|December 18, 2009, 08:22||
Join Date: Nov 2009
Posts: 43Rep Power: 6
I have some trouble with buoyantBoussinesqPisoFoam. I like to run the solver on a tube with several curves.
I tried to adapte the hotRoom tutorial BC on my case and I took the same files for fvSolution and fvSolvers.
Actually I have three patches - inlet, outlet and wall.
My BCs are:
A guess would be, to change the value for p at the wall from uniform 0 to something else, but somehow, I don't expect, that this would solve the problem.
|December 21, 2009, 09:45||
Join Date: Dec 2009
Posts: 1Rep Power: 0
Actually I have tried the Tutorial as well and have some questions about the boundary conditions.
1. What does rhok mean? and why is uniform 0?
value uniform 0;
2. At the wall i do not wanna have zeroGradient or uniform, i wanna have that I can add a Gradient of the temprature.
|December 22, 2009, 10:47||
Join Date: May 2009
Posts: 10Rep Power: 7
I've never run this example but something is wrong with you Pressure
DICPCG: Solving for p, Initial residual = 0.937421, Final residual = 0.0905879, No Iterations 565
time step continuity errors : sum local = 0.187933, global = 0.0077102, cumulative = -0.00408645
DICPCG: Solving for p, Initial residual = 0.822426, Final residual = 0.000737089, No Iterations 1001
1001 iterations is too many
|February 17, 2010, 09:43||
Join Date: Nov 2009
Posts: 3Rep Power: 6
I have the same trouble like Erik!
I have solved a natural convection problem with boussinesqBuoyantFoam of OpenFoam 1.5.
Two weeks ago I have seen that the thermophysical module was modified in 1.6 and that the compressible buoyantFoam (1.5) gives wrong results but buoyantPisoFoam (1.6) is ok.
Now I am in doubt if the incompressible module buoyantBoussinesqPisoFoam (1.6) is similar to boussinesqBuoyantFoam (1.5).
Therefore I tried to transpose my natural convection problem to
There are some different insignificant inputs in the transportProperties
file and a significant modification of the pressure boundary condition.
Namely, if one can use zeroGradient for p at the wall in 1.5, in 1.6 one
have to set the buoyantPressure type.
I do that, and kept the input parameters in the new case for 1.6.
One cannot set "type buoyantPressure" in boussinesqBuoyantFoam (1.5), and "type zeroGradient" for pressure in buoyantBoussinesqPisoFoam (1.6) doese not work.
So, one cannot test a simulation case for comparison.
The main trouble with buoyantBoussinesqPisoFoam (1.6) is that the
pressure solver becomes unstable after a certain number of time steps.
First I have tried with constant time step and than with adjustable time
step but the result is the same. The Iterations of the pressure solver becomes larger reaching the value 1001 and then the run is blowing up.
|Thread||Thread Starter||Forum||Replies||Last Post|
|Compile problem||ivanyao||OpenFOAM Running, Solving & CFD||1||October 12, 2012 09:31|
|Problem with compile the setParabolicInlet||ivanyao||OpenFOAM Running, Solving & CFD||6||September 5, 2008 20:50|
|Compiling problems with hello worldC||fw407||OpenFOAM Installation||21||January 6, 2008 17:38|
|DecomposePar links against liblamso0 with OpenMPI||jens_klostermann||OpenFOAM Bugs||11||June 28, 2007 17:51|
|user subroutine error||CFDUSER||CFX||2||December 9, 2006 06:31|