Historically, motor temperature analysis in electric submersible pumping systems (ESP) attracted the most attention due to the vulnerability of insulation under temperature. For wells with low or moderate downhole temperatures, motor temperature alone is not effective to protect the system against no-flow conditions. This issue has become more critical in unconventional gassy wells, many ESP failure modes are more associated to high temperatures in the pump than the motor. Under gas locking or no flow conditions when production (cooling) fluid stagnates, the pump generates much more heat than the motor and experiences a faster temperature rise becoming a serious issue for the health of the ESP. Traditional pump intake and discharge thermocouples (TC) cannot detect this phenomenon because their locations are too far from the source of heat generation. This paper describes testing where several TCs were placed in an ESP pump. Temperatures were monitored when the pump was operated through different gas volume fractions (GVF) and flow rates. A gas locking condition was also simulated in a test loop to study the transient condition. Subsequently, a thermal model was developed and compared to the testing data.

The test used a fully enclosed, high-pressure gas loop. A 12-stage, mixed flow type with best efficiency point (BEP) at 600 BPD pump was horizontally mounted in a test bench. Ten TCs were installed at the bottom bearing, No.1, 6, and 12 diffuser bearing in both X and Y directions, respectively. Three TCs were attached to the pump housing on bottom, middle, and top locations. Pump intake/discharge temperature and pressure were captured during testing. The mixture volume of nitrogen and water was measured and supplied to the pump intake. Experimental data was acquired continuously for evaluating different operational conditions. The intake pressure, GVF, flow rate and rotational speeds were controlled in the experiments. In a static state, the thermal model started with energy equilibrium and calculated the temperature rise due to the difference between the pump brake horsepower and hydraulic horsepower. In a transient state, finite-element analysis (FEA) was used to predict the thermal profile from the stage bearing to the pump housing.

Based on the thermal testing and modelling results, several ESP failure modes and tear-down examples will be discussed. The concept of minimum continuous thermal flow (MCTF) will be mentioned. A reservoir model was used to understand the difference in the nitrogen/water testing system and to develop the possible strategy to recover from pump gas locking. In summary, the pump temperature study provided a better understanding of the pump gas locking condition, a better method to conduct ESP health monitoring and improve reliability by avoiding overheating the pump.

This paper adds a comprehensive knowledge of pump temperature analysis to the ESP industry. The results will help define the running limitations of an ESP in a gas condition and improve design, application and operation to mitigate the gas locking issue in unconventional oil production.

You can access this article if you purchase or spend a download.