Hi to all,

another question from me:
I've a system with a lateral compensator. The pipeline will be partly removed. The lateral compensator will be forced to have a displacement of 50mm. This will be done by adding Y supports on the pipes on both sides of the compensators. The support on one side will have 50mm less height than the support on the other side.
How can I simulate this in Ceasar II. If I model the compensator with a deltaY = -50, the tension resulting from the forced displacement of the compensator will not be calculated. Is there a way to enter this properly?

Thanks in advance for your replies,
greetings,
Corné

In my humble opinion, i would be using offset feature

I've tried this, but using an offset of -50mm on both the expansion joint element and the corresponding rigid gives me the same stresses as when I don't add the offset.

I also need to add hangers in the system, so the option I mentioned earlier (adding displacements) isn't really on option to, cause of the hanger design.
I'm almost thinking that adding 2 external forces to represent the pretension is the only solution.

Any other suggestions?

EJs are designed keeping in view maximum installaton movements. It is always tried to limit these installation movements to minimum. As standard practice, EJ should not be pre-loaded other than these installation movement. In my opinion, the installation tolerances and clearances should not be modeled.

If I understand your description correctly, you can try either "cold spring" or a "concentrated force". The "force" my be the easier way to go.

Since you know the stiffness of the compensator, you can compute what force will be required to achieve a desired displacement. Define that force in "F1". Then when you want to displace the compensator, include "F1" in your load case.

Thanks all for your ideas.

I don't want to displace the compensator using a force. I enter the compensator as a displaced one (DX = 2000mm, DY = -50mm) in a horizontal pipeline.
I wanted to calculate the stresses enforced by the displacement of the compensator.
I'll add an extra force on the end of the compensator to reflect the reaction force of the compensator.

Be careful with imposing displacments on the compensator. Once you apply displacments, the remainder of the piping can't move it (the compensator), so you lose any interaction that may take place.

I don't add displacements, but a simple downward vector in the piping element input.
In practice the pipes before and after the compensator will both be supported with Y restraints, so the compensator will have no (compensating) function anymore.