struciital. i dames that acenrasii fluctuating strains at-nominal:s! (often-inuch less-than):the sta sires is.thus,well below theiult the: material, “yet ‘still cause cuiilused! with eyeli¢ everlodd, such as the paperclip. can be bent pase (ts yieldwpeink ty without: breaking, ‘but 4 repented - ~bending: int! cause the material tovfail. Fatigue is the catastrophic failure due to. dlysiomie (fluctuating). stresses. Tes can ‘happen’ in ibridges,. aigplanes, machine components, etc. The fi torisli¢s.are: i chi + ‘long.period ofeyclie ‘strain. + the most usual (0%) of migtallic failures (happens alscitin ceramics andi polymers). » is brittle-like!even: in ductile: metals, with little plastic: deformation ‘itocéurs.in'stages involving theititiation: Stages:af res duesto: fatigue: i ‘ . i crack initiation at high séss points (stress raisers). 4 on (inoremented gach cyeley 7 S WE. findlfailuie by fracture ; il propa Ninat’?* Ninitaion + Npronaystpropagition i 1 slow.': : i * along‘erystallographic planes of-high shear stress + Hlatand feuture’ess: fatigue sutice Suuge }- propagation : plastic blunting and sharpening of the crack tip. crack propagates ‘by repetive' + Crack Propagation Rate (not coer) ‘ i i 4 1.1, Fatigue Properties: , ligue. cracking is one ofthe primary dathage mechanisms of. structural svents. Fatigue crackifig results from-cycli¢ stresses that are below the uc tensile sires, or even the yield siréss-of the material. The name ~fatigue”.is based on.the concept that a maierial: becomes “tired” and fails. at tress level below the nominal strength .of (he mateérial: The facts that the ’ it hulk design strengths are not exceeded and the only warning sign-of ai unpending fracture is an,often hard to see crack, makes fatigue damage especially dangerous. |; ti fatigue life of a compcnent can be expréssed as the number of loading required to initiate.a fatigue crack and to propagate the crack to size. Therefore, it can be said. that jfatigue failure. occurs in three ~ crack initiation; slow, stable crack growth; and rapid fracture. éd previously, dislocations play alimajor role in the ‘fatigue crack tion phase. In the first stage, dislocations accuniulate near surface stress init eoneviitrations and: form staictures called persistent slip bands (PSB). after a number of loading cyéles, PSBs are aveas that-rise above (extrusion) I below-Gnirusion) the surface of dhe component due to moverneat of slong slip. planes. ‘This‘leaves tiny steps: in the surface (at serve as ers where’ tiny’ cracks can inifiate, ‘Theso tiny urdek (called ate alonjy plines of high shear stress Whigh is often AS*t0 oo the lodidtng direction, : i t inate }In the second stage of fatigue, some of thg tiny microcracks join together and begin 10 propagate * Uough the material in a direction ‘that is cepenticuhy to the maxintnt tensile stress! Eventually, the growth of one ora few crack-of the larger eraéks:will dominate: over te rest of the cracks: With continued-eyclic loading, te growth ‘of the dominate erick or cracks will continue -until the remy 1g Uneracked section of the componicatcan no longer support the lead. At this point, the fracture toughness is.exceeded and the remmining cross ction ofthe material experiences rapid fracuure. This rapid overload fracture is:the third stage of fatigue failure. | i \ 1.2 Factors Affecting Fatigue In order for fatigue cracks to initiate, thrée basic factors ate necessary: First, the loading pattern: must contain -minimum-and maximum peak values with large’ enough variation or fluctuation. The peak. values may be in tension or picosion and-thay change-over time ibut the reverse loading cycle must nly great for fatigue crack initiation: Secondly, the peak stress miist-be'of sufficiently high value. If the péak stresses are too low, no crack initiation will ‘occur. Thirdly, the material must. experience a sail: kuge: number of cyeles. of uhe applied swess. The number of es required to initiate and-grow:a-crack is largely dependant on the first to Factors, : In addition: to ‘these: three. basic. factors, there are a host of other variables, such as stress concentration, corrosion, temperature, overload, metallurgical siruclure, and residual stressés. which-can.affct-the propensity for fatigue, Since latigue cracks generally initiate at a surface, the: surface condition of the, component being: loadeid’ will-have an: éffect-on its fatigue life, Surface roughness.is important becdusé itis directly related to.thé level and number of stress: concentiations on [the surface. The shigher-the stress concentration the more fikely-a crack ig b. nucleaté: Smooth surfaces increase the time to ucleation. Notches, seratehes, and-other stress risers decrease fatigue life, Matis testdual