MArty,
Do you have any idea as to the cause of the creep damage?
The reason I ask of course is that circumferentials obviously cross over longitudinals all of the time which, which would of course create the same heat regime histories as a weld-o-let on a corcumferential. If there is no creep damage history similar to what you speak of it would lend itself to the argument that its not a welding metallurgy issue but something else.