole ten sat remuncare satcix [aj ‘the problem may be completed in the ">, ¢ system”. The appropriate equations are aotain ed trom eqs. (24), (21), (23) and (25) simply by replacing ali 'r, s,* ty 7p, 0", Thus (%}*Eal{*}° Ba} {4} where the recundents 9 ot 1, (= Gg) are the solutionsANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES Goal {%p> and where Ceal{*} Gea] = Ga) Eis] Esa] ‘The final unit load distribution 1s Ca] = Gs] - Be] Be Gel---- and the matrix of influence coefficients is given by Gal- Bl - BGG = THE MAGNITUDE OF THE REDUNDANT FORCES} SIZE OF ELEMENTS IN [2,,] Inspection of ‘that the redundant corrections to the butions originally eqs. (14) or (29) reveals forces ect in the nature of determinate stress distri~ aseumed (the [s,,] )- Te ts apparent that {f these corrections ars large, shen any inaccuracies in the redundants (ar{sing| frem the difficulties inherent in accurately inverting (%s} or [P.0}) will nave an important effect on the accuracy of the final stress distribution. Thus, as a matter of eral practice, it 1s desirable to keep the magnitudes of the Tedundant forces as small as possible. It follows immediately that one should use for the determinate stress distribution one re~ quiring a minimum of correction, 1.2., Salect as the determinate stress distribution one watch’ ‘dpproximites the tue stress aiserioucion as SHosely ss possible. Tite given in connection with the “choice of redundants" (rule 2, above) te an aid in caking a good selection for the determinate Stress distribution, if, 2s suggested, 2 deter-| minete structure 1s obtained by aking “cuts” whiten Leave & system having properties similar to the original, the stress distribution obtain- eq therein by statics should be a fair epproxi- nation to the final true stress distribution. However, it ts even sore important to realize that’ the determinate stress distribus tion only need be in static equilibrium with the external applied leads and that {t may be deterzined With the aid of any appropriate A830 aux) rules, test {nformation, cr sven Gultive guess-wore wilen leags to § aistrisuvion Hise tothe Fina thie diserioution, Taare 1s no Reed to sét the recundant (cit) member forces equal to zero in establishing the determinate distribution, Instead, reasonable approximate values tay be employed for then. (The ce: tions to these values become the unkxowit Fedundants!) Mathematically, the magnitudes of the re- undant forces are directly dependent upon the magiitudes of the elements tn the matrices [a ine ] or [ajq] a1 the right-hand side of eqs. (21) oF (30). (The right-hand side of such a set of Simultaneous equations 1s called the non-homo- gengous part.) Thus, the relative merits of Several possible determinate stress distributions say de judged by forming the matrix product fa.) with each and comparing results. FOR EXAMPLE, if the doubly redundant struce ture of Example Problen 3, arg. A8.9 ners forme lated in matrix fora the (se] and [ay] matrices would be (see Fig. A8.38 for numbering scheme) Fig. 8,38 a 2 ay Generalized force aursbering scheme in illustrative problem. gg and ag selected as yeu edundanta. 20 0 0 Go-})° = 9 2 * 3 o 402 ° o 9 0 ae 208 -1.884 G2] 1.0 ° several, possible deterainate stress aistri~ buttons (jg) Miah so" be tried, PIEST, he stress distribution obtained by statiss alone in =o) 6 1 ° ° the “cut” structure (a, = aA832 (This ts certainiy a poor approximation to the final stress “istribution). Multiplying ous Fal, EOE D) SCOND, 2 stress distribution in watch loads of 0.40 Ibs were guessed at for a, and ay. (Ina ‘tno times redundant stPucture any oxo" forces may be assigned values arbitrarily walle satis~ fying statis equilibrium.) aq, a were found dy statics. Thus, tt vs Case” {aor} cuss . Note that 2 reasonable guess at a stress distribution resulted in nda-hohdgenous terms only one-tenth as great as those obtained by use of the "out" distribution, The magnitudes of redundants are correspondingly reduced. =1126 ques THIRD, a5 2 matter of tnterest, che true stress distribution, obtained in Example Prab— lem 3, was used, The result: [ 450 zee | asda {f= = 4} eo be ye ‘TRUE +290 ‘The nonhomogeneous temas are practically zero, as they should be". Tue redundant forces would be zero also. ‘+The demonstration here suggests a useful check upon the finsi remult ofa redundant stress calcslation, Alter vb- taining the fia irue stresses [Giz] = [Gyn], one forms (= ggye 7 (eet 2] [4] and compares the regult element-by-element with the matrix previously computed, [era] + [ee] [54] [5] “The "trve-moatris plements ough tobe 2079, oF early # U [Gimp] ta error-tree. az 2245 | BY STATICALLY INDETERMINATE STRUCTURES Exam: Prop! TE the calculation In the SIFed to increase the accuracy of asa of the box team, Example Provler 15. It will ve assured for this purpose that the {nitiel data of that problem Were sufficiently precise to warrant an increase of accuracy, Solus 2 first step taken ms an examination of the unit-reduncant stress ¢istrivutions sbta:ned with the chotes previously rad as redundants, The three un Qistributions were reoresented Pig. 48,59. dant stress phicaily ag in <3 aLTt fe) Fig. 88.29 nitredundant-force stress distributions - az} axial Mange forces. Inspection of the figures stoned thas the following combinations of distributions give new distributions ably less “overlap”, noule Likely to have censider- 20.87 [ 23.38] 1, In matrix form the transformation was Gal-Exl (7° mos 1 0 2 --8330 1 using [2,,"] 48 previously computed, the sulti~ plication gaveANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES NOTE: VOIDS DENOTE ZEROSS. Fig. 48.39a shows plots of the new cistrt- butions, these having greatly reduced "overlap" (coaparé with Fig, 48.38). “anon 23.38) 25.08) Fig. A8.390 New wait redundant stress dstritntions, [s,5] ‘Tre now redundant ¢ ficient matrix [: ] was obtained by multiplying out per eq, (30a). 