Summary A new generation laterolog tool, the Azimuthal Resistivity Imager (ARI) is described. The tool makes deep azimuthal resistivity measurements around the borehole with higher vertical resolution than the Dual Laterolog (DLL) tool. An array of twelve azimuthal electrodes is incorporated into the dual laterolog array so as to provide twelve deep, oriented resistivity measurements while retaining the standard deep and shallow laterolog measurements. To allow full correction of the azimuthal resistivities for borehole effect, a very shallow auxiliary measurement is incorporated on the azimuthal array. Though the full-coverage azimuthal resistivity image has much lower spatial resolution than borehole micro-electrical images, it complements these because of its lower sensitivity to shallow features. Fracture evaluation and computation of structural dip are applications of the tool's imaging capabilities which are discussed and illustrated with log examples. Other log examples cover thin-bed response, Groningen effect and borehole corrections, including that for eccentering of the tool in the borehole. Introduction The Laterolog technique was introduced in 1951, with the Dual Laterolog tool following some twenty years later. Though instrumentation has been upgraded as technology has developed, the Dual Laterolog deep and shallow measurements, LLd and LLs, have remained essentially unchanged since their introduction. Together with induction tools, the laterolog provides the key input for basic formation evaluation. While important advances have been made in the design of induction devices in the past ten years, few comparable developments have been made in the laterolog domain, despite known limitations to the laterolog measurements. Reference electrode effects have plagued deep laterolog measurements since their early days. Though effects such as Delaware and anti-Delaware effect have been overcome by repositioning the measure and current returns, Groningen effect remains a particularly complex problem which has yet to be satisfactorily resolved. It manifests itself as an increase in the LLd reading in conductive beds overlain by thick, highly resistive beds. The vertical resolution of the deep and shallow laterologs is two to three feet, with a typical beam width of around 28 inches. Thin beds are assuming increasing importance as potential reservoirs, and the vertical resolution of the deep and shallow laterologs is increasingly recognised to be insufficient for adequate evaluation of these beds. Development of a pad-mounted laterolog-3 has reportedly improved vertical resolution to two inches, though a consequence is reduced depth of investigation. Paradoxically, pad or skid devices suffer from a larger borehole effect than cylindrical tools. Though the effect of dip is much less severe than for induction devices, whose responses are perturbed drastically, dual laterolog response is affected significantly across dipping bed boundaries. A directional resistivity measurement around the borehole axis would provide a means of correction for the effects of dip. In one sense such measurements are already available in the form of high-resolution electrical borehole imaging tools, which have been shown to be very effective in evaluation of complex reservoirs.

No abstract

A quantitative measurement of the flushed zone resistivity, Rxo, is one of the key parameters for complete log interpretation. When available at very high vertical resolution, evaluation of thin beds is rendered more accurate. To this end, a new device has been developed which not only provides good performance in thick mudcake/high resistivity contrast situations but has also proved reliable in the presence of thin mudcakes and/or low resistivity contrasts. The pad-mounted device uses dual current focusing — vertically unmonitored and azimuthally monitored. This combination provides fine vertical resolution and strong immunity to mudcake or standoff effects. Three measurement buttons, optimally positioned within the main electrode, provide resistivities at three different depths of investigation. The central button reads the flushed zone resistivity, while two others, located towards the pad edges, are predominantly sensitive to mudcake properties. Finite-element simulations were used for design optimization and for construction of the tool response data base. From this data base, the forward model of the tool was obtained by interpolation. The three parameters - Rxo, mudcake resistivity and mudcake thickness - are estimated in real time by a robust inversion algorithm based on the extended Kaiman filter method. This new tool provides superior quantitative measurements of Rxo at a vertical resolution of less than one inch. Log examples show significant improvements in vertical resolution and accuracy in the determination of Rxo. Coherent quantitative estimations of mudcake thickness have been obtained allowing identification of permeable zones.