NASA-CR-215-1965 The fatigue behavior of materials for the supersonic transport《超音速运输材料的疲劳性能》.pdf

1、NASA CONTRACTOR REPORT THE FATIGUE BEHAVIOR OF MATERIALS FOR THE SUPERSONIC TRANSPORT by M. S. Eleuly, C. W. Marschzll, F. C. Holden, and WT. S. Hyler Prepared under Contract No. NASr-lOO-(01) by BATTELLE MEMORIAL INSTITUTE Columbus, Ohio f Or NATIONAL AERONAUTICS AND SPACE ADMINISTRATION l WASHINGT

2、ON, D. C. l APRIL 1965 Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-TECH LIBRARY KAFB, NM 0079807 - NASA CR-213 THE FATIGUE BEHAVIOR OF MATERIALS FOR THE SUPERSONIC TRANSPORT By M. S. Healy, C. W. Marschall, F. C. Holden, and W. S. Hyler Distribut

3、ion of this report is provided in the interest of information exchange. Responsibility for the contents resides in the author or organization that prepared it. Prepared under Contract No. NASr-lOO-(01) by BATTELLE MEMORIAL INSTITUTE Columbus, Ohio for NATIONAL AERONAUTICS AND SPACE ADMINISTRATION Fo

4、r sole by the Clearinghouse for Federal Scientific ond Technic01 Information Springfield, Virginia 22151 - Price $3.00 Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-Provided by IHSNot for ResaleNo reproduction or networking permitted without licens

5、e from IHS-,-,-ABSTRACT Tests were conducted on sheet specimens of two materials, Ti-8Al-lM+lV and AM350 CRT to determine the fatigue lives, rates of fatigue-crack propagation, and residual static strengths at three temperatures, 550F, -llOOF, and room temperature. The effects of prior exposure to 5

6、50F for up to 10 000 hours on the same properties were also studied. Neither material was significantly degraded by the elevated temperature soak. The fatigue characteristics were found to be comparable to those of contemporary aluminum alloys. iii Provided by IHSNot for ResaleNo reproduction or net

7、working permitted without license from IHS-,-,-Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-TABLE OF CONTENTS SUMMARY . INTRODUCTION . EXPERIMENTAL DETAILS . Materials . Specimen Design . Equipment . . . . . . . . . . . . . . . . . . . . . . . . .

8、 2 2 6 CRACK PROPAGATION AND RESIDUAL STATIC STRENGTH . . . . . . 6 Crack-Propagation Studies . Effect of Alternating Stress Amplitude Effect of Orientation . Effect of Cyclic Rates . Effect of Temperature . Effect of Prior Stressed Exposure . Residual Static-Strength Studies Procedures and Equipmen

9、t Effect of Crack Length . Effect of Test Temperature . Effect of Orientation With Respect to the Rolling Direction . Effect of Fatigue Conditions Used to Introduce Cracks Effect of Fatigue-Stress Amplitude . Effect of Fatigue-Cracking Temperature . Effect of Cyclic Frequency . Effect of Exposure to

10、 l-g Stress at 550 F FATIGUE BEHAVIOR . . . . . . . . . . . . . . . . . . . DISCUSSION AND CONCLUDING REMARKS . . . . . . . . . . . . Crack Propagation ., . Residual Static Strength Fatigue Behavior REFERENCES . . . . . . . . . . . . . . . . . . . . . APPENDIX A TENSILEDATA . . . . . . . . . . . . .

11、 . . . . . . . . 69 APPENDIX B CRACK-PROPAGATIONDATA . . . . . . . . . . . . . . , . APPENDIX C STRESS-LIFETIME DATA . . . . . . . . . . . . . . . . . Page 1 1 2 7 9 23 24 24 24 28 28 28 28 32 32 32 32 32 32 39 59 59 60 61 66 71 79 v Provided by IHSNot for ResaleNo reproduction or networking permitt

12、ed without license from IHS-,-,-THE FATIGUE BEHAVIOR OF MATERIALS FOR THE SUPERSONIC TRANSPORT M. S. Healy, C. W. Mars thus the fatigue behavior of potential skin materials is an important factor in de- sign considerations. The skin of the supersonic transport will be subjected to temperatures rangi

13、ng from well below freezing to approximately 550 F. Actually, a major portion of the flight Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-profile is at cruise condition. Consequently, the skin will be subjected to elevated tem- perature for appreci

14、able periods of time. It is because of this stressed,exposure at elevated temperature over a long period of time that the present program included an evaluation of possible metallurgical instability which could affect static or dynamic be- havior of the materials. This Battelle program has been conc

15、erned with the fatigue behavior of two possible skin materials, AM-350 CRT stainless steel and triplex-annealed Ti-SAl- lMo- 1V alloy. The base .fatigue and fatigue crack-propagation behavior of these materials, as well as residual static strength, has been examined with respect to the temperatures