2584 = =,01282 ,0000174 (Eeg] = 3oo |-soe2 280s -.caze © |.c000178 03124 1284 found originally in Sxample Problem 15, viz., A833 T3774 «1782 .0520 Ed si789 2878 .07822 0520 O72 1254. It remained to select a determinate stress Gistribution which would reduce the mgnitude of the redundants, For this purpose, the engin— gering theory of bending was employed to compute Stress Gistrivit 10s satisfying equiltorium for each application of a unit external load. The result was (refer to dxanple Problem 30, Art. av). zi 2018S) 8.13 | Soares 0875] 08 [=o 5 See Ss se | Beet [Eas Ts 2 soos] sora Soa0 [ota [eer fst seuss 0176] 71g 500 | 8.00200. co S22 NOTE: VOIDS DENOTE zeRoas, Next the matrix [on] was cbtained by multiplying out per eq. (30d) S495 1841 1841 00 1167 +1167 1854 -1554 -1450 1450 1082 -1082 1445 -1445 E02 -1802 Bra] <= This result compared very tevorabiy with [2.41] fously obtained, the elements being from one Eait to one-tontn 3 large. ‘The solution may be carrted to completion in the "aa systea” using ene matrices [é0]) [3,5] [ese] ** [Son] sn eoe. (30) ane (22). At8 Thermal zee Calcio ty Matin Mads matrix notation by en extension of presented above. First, consider eq. (21), written in the tornCes}(%}7 Balffape° In Light of the physical interpretations given to [ep] an [4,q] » the equation ts seen to be a matrix-fom statement of the condition for continuity at the redundant cuts, viz., the dis placement at each cut caused by the redundant, forces plus tie displacement at each cut caused by the external loads, aust be equal to zero. To modify this equation for thermal stresses, ‘the appropriate expressions for thermal dis~ placements at the cuts must be aaded. Following the argument used in Art. 42.9 one writes {}* Gal{s}* Eal{} vinoro Gp 18 the displacesent s¢ tne rth out due to thermal straining in the determinate structure, The explicit form tor this term will be derived below. Rewritten, Gals} Balte} Ga} Bq. (33) is 2 modified torm of ea. (21), iving as its solution the redundant forces in an indeterminate structure under the application of doth external loads and a temperature distribution. To derive an oxplictt expression for 5.9, ‘the virtual work concept may be employed to advantage. Thus, following the argument of the ‘virtual work" derivation for deflections (arts. 47.7, 47.8), the thermal deflection at the rt Tedundant cut gust be equal to the total ine ternal virtual work done by the rth-redundant force virtual stresses (due to a unit rire. dundant force) moving through distortions caused by thermal strains. Since the internal stress distribution 1a expressed in terms of the internal generalized forces 4, dy, ft 18 convenient to employ these Q’s tn writing the virtual work of straining. If one lets O,, de the displacement of internal generalized force q, due to thermal straining, ‘then the virtual work done by @ single general- teed force 18 q, dig + The quantity d_ mill de called the mamber thermal distortion. The total Virtual work throughout the structura is ob— tained by suming, giving the deflection at the ri cut as the matrix proquet Sh bey {Sin} where g., 18 the value of the q, duo to a untt (virtual) loa¢ at cut r, Note that the term a. aT STATICALLY INDETERMINATE STRUCTURES mey consist of the sun of several contributtons should q, act on more than one member. Because On 1S dasired at gach of the dundant cuts, sq. (34) 1s expanded by writ: Ke=}- Ea{*s} [arr] #8 of course, the cranspose of [e,. Jy the untt redundant force st: the deterninats structure, (35) Into eq. (33) gives Gal(sb- Balt) Baltes) = Solution of eq. (36) gives the values of the Teduidants q,, after waich the problen aay be completed tn the usual fashion, viz, {}+ Bal{e}- It is obvious that the use of combinatio: of redundants (the "p,0" system of Art. 48,12) is possible. One makes 2 direct substitution or (0) for (6:2 J+ [on] £°F [Soap oP s280 03 is distribution in Substitution of aq. MOR THERMAL OISTORTIONS It remains to establish the torus for 4,5 ‘Thermal strains on an infinitestinal ela- ment of honogenscus mtertal can cause uniform normal extensions only, so that no shear strains develop. Hence only normal (as opposed to shear) virtual stresses need be considered in Computing the internal virtual work. Note thet normal stresses associated with flexure gust be included. Tt follows that only virtual nor tn axially leaded bars and in beams in flexure need Be considered. Hence aq 15 zero zor all q, which are shear flows on panels or torques on shafts, BARS The general expression for the virtual xork done by virtual axial loads u ima bar under varying tenperature T is weferote ‘where a is the materta? thersal coefficient of expansion. ‘+2 will be convenient later to designate by ger the solution to 6a, (96) when the mechanical loads Py are zero, the stresses in such 2 case being purely “thermal”.ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES Several specific cases are now treated. A bar under linearly varying load with Linearly varying temperature is shom in Fig. AB.40, in this case peg, BLY y a TM. Ty Ty tty + et Fig. 48,40 ‘Then, assuming a constant ; of G22) 6 HE). aye? yee son ye BE g, ‘This expression may be put tn the form Wo 4 Sup * ay 8 a woes dy tot TE Ty eon ty Mg ee ED Hote that variation in the cross sectional area of the bar does not affect the distortions Sige Apa The alternate choice of generalized forces for the bar under varying axial load is shown in Fig. Ag4l. Bye derivation similar to chat above one finds 4 Moe ty Moy here Ay tan thy typ eact The simpler ceses of uniform load (q, = 4 jonstant) and unifora ? (7, = 7,) follow im nediately by specialization of tha above forms. Por example, for a bar under constant load a = 44, and constant temperature, Ty = 7, aur. =T, one mas Ay, BEANS For a beam the general form of expression for the virtual work done during thermal strein- ing by a top-to-botton-surface temperature dit Terence Of, varying linearly over the beam deptn, is (see art, A7.a, Ex. Prov. 24), where m= virtual coment (positive for compress ton on top fiber) = Teorro ~ Trop a = beam depth. or Applied to the case of Fig. 48.42 one gets Ly any ony = oy we 8] (aye \(@ . alex |G b 1 L =a (4e #1) a? (& <2) a F on on * M 4 c — constant Fig. 8a2 or Wray don * yy on anere ‘BOT, + SF. te & (29) au (ony + 267 ayn © (eo i) Special forms of the thermal distortion expressions for beams of varying depth may be derived readily as required. ‘Sxample Problen 19 SEES upper Siftace of the bean of Pig. 48.48 1s subjected to 2 temperature 6? above that of the lower surface, varying linearly 28 shown (t.2.,58.36 the temperatures are equal at the left eng and differ by Of, at the right end). Determine the center reactions assuming o 1s constant. Solution: In the illustrative example of art. 48.12 this structure was analyzed by employing as. generalized internal forces the bending moments Q and dy over ths central supports (see Fig. 48.35). "The two central reactions ere denoved dy G5 and q,. ‘The matrix of redundant coettie~ tents, considering qi and qe as requncants (tne better choice, {t will be recalled) was (rer. Art. A8.12) Ga-a& [J] ‘The unit redundant load distribution for qa = 1 and Q, = 1 was r | fa | 2 rpaf_o z[o] a Ba-BEpt 1] 2 ey tlre Momber thermal distortions were computed. (Note that OT was negative according to the convention adopted earlier). 