Correction and evaluation of the effect due to parasitic motion on primary accelerometer calibration

Abstract

Primary accelerometer calibration is carried out under the assumption that a vibration exciter gives a rectilinear motion to an accelerometer to be calibrated. However practical vibration given by the vibration exciter includes parasitic motion such as transverse, rocking, and bending motion. Such parasitic motion would give two serious effects on primary calibration results, transverse motion effect and vibration distribution effect. Transverse motion effect is caused by an inner product of the vectors of both transverse motion and transverse sensitivity. On the other hand, the vibration distribution effect is caused by relative motion between a sensing point of accelerometer and a spot sensed by the interferometer. As these effects have close interaction between parasitic motion and measuring instruments, it would be very difficult to evaluate them by measuring independently each component. In this study, we propose simple methods to correct and evaluate these effects on the primary calibration. The transverse motion effect on primary calibration is corrected and evaluated by turn of fixed transverse sensitivity axis using some spacers with different thickness. The vibration distribution effect is also corrected and evaluated by sensitivity measurements at known different spot positions. The validity of these proposed methods was clarified by experimental results. The proposed methods would enable more practical estimation of uncertainty budget.