Horizontal Displacements

For a straight run of 136 mtrs at 60 Deg c, on a pipe rack between two anchors without a loop absorbing around 65 mm total expansion, results shows almost zero horizontal displacement at intermediate supports.
I was expecting some horizontal displ. created due to buckling of pipe due to absorption of 65 mm expansion.
Also the expansion stress value is around 62 %.
Other than the High Anchor loads, the pipe seems to be not failing. Will this be the true Scenerio.

Thnks.
Rocky

Oh come now THINK!!!!

CAESAR II and the B31 Codes do not deal with unstable columnar buckling. The codes warn you not to do such things.

Try this 10M of pipe Anchor to Anchor heat it up to 350C and check the B31 stress report... it will show no overstress.

YOU MUST THINK AND USE GOOD JUDGEMENT AND BE EXPERIENCED TO DO THIS WORK. The foreword to the B31 codes state this. B31.3 speaks of the qualifications of the designer. It does not say all you need is a degree and a a pipe stress program.

Try reading the online M.E.N. articles on this web site at least! I wrote an article that discussed this.

So go to school do your homework and THINK!

All pipe stress programs (CAESAR II included) utilize the "3D Beam" element. This is essentially an infinitely thin stick. The subsequent analysis assumes:

- small deflections and rotations
- elastic behavior
- no 2nd order effects (such as buckling)
- no local effects (such as shell distortion)

Your model will (analytically) develop extremely high anchor loads. These loads will never be realized in reality, because the pipe will buckle. However, pipe stress programs don't know this. Buckling is not good for your system, which is why you never see a straight run of pipe between two anchors.

At least, not on purpose, hopefully....

Ofcourse it was a hypothetical case for study. Just to know how the system behaves between the anchors.
Thnks for the comments.

Rocky,

You're "hypothetical" case has practical implications. Long straight runs of FRP pipe (and thermoplastic pipe) are often supported in this manner. With its low modulus of elasticity, the loads and stresses developed in the pipe are often well within allowable compressive stress levels. And the loads at the supports are also much more manageable. However, the low modulus also makes the pipe much more susceptible to column-type buckling, and this must be checked by the designer. CAESAR makes no attempt to check this (nor does any other pipe stress program that I'm aware of).

Regards,
Bruce Hebb

Does the nature and importance of column buckling change if you are talking about (displacement based and stress limited)expansion loads as opposed to a (force based) sustained loads?

If so, I think it would be hard to unscramble those eggs without a good review by the designer.

Dave,

Good question. I certainly treat them differently. For the case in question, where the loads are due to restrained thermal expansion, "buckling" of the FRP pipe would probably result in nothing worse than "snaking" of the pipe. Accordingly, most FRP pipe suppliers publish "recommended maximum guide spacings to prevent column failure" which are only marginally greater than the critical length (based on pinned-pinned ends). If the compressive loads were somehow generated by primary loads, a much more generous safety factor would be appropriate to protect against real collapse.

Regards,
Bruce Hebb

Re question on load controlled versus strain controlled buckling, Section III, Subsection NH requires a factor of three margin on load controlled buckling and a factor of 1.67 on strain controlled buckling. B31 codes do not address this.

Re straight pipe between anchors, see B31.3 para. 319.2.3(c). You need to do something different than the default code analysis.