1 2 whee aw (8 an FO) SH SO sretex}e) STATICALLY INDETERMINATE STRUCTURES Cag} U3} of {S} {%}- Bal (}* Ea {s} Solving 287 1983, Bra ote ® sOrmam [i ® © 1} fest| EE seg 1 Lg ct o7E 287 = fact, | ‘333 ape a 1339/L -180/t Tuus the reactions were : Ele oy as = 299 Phe oe a, =-1.60 Ha fy (negative inet Tr Example Problem 20 SESE SIREEFTS sheot-stringor panel of Fig. 48.44(a) 15 to be analyzed for thermal stresses developed ty heating the two outside stringers to a mitorm teuperature T above the center stringer, Assume G = 0.3855. cates DOAN) Solution: ‘The panel was divided for convenience into three bays. The numbering and placing of generalized forces 1s shown in Pig. 42.44(b). ‘Transverse members (ribs, not shown on Figure} were considered rigid in’ their own planes - 2 Satisfactory assumption for symuetric panels, Because of symmetry only one nal? the panel was handled. all menbar flexibility coettictents and thermal distortions were doubled where appropriate. ‘The matrix of flextbinity coefficients was set up as (VOIDS DENOTE ZEROES)ANALYSIS AND DESIGN OF ‘Tae matrix of aember thermal distortions vas (afer 0000888 Here, for example yg = 2 x 2 x 10 = 40 (doubled| once because q, acts on tao stringer ends and doubled again to account for the other kelf of ‘the panel). Fig. AB. ‘The structure was three times redundant. Stringer loads 44 » and d, were selected as redundants. Fig. A8.d5 shows the unit redundant load sketches ror qy = 1, de = Land Ge = 1. Il, iB o | el | [Oi (= | fey} onaine ieaghneine®, LIGHT VEHICLE STRUCTURES aan then = 0 6 o =O 0 0 -8 Lo 0 eo 10 0 0 9 10 ~ 02 0 0 028 - 0280 ° 1025 = .025 Muttiplying out per og. 18, 2.2 B80 8 172.2 1.8 0 18.8 86. ana prays] ‘then the were We.2 13.3 0 k| 13.8 wee we oe 88.2 sexving, wundant equations, per sq. (36), sloaty2 85 Finally, the ‘tion was Fig. A845‘The result compares ravorably with “exact” Solutions mace under the same aesumptions (ret. NACA TW 2240) as rar as the stringer loads are concerned. The staar flow result 13 not very Satisfactory, primarily due te the use of too few “bays” in this analysis. Brample Problem 21 The uniform four-tlangs box beam of Fig, 4€,46e is to be analyzed for the tnermal stresses developed upon heating one flange to a ‘temperature T, unifora spanwise, above tae other three flanges. Fig. A646 Solution: or Tas beam wes divided Into four equal bays giving a four times redundant problem. Four Self-equtlibrating (nero-resultant) independent fi stress distributions mere taken 2s the unknoms, these being shown in Fig. A8,46(b). Such zero Fesultant stress distributions are the only ones possible in a structure having no applied loads’, ‘The zatrix of menber flexibility coeffice dents was formed by collecting cosfficlents iran Gy te rosuelnes ant’ sting Maes eadh ety sucesso ae Gal kgen a eel Sa Where k* = Gt / az (b +c} Mulet plying out: ——e* Since extemal leads are not to he agpited it follows from ‘statics that the generalized forces {or adjacent webs And cover sheets are equal, as are the loads in {coat aad rear soar caps at any given station, Hence an economy of ‘umbering in the generaiizea force scheme 1a possible, labor is saved in the tandling of cata when the same symbol can be employed cn several members whose loads 8 STATICALLY INDETERMINATE STRUCTURES Ly “alma | Ly Multiplying out per eq. (15), ogee sel=[ =] a |S8- ce] om | es = | = [sas vas | me |e For 8 specific case let K*L" = 1. The inverse was computed to 23 i.o70_[= 5203] ~ .2887]- caer] = 8201 | 1.886 [= .3613|- sole Ete +c) [~ 2es7 [> seal Lao? [- ae3e = _.068@7] = 10i8|— 1984 1.752 Member therma fhe Several members, Unit reduncant stress die-| computed ror loads fetdutions were prepared, taking as, Ga, de and | collected from the {2} WONeNONO 1 distortions (449) were Gay Gey Qe and Ge and were one heated flange only. (ote that if two ade Jacent flanges are heated equally, one aust set the corresponding 41 equal to aero; this because the virtue: losas being of opposite adjacent flanges, virtual work aust ‘are known to be equal. ‘Then the solution to sq, (36) was written as.ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES {se}? - Ga Bal (*e} which, when multiplied out gave {x}>- Finally the complete set of thermal stresses vere (Still for the case L*x* = 1) 4.079 Lisge 18562 12413) « Sesh ere) Oceut ~ 16 + oy (a}- a fe)" ‘The result compares favorably with an exact solution (NACA TN 2240) insofar as flange stresses are concerned, an error of less than 1 percent being present at the root. Shear flow values In the sheet are less satisfactory que to the (relatively) crude assumption of constant shear Inseach bean quarter. AB.14 Thermal Deflections by Matrix Methods. A. STATICALLY DETERMINATE STRUCTURES ‘The problem of the thermal deflections of a statically determinate structure was cone sidered earlier in Art, A7.8 in non-natrix fora, It should be apparent from the deriva~ tion, that the mtrix method presented for the esiculation of redundant-cut thermal deflec- ‘tions may be applied equally well to the problem of computing the thermal deflections of external points of 2 determinate structur tus {}- Bia} ores where Ogp t6 the therual deflection of external point m, and [9,,] 18 the transpose of (0,1, ‘the unit-applied load stress distribution (compare with eq. 35). 