transient Thermal Study 0 °K after nearly half of duration

Hey guys,

i have a transient thermal model to calculate.

The rough principle looks like in this picture:

http://ww3.cad.de/foren/ubb/uploads/Evox/Schema.jpg

On each side is a mounting and a peltier element attached to the cube in the center, which has a drilling in the middle for some fluid.

Now i have 2 variations of this model, with a duration of 540 seconds.

1. constant power of 10W per peltier -> good results till 237sec,after that everything is just at 0°K

2. pulsed power of 20W per peltier -> good results till 257sec,after that everything is just at 0°K

I've checked everything.

- Materials are okay..no one is using any graphs for the thermal specifications

- Thermal Loads:

  1. is constant at 10W per each inner surface of the peltier, so 2x 10W

  2. is the same but with 20W changing the side, every 2 seconds over the whole duration of 540sec

- thermal resistance between contact surfaces are all fine

- RAM,CPU,HDD: there are more ressources than i need

I can't figure out where the problem is....i tryed it on 2 different computers, several times...every time i get the same results.

It jumps from one step to the next from good values to 0°K, which can definitely not be possible.

Ther must be sth. wrong with the calculation process, i think.

oh and the pc is calculatin all 540 steps! the simulation doesen't crash, or sth like that...

SolidworksSimulation