Computational Fluid Dynamics (CFD) tools are evaluated for use in industrial design applications by predicting primary control valve performance characteristics. The performance parameter of primary interest to the manufacturer is the flow coefficient, Cv. Valves having relative valve capacity factors between 2.5 and 13 were modeled. The control valve Cv was experimentally measured and numerically predicted. Both equal percentage and linear characteristic valves were represented in the study. The numerical (simulation) study presented in Part 1 showed that the valve Cv and the inherent valve characteristic could be accurately predicted using axisymmetric flow models over most of the plug travel. In addition, the study demonstrates the usefulness of simplified CFD analysis for relatively complex 3-D flows.

No abstract

An experimental study was undertaken to verify an axisymmetric numerical model of a control valve flow field. The numerical model, which utilized Computational Fluid Dynamics (CFD), was formerly developed to be used as a design tool by manufacturers of control valves. In this work the model was first tested by comparing its results to data taken on an axisymmetric flow field experiment. Then the model’s application to actual three-dimensional control valves was tested by studying the pressure and flow field through a three-dimensional control valve. The results showed that the axisymmetric numerical model is accurately modeling an axisymmetric flow field. In addition, the results showed that control valves have a predominantly axisymmetric flow field for most of their plug travel which make them suitable for the model. Finally, the results showed details about the flow field such as where separation and reattachment may occur.