3. REDUNDANT STRUCTURES In the case of the redundant structure, additional strains are present due to the ‘thermal stresses set up; the effect of these strains upon the deflection of external points must be included in the calculation. The appropriate equation is most easily derived by visualizing the action in two stages. FIRST, the redundant structure is 8.39 "out" aking tt determinate, after which the tenperature distribution 1s applied producing therual deflections {ox}- ED} at the external points (compare with eqs. (35) and (37) }, Simultaneously the redundant cuts wAlL experience relative displacenents. SECOND, the redundant cuts are restored to zero displacement by the application of redund ant forces (this problem vas solved in Art. 48.13). The q, are given dy eq. (36); they pro~ duce additional deflections at points = {m}- Beli}, Te total deflection of point m ts then {em} Bal {a} Beal} Ga] GI Ed Bel ‘™eretore {Sa} Galt} * Bed Eig) Boel fee} or, {°sn)+ Beal (fee) Eeai{se}) --- The mtrix quantity in parentheses 1s the totai strain (thermai plus "mechanical"). ‘FOF definite reasons the equation for thermal deflections has beon left in the fora of eq. (38) rather than the nore polished forms Wich mignt de obtained by substitution from eq. (26), FIRST, the dyp, the theraal stresses, will probably have been solved for sreviously and mtl1 de readily available tn explicit fora, SECOND, and far more iavortant from a labor saving stand point, whe init load distr‘eution [e,_] (nose But, transpose ts used tn eq. 36) say be any conven font stress distribution eatisrying Statics im the simplest of "out™ structures. One need ict fen uso the came [eg] cuetrcoution (and seme choice of "cuts") as employed in the redundant ‘thermal stress calculation; a zore convenient choice of cuts may be employed! In principle, any stress distribution statically equivalent to fhe Unit applied iesd(s) may ve used for [a in eq. (38). (See p. a8.3). tm88.40 Example Problen 22 Cespite the rotation cecurring at the right hand end of the beam of Fig, A843. Solution: To compute the rotation 2 unit couple was applied (positive counterclockwise) at the right hand end. an additional generalizea rores, called de, also as added at that point. Then, +267 os 2 (ieee 6 :) and, using some results from Sample Prob. 19, {ea} Note that qs and q4, the intermediate support reactions, were omitted fron consideration. They do not enter into any expression tor the interal virtual work of the structure; or, equally, they are not used to describe the strain energy of the structure. Hence they are Rot included in writing the total strain. (Thedr aq are zero.) The member flexibility ccefficiant matrix salon cao a} tet 2a5 was 12,5 l-& fp 42 Bala o 1 af * frou Example Problem 19, the true thermal stress distribution nas 267 Ere at ay pe EST | eas {u} 8} se. (a. was zero by inspection) fo deteraine [,,]} 2 unit couple a, was applied. Taking as the "cut" structure one where Q, =, 20 one hee siaply fal tndeLe ‘Then substituting tnto eq. (38), {a} “sae ae ests STATICALLY INDETERMINATE STRUCTURES oben eeu ls Tt ts apparent that the values of the re~ dundant moments q., G4 could nave been chosen aroitrarily in (Bab above, without affecting she ronatt. ore, otesrty, [aj,h sor any tect structure visualized will lead to the same result, 4, deing equal to unity in 211 cases. ample Problen 23 “~~ Compute ths axial aovemant of the of the central stringer 48.44, e free end of the panel of Fis. Solution: An additional generalized force, qio, mas. added axially to the tree ond of the central stringer (ref. Fig. 4@.4ab}, Then was Tae [Tessa]es. 33) fs} o000000888 Using the qjp as cbtained in Sxample Problem 20 (with qie = 0), che following product nas tormed: xetANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURE: Then, supstitutt ett sa] into eq. (38), 4 Ur feaints to stad te determinate strate aise EAE ras: f TE ba dotatns cttetution {gh te stresses due to a unit load Gao = 1 were computed tn the "cut" structure obtained when G = Ge = ds = 0, (Note that this ts a attter- gnt choice of redundant cuts from that employed {n computing the thermal stresses in Semple Problen 20.) Thus fs: Tooorrroee and finally, 29.19 26.55 13.34, Lesh ‘ 33146 5.656 4.24 6.855 ‘see 2.402 Cy sat Locotiioce cb =84.5 a7 (ANSWER) As a matter of tnterest, a different untt stress distribution was employed with 2 difter- ent cnoice of "cuts". If the forces a4, a and Qa are set equal to zero ("out"), application of AB. <1 a unit load q,. 21 gives (writing the transpose oa} L&u_} = |_150 +80 50000 woes oot | ‘Then aultiplying out for the theral deflection: Onp = 34.37 (same answer). Th Is apparent rox the above result thet the simplest determinate ("cut") structure should be usec to conpute (g,q5 17 1s completely adequate, CLOSURE ‘The general thernal stress problem is complicated by the fact that the material properties , G-and a vary with temperature. The problem created thereby 1s primarily one of bock~ Keeping - computing the member flexibility caefti— elents (a,,) an¢ the member thermal distortions (4jq) for & structure whose properties vary trom point to point with the temperature, The vari- ations of £, 0 and a with T will, of course, have to ve snow from test data. Two additional complications, not considered here, are the loxering of the yield point with heating (and the attencant increased likelihood of developing Inelastic strains) and the phenom enon of "creap" (the time-dependent development of inelastic strains under steady loading). Should it prove necessary to analyze for ‘thermal stresses under nore than one temperature distribution, the member thermal disterticn, satrix {typ }aay to senersiized cestly Into 8 rectangular form such as Ga) “PoE Where dj 18 the menber thermal distortion associated with force q, from thermal Loading condition R, The matrix (6, "] of ember thera soetticients consists of the constant cosftictents in the 4, expressions previously presented, wntle Tyq NOUlG be the temperature associated with a, tor condition R, (coapere with 2g. (26d), art, A711}.ABAD AB.AS Problems. Note: Problems (1} through () below may be Norked by either the Least work or Dumy Unit Load Methods. The student will de well advised to try some problems both ways for comparison. A sor an 97 ow 100" t > oat, Be soo A Fig. a Fig. > (1) Determins the load in ali the members of the loaded truss shown in Fig. (a), Values in ( } on mambers represent the cross-sectional area in sq. in. for that aember. All members of same material. (2) For che structure tn Fig. (b), deter~ mine the load in each member for 2 700# load at joint B. Areas of members are given by the values in ( ) on each member, All members nade of sane material. 