Laboratory Modeling of Mudcake Application and Erosion for Gravel-Pack Design in Horizontal Wells

Abstract

Summary Successful gravel packing of long horizontal sections depends in part on preventing carrier fluid from leaking off to the surrounding permeable reservoir. This requires correctly designed drill-in fluids to apply a thin filter cake, rendering impermeable the sand formation at the wellbore wall. Filter cake must be embedded sufficiently within the formation interstitial space to resist scouring by flowing sand during the gravel-packing operation. This paper summarizes testing using a large-scale laboratory apparatus and procedure designed to test the effectiveness of drill-in fluids planned for use in horizontal wells recently drilled in the Gulf of Mexico. This paper shows the benefits of modeling a horizontal gravel-pack interval before attempting to install the gravel pack. Gravel-packing techniques for long horizontal well sections were evaluated with a laboratory model of a casing/borehole annulus. In the horizontal section modeled, permeability is sufficient to interfere with the transportation and distribution of pack sand from total depth to dogleg when positive differential pressure is maintained. The goal of this modeling experiment was to predict whether the formation's permeability at the sandface could be reduced enough to allow successful gravel packing without damaging the formation far beyond the sandface. Gravel-packing technology is well established for near-vertical wells, with off-the-shelf equipment and textbook procedures readily available. The exceptionally long producing intervals in horizontal wells present a greater challenge for gravel packing, however, interval permeability is, by design, continuous. This continuous permeability both enhances well productivity and allows increased wellbore fluid flow into reservoir rock. When the permeable formation absorbs the fluid energy necessary to evenly disperse pack sand throughout the section, sand bridging can result, restricting the flow before the pack sand is fully distributed across the entire interval. The laboratory model simulates the axial flow of drilling and/or completion fluids through an annulus. The model has an impermeable confining wall (simulating the casing) and a permeable wall (simulating reservoir rock). Recently, it was used for testing the capacity of a drill-in fluid to reduce reservoir near-face permeability, and allow sand propagation throughout a long horizontal interval. Results obtained both in the laboratory and in subsequent field trials validate the use of a drill-in fluid in reducing near-sandface permeability. The procedure has been used successfully on multiple horizontal wells with typical horizontal section lengths of 1,000 to 2,500 ft. Plans are underway to continue the practice in additional wells.