stress will wlso haye a ficamt effect on fatigue ti Compressive residual stresses from, machining, cold: working, heat treating wilhoppose-a tensile ond arid thus lowsr the. amplitude of cyclic loading ‘secs SEs jong SEN or coy e humemas, in vemteSh ek Greed aSion tracing can sist oe de inating, witch zg sénaiame ta Ee¢ componénts the’ erdck propagation life is, néglected ‘in, design tress levels are high, and/or thé critical flaw size small. For other onenls the crack growth life might be a.substantial portion-of the total 1 ussembly.. Moreover, preexisting flaws or shatp design features untly reduce or-neaily elitninate, the crack initiation portion of «life of a component. The useful life of these components may be by the cate-of subcritical crack propagation, For sam tse ft fuselage structure is a good example of suvcture that is based a slow. crack growth rate design: Many. years ago, the USAF «| a great number of malfunction reports-ftom,.a variety of airerafi. - ts showed that the preponderance of structivral: failures occurred frou: 4) built-in preload stresses, 2) material flaws.and 3) flaw caused by:in- sory ». These facts led to'a-design approach that required the-damage lysis to asstime a material flaw pxists in the woist-orientation. ato’ i! the most undesirable Iocation. Thé ‘analysis helps to ensue ‘that »saare designed. that will support slow stable crack growth until the enches a length where‘it-can reliable be ‘detected using NDT methods. fp sine Fatigue Crack Propagation Hate, 91H or dafd : _ Stress Intensity Factor Ranpo,.ak ne of fatigue crack propagation is deteimined by subjecting fatigue- era etl specimens. like the compact speemen used in fracture toughness: i 1SV SSA deter mest [eb Wik ineneastn, Seches Games sharp mahdfoss ond - owctakine Baws: 35 say nase Save ng by hk Deen ne wwcharising. cindding qexpesing Bask gape) + +Reixnad Bakes od qyeting catiees expansion unt cemtetiva. hence dheomull smesn. Hoctganent is mameined Solution: = aiieinne resents bv design oso matrisls vith low themed cman TR + Comesthn Stee. Chensicgd meactions intteed pits v Sees cies Comesiom Gap eceoss & ti > Gerace cme of matin E Dasinie, s ws pescacdhe sesface coating. = eSd restore comprendive savas, Best ice hein - 5 Haga cna, Sy soso, Inde soe as ER, : st San CeeS~N cuives (stress-number of cycles to-duilure}are obtained using apparatus shown in Fig. 8.21, Different typos of SN curves.are shown in ¢ limit) occurs for some materials (like somo ferrous e, the S—N curve, becomes horizontal al large N uate limit (cndur “Hi allows). ; “Phis: means’ that’ thet a inaximuny stress: amplitude (the: fatigue Jimity below which the: material fever fails, no. matter how large the. number of eyetes For other matgrials.(¢.g,, non-ferrous) ithe, S--V-curve continues to. fall with. NY . Failure by fatigue: shows substantial variability, Fhituie at low loulsris inthe elasti:strain regime, requires a large number-ol cycles. tlyp. 10" 10 10°), Avdigh loads (pkistic regime), one has low-cycle faligue (NV < 10" ~ 10° cycles): a : wo geiteral! types of fatigue tests conducted. One test focuses on nal’ stress required to cause a: fatigue failure in. some number of vycles. This test’ resuills in data presented as a plot of-stress (8): against the numberif eycles:to-failure'éN), which is'known as an-S-N curve. A log scale is almost. always.used:for:Ni, Stress, (S) Niet ssh oe the anes 30" 10810" 408 40% * Nunibor of cyclen (My“ atigue-tife is, influenced by vatiety of factors, st temperature anil surface finish, it conmpligated ways. . Some mates (c.g). dome xteel and altaniuat alloys exhibit au endiranee linit or fatighe Hinitow limit below: which ed siressdocs ‘not induce failure, theoretically. an, infinite number, of cyclés of Idad. Generally spoitciny, 2 ate Hein: componynt being éyeled? at: ubesses: Guloee Ah endurance: Hit will til fron, sane: othe snude"before it fails from fatigue, Most. other non-ferrous. metals (e.g., aluminivnt awl. copper-alloys):extibit no suah limit dnd evensmall sires will eventually causesfailure, . AS i Means to gauge [atigue oluieta istics of nun-forrow. ather alloys that-do-not exhibit an: endurance Limit, a fatigae strength is ‘eeyuently detergrined, and aypically the stress level at whicla componentwill survive 10% londing cycles. Factors that affect fatigue-life he eter ‘hat affectfatigue life OSs. inclunling: stress concentralions: face; sy ‘face’ ‘Toughness; sora, “ele. -cause stress pan lower fatigue life, is applied! Onithi tien ; Surface-stress. cas: saily manipulated to increase fatigue ilife. For example, shoe ‘oducing surface stress are..often referied: 10) generically the mechanisny used:-(o:produce:the stress: Other ‘mare: aser ‘peening and ultiasonie: