The effect of contaminated orifice plates on the discharge coefficient

Abstract

Orifice plates can be contaminated by oil, grease, pipeline sludge or other liquids or solids. Experience shows that sometimes the contamination extends to the sharp edge of the orifice plate, but that on many occasions any possible contamination near the edge is cleaned by the flow. The latter case is investigated here. In some of the existing sets of data the contamination in the form of soft deposits used to gather the data means that interpretation is not straightforward; so Computational Fluid Dynamics (CFD) was used to assist in the interpretation.In the experimental work the contamination was simulated by sticking circular metal discs of defined thickness and radius to the plates so as to leave an untouched region in the neighbourhood of the sharp edge. The effect of this contamination was measured in nitrogen at 63 bar absolute. CFD was also used to investigate contamination of this type.A contaminant angle was defined, and the CFD predictions were plotted against the contaminant angle. The simulations for a diameter ratio, β, equal to 0.6, almost lie on a single curve. The β=0.2 and β=0.4 points lie above this curve and the β=0.75 points lie below the curve. The new NEL experimental data are compared with the CFD and the Advantica data on coated plates with a clean ring as in ISO/TR 12767.It is clear from the experimental data, as indeed from the CFD, that although the ratio of the thickness of the contamination to the distance from the orifice edge is the most important effect on the shift it is not the only effect. There is an effect of the ratio of the distance of the contamination from the orifice edge to the orifice dam height. An equation for the percentage shift in discharge coefficient was derived to fit the experimental data: this has an uncertainty of 0.28% based on 2 standard deviations. The CFD are in remarkably good agreement with the experiments and support the inclusion of the new model in a revision of ISO/TR 12767.