The speed of response of commercial Coriolis meters to a step change in mass flow rate corresponds to a time constant which may range from 0.1 s to several seconds. This response is a result both of the dynamic response of the physical components of the meter and of the electronics and the computational algorithms used to convert that dynamic response into an estimate of the mass flow rate. A comprehensive investigation of the dynamic response is presented with a view to establishing the ultimate limits of the overall meter response. Attention is initially concentrated on a simple straight tube meter and analytical solutions are presented for the response to a step change in flow rate both for an undamped meter and for a meter with internal damping. These results are compared with results from a finite element model of the same meter and then the finite element modelling is extended to geometries typical of commercial meters. Finally, representative results are presented from an experimental study of the response of commercial meters to step changes in flow rate. A study of the essential components of the algorithm used in a meter leads to the conclusion that the time constant cannot be less than the period of one cycle of the meter drive. The analytical, finite element and experimental results all combine to show that the meters all respond in the period of one drive cycle but that the flow step induces fluctuations in the meter output which decay under the influence of the flow tube damping. It is the additional damping introduced in the signal processing to overcome these fluctuations which is responsible for the large observed time constants. Possible alternative approaches are discussed.
展开▼
