For subsea well drilling, the drilling rig is connected to the subsea well by a marine riser and subsea BOP equipped with a remotely controlled wellhead connector latched onto the subsea wellhead profile.
The level of cyclic loading on subsea wellheads is steadily increasing due to use of increasingly larger drilling rigs with larger BOPs, the drilling of wells in harsher environments characterized by high waves.
The remotely controlled wellhead connector forces a series of locking dogs into an externally machined profile on the wellhead. This external profile is generally referred to as a wellhead profile. The fatigue resistance of this safety-critical connector is typically estimated by finite element analyses (FEA). Due to the large size of the equipment, and high cost of testing, very limited fatigue testing, if any, has been carried out. A test method has therefore been developed, where a special test fixture is used to apply realistic boundary conditions and variable tensile loads to a small sector or segment of the wellhead connector. A primary objective is to generate fatigue-critical stress fields in the segments under tensile test load that closely replicates the stress fields in the full-scale wellhead connector. A secondary objective is to evaluate the possibility of using segment testing to determine the fatigue capacity of the full-scale connector. The testing of narrow sector segments allows the use of readily available test apparatus. It is thereby envisaged that the total cost of testing (specimens and test laboratory costs) can be substantially reduced in comparison with full-scale connector fatigue testing.
This paper describes the fatigue testing of wellhead connector segments, and the test results in terms of cycles to failure and the failure modes, i.e. crack initiation point, and final crack geometry. The test scope consists of nine segments tested in-air at ambient temperature (nominal 20 °C), at a frequency of approximately 2 Hz under axial load of R = 0.1. At the time of writing this paper, six out of these nine tests have been performed. These six fatigue tests are presented in this paper.
The test results are compared with estimates achieved by FEA of the test assembly and relevant S-N curves for the materials. It will be determined if the test results can be accurately predicted by the fatigue analysis methodology in Section 5.4 of DNVGL-RP-C203 (C203), including use of the new series of S-N curves for high strength materials in Appendix D.2 of C203. This design approach assumes that other failure modes (e.g. fretting or other local effects in the interface between components) do not govern the fatigue life, as this cannot be predicted by the fatigue analysis method applied here.
The fatigue test set-up and the finite element analysis of the segment test is presented in the OMAE2020-18652 paper.