Author:

Abstract:

The classical transfer path analysis (TPA) is a well established method used for tackling noise and vibration related problems for interior car sound. It has proved to be reliable but a complete analysis takes a long time and relatively large effort. Therefore there is an ongoing research for simpler and faster methods and several have been proposed in the last years. One such method, most often referred to as operational path analysis (OPA), attracted particular attention as it uses only measured operational input and output signals and calculates the transmissibilities between them to characterize the paths. The claim for its accuracy is based on being able to reproduce the original output signal by summing the calculated partial contributions but it has not yet been compared to other TPA methods. This fully operational method is now critically examined and compared to a reference classical TPA measurement. The results of this examination reveal three significant weaknesses. This paper will focus on one of them, namely showing the effect of neglecting a path in the analysis. In classical TPA the error clearly shows up in the summed contribution and thus can be recognized, whereas in OPA this mistake can result in erroneous individual path contributions yet remain unnoticed because it will not affect the summed contribution in some cases.