I am analysing Air cooler inlet line (350F temp) of 42" NB with 24 nos nozzle(12 tube bundle). each cooler header( inlet/outlet, even pass) has supported at 4 points of it's 4 corners with slide pad.
Vendor allowed gap in transvers direction (X) is 2.5" both sides ane
longitudinal movement (Z) of tube bundle is limited to 1/8" by the support at header box.

I am supposed to deliver the final report of this system and i found 2 different approach for analysis.
please suggest me the pactrical/justified approach...

Approach-1

Model the piping and header box (rigid) not tube bundle. End supports of each header is Y,X with 2.5" gap, Z with 1/8" Gap,RZ and RX with 0.1 friction coeff.Ry is free.
1/4 th of the weight of tube bundile has been imposed on the header box near to the support.

System is safe fron nozzle load and moovement of cooler also with the limit.This is analysed at the prelemenery stage and load to structure has been forwarded.

Approach-2

I model (in addition) the tube bundle and rare header with their weight and place 4 supports at proper location per vendor drg. with fric. coeff 0.1 with same supporting as approach-1 (Z is free at the rare end header).
Now the transverse moovement of tube bundle is controled by the frictional force at 4 suooprts and huge moment is accumulating in individual nozzle.
To reduce load we need to modify the layout ( need more flexibility in between piping header to nozzle) , but our design team is not ready to change the IFC lines of that big size. My Stress supervisor also agree with me but, he also telling that it will not effect the system in pactrical. but he did not able to explain how nozzle load will reduce in the field.

I want to know should i be strong on my point of view ?
or let the system go for fabrication which not giving staisfactory result to me? need urjent help.

Approach-1 is relevant. Check again actual boundary condition for X, Y, Z, RX, RY, RZ if it reflects the real situation at the end of header box (see the related cooler drawing). You did not give the information include/not include weight of rigid to us here. I think zero weight is relevant for the header box since it well supported.

Dear,
Samsul

i am supposed to calculate piping load on structure at header box support for structural beam sizing. If i don't consider weight of the tube bundle and friction at the support then how load at support point can be calculated? support type at the cooler is the realistic simulation ( ie bundle can rotated aginst Y axis because of the 2.5 " gap in transvers direction.

Piping expansion is trying to push the cooler in transverse direction at the front end (inlet/outlet)and this movement is restricted by friction and the support ( if moovement exceed 2.5"). But back end of cooler want to stay at the same location as there is no piping force. it only expand in longitudinal direction. So here moment is comming.

In approach-1 there is no back end ...so no excessive moment.
Is it a relevant solution?

For checking load at nozzle connection, Approach-1 is relevant. For me I will not use App-2 to have load on structure. Just give the load on nozzle as boundary condition to structural dicipline for accuracy (the way they consider the weight of the cooler to structure in their model is also more accurate).