Root Cause Analysis of Drilling Lost Returns in Injectite Reservoirs

Abstract

Abstract Identifying the root cause of drilling failures can often be challenging due to uncertainty in whether drilling failures were a result of lost returns leading to borehole instability and pack off, or ineffective hole cleaning leading to lost returns, pack off, stuck pipe, and eventual well failure. Such an analysis is further complicated when losses occur over a wide range of equivalent circulating densities (above and below predicted fracture gradients), well angles, and hole sizes. The work presented combines a fundamental physics approach with field data to identify the root cause of drilling failures in Paleocene and Eocene injectite sand intervals. The field data consisted of image and conventional log data, downhole equivalent circulating densities (ECD), loss rates, leak off tests, and formation integrity tests. Study results showed that unfavorably oriented sand-shale interfaces, which can occur more frequently at the top of an injectite reservoir, can re-open at ECDs below predicted fracture gradients, but above the minimum stress, and result in massive losses. The physical model is shown to correlate well with field data supporting the conclusion of lost returns being the root cause for observed drilling failures. Complimentarily, proprietary transient hydraulics and hole cleaning models showed that fluctuations in drilling parameters (flow rate, ROP, etc.) leading to suboptimal hole cleaning at a specific location along the wellbore were not able to generate significant enough pressures to result in observed losses if ECD at the bit was managed within the anticipated fracture gradient. By identifying the root cause of drilling challenges, potential solutions can be applied to include revised fracture gradients, operational hole cleaning parameters, drilling techniques, and well path planning to reduce overall drilling risk. These results will allow for informed trade-offs between reduced drilling risk and future production volumes.