Best Practices for Designing High-Speed Centrifuge Capillary Pressure Experiments

Abstract Centrifuge experiments are essential for studying capillary pressure and relative permeability in underground geological formations for hydrocarbon production and gas storage applications. Despite the negligible changes in centrifuge equipment design over time, inconsistencies in fluid saturation and pressure interaction characterization are typical. This is due to the variations in test procedures, interpretation, and history-matching across different laboratories or when compared with other standard experiments. This work proposes considerations and best practices for designing a centrifuge experiment. This work utilizes qualitative and quantitative analyses to assess centrifuge tests including the drainage and the imbibition. The study starts with evaluating hardware, trade-offs, and pre-test calibration. Afterward, it delves into the effects of calibration factor (CF) and core-holder deformation due to centrifugal forces on saturation calculation presenting CF calculation methods. Design considerations for rotational speeds and best practices as a function of centrifuge equipment design, and common misconceptions. Other issues discussed include production data analysis and methods for improved processing of the pixel data obtained from high-speed cameras, end-point saturation values validation methods for accurate residual fluid saturations, mass and volumetric calculations, NMR T2 cumulative amplitude, and manual image processing. The findings of this study highlighted and quantified the impact of poor experimental design on centrifuge results based on three carbonate core samples. The findings of this study showed that inadequate pre-test calibrations and high noise levels in high-speed images predominantly influence estimates of residual saturations. These two factors resulted in cumulative errors in residual saturations varying from 1.5-10%. We further assessed the reliability of end-saturation validation techniques using multi-step capillary pressure centrifuge test results obtained for Indiana Limestone core samples, among which saturations calculated from manual pixel analysis and volumetric method were the most reliable. The design of centrifuge experiments is a topic that has received relatively little attention in the literature. This work offers comprehensive guidelines derived from extensive SCAL experience, presented as a critical review. The findings presented in this work can be used to design an accurate experiment and gain reliable capillary pressure data, which are essential for subsequent pore-scale modeling, numerical simulations, as well as production optimization and enhanced oil recovery (EOR) strategies.