Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Fatigue”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,099 records · Page 61

Effects of environment on the fatigue of graphite-epoxy composites

Torsional and flexural fatigue tests were performed on both uniaxial (0 deg) and crossplied (plus or minus 45 deg) graphite-epoxy materials at temperatures of 24 and 74 C in environments of air and water. The results of the torsion testing showed that the number of cycles required to cause an initial decrease in stiffness as well as the rate of stiffness loss was a function of temperature and environment; the most significant losses were noted for tests at the higher temperature in water. The torsional fatigue specimens were subsequently tested in four-point bending to determine the effect of torsional damage on longitudinal properties. This damage caused changes in the flexural stiffness, failure stress, and failure energy, depending on the stress and environmental histories. The flexural fatigue tests also showed a significant effect of water (at 24 C) on the material behavior. These results are compared with the results of previous investigations and are discussed in terms of proposed damage mechanisms.

Sumsion, H. T.↗

Stochastic model for fatigue crack size and cost effective design decisions

This paper describes a methodology for making cost effective fatigue design decisions. The methodology is based on a probabilistic model for the stochastic process of fatigue crack growth with time. The development of a particular model for the stochastic process is also discussed in the paper. The model is based on the assumption of continuous time and discrete space of crack lengths. Statistical decision theory and the developed probabilistic model are used to develop the procedure for making fatigue design decisions on the basis of minimum expected cost or risk function and reliability bounds. Selections of initial flaw size distribution, NDT, repair threshold crack lengths, and inspection intervals are discussed.

Hanagud, S.↗

Early detection of fatigue damage in composite materials

Early detection of fatigue damage in composite materials by nondestructive inspection (NDI) techniques has been demonstrated for glass/epoxy, graphite/glass/epoxy, and graphite/epoxy composites. Modulus and temperature were monitored and a correlation between them observed. Axial modulus and torsional modulus changes were a function of the laminate orientation. Torsional modulus measurements and coin tap tests were performed at 0, 1 million, 5 million, and 10 million cycles, on axial fatigue specimens. Three distinct regions were noted. In the primary region a small but rapid change in stiffness was noted in the first few thousand cycles. This was followed by a secondary region of little or no stiffness change. The tertiary region was characterized by an increasing rate of stiffness change leading to fracture. NDI procedures including holographic interferometry, ultrasonics, penetrant, and X-ray radiography were evaluated for fatigue damage detection.

Salkind, M. J.↗

Environmental effects on the torsion and flexure fatigue properties of graphite-epoxy composites

Effects of torsional and flexural fatigue on the long-time integrity of advanced graphite-epoxy structural composites have been investigated. Reversed (R = - 1) and unidirectional (R = 0) torsional fatigue tests were made on unidirectional, cross-plied and woven graphite fiber materials in air and water at room temperature and at 74 C. Flexural fatigue tests (4-point bending) with R = - 1 were run in air and water at room temperature, and at R = 0 in air. Results show that in torsional cycling, both water environment and higher test temperature contribute to significant degradation of torsional stiffness. The degradation of stiffness from torsional stress cycling was observed to be much greater with R = - 1 than with simple R = 0 cycling. The effect of environment is also greater in the fully reversed cycling.

Sumsion, H. T.↗

Fatigue failure of hydrogen embrittled high strength steels

Results of an experimental investigation are presented concerning the fracture behavior of cathodically charged, quenched and tempered martensitic steels under cyclic load conditions. Introduction of H2 by cathodic charging reduced fatigue life by as much as 60%. It is proposed that subsurface transverse fatigue cracks nucleate simultaneously at multiple sites, such as at microcracks, voids, or inclusions. Fatigue crack growth then occurs on planes perpendicular to the major applied stress axis in the presence of the critical combination of applied external stress and hydrogen.

Kim, Y. G.↗

Fatigue of notched fiber composite laminates. Part 1: Analytical model

A description is given of a semi-empirical, deterministic analysis for prediction and correlation of fatigue crack growth, residual strength, and fatigue lifetime for fiber composite laminates containing notches (holes). The failure model used for the analysis is based upon composite heterogeneous behavior and experimentally observed failure modes under both static and fatigue loading. The analysis is consistent with the wearout philosophy. Axial cracking and transverse cracking failure modes are treated together in the analysis. Cracking off-axis is handled by making a modification to the axial cracking analysis. The analysis predicts notched laminate failure from unidirectional material fatique properties using constant strain laminate analysis techniques. For multidirectional laminates, it is necessary to know lamina fatique behavior under axial normal stress, transverse normal stress and axial shear stress. Examples of the analysis method are given.

