| | A brief summary is given of some prior studies which established the feasibility of using acoustic signatures for inspection of railroad wheels. Experimental and theoretical investigations were conducted to obtain more information on the effects on wheel vibrations of geometrical variations, wear, internal stress, etc. Hardware improvements and interfacing were out for a wayside installation, in addition to software development for real time data acquisition and processing. Field tests were made to evaluate performance, to permit follow-up on certain wheels and to obtain tape recordings from a sample of axle sets in service. These tape recordings were used to optimize the data processing software and to attempt to correlate identifiable wheel conditions with characteristics of the acoustic signature. The greatest signature differences were obtained when one of a pair of wheels was cracked. Differential wear was found to cause the largest differences in the signatures of good wheel pairs. It is claimed that the knowledge gained from this study is sufficient to warrant the installation of a prototype system with a reasonable likelihood of success. Another important finding is that the frequencies of certain resonant modes shift slightly with changes in residual stress. (7912826).-AA |