Elmer Discussion Forum

hello. My experience is that using Picard (default) iteration scheme throughout (without jumping to Newton's) might help. For that, make sure followings are set properly Nonlinear System Newton After Iterations = 30 ! grater than (Nonlinear System Max Iterations = 20) Nonlinear System Newton After T...

Hello Elmer Guys, I understand that there are quite a few options on how and when to activate a solver, by inserting "Exec Solver" command in a relevant solver. "Exec Interval" is also there; but, what interval does it refer to? In fact, I need to activate/deactivate solver as to...

Hi guys, I just wanted to make slight changes in the Naiver-Stokes solver (FlowSolve). I obtained the source codes, subroutine FlowSolve.src and module NaiverStokes.src, and saved them as myFlowSolve.f90 and myNaiverStokes.f90 after making necessary changes. I compiled them locally and obtained the ...

Hi guys, In a 3D time-marching Navier–Stokes problem where the domain (upper) surface evolves in the vertical z-direction (i.e. mesh deforms only in the upper surface boundary), how do I save field variables along a specified line (say sinusoidal) on the surface (evolving) boundary? I tried to speci...

Hi, I am trying to impose a sliding BC for a visco-plastic flow... when I use default Navier Stokes Solver (via GUI) with suitable BC, it works: !--------------------------------------------------------- Solver 1 !--------------------------------------------------------- Equation = Navier-Stokes Pro...

Hi all, I am trying to solve Stokes equation with my own subroutine... the coordinate system is "2-D Cartesian"... in one of the boundaries, I need to set "normal component of velocity" to zero... I tried, but couldn't get it right... A part of .sif for that particular boundary l...

thanks Juha,

actually it works this way--

$elmerf90 -c myModule.f90 /// this creates object file, along with .mod file
$elmerf90 -o mySolver.dll mySolver.f90 myModule.o

suren.

Hi Elmer guys, I am currently trying to link my own module to my own subroutine... but just couldn't get it right! I have a subroutine (say "mySubroutine" in mySolver.f90) which I link to Elmer dynamically as: $ elmerf90 -o mySolver.dll mySolver.f90 It works fine! But, if I try to refer to...

I've got it, now... but with the sort of coding:
DO i=1, dim
Velocity(i,1:n) = FlowSolution((dim+1)*(FlowPerm(NodeIndexes(1:n))-1) + i)
END DO

... getting the nodal solution values directly, rather than using default subroutine GetVectorLocalSolution(Velocity)...

~suren

Hi. I've coded Stokes equation (2-D, Newtonian fluid).. its working well and yielding results (as can be seen in the ElmerPost) same as the ones from the standard NS solver (FlowSolver)... no problem until here... Now I am trying to include the material nonlinearity, which demands the local vector (...