Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Oct 1, 2012, 10:14 a.m. EDT
Hello,
Perhaps the problem is the discontinuity in initial conditions in T in one of the boundaries. Any (in space or time) discontinuity yields convergence problems easily. You could try a ramp function for the boundary conditon, like:
293.15[K]+300[K]*rm1(t/1[s]), where rm(t) has been defined as a ramp function, rising smoothly from 0 to 2 seconds, for instance.
A finer mesh in the 'problematic' boundary is also advised.
Bye,
Jesus.
Hello,
Perhaps the problem is the discontinuity in initial conditions in T in one of the boundaries. Any (in space or time) discontinuity yields convergence problems easily. You could try a ramp function for the boundary conditon, like:
293.15[K]+300[K]*rm1(t/1[s]), where rm(t) has been defined as a ramp function, rising smoothly from 0 to 2 seconds, for instance.
A finer mesh in the 'problematic' boundary is also advised.
Bye,
Jesus.
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Oct 2, 2012, 12:18 a.m. EDT
Hi Ishant,
The problem is related to time steps and mesh size.
Please check this thread:
www.comsol.com/community/forums/general/thread/16808#p46163
--
Tolga
Hi Ishant,
The problem is related to time steps and mesh size.
Please check this thread: http://www.comsol.com/community/forums/general/thread/16808#p46163
--
Tolga
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Oct 2, 2012, 1:06 a.m. EDT
The finer meshing at the bounder layer works well.
I will request you to please give some detail about this ramp function.
Hello,
Perhaps the problem is the discontinuity in initial conditions in T in one of the boundaries. Any (in space or time) discontinuity yields convergence problems easily. You could try a ramp function for the boundary conditon, like:
293.15[K]+300[K]*rm1(t/1[s]), where rm(t) has been defined as a ramp function, rising smoothly from 0 to 2 seconds, for instance.
A finer mesh in the 'problematic' boundary is also advised.
Bye,
Jesus.
--
ishant.jain@tatasteel.com
The finer meshing at the bounder layer works well.
I will request you to please give some detail about this ramp function.
[QUOTE]
Hello,
Perhaps the problem is the discontinuity in initial conditions in T in one of the boundaries. Any (in space or time) discontinuity yields convergence problems easily. You could try a ramp function for the boundary conditon, like:
293.15[K]+300[K]*rm1(t/1[s]), where rm(t) has been defined as a ramp function, rising smoothly from 0 to 2 seconds, for instance.
A finer mesh in the 'problematic' boundary is also advised.
Bye,
Jesus.
[/QUOTE]
--
ishant.jain@tatasteel.com