An accelerated-testing methodology has been developed for measuring the slow-crack-growth (SCG) behavior of brittle materials. Like the prior methodology, the accelerated-testing methodology involves dynamic fatigue ("constant-stress-rate") testing, in which a load or a displacement is applied to a specimen at a constant rate. SCG parameters or life-prediction parameters needed for designing components made of the same material as that of the specimen are calculated from the relationship between (1) the strength of the material as measured in the test and (2) the applied-stress rate used in the test. Despite its simplicity and convenience, dynamic fatigue testing as practiced heretofore has one major drawback: it is extremely time-consuming, especially at low stress rates.

The present accelerated methodology reduces the time needed to test a specimen at a given rate of applied load, stress, or displacement. Instead of starting the test from zero applied load or displacement as in the prior methodology, one preloads the specimen and increases the applied load at the specified rate (see Figure 1).

Figure 1. Applied Load or Stress is increased linearly with time in dynamic fatigue testing. In the conventional approach, one starts from zero applied stress. In accelerated testing, one starts from a preload stress that is a significant fraction of the strength of the material.

One might expect the preload to alter the results of the test and indeed it does, but fortunately, it is possible to account for the effect of the preload in interpreting the results. The accounting is done by calculating the normalized strength (defined as the strength in the presence of preload π the strength in the absence of preload) as a function of (1) the preloading factor (defined as the preload stress π the strength in the absence of preload) and (2) a SCG parameter, denoted n, that is used in a power-law crack-speed formulation. Figure 2 presents numerical results from this theoretical calculation.

For most glasses and advanced ceramics, the values of n are typically greater than 20. In a typical example, on the basis of the curves in Figure 2, preloading a material of n = 20 at 90 percent of its non-preload strength can be expected to result in an increase of only 0.005 in the normalized strength of the material. At the same time, the 90-percent preload would make it possible to perform the dynamic fatigue test for a given rate in only one-tenth the time of a test at the same rate that starts at zero applied load. In other words, testing time would be greatly reduced, without much effect on the test results.

Figure 2. The Effect of Preloading on the normalized strength of a specimen as measured in a dynamic fatigue test has been computed for materials having several different values of n, a crack-growth parameter.

The theory has been verified by extensive experimentation on a variety of brittle materials, including glasses, a glass-ceramic, and various forms of alumina, silicon nitride, and silicon carbide. This accelerated-testing methodology has been adopted as the basis of two standards of the American Society for Testing and Materials for dynamic fatigue testing of advanced ceramics: C 1368 for ambient temperature and C 1465 for elevated temperatures.

This work was done by John P. Gyekenyesi of Glenn Research Center and Sung R. Choi and Ralph J. Pawlik of Cleveland State University and Ohio Aerospace Institute.

Inquiries concerning rights for the commercial use of this invention should be addressed to

NASA Glenn Research Center
Commercial Technology Office
Attn: Steve Fedor
Mail Stop 4-8
21000 Brookpark Road
Cleveland Ohio 44135.

Refer to LEW-17409-1.