Thanks for your replies. Just to clarify, the pipe is horizontal and the growth of the pipe in the +Z direction over many temperature cycles has resulted in the saddle support slipping off the actual support column so that it now acts as a limit stop in the -Z direction when the system cools down. I have modelled the system with a D1 displacement of 250mm at this point in its ambient state and then calculated the expansion during the next cycle and input this as D2 at this point so that my loads cases are as follows:
OPE W+P1+D1+D2+T1
SUS W+D1+P1
EXP L1-L2

Is this an acceptable method of analysing this problem or is there a better way I should be doing this?