The Effect of Entrained Liquid on the Measurement of Gas by an Orifice Meter

Abstract

Abstract One of the simplest and most widely used devices for flow measurement is the orifice meter, Its use in industry is at present limited to measuring homogeneous fluids. This investigation was conducted to study orifice measurement of a two-phase flow system, to determine the magnitude of error induced in the measurements of gas by a liquid phase and to develop correlations which may be useful in predicting and compensating these errors. Introduction The successful operation of modern, large-scale continuous processes requires accurate and reliable measurement of flowing fluids. In most cases, sufficient accuracy can be obtained by measuring a single-phase fluid with one of several devices that are available. One of the simplest and most widely used devices for flow measurement is the flat-plate, square-edged orifice. It is relatively simple to obtain flow rates with an orifice meter for homogeneous fluids such as air, natural gas, various industrial gases, steam. water, oil and all liquids whose physical properties are known. However, serious errors in measurement result when this simple device is used where two or more fluid phases comprise the flowing stream. In many applications, homogeneity is difficult to achieve because of condensation. In others, the presence of more than one phase is unknown or ignored. Regardless, measurement of multiphase flow is complex and very little information is available for determining flow rate of such a system accurately. A standard method for measuring gas flow rates is reported in Orifice Metering of Natural Gas, a report prepared by the AGA. However, this is applicable only to homogeneous gas-flow measurement. A similar method for measuring the flow of liquids is not applicable to multi-phase flow systems. It was the purpose of this investigation to study orifice measurement of a two-phase flow system, to determine the magnitude of error induced by a liquid phase and to develop correlations which may be useful in predicting and compensating these errors. To study flow patterns and behavior in a visual system, this investigation was confined to an air-water system operating at pressures ranging from 0 to 15 psig at ambient temperature, using both horizontal and vertical meter runs. The results obtained are compared to and correlated with data obtained by R.A. Schuster and J.R. Wright in similar studies at different operating conditions. Review of Literature The behavior of compressible and non-compressible fluids in flow channels is well understood, and the pressure drop of a single-phase fluid due to friction in a pipe can be calculated quite accurately. Similarly, the flow rate of single-phase fluids can be calculated from the data obtained with flow-restricting devices such as the orifice meter. Any attempt to combine the equations describing the flow of a homogeneous gas or liquid in a circular duct results in a fundamental discrepancy in the assumptions inherent in deriving the separate equations. The following conditions occur in the simultaneous flow of gas and liquid in a horizontal pipe:1. A reduction in flow area available to the gas phase in a pipe due to liquid accumulation;2. A considerable loss of energy due to the roughness of the liquid surface which forms in waves and ridges; and3. A loss of energy by the gas phase which accelerates and transports the liquid along the length of the pipe, noting that the relative velocity of the gas is the greater of the two. So larger pressure drops are observed in horizontal two- phase flow than are found in the single-phase flow of either phase. Empirical attempts to relate the pressure drops of the individual phases to two-phase pressure drop have been made by several investigators. A study of the literature on two-phase vertical systems reveals the complexity of the flow characteristics, and it is not yet possible to predict the pressure drop due to a flow-restricting device, such its an orifice meter, from these data. JPT P. 657ˆ