Semester of Graduation

Spring 2026

Degree

Master of Science (MS)

Department

Petroleum Engineering

Document Type

Thesis

Abstract

Gas lift valves are critical components in artificial lift systems, and their performance directly influences well unloading efficiency and long-term production. Current industry models used to correct valve opening and closing pressure for temperature rely on assumptions that have not been extensively validated against experimental data. This study experimentally investigates the opening and closing behavior of injection pressure-operated (IPO) gas lift valves under controlled elevated-temperature conditions. A load rate test bench was modified to include a heating system capable of maintaining valve temperatures up to 220°F. Five gas lift valve designs from multiple manufacturers were tested over a range of temperatures and opening and closing pressures were measured using a consistent data-processing methodology. Results show that both opening and closing pressures increase approximately linearly with temperature for all valves tested. Comparison with an industry-standard temperature-correction model revealed significant underprediction of opening pressures at elevated temperatures, with errors exceeding 130 psi in some cases. These discrepancies far exceed experimental uncertainty and are attributed to assumptions such as a constant R-ratio, a fixed dome volume, and the equivalence of dome and closing pressures. A first-principles fixed-volume experiment confirmed that pressure-temperature measurements and Gay-Lussac’s law calculations are accurate within ±30 psi, indicating that observed prediction errors are dominated by model limitations rather than measurement uncertainty. Even though the percentage errors may appear small in isolation, their practical impact is significant when considering the sequential pressure drop accumulated across multiple valves during well unloading, where minimum recommended pressure drops per valve can range from approximately 9 to 52 psi depending on valve port size. In this context, model prediction errors exceeding 130 psi can easily surpass the required pressure loss per valve, potentially causing valve interference and compromising unloading performance. This study proposes a simple experimental correlation of opening and closing pressures as a function of temperature. The proposed model show a disagreement of less than 30 psi with experimental data, while the standard models showed a disagreement of around 50 psi in average. The findings highlight the importance of valve characterization at temperature and suggest that the proposed correlation offers a practical alternative to existing correction models for gas lift design under downhole conditions.

Date

4-7-2026

Committee Chair

Waltrich, Paulo

LSU Acknowledgement

1

LSU Accessibility Acknowledgment

1

Download

Share

COinS