Discussion Forum

Hi

well I would not say I understand everything either, but wen I read p347 CMPrefGuide v4.3:

Error Estimates for the Time-Dependent Solver:
... "A solution on the coarse base mesh is computed in the next time interval, and the error indicator is evaluated at the points specified in the Sample points field. In this way a new adapted mesh appropriate for the next time interval can be generated. The sample points must be specified as a number between 0 and 1 because they are interpreted as
being relative to the time interval under consideration. Entering a scalar value of 0.5 means that the error indicator is evaluated at the midpoint of the interval. The default value is range(0.0,0.1,1.0), which gives 11 sample points from 0 to 1." ...

The "Sample points" are to be defined in the time interval, when the solver will evaluate the mesh and to decide if it need to refine before taking next full time step step. Hence you define a number or a series of number between 0 and 1 (times the time step interval) and the solver sill evaluate the results at these time sub-steps based on current mesh and if L2 norm is too bad it mihght decide to refine the mesh before solving for the next step.

Sound reasonable to me, still need to see how it behaves, for that one would need to study further the log file, and perhaps put it in "detailed log" mode. Now I'm no user ofthe refinement feature, mainly because in3.3 to 3.5 I was not very happy with the convergence, it refined everywhere, while I was used to refinement in critcal regions (for me it was high stress rgions in Solid structural models). But in the latest releases the automatic mesh refinement and the L2 norm set up has been changed, hence it would be worth for me to try it again ;)

Hope it helps on the way

--
Good luck
Ivar

Error Estimates for the Time-Dependent Solver:
... "A solution on the coarse base mesh is computed in the next time interval, and the error indicator is evaluated at the points specified in the Sample points field. In this way a new adapted mesh appropriate for the next time interval can be generated. The sample points must be specified as a number between 0 and 1 because they are interpreted as
being relative to the time interval under consideration. Entering a scalar value of 0.5 means that the error indicator is evaluated at the midpoint of the interval. The default value is range(0.0,0.1,1.0), which gives 11 sample points from 0 to 1."

The last three sentences were missing in the 4.2a reference. I'm glad they cleared that up because it also took me a while to figure out that "sample points" are points in time, not space.

In my models, I often enter "0" in the sample point field, so as to avoid all time-stepping on the coarse mesh.

Hi

still perhaps 0.5 is better to give it a kick in to the future ;)

--
Good luck
Ivar

Guys, it is a little bit clearer, thank you! But, when I put there whatever (0, 0.5 or the whole interval range), even don't fill it at all!!, the result is still the same, I mean the time steps which COMSOL takes.
Is there still some mysthery for me how to use it correctly and efficiently.
K.