ERROR IN BENDING WAVE POWER MEASUREMENTS RESULTING FROM LONGITUDINAL-WAVES

Bending wave power in the ''far field'' of a beam is often estimated w ith two closely spaced accelerometers. When measuring power in flows c onsisting of bending waves only, this method has been shown to provide reliable results. However, in most practical measurement cases some l evel of longitudinal wave flow will be present. Measuring bending wave power flow in a beam that has a significant level of longitudinal wav e flow can provide inaccurate results, particularly when techniques us ing measurements on only one side of the beam are used. To determine t he limitations of one of the most commonly used one-sided bending powe r measurements (the two-accelerometer two-position linear finite diffe rence estimate), a measurement on a beam is modeled for combined longi tudinal and bending wave power flow. This information is used to predi ct the sensitivity of the basic two-accelerometer technique to longitu dinal wave errors in bending wave measurements on a 1.524 mx0.1016 mx0 .1016 m 7075-T651 aluminum beam. (C) 1996 Institute of Noise Control E ngineering.