Capillary Pressure and Resistivity Measurements Using the Centrifuge: a New Interpretation Method

Simultaneous in-situ measurements of resistivity and capillary pressure in a centrifuge up to speeds of 10000 rpm are reported. The setup can easily be automated for routine measurements and the electrode design takes advantage of the centrifugal force to maintain good contact between the sample and the electrodes. For converting the experimentally measured average saturations and resistivities into state properties, a new perturbation method has been developed and compared with existing methods. Capillary pressure data obtained by the centrifuge and microporous-membrane-based method show good agreement for a synthetic ceramic sample. Introduction While measurement of capillary pressure curves of rocks using the centrifuge is fairly common,1-3 performing resistivity measurements in centrifuges has required the sample to be removed at every equilibrium step to make external measurements.4 Besides being inconvenient, this may introduce errors because of saturation reversal in parts of the sample. Fluid losses resulting from sample handling are also likely. Simultaneous in-situ measurements of capillary pressure and resistivity have been reported recently.5,6 Durand and Lenormand5 report data up to a maximum speed of 3000 rpm using electrodes in a polymer casing on the lateral surface of the core. Because of the shape and arrangement of the electrodes it is difficult to determine a geometric factor for this setup and the analysis is further complicated by the presence of saturation gradients. Raghuraman and Ramakrishnan6 have proposed a design where both the average resisitivity and saturation are measured over the entire sample volume and the cell constant is known exactly. The advantages of their four electrode configuration are that it uses the centrifugal force to maintain good electrode contact with the rock and measures the average resistance of the sample over the entire length. The setup can be easily automated to make routine measurements of three samples up to speeds of 10000 rpm. Unlike the time-consuming porous plate/membrane-based methods, the faster centrifuge technique is an indirect procedure and therefore requires additional data processing to yield the capillary pressure curve.2,7 In this paper, we describe a new perturbation method for centrifuge data interpretation. We compare this method with other techniques *author for correspondence such as the Forbes’ Sαβ method3 and Rajan’s method8 using various model capillary pressure curves. The design of the electrodes employed here causes some bending of potential lines near the sample end faces. A commercial simulator (Maxwell EM 2D Field Simulator, ANSOFT) was used to generate the current and voltage lines in the sample and determine the error in resistivity measurements at different speeds resulting from the nonuniform electric field. For comparative purposes we built a two-ended membrane setup similar to that described in the literature.9 This setup gave the capillary pressure curve directly and was used to validate the data-processing algorithms for the centrifuge.