1、NASA TECHNICAL NOTECOr.NASA TN D-7173EFFECT OF SPECIMEN THICKNESSON FATIGUE-CRACK-GROWTH BEHAVIORAND FRACTURE TOUGHNESS OF 7075-T6AND 7178-T6 ALUMINUM ALLOYSby C. Michael Hudson and J. C. Newman, Jr.Langley Research CenterHampton, Va. 23365NATIONAL AERONAUTICS AND SPACE ADMINISTRATION WASHINGTON, D.
2、 C. APRIL 1973Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-1.4.7.9.12.Report No. 2. Government Accession No.NASA TN D-7173Title and SubtitleEFFECT OF SPECIMEN THICKNESS ON FATIGUE-CRACK-GROWTH BEHAVIOR AND FRACTURE TOUGHNESS OF7075-T6 AND 7178-T6
3、ALUMINUM ALLOYSAuthor(s)C. Michael Hudson and J. C. Newman, Jr.Performing Organization Name and AddressNASA Langley Research CenterHampton, Va. 23365Sponsoring Agency Name and AddressNational Aeronautics and Space AdministrationWashington, D.C. 205463.5.6.8.10.11.13.14.Recipients Catalog No.Report D
4、ateApril 1973Performing Organization CodePerforming Organization Report No.L-8731Work Unit No.501-22-02-01Contract or Grant No.Type of Report and Period CoveredTechnical NoteSponsoring Agency Code15. Supplementary Notes16. AbstractA study was made to determine the effects of specimen thickness on fa
5、tigue-crackgrowth and fracture behavior of 7075-T6 and 7178-T6 aluminum-alloy sheet and plate. Spec-imen thicknesses ranged from 5.1 to 12.7 mm (0.20 to 0.50 in.) for 7075-T6 and from 1.3 to6.4 mm (0.05 to 0.25 in.) for 7178-T6. The stress ratios R used in the crack-growthexperiments were 0.02 and 0
6、.50. For 7075-T6, specimen thickness had relatively little effecton fatigue-crack growth. However, the fracture toughness of the thickest gage of 7075-T6was about two-thirds of the fracture toughness of the thinner gages of 7075-T6. For 7178-T6,fatigue cracks generally grew somewhat faster in the th
7、icker gages than in the thinnest gage.The fracture toughness of the thickest gage of 7178-T6 was about two-thirds of the fracturetoughness of the thinner gages of 7178-T6.Stress-intensity methods were used to analyze the experimental results. For a giventhickness and value of R, the rate of fatigue-
8、crack growth was essentially a single-valuedfunction of the stress-intensity range for 7075-T6 and 7178-T6. An empirical equation devel-oped by Forman, Kearney, and Engle (in Trans. ASME, Ser. D: J. Basic Eng., vol. 89, no. 3,Sept. 1967) fit the 7075-T6 and 7178-T6 crack-growth data reasonably well.
9、17. Key Words (Suggested by Author(s)Fatigue-crack growthFracture toughness7075-T6 aluminum alloy7178-T6 aluminum alloyThickness effect18. Distribution StatementUnclassified - Unlimited19. Security dassif. (of this report)Unclassified20. Security Classif. (of this page)Unclassified21. No. of Pages32
10、22. Price*$3.00For sale by the National Technical Information Service, Springfield, Virginia 22151Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-EFFECT OF SPECIMEN THICKNESS ON FATIGUE-CRACK-GROWTHBEHAVIOR AND FRACTURE TOUGHNESS OF 7075-T6AND 7178-T
11、6 ALUMINUM ALLOYSBy C. Michael Hudson and J. C. Newman, Jr.Langley Research CenterSUMMARYA study was made to determine the effects of specimen thickness on fatigue-crackgrowth and fracture behavior of 7075-T6 and 7178-T6 aluminum-alloy sheet and plate.Specimen thicknesses ranged from 5.1 to 12.7 mm
12、(0.20 to 0.50 in.) for 7075-T6 andfrom 1.3 to 6.4 mm (0.05 to 0.25 in.) for 7178-T6. The stress ratios R used in thecrack-growth experiments were 0.02 and 0.50. For 7075-T6, specimen thickness hadrelatively little effect on fatigue-crack growth. However, the fracture toughness of thethickest gage of
13、 7075-T6 was about two-thirds of the fracture toughness of the thinnergages of 7075-T6. For 7178-T6, fatigue cracks generally grew somewhat faster in thethicker gages than in the thinnest gage. The fracture toughness of the thickest gage of7178-T6 was about two-thirds of the fracture toughness of th
14、e thinner gages of 7178-T6.Stress-intensity methods were used to analyze the experimental results. For agiven thickness and value of R, the rate of fatigue-crack growth was essentially a single-valued function of the stress-intensity range for 7075-T6 and 7178-T6. An empiricalequation developed by F
15、orman, Kearney, and Engle (in Trans. ASME, Ser. D: J. BasicEng., vol. 89, no. 3, Sept. 1967) fit the 7075-T6 and 7178-T6 crack-growth data reasonablywell.INTRODUCTIONFatigue cracks of various sizes have been discovered during the service life ofmany aircraft structures. As a result, the predictions
16、of fatigue-crack-growth ratesand fracture toughness of parts containing fatigue cracks have become of considerableinterest to aircraft designers and operators. In order to make such predictions, theeffects of a wide range of parameters must be understood. Many of these parameters,such as component c
17、onfiguration, stress ratio, loading sequence, and environment, havealready been investigated at NASA Langley Research Center and are reported in refer-ences 1 to 7. However, relatively little research has been conducted on the effects ofProvided by IHSNot for ResaleNo reproduction or networking perm
18、itted without license from IHS-,-,-material thickness on fatigue behavior. Consequently, a series of axial-load fatigue-crack-growth and fracture-toughness experiments were conducted on 7075-T6 and7178-T6 aluminum-alloy specimens ranging in thickness from 5.1 to 12.7 mm (0.20 to0.50 in.) and from 1.
