Dear milist'r & admin,

Recently i still confused how to model seabed support on offshore pipeline? Could like to help me to recommend what the right model of seabed support close to the actual condition of seabed? is it true if i model the seabed support as Y+ with friction ? or with soil model with zero buried depth?

Thanx for your help

Regards,
Andriksyah

Dear F.Andriksyah,

If you don't have any clamp for the pipe laying on seabed you mentioned, +Y support per 5D is quite accurate. But please note as in actual condition pipe laying on seabed in design/construction stage may will be burried by soil due to current. Please consider this. Typical value of the burried depth near to existing line pipe may help. Please search the related document. May be your client has it.

Dear all,

Thanks Samsul for your help
(I wish you will be success in the new company )

Actually i have considered to analyze unburied offshore pipeline along its route with Y+ (with friction) support. I would like to know your opinion whether it is unnecessary or not because during analysis i found unconvergence error in my model.
As in this forum (discussion topic about unconvergence), this type of unconvergence is due to existence of many Y+ support and its friction coefficient. I doubt to change Y+ to be Y without friction coefficient (based on recommendation as stated in the discussion forum). Is it quite accurate?

May i change the seabed support to be soil model with "zero" burial depth instead of Y+ support?

Please enlighten me, again
Thanks.

You can change your +Y to Y supports as long as you carefully check your results. The one(s) you change should have a very small upward force only to get representative results.

Regarding to non convergence, hope using the attached file will help.

For friction problems, go to Tools-Configure Setup and change the Friction Stiffness to a lower value, increase the Friction Angle Variation, and increase the Friction Normal Force Variation.

You can't do that because it doesn't reflect the real condition except only to get the soil stiffness.

Here's what the American Lifelines Alliance (ALA) says about offshore pipelines in "Guidelines for the Design of Buried Steel Pipe" Appendix B:Soil Spring Representation -

"Horizontal soil loads of offshore pipelines resting on the sea floor increase more gradually with displacement due to the formation of a soil mound in front of the pipeline. Determination of the soil spring characteristics for this condition requires special treatment by experienced practitioners and is not covered in these guidelines."

COADE will be adding the ALA soil spring representation to CAESAR II Version 5.20.

Samsul,

That's a pretty good writeup on convergence and CAESAR II. Thanks fo rsubmitting it here.

I'll second your point on reaching convergence - that it's more art than science.

My art is a little different. I try not to adjust friction tolerance but instead turn nonlinear restraints into linear restrants. I can see the "error" in the results in terms of pipe deflection or restraint load. When you fool with tolerance, like you say, you are adding more uncertainty to the analysis.

Dave,

That's a re-print of my article on non-convergence. But I am glad you like it.

Many discussions about non con-convergence takes place in this forum. Before I attached it yesterday, I tried to find the previous posts contains it. I believe I've ever found it somewhere in page xx. But could not find yesterday. So I decided to attach it once again

"COADE will be adding the ALA soil spring representation to CAESAR II Version 5.20"...we can't hardly wait...:)