Improved Density Log Lithology Identification Using a Borehole-Compensated Photoelectric Factor

Abstract A photoelectric factor (Pe) measurement is now commonly available with density tools. By aiding in lithology identification, Pe information provides better estimates of matrix densities, thereby improving porosity interpretation from density logs. Pe measurements, however, are adversely affected by borehole rugosity and mudcake. In order to minimize these effects, a borehole compensation method has been developed that yields improved Pe accuracy in non-barite muds. Prior Pe measurements normally were determined from the density detector farthest from the source (PeL) and were not compensated for borehole effects. However, the Pe obtained from the more tightly collimated short-spaced detector (PeS) is less affected by non-barite mudcake, tool standoff, and borehole diameter than PeL. Thus, unlike the compensation scheme used for density measurements, the primary Pe measurement is obtained from the short-spaced detector. The densities measured by the short-spaced and long-spaced detectors are both affected by the same borehole factors, although with different sensitivities. It is appropriate, therefore, to derive the density correction term from the difference between the two densities. However, PeL is sensitive to several borehole parameters that do not effect PeS. Thus, a correction based on the difference between PeS and PeL, although partly successful, can lead to inaccuracies. Therefore, a better Pe compensation method was developed. The primary borehole parameters that affect PeS are the mudcake thickness, mudcake density, and mudcake Pe. (Standoff is included in this analysis by considering the mud between the pad and the formation to be mudcake.) In order to reduce the complexity of the problem, the mudcake Pe is assumed to be small, which means that this model is only appropriate for muds that do not contain barite or hematite. Under this condition, it is shown that the required Pe correction can be determined from the density correction and an estimate of the mud weight. The correction term is added to PeS to obtain a compensated Pe. A correction term is also obtained for PeL. However, the compensated PeL is less accurate than the compensated PeS, so it is only used in conjunction with PeS to produce a quality curve. Verification of the Pe compensation is demonstrated with test formation data obtained using various borehole fluids, tool standoffs, and simulated mudcakes. Shop calibration procedures and wellsite check tolerances are presented for density and Pe logging. Finally, log examples are presented to illustrate improvements in lithology identification using the compensated Pe.