Search found 196 matches
- 26 Jan 2015, 13:21
- Forum: ElmerSolver
- Topic: GetVariable used on components is not uptodate
- Replies: 3
- Views: 2261
GetVariable used on components is not uptodate
Hello dear Elmer community, i have the following situation I use a solver to calculate some unknown "pot" with 2 degrees of freedom. After the solver finished i can use VariableGet(Solver%Mesh%Variables,"pot") on its Variable and there i get the two components of the unkown as ca...
- 26 Jan 2015, 13:17
- Forum: ElmerSolver
- Topic: Point source for StatCurrentSolve
- Replies: 23
- Views: 12291
Re: Point source for StatCurrentSolve
Exactly, and when u think through the units of Ax=b arising from the PDE of OHms law ( -div.(sigma*grad_pot)=rhs) then the b has to have units of Ampere. which gives to possibilities for the rhs: 1) it is a source term in the volumetric meaning (-div.(sigma*grad_pot)= f) then f would have to have un...
- 23 Jan 2015, 12:35
- Forum: ElmerSolver
- Topic: Point source for StatCurrentSolve
- Replies: 23
- Views: 12291
Re: Point source for StatCurrentSolve
HI, setting the "potential load" is more of Neumann BC than a Dirichlet one. And setting the nodal load at just one node is a current source that is coming from one point. The current source doesn't actually have a direction. Static Potential and current work quite analogue to a static hea...
- 21 Jan 2015, 11:47
- Forum: ElmerSolver
- Topic: Point source for StatCurrentSolve
- Replies: 23
- Views: 12291
Re: Point source for StatCurrentSolve
Aren't the potential load and the current density the same thing?
- 10 Dec 2014, 13:06
- Forum: ElmerSolver
- Topic: Evaluating equation on different bodies
- Replies: 3
- Views: 2103
Re: Evaluating equation on different bodies
very smart, very true,
i was not thinking of iterative solvers enough,
thanks peter
i was not thinking of iterative solvers enough,
thanks peter
- 09 Dec 2014, 14:48
- Forum: ElmerSolver
- Topic: Evaluating equation on different bodies
- Replies: 3
- Views: 2103
Re: Evaluating equation on different bodies
Hello, Thinking about my own post i just had the idea that maybe the passive/active elements feature could be a solution to my problem. So i could give every subdomain its own body index and then in the solver switch only elements of one body to active somehow. Would that actually make the system ma...
- 09 Dec 2014, 11:48
- Forum: ElmerSolver
- Topic: Evaluating equation on different bodies
- Replies: 3
- Views: 2103
Evaluating equation on different bodies
HI dear elmer community, after a long time i ahve some questions again. So my situation is the following: 1) I solve the heat equation (my own implementation of it, but very standard) on my whole model domain. 2) The heat sources come from another solver that calculates some kind of joule heat, so i...
- 11 Aug 2014, 11:37
- Forum: ElmerGUI
- Topic: Thermal simulation - multiple bodies
- Replies: 12
- Views: 12553
Re: Thermal simulation - multiple bodies
Hello, i had a typo in my earlier post, i meant to write "direct solver" not "directz" solver. So instead of using an iterative solver, using a direct one. And regarding mzenkers last post: Don't hesitate to correct me here if i am wrong but isn't the Dirichlet boundary a heat si...
- 05 Aug 2014, 15:41
- Forum: ElmerGUI
- Topic: Thermal simulation - multiple bodies
- Replies: 12
- Views: 12553
Re: Thermal simulation - multiple bodies
HI there,
have you tried using a directz solver and see if that gets a solution?
And i think the initial condition is also influencing the steady state simulation.
It is the initial guess for the iterative linear solver. So maybe putting it to 120 improves the convergence.
best regards
Franz
have you tried using a directz solver and see if that gets a solution?
And i think the initial condition is also influencing the steady state simulation.
It is the initial guess for the iterative linear solver. So maybe putting it to 120 improves the convergence.
best regards
Franz
- 29 Jul 2014, 11:33
- Forum: ElmerSolver
- Topic: Symmetric coordinates and User defined Solvers
- Replies: 2
- Views: 2633
Re: Symmetric coordinates and User defined Solvers
Hello Peter, Thanks for the answer. Okay so as i understand the metric information gathered by "CALL CoordinateSystemInfo( Metric,SqrtMetric,Symb,dSymb,x,y,z )" is taken care of the coordinate trasnformation for cylindrical coordinates so for example simple diffusion in cartesian coordinat...