Effects of Mandrel, Borehole, and Invasion for Tilt-Coil Antennas

Abstract

Several recently published studies discuss the concept of inductive resistivity logging devices with oblique transmitting and/or receiving coils. Both wireline induction and LWD propagation resistivity tool concepts have been considered. Directional resistivity measurements and improved anisotropy measurements are among the benefits promised by this type of device. Analyses based on point-magnetic dipole antennas were used to illustrate these potential benefits. The effects of a metallic mandrel, borehole, and invasion were not considered due to the absence of a suitable forward model. This paper characterizes mandrel, borehole, and invasion effects for a variety of candidate tilt-coil devices with antenna array parameters similar to those of the previous studies. The characterization is based on calculations from a new forward model that includes tilted transmitting and receiving coils of finite diameter embedded in a concentric cylindrical structure. Important details of the forward model used in the calculations are also provided. Introduction Conventional propagation resistivity devices are routinely used for geo-steering applications. Since data from these devices have essentially no azimuthal sensitivity, the LWD engineer is greatly aided by a priori information regarding the proximity of the target bed relative to other geologic features such as shales and water-bearing zones. Suitable a priori information is often available from offset logs. In cases where offset logs are not fully useful due to changing depositional environments or different tectonic settings, azimuthally sensitive resistivity data would improve the quality of the geo-steering effort. One way to achieve azimuthal sensitivity to benefit geo-steering (and to use for imaging) is to construct a tool similar to a conventional propagation resistivity device, but with the transmitters and/or receivers tilted with respect to the axis of the drill collar. In fact, DRTs (Directional Resistivity Tools) have been proposed in the literature for this purpose.1–3 To the knowledge of the authors, DRTs have only been analyzed using point-dipole models which ignore both the drill collar and the finite size of the antennas. For apparent lack of a suitable forward model, mandrel, borehole, and invasion effects have not been considered in the literature. A model has been developed which accounts for tilted transmitters and receivers embedded in arbitrary layers of a concentric cylindrical structure.4 Many important details of this model are discussed in Appendix A. The term mandrel effect is used here to denote the difference between values calculated using a point-dipole model and the model that accounts for the mandrel encompassed by the antennas. Mandrel effects on DRT measurements will be grouped into three categories:absolute effects where the mandrel primarily attenuates the signals due to a reduction in the magnetic moment of the antennas;residual effects that remain after an air-hang calibration is applied to the data; and,perturbations to the azimuthal sensitivity of the measurement caused by the finite size of the antennas and the drill collar. Algorithms that transform raw tool measurements to resistivity values can be based on computationally simple point-dipole solutions without significantly degrading the accuracy of the results if mandrel effects associated with categories I and II can be suppressed. For conventional LWD propagation resistivity measurements, mandrel effects of type I are addressed by air-hang calibration. Algorithms that suppress type II mandrel effects are discussed in the literature.5 Type III mandrel effects are not discussed here. From the examples discussed below, one may conclude that currently practiced methods for removing mandrel effects from conventional LWD measurements are suitable for application to DRT data provided that:the transmitter and receiver are tilted equally in opposite directions; or,one of the antennas is not tilted.