19、3 to 6.4 mm (0.05 to 0.25 in.), respectively. These materials wereselected because of their frequent use in aircraft construction.Stress-intensity methods were used to analyze the data because these methods haveshown great promise for predicting fatigue-crack propagation and fracture in complexstruc
20、tures. For example, Poe (ref. 8) showed that fatigue-crack growth in stiffenedpanels can be predicted from stress-intensity parameters and the data from tests ofsimple sheet specimens.An empirical equation developed by Forman, Kearney, and Engle (ref. 9) was fittedby least-squares techniques to the
21、fatigue-crack-propagation data. This equation fit thefatigue-crack-growth data generated in a previous study of stress-ratio effects reason-ably well (ref. 3).SYMBOLSThe units used for the physical quantities defined in this paper are given in both theInternational System of Units (SI) and the U.S.
22、Customary Units. The measurements andcalculations were made in the U.S. Customary Units. Factors relating the two systemsare given in reference 10 and those used in the present investigation are presented inappendix A.a half-length of a central symmetrical crack, mm (in.)a: half-length of crack at s
23、tart of a fracture-toughness test, mm (in.)C constant in fatigue-crack-growth equationda/dN rate of fatigue-crack growth, nm/cycle (in./cycle)E Youngs modulus of elasticity, GN/m2 (psi)e elongation in 51-mm (2-in.) gage length, percentKcn critical stress-intensity factor, MN/m (psi-in 7/ 3/2 ( l/2Km
24、ax maximum stress-intensity factor, MN/m psi -in JProvided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-Kmin minimum stress-intensity factor, MN/m (psi-in J/ 3/2 ( l/2AK stress-intensity-factor range, MN/m Ipsi-in /N number of load cyclesn exponent in fati
25、gue-crack-growth equationPa amplitude of load applied in a cycle, N (Ibf)Pj maximum load applied to specimen during fracture-toughness test, N (Ibf)Pm mean load applied in a cycle, N (Ibf)pmax maximum load applied in a cycle, Pm + Pa, N (Ibf)Pmin minimum load applied in a cycle, Pm - Pa, N (Ibf)R ra
26、tio of minimum stress to maximum stressSa alternating gross stress, Pa/wt, MN/m2 (psi or ksi)Sf maximum gross stress applied to specimen during fracture-toughness test,Pf/wt, MN/m2 (psi)Sm mean gross stress, P/wt, MN/m (psi or ksi)maximum gross stress, Pmax/wt MN/m2 (psi)minimum gross stress, Pmin/w
27、t, MN/m2 (psi)t Specimen thickness, mm (in.)w specimen width, mm (in.)OL secant correction factor for stress intensity in a finite width panel, I/sec -|au ultimate tensile strength, MN/m, (ksi)ff yield strength (0.2-percent offset), MN/m (ksi)Provided by IHSNot for ResaleNo reproduction or networkin
28、g permitted without license from IHS-,-,-SPECIMENS, TESTS, AND PROCEDURESSpecimensThrough-crack test specimens were made from three thicknesses each of 7075-T6and 7178-T6 aluminum alloys. The thicknesses and tensile properties of these alloys arelisted in table I. The tensile specimens used to obtai
29、n these properties met ASTM Stan-dards (ref. 11). The nominal chemical compositions of the two alloys are shown intable H.The specimen configuration used in both the crack-propagation and fracture-toughness tests is shown in figure 1. These specimens were 292 mm (11.5 in.) wide and889 mm (35.0 in.)
30、long. The longitudinal axes of all specimens were parallel to the roll-ing direction of the material. A notch 2.54 mm (0.10 in.) long by 0.25 mm (0.01 in.) widewas cut into the center of each specimen by use of an electrical discharge machining pro-cess. The heat-affected zone resulting from this pr
31、ocess is less than 0.25 mm (0.01 in.)wide. Consequently, after crack initiation, all of the material through which the fatiguecrack propagates is unaltered by the cutting process.A reference grid (ref. 12) was photographically printed on the surface of the spec-imen for crack-propagation monitoring.
