Learning Forum

[Sung ki Son]

In 2 way coupling how we can identify the correct value of coupling iteration? Does it mean that increasing the number increases the accuracy?

If we suppose that we use Icepak Q3D 2 way coupling (transient) simulation with iteration number of 6, then our expectation is that for each time steps Ansys shall calculate the EMLoss value in Q3D sends to Icepak then Icepak calculate the temperature and this shall be done for 6 times for one time steps till the iteration reaches the conversion value.

 

However, based on the screen shots and analysis, it appears that ANSYS Q3D is calculating the EMLoss value first and then initiating the iteration process in Icepak based on this calculated EMLoss value. While this approach may be suitable for steady state simulations, it may not be appropriate for transition solution simulations, where coupling iterations need to be performed for each time step.

Also, in monitor graph also it seems only the last temperature value is updating its value.

Also, in this particular case, the coupling iteration is set at 6 and the transition time is 0 to 1 time steps with an interval of 0.25. In case of real time two way coupling, 6 coupling iterations should be performed for each time step, which means a total of 30 coupling iterations for 5 time steps, and therefore 30 completion messages shall appear on the message manager of AEDT, not only 6 times.

 

I wonder is there any other specific way to do the 2 way coupling simulation for transition mode?

 

[Pdev]

Hi Sung Ki Son, 

EM-Icepak workflow does not support transient in a bidrection coupling. It is only for steady state. For each iterations you can start the thermal solve from the previous iteration results that speed up the process of convergence. 

Thanks,
Pdev

[Sung ki Son]

Dear Pdev,

Thank you for your response, I didn't get your instruction properly, would you please to explain more in detail?

Also, is the Icepak transient joule heating reliable? I meant does it consider the thermal modifer to in simulation?

I look forward to hearing from you.