You can stop your simulation after a certain amount of time, adjust the boundary conditions, and continue using the previous results. So what I would do is:
1. Enable your heat boundary condition and run the simulation for a set amount of time.
2. Create a copy of the simulation project along with the results (as a backup)
3. Turn off the heat boundary condition and continue the simulation for a certain amount of time.
I didn't know that we could stop and change the parameters during a transient study. Interesting. I will try.
In the above case, I run the first transient study till a certain point, for example till 10 seconds, 10 steps of 1 second.
Then duplicate (right mouse button on the below tabs->Duplicate) the study.
Define in the properties, the initial temperature to be imported from the previous study, and at step 10 (seconds).
In that second study, in the Thermal Loads, suppress the heat boundary condition.
You can also duplicate the study, and import the initial temperature from the 1st study at other steps earlier, for example at step 9, or step 8, and see the different results.
Hope this helps,
These suggestions do seem to work.
Of course the heat staking head (hot part) has to be suppressed in its entirety to allow the (formerly hot) contact surface to be available for convection cooling, so my concern is that the mesh would be different in the second simulation and therefore the initial results would not be carried over correctly. Strangely though, it does seem to work, and the temperature and internal pressure of the enclosed plastic part DO transfer nicely over to the second part of the simulation without error.