Titanium is increasingly being applied to offshore installations due to its high strength, low modulus, environmental resistant properties and high fatigue capacity. These combined properties make certain titanium alloys ideal in the application as a stress joint (TSJ) where flexibility and fatigue capacity are the main design limitations. Over 80 TSJs are currently in service worldwide and are now being designed for more arduous fatigue service involving high motion floaters and locations with more severe sea states. This paper explores the application of BS7910 level 2 & 3 ECA for an ASTM Grade 29 titanium production riser TSJ to a 100 year storm operational limit, and a 1000 year storm survival event for a high motion floater in the GOM. Two critical locations are addressed: 1) A section at the top of the tapered wall immediately below where the TSJ terminates at the platform in the stress joint and where the base material properties apply, and 2) the girth weld location farther down the tapered section where the wall is thinner. Typical measured J-R curves together with Paris-law constants with mean stress effects incorporated are presented and used in the ECA. Comparisons between the BS7910 level 2 and 3 show the increase in allowable defects due to the more elaborate procedure at level 3. However, since the fatigue loading (and S-N endurance) of this particular case is high, the allowable initial defects turn out to be small for the weld location, and the weld location was more critical than the base metal location in spite of lower fatigue loading. This is due to the weld metal having a higher crack growth rate in the Paris region, a reduced crack growth threshold, and a reduced J-R curve. Other approaches to fatigue performance integrity for cases where allowable defects may be too small to detect are suggested. These include the local thickening of the wall at the weld and the redesign of the taper profile optimized for fatigue rather than extreme load only. It is concluded that: a) Level 3 analysis gives larger allowable initial defects level 2; b) It is important to carry out ECA at an early stage in the design process to identify potential problem areas; c) Non-conventional NDE methods should be investigated for future TSJs which are designed to operate under demanding fatigue loadings.

This content is only available via PDF.
You do not currently have access to this content.