can also produce this surfagecompressive’ stress and-can igue life of the coniponent. This - improvement is. normally: : high-eycle fatigue. Little improvement ‘is obtained “in the 2 régime. Sertain materials, such-as steel, will aever fail due to farigud * senutin below a certain Jovel. Gthor materials. such as “eventually fail due to fatigud regardloss of the stiesses the t a i ale. a“8eFeCt geometry aad decation, uch: as ser ‘aches: gous dUguctifie, ste ‘The size, shape, and, location “of s. and dents can have a significant aL uneven ‘cooling, leading. to-a heterogeneous distribution. of Properties: such as ‘hardness.and ductility. and,.in:-the case’ of alloys, npoesition. ‘Uneven cooling of castings, for example, can ple. 2 giv levels of tensile residual gtr which. will encottrage crack ney; and location:of internal defeéts:. Casting défects such as gas; * sud shritikage voids, for example, can significantly impact fatigue ? i $ : vhere stwain-rate sensitivity is observed (ferrous inétals, copper, +. wtc.) stain’ rate also affects fatig tropic inaterials, ‘the ditection of tl tl to'a residual strain, Viscoelastic creep data can be'presented in one.of two ways: Total strain can be plowed as a function of time for a given. temperattire .or-temperatres.; Gclow .a critical value of -ajyplied: stress, a. material: “may -exliibir linear viscoelasticity. Above this" -criticdl stress? the grows sisptoportionately . faster. ‘The second way} of ‘graphically: presenting ic creep in a material is by plowing the dulus ~oeuelustie creep" 2 creep. modulus (const. pplicd stress divided by total strain: at a particular, time).as a-funetion ofWe Below 4 ft | time , ey is critical’ stress, wlependen es i i wig response cm A fonily sof ques. describing strain orcs Viscoclastie ores nets plied: st088. may -be represisited bya’ single ieepel © steep. modulus versus: time: cni've iit Whe appilied stresies, are Delow the thatevialls critical stress value, ¢ ms time, the viscodladtic fereep module’ is Additionally, the snieleeular: welghv of the polymer of interest is.known to altect its creep ‘behavior, The effect of ihereasing molecular weight tends to ProMets secondary Bonding between: polyrer chains”. hug stake the polymer. more créep resistant, Similarly; atématié;polymers are-even, more ‘ereep resistant due 16 the’ ydded stiffness ‘from the sings. Both sn seight and aromatic cings aéld.1@, polymers} chormal stabi creep resistance-of a.polymek. (Meyers/and:Ghayla, 199! } eer : wot? Both: polymers and metal Jean: creep?) Polymers . experience: significant cfvep at all temperatures: cibeve ~~ here.-are see a differences. between polymettic.andimetalli Actallic ercep: . ismotdineal + | -innetrecoverible «only significati 1 z 3. ‘Therinal shock a et . ‘ . t * Thermal shock: is’ the naine giver to eiacking: Aemperature'change, Glass,and:ceramic, object ie this form of failure, duc to. their tow toughingss, Tow ith F and high thermal.expansion ,cocilicients; However, they are used in Nigh temperauure applications duetoitheir highkinehing points > ' “ }te The data is i aut is obtained by evel: ay yy eye. § Y notches ¥ The aistul procedure ig ia fies i sively or notelied specimens until failure: failure is expected tn a ine St specimen ata high peak stress where Wooroe sil far euch et aed shutt numberof cycles, ‘The test stress is succeeding speci i ‘i : a Fas aic-aptel oa ae pecimen until one of two specimens do not | fuhbe ee numbers of cycles, which. is usually-at least 107 eyeles. iutewe teen at which & runeut (nun-fdilure) o¢eurs is taken as the ne ean hold, Not all puuerials have 4 fatigue threshold (most nonferrous, neta al ys do hot} and for these materials: the test is usually terminated - att SLO" eyeles. : Since che amplitude of the cyclic loading has'a major effect on the fatigue snee, the S$-N relationship is determined for one specific loading sic. ‘Thé amplitude is express as the 'R ratio value, which is the 4 peak sivess divided! by the maximum peak stress. (R=Ginin/Ginas)- Tt st common lo testat an Rratio of 0.1 but families of curves, with cack «a different R ratio, arb often developed. tion to the cyclie stvess controlled fatigue test is the cyclic strain _ Incthis test! the strain ampliude is’ held constant during Strain controlled: éyclic loading is! more representative of the “e found in thermal cycling, where ti coriiponent expands and contacts aut soase to (Muctuations in the operating temperature. Ib show First. § be ‘noted that tiere are several short comings of S-N fatigue data. < conditions of the-test specimens d¢ not always represent gomual Hons. For example, camponents avith surface conditions. sach from. corrosion, which differs from the condition of the test navns