Mclaughlin, P. V., Jr.↗

Experimental and theoretical investigation of fatigue life in reusable rocket thrust chambers

During a test program to investigate low-cycle thermal fatigue, 13 rocket combustion chambers were fabricated and cyclically test fired to failure. Six oxygen-free, high-conductivity (OFHC) copper and seven Amzirc chambers were tested. The failures in the OFHC copper chambers were not typical fatigue failures but are described as creep rupture enhanced by ratcheting. The coolant channels bulged toward the chamber centerline, resulting in progressive thinning of the wall during each cycle. The failures in the Amzirc alloy chambers were caused by low-cycle thermal fatigue. The zirconium in this alloy was not evenly distributed in the chamber materials. The life that was achieved was nominally the same as would have been predicted from OFHC copper isothermal test data.

Hannum, N. P.↗

Ultrahigh vacuum, high temperature, low cycle fatigue of coated and uncoated Rene 80

A study was conducted on the ultrahigh vacuum strain controlled by low cycle fatigue behavior of uncoated and CODEP B-1 aluminide coated Rene' 80 nickel-base superalloy at 1000 C (1832 F) and 871 C (1600 F). The results indicated little effect of coating or temperature on the fatigue properties. There was, however, a significant effect on fatigue life when creep was introduced into the strain cycles. The effect of this creep component was analyzed in terms of the method of strainrange partitioning.

Kortovich, C. S.↗

Fatigue performance of metal-lined graphite/epoxy pressure vessels

Using an ultrahigh-strength graphite fiber, a program was started to develop a thin metal-lined fiber/epoxy pressure vessel that would have a fatigue life of over 1000 cycles. First, the performance factor of the fiber/epoxy composite was found to be 351 kN m/kg from the average of 18 rubber-lined pressure vessels. Then, both aluminum- and titanium-lined vessels were filament wound with the graphite fiber in an epoxy matrix. Several of these metal-lined vessels were subjected to hydraulic fatigue testing to about 50% of their expected burst pressures. The average fatigue life of the aluminum-lined vessels was 462 cycles; the average for the titanium-lined vessels was 2190 cycles.

Hamstad, M. A.↗

Axial fatigue of /0, +30, and -30 deg/6s graphite/epoxy

Results are presented for an investigation carried out on a /0, +30, and -30 deg/6s graphite/epoxy angle-ply laminate in order to characterize the overall fatigue behavior, to evaluate strength/moduli degradation due to cyclic loading, and to identify the failure modes and mechanisms. The specimen life determined from fatigue tests is measured as a function of cyclic stress amplitude for various stress ratios. Fractured specimens are sectioned and examined by conventional metallography and SEM to reveal failure modes and mechanisms. It is shown that the laminate studied is susceptible to axial fatigue failures under cyclic loading which involves both tensile and compressive mean stresses. Fracture surface observations indicate the occurrence of complex failure modes. A variety of local failure mechanisms are identified.

Ramani, S. V.↗

Fatigue damage of notched boron/epoxy laminates under constant amplitude loading

Fatigue damage in (0, + or - 45) and (0, + or - 45,90) boron/epoxy laminates was studied with X-ray radiography and scanning electron microscopy. In addition, limited tests for residual strength and stiffness were performed. The results of this study suggest that in boron/epoxy laminates the 45-degree plies play a key role in the fatigue process of boron/epoxy laminates that contain them. The fatigue process in the + or - 45-degree plies starts as intralaminar matrix cracks.

Roderick, G. L.↗

Biaxial fatigue loading of notched composites

Thin walled, 2.54-cm (1-in.) diameter tubular specimens of T300/934 graphite/epoxy were fabricated and fatigue cycled in combinations of axial, torsional, and internal pressure loading. Two different four-ply layup configurations were tested: (0/90)S and (+ or - 45)S; all tubes contained a 0.48-cm (3/16-in.) diameter circular hole penetrating one wall midway along the tube length. S-N curves were developed to characterize fatigue behavior under pure axial, torsional, or internal pressure loading, as well as combined loading fatigue. A theory was developed based on the Hill plane stress model which enabled the S-N curve for combined stress states to be predicted from the S-N data for the uniaxial loading modes. Correlation of the theory with the experimental data proved to be remarkably good.

