Hello Mathias,
What type of analysis system are you using? Presumably coupled field transient?
Also, what version of Workbench are you using?
We sometimes see non-physical "thermal undershoot" or "spurious oscillations" in the transient thermal response when heat is applied very suddenly. Calculated temperatures in these first order systems can briefly drop below those that are physically possible.
But the behavior you are reporting does not appear to be related to this. It appears that the device temperature is dropping below the 60 C environment very gradually... some 1300 seconds into the transient. It is not immediately obvious to me why this is happening.
It might be helpful to compare the net internal energy (heat in Joules) added over time that is attributable to Joule heat to the net heat leaving the system due to your convection boundary condition(s) and/or any others that you may have. The thermal power in Watts leaving the system is easily obtained with post processing features that are natively supported in the Mechanical GUI. This would need to be integrated over time to get net heat removed from the system.
It is less straightforward to get the net heat added to the system by Joule heat. There may be an easier way to do this, but for now I can only think of using an APDL command object under the Solution branch, something along the lines of the following (untested):
finish
/clear
resume
esel,s,ename,,226,227 ! SELECT COUPLED FIELD ELEMENTS
*get,nelems,elem,,count ! nelems = # OF SELECTED ELEMENTS
/post26
elm=0 ! INITIALIZE ELEMENT ID#
*do,i,1,nelems ! LOOP OVER SELECTED ELEMENTS
elm=elnext(elm) ! elm NOW EQUALS NEXT HIGHEST SELECTED ELEMENT ABOVE elm
esol,3,elm,,jheat ! READ JOULE HEAT PER VOLUME FOR ELEMENT elm INTO VARIABLE #3
*get,volu_i,elem,elm,volu ! volu_i = ELEMENT elm's VOLUME
add,2,2,3,,,,,,volu_i ! ADD VARIABLE #2 AND volu_i*#3 (NET JOULE HEAT FOR ALL ELEMENTS ACCUMULATE IN VARIABLE #2)
*enddo ! END DO LOOP
int1,4,2,1,,Qin ! VARIABLE #4 = INTEGRAL OF VARIABLE #2 WITH RESPECT TO TIME
/show,png
plvar,4 ! PRODUCE AN XY PLOT OF VARIABLE #4
/show,close
This requires that you set "Save MAPDL db" = "Yes" in Details of Analysis Settings prior to solving.
--Bill
You are navigating away from the AIS Discovery experience