Marlin Failure Analysis and Redesign; Part 3, VIT Completion With Real-Time Monitoring

Abstract

Abstract Following the failure of the initial well in the Marlin field1 the preferred solution for completing the remaining wells was to control wellbore temperature via vacuum insulated tubing (VIT)2. Implementation of VIT required a number of computational and experimental innovations, including:Provision for insulating the tubing couplings, the source of up to 90% of VIT heat loss;Detailed flow loop temperature profiles using both axial probes and radial probes traversing the annulus outside the VIT. These profiles supplemented the conventional values of overall heat transfer coefficient and thermal conductivity obtained from the flow loop measurements;VIT performance, as measured experimentally, must exceed both thermal and mechanical design bases. As well survival depends on proper VIT performance, a distributed temperature monitoring system was developed and evaluated during full scale testing. On the Marlin TLP, fiber optic is run in each well along the length of the VIT to continuously monitor the production annulus temperature profile. A software system was also developed to take binary fiber data and feed an integrated thermal simulator-casing design software package that calculates safety factors for the B and C annuli. These real time safety factors interface with the platform alarm system and are continually monitored by operators. If a low safety factor is calculated, a well will be shut-in. In addition to feeding the platform alarm system, the software provides data to a web based plotting program. If a single joint of VIT loses its insulating properties, this specific joint can be identified and appropriate action taken. The monitoring system has also proved to be a valuable quality assurance measure for special annular gels used to minimize conduction and natural convection in the production annulus. This paper focuses on the value of the combined VIT and fiber/software monitoring system as a means of both controlling and observing well thermal behavior. Typical temperature vs. depth curves are used to illustrate the detailed information retrieved. Introduction Despite an extensive effort to ascertain a root cause of the collapse of the (10–3/4 in. × 8–5/8 in.) production tieback and ensuing deformation of the production tubing in Marlin Well A-2, a singular mechanism could not be discerned. This lack of resolution is primarily due to the fact that the status of the (13–3/8 in. × 10–3/4 in.) intermediate casing is unknown. Nevertheless, all possible root causes which the investigation team deemed reasonable had as a component an increase in temperature of the wellbore. This fact, coupled with the limited options available from the batch drilled wellbores, led the team to focus on controlling wellbore temperature by insulating the production tubing. Vacuum insulated tubing (VIT), even apart from its cost, is, unfortunately, a solution with its own set of design challenges. Thermally, it must, of course, be substantiated that the thermal characteristics are sufficient to solve the problem - that is, to keep the annuli sufficiently cool to render the well safe. For Marlin, this requirement was satisfied in two ways - by experimentally determining the overall heat transfer coefficient of VIT targeted for Marlin wellbores and by numerically modeling the resulting thermal performance of the VIT under a production scenario. The discussion below details the latter exercise, the former being addressed in the preceding paper in this series2. Further, validation of adequate thermal performance must be supplemented by checking the mechanical integrity of the VIT under a variety of burst, collapse and tension load cases.