16、and stressed exposure to which the supersonic transport skin will be subjected. Because of the large scope of this research effort, some of the variables were ex- plored to a very limited extent. These studies were intended to cover a wide range of potential problem areas in order to determine those

17、 that require further detailed study. Since the program has covered a 2-year period, generally only experimental data obtained during the second years research are reported. The data from the first year is presented in NASA CR-28, April 1964 Reference (l) . In the discussion and conclu- sions refere

18、nce is made to the results of both programs. EXPERIMENTAL DETAILS Materials Three sheets of AM-350 CRT stainless steel and of triplex-annealed Ti-SAl- lMo-1V alloy were provided by Langley Research Center from a large quantity of mate- rial specifically obtained for NASA programs. Manufacturers chem

19、ical-analysis and mechanical-property data appear in Appendix A. During the first year of the Battelle program, tensile data were obtained from specimens of each of the three sheets of each material that were provided. A discussion of the results and the detailed data appear in Reference (1). The da

20、ta are summarized in Table 1, together with tensile data on edge- notched specimens of the same design as the fatigue specimens. Specimen Design Four types of specimens were employed in the program. These specimens include an unnotched specimen, an edge-notched specimen, a Z-inch-wide center-notched

21、 speci- men, and a 4-inch-wide center-notched specimen. The latter specimen was made only from the titanium alloy; only a few specimens were made. Drawings of each specimen are shown in Figure 1. Details concerning the fabrication of the specimens are given in Reference (1). Provided by IHSNot for R

22、esaleNo reproduction or networking permitted without license from IHS-,-,-TABLE 1. SUMMARY OF TENSILE DATA FOR AM-350 CRT AND TRIPLEX- ANNEALED Ti-SAl- lMo- 1V SHEET SPECIMENS 0.2 Per Cent Ultimate Offset Tensile Elongation Modulus of Temperature, Yield Strength, Strength, in 2 Inches, Elasticity, F

23、 Orientation ksi ksi per cent lo6 psi AM-350 CRT -100 RT 550 Ti-SAl- lMo- 1V - 100 RT 550 Unnotched-Specimen Data L 221 273 20. 2 28. 1 L 221 233 21. 1 27.8 T 201 236 13. 1 29. 5 L 184 201 3. 7 25. 2 AM-350 CRT RT L 550 L L 166 179 12.0 20. 0 L 140 152 12.9 18.8 T 132 144 11. 8 17.4 L 98 125 9. 5 16

24、. 7 Notched-Specimen Data, kt = 4. 0 Ti-SAl- lMo- 1V RT L - Notched Tensile Strength, ksi 242 200 - 166 - 550 L - 125 - me 3 Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-I 4 - R=4 / D/2 * .001 -_t- - D/2 A.001 -0 t a. Unnotched specimen 0) -I 3 -7

25、 e gdrill, reom tosdiom / 1 1 0.275 * 0.002 t 0.560 * 0.001 t - - + - - - 0.550 l 0.001 0.041 R 1 y7, *0.02 b. Edge -notched specimen FIGURE 1. FATIGUE SPECIMENS 4 Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-II -f 3.250*“OO + V Slot t- II Es dril

26、l ,reom tosdiom gdrilI,ream to the center notches had radii of 0. 0045 inch and lengths of 0. 125 f 0. 005 inch. The theoretical stress-concentration factor for the edge- notched specimen is 4.0, which is the same notch severity as that used in several Battelle and NASA 293) ported by Gerard 4 rogra

27、ms. ( Using the relations of Inglis14) and Dixon(5), as re- the stress-concentration factor for the 2-inch-wide center-notched specimens is approiimately 7.9. Similarly, Kt is about 8. 1 for the 4-inch-wide center- notched specimens. Equipment All fatigue testing was conducted in Krouse direct-stres

28、s fatigue machines of 5,000 or lO,OOO-pound capacity. These machines operate at 1725 and 1200 cpm, respectively. The smaller capacity machines are equipped with automatic hydraulic load maintainers that monitor test loads on each specimen. Accuracy of load setting and maintenance is about 3 per cent

29、 of the maximum test load. For specimens that were stressed in compression for a portion of the cycle, anti- buckling jigs, similar in design to those used at Langley Research Center, were em- ployed. These jigs consist of graphite plates held in contact with the specimen by steel pressure plates an

30、d pressure screws mounted in a rigid frame. The means of heating the elevated-temperature fatigue specimens has been modified as a consequence of some difficulties experienced occasionally with the antibuckling jig- heater method described in Reference (1). The present equipment consists of two nick

31、el-chromium coiled resistance-wire heating units fixed to opposite sides of a closed section of asbestos pipe. This furnace type of heater encloses the antibuckling jig and specimen grips. The specimen temperature is indicated by a Chromel-Alumel thermo- couple located in a hole in the graphite plat

