residual norm and unconverged equation error

59 views (last 30 days)
Andreas
Andreas on 29 Oct 2012
Answered: PENG SHI on 21 Dec 2018
hi. i use a simulink and simscape model to buildd an aplifier for audio use. class d.
when i start simulation I get this error:
_Steady state solve failed to converge. ...Nonlinear solver: failed to converge, residual norm too large. ......Here is the set of components with unconverged equations:
'mosfet_driver/N-Channel MOSFET' 'mosfet_driver/gate driver 2/NPN Bipolar Transistor'_
the strange about this error is that i did not change actually anything on my model since the previous time, that the simulation worked! so strange and wuote annoying I'd say. I suppose it is some kind of numerical linear algebra error inside the simulink computation engine.. but i cannot tell what the problem exactly is! so any help!?
thanx. andreas

Answers (2)

Guy Rouleau
Guy Rouleau on 29 Oct 2012
Here are a few tips to make your model more robust to this error:
- Try increasing the Consistency tolerance in the Solver Configuration block. The default is 1e-9. Try something like 1e-6.
- Some components in your circuit probably have parameters like snubbers or "Open resistance" or "Closed Conductance". Those are there for numerical reasons. Making them less ideal should help.
  2 Comments
Andreas
Andreas on 31 Oct 2012
reduced the solver tolerence but did not help. actually it showed another error when setting it too low (e.g. 1e-3) as expected.
-when talking about resistence and conductunce do you mean those of a capacitor (which has parallel conductance)? also the ohmic resistence inside a mosfet of npn has to play some role i suppose. note that for my n channel power mosfet i use irf 6665 and i cannot find the correct ohmic resistance info inside the datasheet...
Andreas
Andreas on 31 Oct 2012
well... it seems i had made a mistake. i had set in the n channel mosfet ids=20 uA instead of 19A (due to a misunderstanding of the datasheet). it performs the simulation now( 5 minutes or more for 0.004seconds for the whole system on a i3 2.2Ghz system!!!)
i wonder if the error i got sufficiently described the (obviously) extremely low value in the ids current, which caused the error to show. i believe that it could be more implicit on what to do.
anyway Mr.Rouleau thank you for helping me.. again!
Andreas

Sign in to comment.


PENG SHI
PENG SHI on 21 Dec 2018
I have encountered this problem as well. It is resolved by reducing min time step in my case.

Products

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!