CT-Failure Causes: A Decade of Experience

Abstract

This article, written by Assistant Technology Editor Karen Bybee, contains highlights of paper SPE 113676, "CT-Failure Monitoring: A Decade of Experience," by A.R. Crabtree, SPE, BJ Services, prepared for the 2008 SPE/ICoTA Coiled Tubing and Well Intervention Conference and Exhibition, The Woodlands, Texas, 1–2 April. The paper has not been peer reviewed. The full-length paper reviews a decade of tracking coiled-tubing (CT) failures and their causes and sets the results against the perspective of the changing nature of the business over the same time period. Critical factors for reliable CT service-life performance are identified, and whether reliability is higher today than a decade ago is examined. Introduction Since the beginning of 1996, all early CT-string failures experienced in company operations have been examined to identify cause of failure. The results of these examinations are recorded in a database to enable historical tracking, so trends can be observed and appropriate policy or procedural changes made. The full-length paper presents data captured from 1997 through 2007. Early failures are those that are experienced before the safe working life of a CT string has been fully consumed. Very few strings are used until they reach their maximum-allowable safe working life, and, therefore, this database basically captures all string-failure data. The database contains information on more than 250 failure investigations. During the same period, more than 2,200 CT strings were in service, performing approximately 60,000 jobs, using an average of 138 CT units, with both land and offshore operating locations. Research into the low-cycle fatigue performance of CT conducted since the early 1990s has resulted in predictive algorithms and fatigue-management software. This, combined with improved string-manufacturing techniques, has all but eliminated most pure cyclic-strain-induced fatigue failures (i.e., fatigue not associated with secondary mechanisms such as corrosion). Data Overview Examination of all the data (Fig. 1) shows that the leading reasons for early failure are dominated by five main classifications: corrosion, mechanical damage, manufacturing flaws (or defects), operator error, and hydrogen sulfide (H2S) cracking. These groups account for approximately 94% of all failures. Reviewing the overall data and superimposing the number of jobs per year (Fig. 2), it can be seen that the number of failures appears to be proportional to jobs performed. This would seem to indicate that the failure rate (jobs performed/number of failures) has not improved. However, this is not the case because the chart belies the underlying numbers. In fact, the failure rate from 2000 through 2007 improved 47%. The year 2003 was anomalous and is a result, in part, of an increase in the CT fleet in the later portion of 2002, leading to the addition of several new CT strings being used in 2003.