Community > Discussion Forum > time dependent chds simulation doesn't converge

Discussion Forum

time dependent chds simulation doesn't converge

Topics: Geometry, Solving, Microfluidics, 4.2a

Thread index  |  Previous thread  |  Next thread  |  Start a new discussion

RSS FeedRSS feed   |   Email notificationsTurn on email notifications   |   1 Reply   Last post: January 25, 2012 7:52am UTC
Mikiyas Tsegaye

Mikiyas Tsegaye

January 24, 2012 9:00pm UTC

time dependent chds simulation doesn't converge

hey all,

I've done a very simple geometry swap for the electroosmostic micro-mixer example. I deleted the original geometry from the model that comsol distributes and put in my own simple 2D geometry. I then reassigned all the boundary conditions, and hit run.

It didn't .

It crashes saying that the last time step didn't converge in the time-dependent solver. I tried changing the solver (MUMPS, SPOOLES,...), but that didn't help either. Everything is as it was originally, except for geometry specific reassignments. What am I doing wrong?

thanks

Mikiyas

Attachments:   mixer-cosntr.mph  

Reply  |  Reply with Quote  |  Send private message  |  Report Abuse

1 votes
Ivar Kjelberg

Ivar Kjelberg
Moderator

January 25, 2012 7:52am UTC in response to Mikiyas Tsegaye

Re: time dependent chds simulation doesn't converge

Hi

I believe you have still a little homework to do ;)
You are not using the default COMSOL physics boundaries and you have not used boundary layer meshes so you do not resolve the flow along the walls correctly. And have you checked your Reynolds number and Reynolds Cell number ? check the doc on mesh size related to physics and to time dependent solutions

then I suppose you want the stationary to not solve for the CHDS and leave that for the time series or ? (not the most important) when you do not solve for all physics, be sure you define correctly the values to use for the dependent variables "not solved for"

And when you mix several solvers, do not forget that the last Data set is the first Store solution and the first Data set is the final result, this is easy to get wrong (probably not your case anyhow)

Interesting model ;)

--
Good luck
Ivar

Reply  |  Reply with Quote  |  Send private message  |  Report Abuse


Rules and guidelines