300 Fig. * 10004 (3) For the loaded trues ta Fig. c, determine the axial load in all members, Values in parenthesis adjacent to members represent rela tive areas. § is constant for all members. (4) For structure in Fig. 4, calculate the axtal loads in all the members, ‘Values in parenthesis adjacent to each meaber represent relative areas. © ts constant or same for all members. Fig ¢ STATICALLY INDETERMINATE STRUCTURES (8) For the "King post" truss in Fig. ¢, calculate the load in meuber 6D. Memoers a3, B80 and BD have area of 2aq. in. sack, The con tinuous member ADC has an area of 2.25 eq. in. and aoment of tnertia of 21¢ in. sane for all aembers. 40008 0 | 60 60 Fig. & 4. (6) In Fig. £, AB is a steel wire botn 0.50 99. in. area. The steel angle frane CBD has 4 4/3q. in. cross section. Determine tae load in member AB. E = 30,000,000 pst. (7) in Fig. g find the loads in the to tie Tods BD and CE. [ae = 72 1n.*; Ang = 0.05 aq. In. Age = 0.35 SQ tn. 8 18 sad ror all Eeabers® 5 fe bo or ” / | ‘000 4 je soln wt FE sco Fig. 8 Pies (8) For the structure in Fig. h, deternine the reactions at points a, 5. Members CE snd ED are steel tie rods with ateas of 1 sq. in. each, Member AB 1s @ wood bean with 12" x 12" cress section, Esteei = $0,000,000 P51. Sxood = 1,300,000 pst. (9) For the structure in Fig. 1, determine the axtal leeds, bending monents and’ shears in the various members. The structure is continuous at joint D. Members AB, BC are wires. The member areas are AB = 1.2, BC = 0. 6.0; BDE = 10.0. The moment of Inertia for members CD = 60,0 in.t; for EDE = 140 ins (20) Re-solve Sxample Problem 10, using as r2dundants the two restraints (couple and transverse force}, Solved way the problem is doubly redundant as no advantage is nade of the symmetry of the structure. (11) Add two additional members, diagonals FB and 20 (each with areas 1.0 in*) to the truss of Pig. A7.85, chapter A-7. Find the matrix of ag.15 one end tn this Ds atANALYSIS AND DESIGN o influence cceffictents, aos. ae o.2 as 18 1 | a2 68.5 65.4 288 [o]* 2 | ars ese cee a62 18.3 25.4 25.2 20.5 (22) Resolve the doubly redundant beam of Exemple Problem 8, page AB.3 dy matrix netacds. The redundant reactions should be given "4" symbols, (See Bxemple Problen Ida, page 46.20). (23) Re-solve Sxample Problem $, age A8.12 by matrix methods. For simpiletty, make your choice of generalized forces including those designated as X and ¥ in the example so that Figs. A.21 and A8.22 can be used to give ‘the gyp loadings. (14) By matrix methods re-solve Brample Problen 4, p. A6.12 using 3 equal bey divisions along the panel (3 times redundant). Use the same structural dimensions as in Example Prob- lem 20, p. 46.6. Compare the results with those ébtained from the formulas developed in Example Problem 4. (15) Show that the matrix equation eq. (21) ts modified to cover the initial stress| VESICLE STRUCTURES Aes problens of Art. A8,8 by writing (o=s] {a} + fol Cr} - Gel} nore a {2 the initial imperfection assoctated Mth force qy. Refer to the argument leading to eq. (11) of art. a8.d, (16) Using the equation of problem (15), above, resolve Sxample Problem 5, p. AG.1¢. (27) Using the equation of problem (15) above, re-sclve Example Problem 7, p. A6.14. (18) Using the matrix methods of Art. 48.13, re-solve Example Problem 9, p. A.15. (29) For the doubly symmetric tour flange box beam of Example Problen 15, p. 48.24, de~ termine the redundant stresses qa, Q and Qua if one flange is heated to 2 tenperaturs T, uniform spanwise, above the renainder of the structure. ans. Qa 2081 a 1608 aad ed perms ‘Douglas DC-# airplane. Photograph showing simulated aerodynamic toad being applied to ‘main entrance door of fuselage test section.AB AA STATICALLY INDETERMINATE STRUCTURES DOUGLAS DC-8 AIRPLANE, An outboard engine pylon mounted on a section of wing for static and flutter tests. The steel box represents the weight and moment of inertia of the engine,CHAPTER AQ BENDING MOMENTS IN FRAMES AND RINGS BY ELASTIC CENTER METHOD. AS. Introduction In observing the inside of an atrplane fuselage or seaplane null one seas 2 large munder of structural rings or closed ‘rames. Sone appear quite Light anc are essentially used to saintain the shape of the body metal shell and 9 provide stabilizing supports for the Longi- tudinal shell stringers, At points where large load concentrations are transferred between bedy| and tail, wine power olant, landing gear, ate., relatively neavy frazes will de observed. In null constriction, the bottom structural crem~ ing transfers the water pressure in landing to tha bottom portion of the mull cranes wnien tn turn transfers the losd to the Aull shell. In general the frames are of such shape load distribution of such charecter that these franes or rings undergo bending forces in transferring the applied loads to the other resisting portions of the airplene boay. ‘These vending forces produce frame stresses in general waiea are of major importance in tne Strength proportioning of the frame, 2nd thus 3 reasonable close approximation of such cenc- ing forces 13 necessary. Such frames are statically indaterninate relative to internal resisting stress and thus consideration must be given to section and phys(c2l oroperties to obtain 2 solution of the aistribution of the internal resisting Zorcas. There are any aothods of applytns the princtples of continuity to obtain the selutten sdundant forces in closed rings or ‘Tne author prefers the one he "Elastic ny years tn] netnod nas orizt- ferred to as Sten Muller-Bresiau®, The sain difference nethod as comnared to aost other aetnods| ution 15 thet the redundant for¢ acting at a szeclal are as- ‘int called the slas- edunaants which are ind Syjutter-Bresiau, H., Die Neueren Hethden der Festigkeltslenre und der stank der Saukon- steuktionen, Leipsig, 1886, a9. Assumptions In the derivations which follow the dtstor— tions due to axial and shear forces are noglect- 24. In general these distortions are small compared to frame bending distortions