Optimization of Rotary-Shouldered Connection Reliability and Failure Analysis

Abstract

IADC and SPE Members Abstract This paper presents a case history of drillstring failures in some ultradeep wells drilled in Italy. A new method to optimize the selection of the most suitable Rotary Shouldered Connections (RSC) and failure analysis is developed by using a new factor identified as Fatigue Strength Ratio (FSR). The concept of FSR is mainly based on information available from full scale fatigue tests of RSC's. On the basis of A.8.3 paragraph of API RP 7G this work suggests a wider discussion about RSC's operating limits under combined various static and dynamic loads. A better understanding of static and fatigue characteristics of connections is provided along with plots and equations. Introduction Due to the increase in well depths and the complexity of drilling operations to be faced in northern Italy, the control of drillstring element serviceability is becoming more and more important. In fact, apart operating risks, the cost of a complete drillstring can be an important economical factor. RSC's are the weakest items in terms of reliability especially the drill collars' ones; thus the necessity to have a better understanding of their behaviour under applied static and dynamic loads. Early in the eighties, the discovery of Villafortuna-Trecate field, - north west of Italy, lead to the drilling of ultradeep wells. Operating conditions are extremely severe, since along with the considerable depth of the wells (from 6000 m to 7000 m), high temperatures (up to 180 C) and pressures (over 1000 bar) have to be faced. Furthermore, the drilled formations are very hard and abrasive, while the bottom hole environment is very aggressive due to the presence of H2S and CO2. In such operating conditions an abnormally high occurrence of drillstring failures was recorded. A survey based on the first 7 wells showed that most of the failures took place in the BHA (77%, not including HWDP). A significative number of failures was detected in the drill collars (41% of the total) and the combinations 8" DC with 6 6 / Reg and 9" DC with 7 / Reg gave the highest number of failures (50%), mainly in the last engaged thread of the box (see Fig. 1). So, in order to decrease the number of drillstring fatigue failures non standard inspection procedures were introduced. Non destructive tests (NDT) of drillstring elements were set at 300 hours of rotation for the BHA, 600 hours for HWDP and the first 20 DP's and 1500 hours for the remaining drillstring. A further survey demonstrated that, after the NDT's procedures' were introduced, the number of failures was actually decreased. However, a high number of cracks -was detected during NDT's (Fig. 2). P. 935^