will shave. signiticantly different fatigue _ performance. ore, there is often a considerable ariount of scatter in fatigue datt 1 eurefully machined standard specimens out of the same lot oF ini are used, Since there is considerable seuttei in the data, 0 rectietion oor alten applied to-the'S-N curves to piovide conservative valacs for +4, af components. ‘ as gue Crack Grewtl Rate PropecticsFar sume components th heculbse Suess love ceompouents. Une } file ol the ass = eraek wap Sate Hah, anufoy the Cluck piowth lite might by nbly. Motcuve } uy significantly reduce or ay i ihe Jaligue lite of % yoverted by the r aaah Far ext HbsuniGal portion ot the ws B DICE ing This oF dhagp dewea Fe curly clinics the crack 1 @cCoMpORCHL The usetul lite of th ale ot suberiti itiation sestl: hone weriporicnts may bis rack propagation, Aircraft fselage structure -ig a Reod exumple of structure that ix te fargely onea slow erick Biow! th rate dusiyn. Many yetirs azo, the Mis & reviewed a great number of malfunction reports from a ise ke Barras he.reporis showed that the preponderance of structural fail ii wee reloud stresses, 2) inaterial Taws and 3) flaw cuusett by «is Durltsn preload stresses, 2) material es semive usage. These facts led to a design approach that rear ont wierance analysis to assume a material Mawiexists i bi : te aire has Wout the most undesirable location. The See eecrack bisa vemuil flue trees are designed that Will support sour st CS tn sack leaches a Jength where it can reliable be dete: fs No Wc Crane, Peepagation Rate, OSH or dahil 3 Latipys ity sta oe trose tntontity Faster Rata toni ined by subj iin sty of fatigue cick phepapution is étovestutes! ty # ies inns, Hike thy compact spreincu wsedt Ry FisPonts 5 i . 10. consiant-amplitudé cyclic loading.-‘The incremental’ inérease. i length is recorded along with the corresponding ‘number of clay sd fos aieqittre stress iritensity (IK), crack length (a), and eyele cour a). ng.the test, The data-is.presented‘in an.“a versus N” curve as shown ¢ 10 the sight, Various versis’N curves ‘can-be ‘generated by the maguitude ‘of :the. cyclic Joading and/or the. size of the initial an be reditced:t6 a single:curve.by:presenting the data in terms of wowth*rate per cycle of loading (Dal DN or da/dN) versus the ion of the stress-idtensity factor at the tip-of the ctack (DIK)..DK, is stive-of the mechanical driving force, and it' incorporates the effect ng cick length andthe magnitudeiof the cyclic loading. (See the iraclure toughness. for more’ information on. the stréss-intensity J. The most common: forin-of presenting fatigue crack growth data is a xe crack. propagation béhaviok of thany materials can be divided regions as shown'in-the image. Region 1is.the fatigue threshold s to lew to propagate'd crack, Region M-encompasses tw ol crack growth changes roughly Jineatly with a change y fluetuation. In region 1H eases in the stress. niplitude, produce relatively large increases in crack growth rate: material:isnearing the point of unstable fracture. : to iv sitme 2 th Cuaractenisticsoltatigue;failures, ving, characteristi . ‘The -pr initiation. site, movement, a Shenomehon: ‘anal inechanies, . Failure is essentially prob required for failire Waries samples, Anal lysed: i i the-applicd nytt sduniulative: Materials.do: net recover “when Damage! sled. E ‘ a iThermal shock occurs when‘ asthermal object to expand by differerit amounts, Thig differential éxpansion ean:-by understood in terms of stress or-of strain, equivalently, At sonie point, his sites’ overcomes the-strengih. of the material; enusing @ crack 1o.ferm. If aothing: stops this crack front Propagating thréugh the-matctiak, gradiont ‘causes: enkeparts iol ay the object's structure to-fail,. « Thermal shock canbe prevented iby: . ‘Reducing the thermal gradient seensby the object, by. changing :its temperature moreslowly 2 increasing. the:material'sithermaliconduceiity, re 2. Reducing the mateiial's.coéificiént OF thetmaliexpat & Tncreasing its strength ‘ ‘ 4. Anereasing iistoughtiess «+ Borosilicate glass suchas Hyréx issmade to, wa than most other ‘glass through a -comibiiiati coefficient and greater strength, though fused th ‘espects, Some glass-ceramic materials.ir of material with. a negative expansion: coeffi coelliciont can :be réduced to almost exactly: ze1 range of. temperatures. Wear . i An ‘materials science, wear is:the-erosion-of mat the action of another solid. ‘fhe-study of the discipline of tribology. Ther Surface fyugue + a Adhesi: eat a ‘Abrasive wear Corrosive: wear up e "Fhe definition of wear doés not inelude: 1 deforiation, although wear has-occured: de: i fi fai ‘include i this définition. fails to inelug 5 " or eavitation (also. corrosion) where: counterbes y ;