anc thus ‘the errer is small. In computing distortions plane sections are assumed to remain plane after bending. Tals 1s not strictly true ise the curvature of tne frame chanzes this linear distribution of bend ing Stresses on 2 frame cross-section. (correo~ ‘tions for curvature Influence are given in Chapter 415. Furthermore it is assused that stress is proportional to strain. Since the airplane stress analyst must calculate the ultimate strength of a frame, this assumption obviously does not hold with heavy frames where the rup- turing stresses for the frame are above the ora portional limit of the frane material. ‘Tis chapter wil] deal only with the the= oretical analysis for bencing moments in frames and Tings by the elastic center method. Prac~ tical questions of body frame design ars covered ina later chapter. ‘The following photograshs of 2 sortion of the structural framing of the hull of a s3a~ plane illustrate both light and heavy frames.Age BENDING MOMENTS |A9.2 Derivation of Equations. Unsymmetrical Frame, Pig. A9.1 shows an unsymmetrical curved beam fixed at ends (4) and (8) and carrying some external Loading Pi, Pa, otc. This structure 1s statically indeterminate to the third degree because the reactions at (A) and (8) have three unknom elenents, nanely, megnitude, direction and line of action, making a total of six unimoms with only tnree equations of static squilibriun evalladie. Fig. 9.2 In Pig. A9.2 the reaction at (4) has dean replaced by its 3 components, namely, the forces X, and Y, an the moment My and the structure is now treated as 2 cantilever beam fixed at end (8) and carrying the redundant loads at (4) and the known external loading P,, Pa, etc. Because joint (4) 1s actually fixed {t does not suffer translation or rotation when structure 1s loaded, thus the movement of end (4) under the loading system of Pig. A9.2 aust be zero. Therefore, tares equations of fact can be written stating that the horizontal, vertical, and angular deflection of roint (A) must equal zero. IN_FRAMES AND RINGS consider @ suall element ds of the curve bean as shown in Fig. A9.2. Let My equal the tending manent on this stall eleneit due to the given external lead system. The total >ending mouent on the element ds thus equals, MEM tM > ar + Tax (Moments which cause tenston on the fibers of the frame are regerded as positive moments.) ‘The following deflection equations for point (A) must equal zero: ° (angular rotation of (8) = oy * 3, (movement of (A) in x direction = oj ay = 0% (movement of (A) in 7 direction = 0) Froa Chapter A7, which dealt with defisetion theory, we have the following equations for the movement of point (a)i~ In equation (2) the term m is the dencing moment on a element 4s due to a unit moment applied at point (4) (See Fig. 9.3). {mg moment 1s thus 4 (ose equal one or unity §eit moment Fig. A9.3 ‘The bend= for all ds elements of treme. ‘Then substituting in equation (2) and using value of M from equation (1) we obtain - 9 = gE + abet — cgrht when: In equation (3) the tema a represents the bending monent on 2 element ds due to a unit load applied at point (A) and acting in the x direction, as {Llustrated in Fig. AS.4. + tgs + PEP so (5) The applied unity road tas a posisive | as sigs ao tt has ben A assimed acting toward wT, the right. The * 8 distance 7 vo te cs slanent <9 2 piis Fig, 8.4 distance as it IsANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES measured upward from axis x-x through (4). However the bending moment on the ds elenent shown is negative (tension in top fibers), thus the value of m= (1) y= -y. ‘The minus sign 4s necessary to give the correct bending aonent Sign. Substituting in Equation (3) and using trom Equation (1):- ae + DIES - gai - nail In equation (4) the term a represents the bending moment on a element ds due to 2 unit load at point (A) acting in ¥ direction as EGE os) sllustrated in Fig. 49.5, Hence, a = L(x) = x Substituting tn equation (4) and y ds using ¥ from equation (1), we obtain, PHEEE 6 yr _ ay nest =o Equations 5, 6, 7 can now be used to solve for the radundant forces My, Xq and Yq. WLUR these values nom the true beiding monent at any point on structure follows from equation a. REFERRING REDUNDANTS TO ELASTIC CENTER For the purpose of simolitying equations 5, 6, 7, let it be assumed that and A ts attached to @ inelastic arm terminating at 2 point (0) as {llustrated in Fig. 49.8. The point (0) coincides witn the centroid of the 3/EI values for the structure. Reference axes x and y will now be taken with point (0) as the origin. The redundant reactions will now be placed at point (0} the end of the elastic Fig. 49.6, inelastic bracket, 28 show in Fig. A9.6. Since point A suffers no movenent in the actual structure, then we can say that point (0) must undergo no movement since (0) 1s connected to point (A) by @ rigid arm. ‘Thus equations 5, 6, end 7 can de rewritten using the redundants Xj, Yj, and Mp in place of Xg, Yq and My respectively. ‘te axes x and y through the point (0) are centrotdal axes for the values ds/ai of the structure, This fact means that the sumations- yas xds reo ane 2SBe0 ‘The expressions Z x*ds/EI, Z y*ds/EI and 2 xyds/Et also appear in equations 6 2nd 7~ Thoge Coma nill be referred to a8 eleatte Renente of inertia and proqiot of inertia of the frane about y and x axes through the elastic conver ef the. trane, and for siapiictty wilt be Given the following’ syabol. xtds Mas = poy Ey ay Equations 5, 6 and 7 will now be rewritten using the redundant forces at point (0). e2 2% + mrs 0 hence, 83 Bo Xyte = Tol 0 by F2 = = (8) by 2 ‘Te term Mol ¥as/EI ts zero since 2 yds/E1 {5 ero, thus M drops out wnen substituting tn Equation (6). = TREE - xgtyy + toly 20 = = = (20) The term Mgds/OI represents the angie change between the end faces of the ds element when acted upon by a constant static menent My. ‘his angie change whicn actually ts equal in value to the area of the M,/BI diagram on the alonent ds will be given the symbol Bg, that ts, $, = Nqds/EI. itn this symbol sudst{tution, equatians 8, 9, 10 can now be rewritten 2 rellows:~ WE Bey * Xly = Yoley FO === = = 2112) 2 Bx ~ XKolay + Yoly Solving equations (12) and (13) for the redundant forces Xo and Y, we obtain,Aas BENDING MOMENTS