Untied Expansion Joint Modeling

Good morning,

I'm currently modeling the single rubber expansion joint below. There is a pump anchor on the left and a tank nozzle on the right. I'm having trouble with the correct modeling of the limit rods.



If the limit rod restraint is placed in the +X direction as shown above with a cnode to that flange node, the joint is compressed and the nozzle loads are as expected. But doesn't this restraint not allow for this point to move in the -X direction? And this is the opposite of where I would expect it to go.

My thought was that the limit rod restraint should be in the -X, as shown below. This is because the limit rod is only to stop expansion, not compression.



But placing the restraint in the -X results +300,000lb of axial force at the anchor, nozzle & rigid restraint. What am I missing here? Are these forces due to the pressure thrust release during compression of the expansion joint? If so, why is the rigid/limit rod taking these forces when in reality it will not be in effect due to the joint compressing?

I'm also thinking maybe I am misunderstanding how the CNode is acting in this case.

Any opinions?

Thank you.

The best solution is to not even model the limit rods. They are intended to only activate in the case of a main anchor failure which is NOT what your analyzing. If you must model them, then the restraint sign should be in the direction of the allowed movement, in your case -X. Pressure thrust (P x A) would be applicable. I'm not sure why you didn't get a huge load when you modeled it as +X.