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.
what is the reason for the error of conditioned preconditioner. Increase factor in error estimate to 532.6030621254678
Posted Mar 29, 2011, 2:30 a.m. EDT 1 Reply
Please login with a confirmed email address before reporting spam
Hi guys,
I am facing difficulties in solving my model, always comes am error by showing a Warning: Ill-conditioned preconditioner. Increase factor in error estimate to 532.6030621254678. What should i do to over come this error? Another thing, when mention less value or current (A/m²), this warning doesn't come but when i give a higher value of current (A/m²), this message is coming. Please give me some idea.
Regards,
Rashed
I am facing difficulties in solving my model, always comes am error by showing a Warning: Ill-conditioned preconditioner. Increase factor in error estimate to 532.6030621254678. What should i do to over come this error? Another thing, when mention less value or current (A/m²), this warning doesn't come but when i give a higher value of current (A/m²), this message is coming. Please give me some idea.
Regards,
Rashed
1 Reply Last Post Mar 29, 2011, 3:18 p.m. EDT
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
It can (but not necesarily) come from a scaling issue, your dependent variables are too different and needsome prescaling that COMSOL is not capable of detecting (although mostly it manages to scale "automatically"
try to (in V4) to set the solver log to detailed. Often then you get the scaling values and compare them to what you expect.
Again this could be an issue, but there are other reasons possible
--
Good luck
Ivar
It can (but not necesarily) come from a scaling issue, your dependent variables are too different and needsome prescaling that COMSOL is not capable of detecting (although mostly it manages to scale "automatically"
try to (in V4) to set the solver log to detailed. Often then you get the scaling values and compare them to what you expect.
Again this could be an issue, but there are other reasons possible
--
Good luck
Ivar