A strange question asked by a Client and I'm trying to find a convincing answer for my understanding. The question is, the support forces (guides/ line stop) determined by pipe stress analysis (PSA) are at the node points, i.e., at centre of pipeline. But in reality the reactions are exerted by the steel structure which may be at significantly below the centreline (pipe radius + shoe + steel member dimension etc). Does that mean that the piping is actually subjected to a moment/ torque (= reaction force x moment arm height), that is ignored in PSA?

For large pipelines (say >2m dia), the height between the pipe CL and the actual steel which acts as guide or line stops could be significant (say, >1m). does that mean the piping system is subjected by an additional Moment = Support Force x Moment arm height, that the analysis does not take into consideration? Must be something basic I'm overlooking - will appreciate others' opinions.

Yes, all pipe elements are simply 3D beam elements - infinitely thin sticks. You're right, the support is at the node point, on the centerline. Therefore you're correct in that there could be a moment or torque acting at the support location that your analysis isn't accounting for.

One way to account for this effect would be to drop a dummy rigid element from the pipe node down to the support location, and put the support at the bottom. Granted this is a lot of work, test it out to see if it resolves your concerns.

Client's concern is right. For big dia pipes, actual support distance is more. As Richard told, rigid element can be modelled and restraints can be put at the bottom where actual support will come.
Here, additionally, you can restrict rotational moment in axial direction (torsional) to check the moment. this moment will be of significant value for large size pipes.
One of the FCC unit I saw, the pipe dia was around 158 inch (around 4000 mm.). In stress analysis, we restricted torsional moment to check the value. It was considerably high.

The other thing is the stiffness of the support. Caesar II uses 1E12 lb/in for the default stiffness of all supports. From what my structural engineers tell me, that is a bit too stiff for most supports. So unless you are also modeling the support steel or at least adjusting the stiffness at every support point there will be more errors.

The model is just that, a model. It would take so long to try and cover all the real life values that you could never get them all. So we have to trade accuracy for efficiency. It can be tricky to find the balance.

Thanks,
Duncan

Damcewen,
support stiffness you can get from structure engineer in your company, but remember this stiffness may not be constant,
structural engineer's stiffness is based on steel at initial state data, but after pipe sit on steel, steel had been bended, the stiffness may not be initial state number,
and if you have multiple pipe on one beam, each beam has own stiffness, but those stiffness will be interacted. you can imagined only one pipe on beam, two pipes on beam at the same time?
all those should use your and structural engineer judgement.
Good luck
Limin