【正文】 stently provide the failure discrimination necessary for all types of bearing life tests. It is then necessary to select a system that will repeatedly terminate machine operation with a consistent minimal degree of damage. The rate of failure propagation is therefore important. If the degree of damage at test termination is consistent among test elements, the only variation between the experimental and theoretica l lives is the lag in failure detection. In standard through hardened bearing steels the failure propagation rate is quite rapid under endurance test condit ions, and this is not a major factor, considering the typical dispersion of endurance test data and the degree of confidence obtained from statistical analysis. T his may not, however, be the case with other experimental materia ls or with surface hardened steels or steels produced by experimental techniques. Care must be used when evaluating these latter results and particu larly when paring the experimental lives with those obtained from standard steel lots. The ultimate means of ensuring that an endurance test series was adequately controlled is the conduct of a post test analysis. T his detailed examination of all the tested bearings uses high magnif ication optic al inspection, highermagn ificat ion scanning electron microscopy, metallurgical and dimensional examinations, and chemical evaluations as required. T he characteristics of the failures are examined to establish their origins and the residual surface condit ions are evaluated for indications of extraneous effects that may have influenced the bearin g life. This technique allows the experimenter to ensure that the data are indeed valid. The “Damage Atlas” piled by Tallian et al. [] contain ing numerous bl ack and white photographs of the various bearin g failure modes can provide guidance for these types of determinations. T his work was subsequently updated by Tallian 河南科技大學(xué)外文 翻譯 11 [], now including color photographs as well. The posttest analys is is, by definit ion, after the fact. To provide contro l throughout the test series and to elim inate all questionable areas, the experimenter should conduct a prelim inary study whenever a bearing is removed from the test machine. In this portion of the investigation each be arin g is examined optically at magnif ications up to 30 ? for indications of improper or outofcontrol test parameters. Examples of the types of indications that can be observed are given in Figs. . 6. Figure 19. 2 illustrates the appearance of a typical fatigue originated spall on a ball bearing raceway. Figure 19. 3 contains a spalling failure on the raceway of a roller bearing that resulted from bearing misalignment, and Fig. 19. 4 contains a spalling failure on the outer ring of a ball bearing produced by frettin g corrosion on the outer diameter. Figure 19. 5 illustrates a more subtle form of test alteration, `where the spallin g failure originated from the presence of a debris dent on the surface. Figure 19. 6 gives an example o f a totally different failure mode produced by the loss of internal bearin g clearance due to therma l unbalance of the system. The last four failures are not valid fatigue spalls and indicate the need to correct the test methods. Furthermore, these data poi nts would need to be elim inated from the failure data to obtain a valid estimate of the experimenta l bearing life. 2 .AVOIDING FAILURES The best way to handle bearin g failures is to avoid them． This can be done in the selection process by recogniz ing cr itica l performance characterist ics ． T hese include noise， starting and running torque， stiffness， nonrepetitive run out，and radial and axial play． In some applications, these items are so crit ical that specifying an ABEC level alone is not sufficient． Torque requirements are determined by the lubricant ， retainer ， raceway quality(roundness cross curvature and surface finish)， and whether seals or 河南科技大學(xué)外文 翻譯 12 shields