A New Method to Significantly Simplify the API Flow Tests for Gas-Lift Valves

Abstract

Abstract Accurate gas lift valve (GLV) performance is crucial for precise gas lift system design, but complex and time-consuming testing techniques are required to obtain valve performance correlations. Therefore, this paper presents a modeling approach that aims to reduce the testing effort and obtain the coefficients required for valve performance correlations. Specifically, the model replicates the flow capacity test (FCT) described in the API 11V2 (2001) and API 19G2 (2010) to determine the flow coefficient (Cv) and critical pressure ratio (Rcp) for gas lift valves (GLV). Technically, the FCT requires a modified GLV with an adjustable stem positioning system to obtain flow rate as a function of pressure drop and calculating Cv and Rcp for each stem travel. Among the drawbacks of the FCT are the valve modification itself and the large number of tests necessary. The proposed approach employs a 1D mechanistic model that considers equivalent diameters of the GLV orifice port and the check valve opening area to calculate the gas flow rate. The pressure drop across the restrictions is derived from Bernoulli's equation and used to calculate the flow rate, Cv and Rcp for several stem positions of the gas lift valve. Experimental data from a dynamic flow test (API 11V2) is used to calibrate the model. Simulation results of 12 different valves are compared to the Valve Performance Clearinghouse (VPC) database to validate the modeling approach. The VPC database was created based on GLV flow tests for valves from several manufacturers with different models and configurations. Results of flow capacity using the 1D mechanistic modeling show a consistent agreement with the VPC database. Overall, the error was always below 15% for both Cv and Rcp, which is a positive result considering that the current most accurate method shows errors of up to 25%. Even though the 1D modeling may be oversimplified given the consideration of 3D area changes as equivalent circular diameters, the method predicts Cv and Rcp with considerable accuracy. Moreover, the calibration using only one dynamic flow test result significantly reduces the time required to perform an FCT and eliminates the need to modify a GLV.