Search found 46 matches

by fgillet
28 Aug 2014, 09:36
Forum: Bug reports
Topic: Linear System Direct Method automatically set to "umfpack"
Replies: 1
Views: 3734

Linear System Direct Method automatically set to "umfpack"

Hi, in the CheckSolverOptions subroutine,in MainsUtils.src, the direct method solver is automatically set to "umfpack" in sequential: Lines 86-90 IF ( str == 'mumps' ) & CALL Warn( 'CheckSolverOptions', 'Currently no serial interface' // & ' to the MUMPS solver implemented, trying ...
by fgillet
18 Nov 2013, 16:29
Forum: Bug reports
Topic: FUNCTION MeshExtrude
Replies: 0
Views: 2468

FUNCTION MeshExtrude

Hi, There is a bug in the numbering of the base line parent elements in the function MeshExtrude for the cases where the BaseLine is preserved. For the Mesh_out elements corresponding to the 2D baseLine, the Mesh_out parent index is "Mesh_in % NumberOfBulkElements*(in_levels+1)+(in_levels+2)*Me...
by fgillet
17 Jul 2013, 16:56
Forum: Elmer/Ice
Topic: How to use AIFlow Solver ?
Replies: 1
Views: 5081

Re: How to use AIFlow Solver ?

Hi, sorry, the AIFlow solver use the residual free bubbles method described in Baiocchi et al. (1993). Bubbles are not implemented for prismatic elements (706). So for the moment you can not use the AIFlow solver with 706 elements. The fabric field is initialised through the values of the 2nd order ...
by fgillet
29 May 2013, 14:02
Forum: Elmer/Ice
Topic: Problem with solving temperature after inverse method
Replies: 15
Views: 21726

Re: Problem with solving temperature after inverse method

Hi,

I didn't know about this keyword; Thanks Rupert
If it works as expected that would be an elegant solution.

cheers

Fab
by fgillet
28 May 2013, 20:30
Forum: Elmer/Ice
Topic: Problem with solving temperature after inverse method
Replies: 15
Views: 21726

Re: Problem with solving temperature after inverse method

Hi, I think there is no problem with your relaxation. When starting from a given observed topography, the incompressibility equation usually leads to very high and unphysical ds/dt (i.e. vertical mesh velocity). This is why relaxation is needed, and at the beginning there is very high mesh velocitie...
by fgillet
12 Mar 2013, 12:50
Forum: Bug reports
Topic: StructuredProjectToPlane.src
Replies: 1
Views: 2888

StructuredProjectToPlane.src

Hi,

In the StructuredProjectToPlane.src solver,
the variable "nsize" should have the save attribute.

best regards

Fabien
by fgillet
19 Oct 2011, 13:49
Forum: ElmerSolver
Topic: Stokes and 2nd order elements
Replies: 2
Views: 2205

Re: Stokes and 2nd order elements

Hi Juha,

Thanks, I will try the flux Solver;

I forget to say that p2/p1 does the same as Bubbles.

Cheers

Fabien
by fgillet
19 Oct 2011, 10:47
Forum: ElmerSolver
Topic: Stokes and 2nd order elements
Replies: 2
Views: 2205

Stokes and 2nd order elements

Dear all, attached a small test case of the Stokes Solver with second order elements; There is a sharp discontinuity on the bottom BC orientation (but this is common in ice flow modelling and real applications). This results in noisy results when using the "Bubbles" stabilisation flag (as ...
by fgillet
18 Oct 2011, 18:21
Forum: Bug reports
Topic: Periodic BC Use Lagrange Coefficient
Replies: 1
Views: 3049

Periodic BC Use Lagrange Coefficient

Dear Elmer developers,

"Periodic BC Use Lagrange Coefficient" results in a seg. fault in recent Elmer revisions.

The pb comes from line 2153 of SolverUtils.src
"DO WHILE(ASSOCIATED(A))"

which, I think, should now be
"DO WHILE(ASSOCIATED(F))"

best regards,

Fabien
by fgillet
05 Aug 2011, 10:27
Forum: ElmerSolver
Topic: Sliding and non penetration
Replies: 5
Views: 3856

Re: Sliding and non penetration

Hi, I'm coming back to my problem. I don't understand exactly how Dirichlet conditions are applied with the Normal-Tangential system, but I assume that a normal is computed for every BC nodes and that at each nodes the normal is the sum of the normals to the elements the nodes belong to? It seems fo...