IN FRAMES AND RINGS = Mae = Bex Ge) te AG) ee. ca & G- EE) ~ Losec = yy (2) rE GeEy tot a ‘A9.3. Equations for Structure with Symmetry About One Axis through Elastic Center If the structure ts suen that either the x or y axis through the elastic center Js a axis of symmetry then tne product of inertia Dxyds/Sl = Ixy = zero. Thus making the term Igy = 0 in equations 11, 14 and 15 we obtain, A9.4 Example Problem Solutions, Structures with at least One Ass of Symaetry. auple Problem 2 1016. Fig. 49.7 houdrer_o shows a rectangular frame with fixed supports at points 4 and D, and carrying a single load as shown. The problen {3 to determine the bending moment aia~ graa under this 1 Toading. The tirst Fig. 9.7 step in the solution 1s to find the location of the elastic center of the frame and the elastic aoents of inertia ly and Ty. Due to symetry of the structure about the ¥ axis the centrofdal T axis 1s located midwey betseen the sides of the frane, and thus the elastic center (9) lies on this axis. Table A9.1 shows soe of the necessary calculations to determine the location of the elastic center and the elastic soments of inertia. The reference axas used ars x*-x* and y-y. Table Ag. Pore| ade ft. dst. | yg ton | Ea AB |2si0) 8 | -12 |-120 12} 129 Tre temas t; and ty are she elastic soaent of inertia of each portlon of the crane ebout its centrofdel x and y axes. Since 1 ig con- Stant over each portizn the centroidal moment of inertia of each portion ts identical to that of a rectangle about its cantroidel axis. To explain tor aenber 48 dsdes Referring to Fig. a, lone 2 2 by gos spe = bx Fx co Referring to Fig. 0, ty eee tin ctecmes toate Slastie senteraes aowee ff calculated: - | | y Having the aoment of inertia about exis ctx? we can now find its value about the centroidal axis xx of the frane, by use or the parallel, axis theoren. tw Tx = an(F*) = 29800 = 52 x 20.605%= sles Ly = Ty ~ IW(R*) = 3456 ~ 32(0) = 3456 ‘The problem aow consists in solving equa tions (16), {17} and (18) for the redundants at the elastic center, nenelyoR ANALYSIS AND DESIGN OF FLIGHT VEHICLE STRUCTURES = Bq _ Area of static M/I diagram My = avi * Tiger slistic welsh or Seniotae Moment of static M/I diagram about ony x axis Xo = “TO * Elaseis uouent of inertia about x axis Moment of static M/I diagram about y axis BUSStie oaeat or inertia about y axis to > Ofte Thus to solve these three equations we sust assume 2 Static frame condition consistent with the given frame and loading. In general there are a number of static conditions that can de chosen. For exanple in this problem we might select one of the statically determinate conditions illustrated in Pig. A9.8 cases 1 to casei 108 TALE ‘ae OS at case 2 “case $ Fig. 9.8 Tq Illustrate the use of different static conditions, three solutions will be presented with each Using a different static condition. Solution Now 1 In this solution we Will use Case 3 as the static srane concition, The bending moment on the frame for this statte frame condition is given in Fig. 49.3. The equations ase el dogram 2-SNg/t diagram Ms is positive, tension fon inside of frame, fe, Thee ig. 9.10 for the redundants require @g the area of the Mg/I diagram. Fiz. 13.10 shows the Me/T curve natch is obtained by dividing the values in Fig. AG.9 by the ter 2 which {s the mment of Inertia of member BC as given {n the problem. Sincs the equations for Xp and Yp require the Ronent of the Mg/I dfagran as 2 foad about axes through the elastic cencer of frame, the area of the Mo/T dlasram will be concentrated at the centroid of the diagram and along the centerline of the frame, or more accurately 49.5 along the neutral axis of the frame members. In Pig. 49,10 the area of the M,/I diagram pow EES secon > -L x ana y troa the axes ms, My = He: ty = = wots Fig. A9.12 shows these values of the radundants acting at the elastic center. = (270) Tiron Table AS.1 * 270(9.275) eee =. =8.497 in.1D. 7989 1b. HEROUCE) = 9.1582 1b. San Fig. A913 ‘The bending monents que to these redundant forces will now be calculated. My = - G.497 - 1562 x 12 + .7989 x 20,625 > 3.06 in.1b. Mg = - 6.437 - .7089 x 9,875 - .1862 x 12 = 17.75 tn. Mo = - 9.87 - ,7989 x 9.376 + -1562 x 12 = 14.00 tn.tb. 8.487 + .7989 x 20,625 + -1862 x 12 = 9.81 tae. ‘These resulting values are plotted on Fig. 49.12 to give the tencing moment diagram due to 12 redundant forces at the elastic center,49.6 BENDING MOMENTS Adding this bending moment diagran to the static dending dtagran of Pig. A9.9 we obtain the final bending moment diagram of Fig. aS.15. ‘The final bending moments could also be obtained by substituting directly in equation (1) using sudscript (0) instead of (4), Thus, Meaty +My hy + ox For example, determine bending acment at point 8. For point B, x = -12 and 7 = 9,375, Ms = 0 substituting in (19) Mg = 0 + (-8.487) - .7589 x 9.375 = 11582 (412) = 17.75 as previously found AT POINT D. x= 12, y = 20.625, My #0. Mp = 0 + (8.497) ~ .7989 (20.625) + 21582 x12 = 9.81 in.Jb. Solution No. 2 In thig solution we witl use Case 4 (See Fig, 9.8) as the assumed static condttion, ‘that 18 two centilever beams with half the externa loed or 5 2b. acting on each cantilever. Fig. 49.14 shows the static bending nonent diegran and Ftg.,a8.15 the ert diegren, we 405 -20) Lso watts Fig. A914 0 “0 ag . So Dat srbtin aris fart et Se © ig, 49.18 (452405)9.375+(-00-300) ( 25) Bey Xo = Bee = 0 = ae = 0.7939 1b. IN FRAMES AND RINGS mix, [15 {-10)~105%5-200(~12) 00x12} 7 SSS 259180) _ GRE! = 2.6 we. Yo ‘The final moments at any potnt can now be found by equation (19). Consider paint 3:~ Ms = 250 trom Fig. 49.14 x 912, y = 8.875 Subt. in (19) Mg = ~30+51.55-,7309x9,975+2,656(~12] 17.95 tn.ib. which checks £1 solution. Consider point D:~ Mg = =90, x = 12, y = ~20.625 supe. in (19) Mp = ~90+51.56 ~ 7969(-20.625) + 2.656 x 12 = 9.30 tnt, Soles 3 In this solution we will use Case S (Pig. 29.8) as the assumed static condition, namely a frame with 3 binges at points A, D and B, as {llustrated in Pig. 49.16. Before the bending moment dia~ gram can be calcu lated the reactions at A and Dare necessary. ‘so To fina ¥p take meaents about point a. pst 5 My = 10x5 = 24¥p ° hence, Vp = 2.8 To find Ya take 2Fy = 0 =o hence %y = 7.3. 