No change in a time dependend model. Where is the fault?

I tried a very simple model:
= 3-D cylinder from
= copper (from the builtin material library) with an
= initial temperature of 293.15 K
= insulated at all surfaces, with one
= heated bottom surface at 1W/m^2 (overwriting the all-sided insulation for this surface.)

I used the study
= time-dependend
= time span= 0 to 100 s in 1 s time steps

I visualized the result at a surface vertically reaching from the heated bottom of the cylinder to the top surface.

I. e. there is heat input from one side and there are non-zero thermal parameters for the copper.
But surprisingly, no change in time can be found in the model result ! As You see in the attached file, after 100 s the whole cylinder is still at 293.15K.

Any idea, what I might have overlooked?