32、 The spacing between grid lines was 1.3 mm(0.050 in.). Metallographic examination and tensile tests conducted on 7075-T6 speci-mens bearing the grid indicated no detrimental effect on the material.Testing MachinesThree axial-load fatigue-testing machines were employed in this investigation. Thecapab
33、ilities of these machines are listed in the following table:Machine typeSubresonantHydraulicCombination:As subresonant unitAs hydraulic unitMaximum loadcapacitykN891334467587Ibf20 000300 000105 000132 000Operatingfrequency usedHz301 to 5140.7 to 1.0cpm180060 to 30084040 to 60Machinedescribed in Ref.
34、 13App. BRef. 14The 1334-kN (300 000-lbf) tester described in the preceding table was also used forfracture-toughness tests requiring loads in excess of 534 kN (120 000 Ibf). A hydraulicProvided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-axial-load unive
35、rsal testing machine was used for fracture-toughness tests requiringlower loads. This universal machine had a load capacity of 534 kN (120 000-lbf).Test ProcedureAxial-load fatigue-crack-propagation experiments were conducted at stress ratiosR of 0.02 and 0.50. The maximum gross stresses in these ex
36、periments ranged from 69to 276 MN/m2 (10 to 40 ksi) for 7075-T6 and from 52 to 155 MN/m2 (7.5 to 22.5 ksi) for7178-T6. The alternating and mean loads were kept constant throughout each test. Thefatigue-crack-growth data were obtained by observing crack growth through 10 powermicroscopes. The number
37、of cycles required to propagate the crack to each grid linewas recorded so that crack-propagation rates could be determined.Fracture-toughness data were obtained two ways. Most of these data came fromstandard toughness tests in which fatigue-cracked specimens were monotonically loadedto failure at a
38、 load rate of 2.2 kN/sec (30 000 Ibf/min). The remainder of these datacame from fatigue-crack-propagation tests which were continued up to specimen failure.In these tests, the maximum load in the fatigue-crack-propagation test was assumed to bethe load at failure.When a centrally cracked sheet speci
39、men is loaded in axial tension, transversecompressive stresses are generated near the crack surface (ref. 15). These compres-sive stresses can buckle thin specimens out of the plane of the sheet near the crack. Theincrease in stress-intensity factor due to this buckling is difficult to calculate; co
40、nse-quently the thinner gage specimens (t = 5.1 mm (0.20 in.) for 7075-T6 and t = 1.3 and4.1 mm (0.05 and 0.16 in.) for 7178-T6) were clamped between oiled guide plates(ref. 16) to restrain buckling. The thicker specimens did not buckle; therefore guideplates were not used.RESULTS AND DISCUSSIONFati
41、gue-Crack-Growth ExperimentsThe results of the fatigue-crack-growth experiments on the 7075-T6 and 7178-T6specimens are presented in table III. This table gives the average number of cyclesrequired for a through-crack to propagate from a half-length of 2.54 mm (0.10 in.) tothe listed half-lengths. F
42、atigue-crack-growth rates were determined graphically fromcrack-growth curves which were faired through the data of table ffl.The fatigue-crack-growth curves for the 7075-T6 specimens of different thicknessesare presented in figure 2. At eight of nine stress levels, fatigue cracks propagated fastest
43、in the 5.1-mm-thick (0.20-in.) 7075-T6 specimens. However, for a given stress level,Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-the ratio of the maximum to the minimum number of cycles required to reach a givencrack length never exceeded 1.7, the
44、reby indicating a relatively small thickness effect.The fatigue-crack-growth curves for the 7178-T6 specimens are presented in fig-ure 3. At six of seven stress levels, fatigue cracks propagated slowest in the 1.3-mm-thick (0.05-in.) 7178-T6 specimens. For a given stress level, the ratio of the maxi
45、mumto the minimum number of cycles required to reach a given crack length never exceeded2.7, thereby indicating a moderate thickness effect.Fatigue-crack-growth curves for 7075-T6 and 7178-T6 specimens of about the samethickness (5.1 and 4.1 mm (0.20 and 0.16 in.), respectively) and tested at the sa
46、me valuesof Smax and R are shown in figure 4. For a given stress level, the ratio of the max-imum to the minimum number of cycles required to reach a given crack length neverexceeded 1.7. In two instances fatigue cracks grew fastest in 7075-T6, and in the twoother instances, fastest in 7178-T6. Thus
47、, in the thickness range of 4 to 5 mm (0.16 to0.20 in.), the two alloys appear about equally resistant to fatigue-crack propagation.Inspection of the fracture surfaces of the specimens (fig. 5, for example) indicatedthat intermittent bursts of crack growth (referred to hereinafter as “pop-in“ (ref.
48、17)occurred in the interior of specimens having thicknesses as small as 4.1 mm (0.16 in.).The dark areas in figure 5 indicate pop-in. The light areas indicate normal, microscopicfatigue-crack growth. The reason for this pop-in is not understood at this time.The fatigue-crack-growth data in table III
49、 were analyzed by using stress-intensitymethods (see appendix C). For a given thickness and value of R, the rate of fatigue-crack growth was a single-valued function of the stress-intensity range for 7075-T6 and7178-T6 (fig. 6).An empirical fatigue-crack-growth equation developed by Forman, Kearney, andEngle (ref. 9) was fitted to the test data. This equation
