flexMen,

I think we're getting closer. Let's say that your occasional load case looks something like this:

W+P1+SEISMIC1

Now, unless you've got lifted supports that need to be removed, you are going to get the same calculated values of stress no matter what T you use in an operating case. You agree that the load needs to be checked against the lower Sh value due to the higher operating temperature.

So you then ask why not use the higher value of Sh when you run, possibly in a different file, with a lower operating temperature.

My question is - why run this case at all? Since you have to run this case against the lower Sh value, running it again against a higher Sh value adds nothing to your analysis and gives you no useful information.