Davies' bar is usually employed for dynamic characterization of vibration and acceleration measurement systems to determine the time history of the strain pulse. It is usually instrumented by electrical resistance strain gauges whose longitudinal extension can introduce limitations in the frequency response of the device. Mathematical models and experimental investigations can be found in the literature. The method proposed in this paper focuses on the smoothing effect introduced by the measurement technique. It is based on the processing of two measurement signals from two couples of strain gauges, having suitably selected different gauge length on the ground of the assumption lambda/L = n, by which the bar has been instrumented. The study of the direct problem, modeling the strain pulse as a Gauss bell curve, provides inforrnation about the function epsilonmax(L), defined as the locus of the points to which belong the maxims of the recorded strain pulse, expressed as a function of the longitudinal dimension of the S.G. active grid. The study of the inverse problem, defines the parameter sigma as a function of measured values from two different strain gauges that enables to obtain by extrapolation epsilonmax that represents the maximum strain that could be measured by means of an ideaI strain gauge with zero grid length. The equations describing the propose procedure appear simple and their application results easy and immediate. The proposed procedure proved to be valid and was confirmed by the comparison of its results and experimental results reported in literature. The overall quality of the obtained results is confirmed by the related uncertainty value obtained in the uncertainty evaluation perforrned.
Elastic wave characterization along Davie's bar
VALLASCAS, RINALDO;
2004-01-01
Abstract
Davies' bar is usually employed for dynamic characterization of vibration and acceleration measurement systems to determine the time history of the strain pulse. It is usually instrumented by electrical resistance strain gauges whose longitudinal extension can introduce limitations in the frequency response of the device. Mathematical models and experimental investigations can be found in the literature. The method proposed in this paper focuses on the smoothing effect introduced by the measurement technique. It is based on the processing of two measurement signals from two couples of strain gauges, having suitably selected different gauge length on the ground of the assumption lambda/L = n, by which the bar has been instrumented. The study of the direct problem, modeling the strain pulse as a Gauss bell curve, provides inforrnation about the function epsilonmax(L), defined as the locus of the points to which belong the maxims of the recorded strain pulse, expressed as a function of the longitudinal dimension of the S.G. active grid. The study of the inverse problem, defines the parameter sigma as a function of measured values from two different strain gauges that enables to obtain by extrapolation epsilonmax that represents the maximum strain that could be measured by means of an ideaI strain gauge with zero grid length. The equations describing the propose procedure appear simple and their application results easy and immediate. The proposed procedure proved to be valid and was confirmed by the comparison of its results and experimental results reported in literature. The overall quality of the obtained results is confirmed by the related uncertainty value obtained in the uncertainty evaluation perforrned.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.