I have a task to analyse steam (427degC, 2000kPa) piping from the battery limit (BL) at the bottom of a flare stack to its tip.
I am in possession of the OBL Caesar model on the other side of the BL. OBL model has +/-30 analysed cases including all sorts of structural displacements and wind loads.
The first and only anchor is inside my scope , attached to the stack, which means includes wind displacements which are significant. OBL model does not have anchors anywhere close to the BL, as it would create overstress in their system :rolleyes: . Their piping is already fabricated and mostly installed and I have to find a way to live with what ever is coming from their side.
My problem is to do analysis correctly which means to consider all possible operating scenarios.
In OBL analysis they have already used D1-D8, and I need 6 displacements in my portion of piping. These displacements come from movements of the stack due to wind (again separate analysis done by a third party).
What is alternative to join these two analysis correctly, and check all combinations of loadings.
All ideas will be appreciated.
Regards

First of all pick your boundary.
Is it the tie in at the battery limit or is it the anchor?

If its the anchor(pref) then model upto same.
don't forget to review the pipe at the anchor for the effects of the worst combinations of loads from both sides of the anchor.


If it is the battery limit, then review the piping between the anchor and the battery limit,
if it is simple enough, then model upto the battery limits and use if possible use displacements.
if it is complex,and you cannot use displacements then model the whole thing
(this can be done using the merge commands)

this might sound hard work but i find it can
be usefull in a sticky corner.

Another idea, might be to find a position on the existing analysis where the piping does not move.
at this point,position an anchor and treat the rest of the piping as 'yours'

Or......... Model the stack............

you might want to give out to the other analysist for not properly looking over the fence.

BL is just a flange, there is no anchors anywhere near. That is the problem: there is no fixed point neither in my system nor in theirs.
In the OBL model even other supports (rests, guides, stops) are moving for whatever reasons in number of the cases considered.
My 80m tall stack is moving due to wind and thermal. I have modeled the stack for the thermal displacements, but instead of applying the wind load I have used the displacements supplied by the specialist. Reason for this is that I cannot get the same results of movements due to wind to match the results received from the "stack guru".
I have tried to extend the OBL piping to my first stack anchor (modeled as fixed point) and then take the loads at the BL and input them in my IBL model at the "free" BL flange. This might be good enough to check the piping on my side, but I do not believe that this is giving realistic results with regards to OBL, as the impact of the horizontal movements of the stack is omitted.

To resolve this problem the first price would be if there is a way to trick Caesar and specify more then 9 displacements at one node

The second price would be to input a load instead of displacement, but how to calculate this load? Is that the difference in load between fixed anchor and displaced anchor load? That would mean a lot of work…

I hope this will deserve a consideration by Coade to include for more displacements in future upgrades of Caesar.

Sorry, I should have said:
To resolve this problem the first price would be if there is a way to trick Caesar and specify more then 9 displacements IN ONE MODEL (STRESS FILE)

I can't really think of a way to trick the software here.

If you have any leftover thermal sets, you can use those to set a coefficient of expansion on short rigid elements at the terminations that need more displacement sets. The rotations would be hard to set and the whole exercise would be hard to justify labor-wise.

Unless you are doing cummulative damage (fatigue) calcs (where all load cases are evaluated together) I don't see why two separate jobs wouldn't do the trick. Use Save As and then reset the displacement sets. A little more juggling on your end but it's straight forward.