10+ 2.5 + Va To find Hp take moments about hinge at = of all forces on frame to right side of = and equate to zero, Wp = 2.5 x 12 + SOHD= 0, hence Hp = 1. ‘Then using ZF, ory, 0 for entire frane, we obtain 19900 Tue next step tn the solution ts to assume a static ring condition and determine the static (tis) diagran. In general tt 1s good practice to try and assume a static concttioa such that the Mg diagram ts symmetrical about one or if possisie about both x and y axes through the Blastie center, thus making one or both of the Tedundanta Xj and Yo zero end thus Tedueing consideraply tne aucunt of nmerical calcula tion for the solution of the problen. In order to obtain symmetry of the Ms dtagran snd also the Mj/I diagram since I 1s constant, the static condition as show tn Fig. ag.24 is assumed, namely, 8 pin at (e) and roliers at (£), The static bending moment at points (a), (b), (c) and (d) are the same nagnitude ani equal, Mg = 50(18 ~ 18 cos 460) = 265 tn.10. ‘The sign 1s positive because the bending monent produces tension on the inside of the ring. ‘Tne next step 1s to determine the Js, dgx and aay values. dg is the area of the Me/T diagram, how ever since I 1s unity 1t is the area of the Afagrem. Tha static My disgram of Fiz. A9.24° 45 divided into similar portions labeled (1) and (2). Hence area of portton (1) = Pr#{e -in c), gn nie 50 1d. and a = 459. where P Substituting and multiplying by 4 since there ara four portions labeled (1).» Bs(x} = 4[50 « 188(0.785 - 0.707)] = sos2 ‘he area of portion Labeied (2) equals, pre(1 - cos a). Since there are te areas (2) we obtain, Wy (a)= 2[50 x 18° x F (2-0.707)] = 15000 Hence, total J = 16000 + S0s2 = 20082 Since the centrotd of the My diagram due to symetry about Doth x and y aes coincides4.10 IN FRAMES AND RINGS BENDING MOMENTS with the center or elastic center of the trane, the teras Zigx and Udy will be cero. Substituting te determine the value of the redundants at the elastic center ne obtain, = gis, = = [2e052) Mo = est “TIS. 2 Tay - 20082(0) 2 Xo ST = To = 177 and. Bax « (200521 (0) Yo Sp = —Texao =O Fig. A9.25 shows the values acting at the elastic center an¢ the bending moment diagram produced by these forces. adding the bending mouent diagram of Fig. 49.25 which {s a constant, value over entire frame of ~177 to tha static moment diagram of Fig. A9.24 gives te final Dending mouent diagram as shown in Fig. 49.26. Example Problen 4. Hull Frame Fig. A9.27 shows a closed frame subjected to the loads as show. Tne problem is to determine the bending moment diagram. statle moment Fig. 49.29 Fig, 49.20) Solution The first step 1s to determine the elastic center of the ‘rame and the elastic aonents of Inertia, Table AG.4 shows the calculations, A reference axis x'-x' nas seen selected at the micpoint of the side 4B, Since a sta’ condition has been Selected to maxe the diagram symmetrical about y axes through the elastic contar (sea Fig. 29,28), {t ts not necessary to determine ly since the redundant Yo mill be zero due to sis yan Table 9.4 [Mester “aa |t wet] y iy | tear tixe mn? Bow [Pea f.5] Sa af , aad oonisiz00= 158600 Fas. ‘aa _[e0.0 fo] 0.0] 0 [0 liecoowo = se000 we [ooo [id so.o] 0 [0 liso00+0 = sano ‘Ac _[onca [nd ance] waa fooslona = so0s0 = oer car_fasca [nd v0.2] aa feoeiean » aneso > oarm Teale fal (ra ea In the last column of Table A9.4 the ters iy 18 the moment of inertia of 4 particular mamber about its om centroidal x axis. for member 803; 23F” = 13 x 30°/1.5 = 5400 T teeoun "Ng py + debe acestcaee Let } = distance trom X'x' rer. axis to controidal elastic axis X-K. BY 2 BBs sa tn, By parallel axis theoren, Txt = 5.84*(2#) = 262140 ~ 6.54% (221.2) = 252400 lx The next step in the solution 1s to com pute the static moment elastic weignts ds and their centrotd locations. In Fig. 49.29, the static frame condition assumed is a pin 2t point Aand rollers et point a’, wnich gives theANALYSIS AND DESIGN OF FLIGHT VESICLE STRUCTURES general shaze of statis moment curve as shom tn the flgur consider whe bending toent curve will be con sidered in tno parts, manety (1) anc (2). tern @5 represents the area of the Ms/L dlagran. The ‘Thus for portion (1) and (1'} Ba(aj * Bacar) * cis 2[E000-£ 590,524 - o4s]] # 172800 qe sin ‘The vertical distance from the Line 3B! to centrolt of Mg curve for portion (1) and (1') ts, rae _ sinte ril = cos a = 2 ye——alae cola = ovae7 - 25) TEE OS 210 1n For portion (2) of the Ms diagram the area of the He/I diagram wnich equals dy 1s dg¢x) # FEE (1 = coe a)x £O002S0* HE 0 367 = 1,007,000 Consider menber AC, Sron free body clagren of bottom portion of trane (Fig. 49.31) the equation for bending oaent 8000 a aete on member AC equais: Mq = 4690 x -100 x* and ¢ when t= 2 area of equals, 38.4 we (4890 x ~ 100 x#)ex ° seus cox 100x874 Teg et 10927) sass , curve between & Distance to centrota of M/T om A vurve along Ine aot ‘38.4 way (3690 x 100 x" }xax. a 28.4 4620 _x* _ 100 x* sve Es - 3g) Fea05 Re = 21.77" Vertical distance from line AA" to centroid = 21,77 x 24/28.4 13.4", The static noment welznt for Alc! ts same as for Ad, thus Sag + Dyrqt = 2 x 784000 = 1568000 Fig. A9.29 shows the frame with the manent wolgnts 23 located at the centroids, together with the Fedundant forces My and Xq at the elas tic center. It makes no difference where the frame {s cut te form our residual cantilever, Af one of the cut faces ts attached to elastic center and the other {3 considered fixed. With the elastic properties and moment weights known ndants can be solved for: __ (4997000 + 172800 + 156800) | 12880" (-1007000 x 48.26 + 172600x 33.36 + 156500 x-5.26 ‘B5E400 =98.Se Yo ts zero because of the symmetrical trane ana iofaing. the final or tove bending soxont at any poine equals ME My +My = Ky Tus for point B Mg = 0 + (-12430) ~ (-96.5 x 23.36) = -10130"# For point Mg at point © = 4690 x 38.4 ~ 200K(38.4)*/2 = 32700 $2700 = 12480 = (498.5 x ~60.54) = 14800" Hence Me Fig. A9.S0 shows the ganeral shape of the true rraze bending moment diagran. Example Proplem 3 Fig. Ag/S1 shows the general details of one

Analysis and Design of Flight Vehicle Structures by E F Bruhn PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Analysis and Design of Flight Vehicle Structures by E F Bruhn PDF

Uploaded by

Copyright:

Available Formats

You might also like