Elimination of transducer mass loading effects from frequency response functions

It is well known that the quality of measured frequency response functions (FRFs) is adversely affected by many factors, most significant sources being noise and systematic errors. It is also known that the accuracy and the reliability of various analyses using the measured FRFs depend strongly on the quality of measured data. This paper aims to remove one of the major systematic errors in measured FRFs, namely the mass loading effects of transducers. This paper presents a new method based on the Sherman-Morrison identity for the elimination of mass loading effects of transducers from measured FRFs. The formulation presented here is general in the sense that it can be applied for both fixed- and moving-transducer cases. In the case of moving-transducer type of tests, the use of dummy mass is utilised. However, one of the distinct features of the new method is that it avoids the need for the measurement of cross FRFs of a structure with a dummy mass attached to the structure, hence avoiding further contamination of cross FRFs. The applicability of the method is also assessed using experimental as well as simulated data. (C) 2003 Elsevier Ltd. All rights reserved.