32、e of the antibuckling jig. The hole is drilled in the graphite plate on the side away from the specimen. The thermocouple is inserted into this hole to within about l/32 inch of the specimen surface. The thermocouple is connected to a potentiometer controller which regulates the electrical power to

33、the heat- ing units. The manner in which the fatigue specimens are cooled to - 110 F and the equipment for stress-elevated temperature exposures are discussed in Reference (1). CRACK PROPAGATION AND RESIDUAL STATIC STRENGTH - The flight profile of the supersonic transport indicates that there are a

34、number of environmental factors that may influence both the manner in which a fatigue crack will propagate and the strength of a structural element that contains a fatigue crack. As a result, a wide variety of variables were studied to determine which ones may have a significant effect upon the crac

35、k-propagation behavior and residual static strength of the two materials. The factors that were explored were the following: alternating stress amplitude, cyclic frequency, specimen orientation with respect to principal rolling di- rection, temperature, and prior stressed exposure at elevated temper

36、ature. The studies 6 Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-reported in Reference (1) were devoted to the AM-350 stainless steel. The work that is reported here is primarily concerned with the titanium alloy; however, the behavior of exposed

37、 stainless steel specimens has also been examined and is included in this report. Crack-Propagation Studies The fatigue-crack popagation studies were conducted in conjunction with the residual-strength investigations. In the latter effort, fatigue-cracked specimens with nominal tip-to-tip crack leng

38、ths of 3 / 16-, 3/8-, 3/4-, and l-inch were evaluated. Crack-propagation data were obtained from those specimens cracked to 3/4- and l-inch crack lengths in addition to several specimens in which the cracks were propagated until specimen failure occurred. All of the residual-strength tests were cond

39、ucted on the 2-inch-wide specimens and therefore the majority of crack-propagation information per- tains to the 2-inch-wide specimens. A very limited number of 4-inch-wide Ti-8Al- lMo-1V specimens were also employed in the crack-propagation studies. Data on the crack-propagation behavior of the tit

40、anium alloy were obtained at a mean stress of 25 ksi, and the data for the AM-350 were obtained at a mean stress of 40 ksi. All of the stainless steel specimens were cracked in a lO,OOO-poundKrouse fatigue machine at a cyclic rate of 1200 cpm. The titanium-alloy specimens were cracked in a 5000-poun

41、d Krouse machine at frequencies of 1725 cpm and 34 cpm; however, the larg- est stress amplitude employed in the program necessitated the use of the lO,OOO-pound machine. During cracking, the net-section stresses were held essentially constant with an allowable variation of *lo per cent for the steel

42、 as in Reference (l) and 0 to t 10 per cent for the Ti-8Al- lMo- 1V alloy. The net-section stresses are based on the untracked portion, regardless of whether the specimen went into compression or not. A few titanium-alloy specimens were cracked at constant gross- section stress for purposes of compa

43、rison with the data obtained under constant net-section stress. Crack-propagation measurements were made by stopping the machine and measuring crack length with a measuring microscope mounted on a traverse. Measurements were taken with the specimen subjected to the maximum stress of the particular t

44、est. At - 110 F and 550 F, the data were obtained in the same manner by observing the crack through a window in the enclosure. From the experimental data, total crack length was plotted against total number of cycles. As was indicated in Reference (1) the effect of the *lo per cent variation in stre

45、ss is readily apparent in these plots for the steel; however, there seems to be little evidence of the 0 to t 10 per cent stress variation in similar curves for the titanium alloy. In order to determine the effects of the several variables in terms of the crack- propagation behavior, curves were dra

46、wn on the crack length versus cycles plots to minimize the small effect of the load changes required to maintain a constant nominal net-section stress. Crack-propagation rates were determined by evaluating the slope of the curves at specific crack lengths of 0. 250, 0. 350, 0.500, 0.625, 0. 750, and

47、 0. 950 inch. For purposes of this investigation, “initiation” has been defined as a total crack length of 0. 150 inch for the titanium and 0. 130 inch for the steel. Although most of the specimen tests were stopped at nominal crack lengths of 314 and 1 inch, two specimens, under each set of conditi

48、ons, were tested to failure from a nominal crack length of 1 inch without further adjustment of the net-section stress. The propagation 7 Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-0.E Jz 0 .E ; 0.6 s? s Y : 3 0.4 G ; I- 0.2 0 0 IO 20 30 40 / Total Number of Cycles X low3 6 d=0.38 I I 70 a A-48097 FIGURE 2. TYPICAL FATIGUE-CRACK PROPAGATION CURVES FOR UNEXPOSED, LONGITUDINAL 2-INCH-WIDE Ti-8Al- lMo- 1V ALLOY SHEET SPECIMENS AT ROOM TEMPERATURE AND 1200-1725 CPM