Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Extension parametric - step size computation

Please login with a confirmed email address before reporting spam

I am doing using the non-isothermal flow physics to calculate convective heat transfer in a power electronics package.

In the log, I can read after almost each parametric step:
"Breakdown in parameter step size computation. Parameter step size unmodified."

Besides, the computation stops before it reaches the last parameter step (about halfway). I am using a extension-continuation parametric study in order to increase progressively the heat dissipated.

What does this error mean? How can I solve this problem and get the computation to work for all steps?

Thank you for your help.

2 Replies Last Post 19.07.2012, 01:32 GMT-4

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 18.07.2012, 19:01 GMT-4
Hi Antoine.

I'm also doing a nonisothermal model and having the same error. I believe what it means is that the amount the parametric solver changes per calculation is held constant. So, for instance, if I am screening from an inlet pressure parameter of:

P = [10,100,1000,10000] Pa

Comsol initially calculates solutions by stepping the inlet pressure by an intermediate value of the first two steps, lets say 50 Pa. With this error, Comsol continues to step by 50 Pa, instead of increasing its step size, which may be more efficient for latter calculations such as going from 1 kPa to 10 kPa. So, it may just take the simulation longer but it will eventually reach the answer.

Does that make sense? I'm going to look at some parametric solver parameters and see what I come up with.

Ramsey
Hi Antoine. I'm also doing a nonisothermal model and having the same error. I believe what it means is that the amount the parametric solver changes per calculation is held constant. So, for instance, if I am screening from an inlet pressure parameter of: P = [10,100,1000,10000] Pa Comsol initially calculates solutions by stepping the inlet pressure by an intermediate value of the first two steps, lets say 50 Pa. With this error, Comsol continues to step by 50 Pa, instead of increasing its step size, which may be more efficient for latter calculations such as going from 1 kPa to 10 kPa. So, it may just take the simulation longer but it will eventually reach the answer. Does that make sense? I'm going to look at some parametric solver parameters and see what I come up with. Ramsey

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 19.07.2012, 01:32 GMT-4
Hi
The solver tries to use intermediate stepping dependent on the results and the Jacobian it derives from the model, if you give large steps for a non linear problem, COMSOL might decide to do many intermediate steps. This behaviour can be adapted and changed in the solver sub node settings. But it is not always obvious what to change ;)

--
Good luck
Ivar
Hi The solver tries to use intermediate stepping dependent on the results and the Jacobian it derives from the model, if you give large steps for a non linear problem, COMSOL might decide to do many intermediate steps. This behaviour can be adapted and changed in the solver sub node settings. But it is not always obvious what to change ;) -- Good luck Ivar

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.