All fields not solved

fabianpk · November 9, 2005, 04:46

Hello,
I have a bit of a problem with Foam not solving everything I want it to solve. It used to be that it didn't solve for density, but I could solve that by taking a very small timestep and then restarting the calculation (of course, this was only done for testing purposes). However, now it solves for rho (which I initialize together with p using a version of potentialFoam), but it doesn't solve for p!
I use pressureInlet conditions for the velocity, and fixed values for p and rho (except at the outlet for rho I have zerogradient).
It's a compressible case.

Does it skip the solving because it thinks that the difference will be too small? My timesteps are of order 2e-10s, and my time of interest about 1e-3.

/Fabian

hjasak · November 9, 2005, 06:24

The solver is always called in the right sequence so the only thing that may have happened is that the code thinks that the equation is already solved. If you really want it solve, try tightening the tolerances; for more info, switch on the lduMatrix debug switch.

Enjoy,

Hrv

November 9, 2005, 04:46	Hello, I have a bit of a pro	#1
fabianpk Member Fabian Peng Karrholm Join Date: Mar 2009 Posts: 61 Rep Power: 17	Hello, I have a bit of a problem with Foam not solving everything I want it to solve. It used to be that it didn't solve for density, but I could solve that by taking a very small timestep and then restarting the calculation (of course, this was only done for testing purposes). However, now it solves for rho (which I initialize together with p using a version of potentialFoam), but it doesn't solve for p! I use pressureInlet conditions for the velocity, and fixed values for p and rho (except at the outlet for rho I have zerogradient). It's a compressible case. Does it skip the solving because it thinks that the difference will be too small? My timesteps are of order 2e-10s, and my time of interest about 1e-3. /Fabian

November 9, 2005, 06:24	The solver is always called in	#2
hjasak Senior Member Hrvoje Jasak Join Date: Mar 2009 Location: London, England Posts: 1,905 Rep Power: 33	The solver is always called in the right sequence so the only thing that may have happened is that the code thinks that the equation is already solved. If you really want it solve, try tightening the tolerances; for more info, switch on the lduMatrix debug switch. Enjoy, Hrv __________________ Hrvoje Jasak Providing commercial FOAM/OpenFOAM and CFD Consulting: http://wikki.co.uk

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
navier-stokes solved?	marraco	Main CFD Forum	4	February 18, 2008 07:30
Are the quations solved in FDM or FVM ?	anon	Main CFD Forum	0	December 12, 2005 22:29
Solved examples	reda	Main CFD Forum	0	April 6, 2003 17:36
looking for a problem could be solved by CFX5.5.1	Hason	CFX	3	October 24, 2002 09:25
looking for a problem could be solved by CFX5.5.1	Hason	Main CFD Forum	1	October 16, 2002 16:01