Francis, P. H.↗

Application of strainrange partitioning to the prediction of MPC creep-fatigue data for 2 1/4 Cr-1 Mo steel

Creep-fatigue tests were conducted on annealed 2.25Cr-1Mo steel in air at 538 C involving interspersion of rapid strain cycles (0 to 22 cycles) between constant tensile stress-hold periods (23 or 47 hours). Life predictions were made on the basis of the strain range-life relations obtained from short-time material characterization tests on tubular specimens taken from the same heat as the specimens for the creep-fatigue tests. The creep-fatigue tests lasted from about 500 to 5000 hours, while the short-time tests used as the prediction basis lasted from 0.1 to 100 hours. By partitioning the creep and plastic strains within every cycle and summing the damage on a cycle-by-cycle basis, the predicted lives agreed with the observed ones within factors of two.

Saltsman, J. F.↗

Rolling-element fatigue life of AISI M-50 and 18-4-1 balls

Rolling element fatigue studies were conducted with AISI M-50, EFR 18-4-1, and VAR 18-4-1. Groups of 12.7 mm (1/2-in) diameter balls of each material were tested in the five ball fatigue tester. Test conditions included a load of 1540 N (347 lbf) giving a maximum Hertz stress of 5520 MPa (800 000 psi), a shaft speed of 10,700 rpm, and a contact angle of 30 deg. Tests were run at a race temperature of 339 K (150 F) with a type 2 ester lubricant. The rolling element fatigue life of AISI M-50 was not significantly different from that of EFR 18-4-1 or VAR 18-4-1 based on a statistical comparison of the test results.

Parker, R. J.↗

Development of an improved method of consolidating fatigue life data

A fatigue data consolidation model that incorporates recent advances in life prediction methodology was developed. A combined analytic and experimental study of fatigue of notched 2024-T3 aluminum alloy under constant amplitude loading was carried out. Because few systematic and complete data sets for 2024-T3 were available in the program generated data for fatigue crack initiation and separation failure for both zero and nonzero mean stresses. Consolidations of these data are presented.

Leis, B. N.↗

Biaxial fatigue loading of notched composites

Thin-walled, 2.54-cm diameter tubular specimens of graphite/epoxy were fatigue cycled in combinations of axial, torsional, and internal pressure loading. Two different four-ply layup configurations were tested: (0-90)s and (+ or- 45)s; each tube contained a 0.48-cm diameter circular hole penetrating one wall midway along the tube length. S-N curves were developed to characterize fatigue behavior under pure axial, torsional, or internal pressure loading, as well as combined loading fatigue. A theory was developed based on a plane stress model which enabled the S-N curve for combined stress states to be predicted from the S-N data for the uniaxial loading modes. Correlation of the theory with the experimental data proved to be remarkably good.

Francis, P. H.↗

Ferrographic analysis of wear debris generated in accelerated rolling element fatigue tests

Ferrographic analysis was used to determine the types and quantities of wear particles generated during accelerated rolling contact fatigue tests. The NASA five-ball rolling contact fatigue tester was used. Ball specimens were made of AMS 5749, a corrosion-resistant high-temperature bearing steel. The lubricant was a super-refined naphthenic mineral oil. Conditions included a maximum Hertz stress of 5.52 billion Pa and a shaft speed of 10,000 rpm. Four types of wear particles were observed: normal rubbing wear particles, fatigue spall particles, spheres, and friction polymer.

Jones, W. R., Jr.↗

Effect of preload on the fatigue and static strength of composite laminates with defects

The effect of a preload cycle on the structural performance of three graphite/epoxy composite laminates was studied. The layups studied were a laminate typical of general purpose structures (L1), a laminate representative of a filament wound tank (L2), and a laminate representative of turboengine fan blades. The effects of three sizes of simulated initial defects were studied. The tests developed static strength data, fatigue to failure data, and residual static data after application of a predetermined number of fatigue cycles. For L1 specimens, there was a slight trend for the static strength to be greater for preloaded specimens. After application of cyclic loading, however, the influence of preloading was insignificant. In L2 and L3 specimens there was no consistent difference in the static or fatigue results between preloaded and nonpreloaded specimens.

Porter, T. R.↗