Search found 4828 matches

by raback
23 Feb 2012, 13:11
Forum: ElmerGUI
Topic: Help with my acoustic anylize
Replies: 6
Views: 5102

Re: Help with my acoustic anylize

Or perhaps the absolute value: |z|=sqrt(Re(z)^2+Im(z)^2). Hard to say without knowing the objective. -Peter
by raback
22 Feb 2012, 22:38
Forum: ElmerSolver
Topic: Reactive transport problems
Replies: 2
Views: 2767

Re: Reactive transport problems

Hi Elmer has an advection-diffusion solver with a namespace. You can have multiple instances of the same solver for different variables. Of course with, say more than 3 variables, this becomes cumbersome. Also there is no mechanism to easily account for the binary diffusion. To have a proper reactiv...
by raback
22 Feb 2012, 20:40
Forum: ElmerGUI
Topic: Help with my acoustic anylize
Replies: 6
Views: 5102

Re: Help with my acoustic anylize

Hi

The difference is sqrt(-1) ;-) So the 1st component is the real part and the 2nd component is the imaginary part.

-Peter
by raback
22 Feb 2012, 20:32
Forum: ElmerSolver
Topic: Error using Hypre
Replies: 10
Views: 7627

Re: Error using Hypre

Hi

Maybe you could add also just the sif. The estimated upload time is 12 minutes.

Generally Hypre is one option in the parallel toolbox. Most often it is not needed but there are cases when it provides the best strategy.

-Peter
by raback
22 Feb 2012, 20:29
Forum: ElmerSolver
Topic: ParticleFalling
Replies: 12
Views: 8965

Re: ParticleFalling

Hi Auray

Well, I tweaked a little the output. May be a compiler-dependent issue in rev. 5598. Haven't much tested this version after the vtu format was implemented.

-Peter
by raback
22 Feb 2012, 19:13
Forum: ElmerSolver
Topic: Saveline ordering error
Replies: 7
Views: 5222

Re: Saveline ordering error

Hi

Copy SaveData.so (or SaveData.dll in Windows) to your working directory with a different name, say MySaveData. Then in the Procedure statement refer to this module.

-Peter
by raback
22 Feb 2012, 19:12
Forum: ElmerSolver
Topic: AdvectionDiffusionSolver - bias of overall mass
Replies: 3
Views: 2394

Re: AdvectionDiffusionSolver - bias of overall mass

Hi

Sorry, that should be fixed in rev. 5597. Thanx for reporting!

-Peter
by raback
22 Feb 2012, 16:54
Forum: ElmerSolver
Topic: AdvectionDiffusionSolver - bias of overall mass
Replies: 3
Views: 2394

Re: AdvectionDiffusionSolver - bias of overall mass

Hi

Try to switch the "Element Degree" to 2. Also you might set "BDF Order" to 2. Of course decreasing the space/time resolution should have a similar effect, but using second order discretization might be more cost effective.

-Peter
by raback
22 Feb 2012, 10:48
Forum: ElmerSolver
Topic: Multiple Domains with different physics: multimesh?
Replies: 3
Views: 3619

Re: Multiple Domains with different physics: multimesh?

Hi One thing is that you cannot have "Temperature" being solved by two different solvers. You should name them differently, say "TempLoc" and "TempGlob". Then if you would solve fist for TempGlob you could set as BCs TempLoc = Equals TempGlob Now this gets you the hiera...
by raback
21 Feb 2012, 23:47
Forum: ElmerSolver
Topic: Strain in
Replies: 5
Views: 5148

Re: Strain in

Hi

Look at the recent consistency test "StrainComputation03". That provides with an example.

-Peter