XANSYS Message: 18496 [Go back to message list]
[bookmark on del.icio.us]
No rating yet
Subject: Re: how many substeps?
Author: Martin Liddle
Date: 2000-11-30 10:19:00In article ,
Bryce Katherine Ann writes
>I am doing a nonlinear magnetic analysis, and am getting a warning that
>says "using 1 iteration per substep may result in unconvergent
>solution....".
>I don't have much experience with this, and am just wondering how many
>substep iterations you would recommend.
Sorry for late reply. Still catching up on an enormous back log of news
group postings. Is this a vector potential or a scalar potential
solution? If its scalar potential is it reduced, difference or general?
If its the first load step of difference or general then its probably OK
otherwise its probably wrong. The answer to the question how many
equilibrium iterations (not substeps as per your question) is probably
as many as it takes to converge to your desired accuracy level. 10
might be typical but I have problems where whatever I try I seem to end
up with 30 to 40 which everybody says is too much but I can't find a way
to reduce them.
Martin Liddle, Tynemouth Computer Services, 27 Garforth Close,
Cramlington, Northumberland, England, NE23 6EW.
Phone: 01670-712624. Fax: 01670-717324.