I have an analysis that includes pipe supported in two independent structures. These structures move relative to each other in some loading cases (eg seismic). I plan to impose deflections at the supports in one of the structures to model the relative displacements.

Is there a method where I can enter the displacements only once, or do I need to input these as displacements at all supports?

If you are doing a dynamic calc, the independent support displacements are input with the Spectrum Load Cases.

There is an example in the Caesar manual. (Example 4 Cryism ?)

In static analysis we can model support displacements by putting a cnode at restraint node and defining displacements for cnode .But these displacements should be treated properly while making analysis cases. These are not thermal displacements.
There is one more related query about modeling support displacements in static analysis using Cnodes at support points. Take a case where we are modeling seismic sway induced displacements or ship movement induced displacements .If there is a hager ( not can type) , either rod hanger or spring hanger do we need to model the support induced displacements at the hanger ? Some consultancies are modeling , while some or not. In my opinion if it is a can typr hanger displacement needs to be modeled at cnode , but not for other hangers. Anybody can help me ?
Regards,
Godbole

Regarding your question on giving support displacements for hanger supports, my opinion is like this:

There is no need for doing so for either type of hangers as if we attach a structural steel model with the pipe model, we usually find ( at least I found so) the differential movement effect is really not there ( both move similar to rigid body motion ) as it would be if the models are considered separately. Conventional analysis by feeding support displacements make the analysis complex and increases cost.

Anindya Bhattacharya

For support displacements, cnodes are to be used. However, we have to specify displacements only in the direction of the restraint at Cnode point. For example, if the support is x restraint, give displacement in x direction only leaving other displacments blank. Similarly, if the support is spring hanger, give displacement only for y direction at c node.
I have checked and found that even if uou give displacements at other directions, caesar ignores
those displacements and takes displacenment only in restraint directions.

CNODES create an association of degrees of freedom between two nodes. So, if you define a restraint at node 200, with a CNODE of 1200, in the "X" direction, with a rigid stiffness, then:

• Both 200 and 1200 will have exactly the same displacement in the "X" direction.
• The other 5 degrees of freedom between these two nodes are not associated. Therefore applying "Y" or "Z" displacements to node 1200 has no effect on the piping model.

Thanks for the responses
I have now also realised that by switching off the “Connect though Cnodes” OFF (in the configuration setup) that I can Cnode several supports to the same node. This is saves a lot of repetitive input.
In a two level structure, I have used nodes 9010 and 9020 to define the displacements at the two elevations. These nodes do not exist in my model. I then Cnode all supports at level 1 to node 9010 and Cnode all supports at level 2 to node 9020.