Brown University, Providence, R. I.
Brown University, Providence, R. I.
Pages: 8 Published: Jan 1957
Experiments conducted in this laboratory over the past several years have shown that appreciable changes occur in the propagation and recovery behavior of ultrasonic waves in materials during stress or fatigue cycling. Although most of the exploratory work has been done on three types of aluminum, the method is not restricted to aluminum. The following is a brief survey of the effects observed, and no explanation of the observed behavior is offered. Some explanation is offered in an earlier report by the authors. The results of the work show that in all cases studied the ultra- sonic attenuation increases as a function of the number of cycles of loading and unloading. The loading in these experiments was done in tension and in tension and compression. The form of the attenuation-cycles curve depends among other things on the magnitude of the load and the speed of cycling. There are several areas of interest—areas from which new information is derived: 1. The initial or early stages of cycling where (in 2S aluminum) (see Fig. 1) there is an initial increase in attenuation followed by a settling down or leveling off of attenuation in the first thousand cycles of operation. This result was expected as the result of previous work on plastic deformation. 2. An increase in attenuation as a function of the number of cycles in tension or in tension and compression depending on the particular experiment. The results of these measurements in 2S, 24ST-4, and 75S aluminum are shown in Figs. 3(a) and (b), 5, 6, 7, 8, and 9. These results are presented only for the purpose of giving a rough idea how the attenuation changes as cycling proceeds. In all cases the loading and speed of cycling have a strong effect on the results. There is strong evidence that most of the early attenuation changes occur without any visible crack formation on the surface of the specimen. 3. There is an impressive change in the attenuation-recovery behavior in these aluminum samples. 2S aluminum in particular has a 100 per cent attenuation recovery at the start of the cycling, but at the end of 107 cycles it was found that there was zero attenuation recovery. Figure 2, taken from an earlier paper, shows an example of initial recovery in 2S aluminum. In this case the deformation was carried further than in most fatigue experiments but this did not alter the recovery behavior. Two types of aluminum, 2S and 24ST-4, show practically complete recovery of attenuation when held at constant strain or immediately following unloading. On the other hand 75S aluminum shows very little recovery at any time. It has been found that the attenuation recovery for 2S aluminum is considerably more rapid in the initial part than is the recovery of 24S or 75S aluminum. The attenuation recovery of 24S aluminum is in turn more rapid than the recovery of 75S aluminum. As mentioned above, it was found that after fatigue cycling recovery was gone in 2S aluminum. A similar result was observed in 24S aluminum, and it is expected that the same type of change would occur in 75S aluminum. It is not known at present just how the change in recovery proceeds as a function of cycling in 2S and 24S aluminum, but it is established that the recovery is entirely gone after appreciable cycling. Figures 3, 5, 6, 7, 8, and 9, showing the manner in which attenuation depends on the number of stress cycles, will be discussed separately.
Paper ID: STP48492S