[andrearimo]

Hi everybody, i've been trying to simulate a compressible (M=0.5->0.8) intake for a month now without any results, so I post hoping to find some help.

After a lot of trials, i finally ended up simplifyng the case to a simple tube, constant diameter, with a 90deg turn and low speed.

i used inlet and outlet (velocity 10m/s at the inlet and pressure 101000Pa at the outlet) and tried to run it in simplefoam, rhosimplefoam.
both gave me problems after about a hundred iterations (pressure becoming negative or max nuimebr of iter exceeded)

so i tried rhoturbfoam, with the aim of seeing what was the problem, and after some trials i saw this:
1 - the pressure front developing at the inlet travels through the tube, reaches the end and bounces back, doing it severa times until error (time step 1e-5s simulation error at about 0.012sec). max number of iteration exceeded. the same happens for speed, it bounces back and forth.
2 - reduced the velocity to 1m/s and the simulation reaches endtime, but same thing as above.
3 - just for fun, tried potential foam. here the result was interesting... for v=1m/s gave honest results, for v=10 the velocity at the outlet exceede 7000m/s...

i assume that i'm having some problem in the phisics, but i can't figure out what.

if anyone want to see this the case is downloadable from :
ftp://andrearimo.homedns.org/pub
user : anonymous
pass : <your>

it's in the OpenFOAM folder. there are some zip files, one for every solver i tried. the mesh is always the same, it's a netgen mesh posted on this forum as a tutorial for rhoSimpleFoam solver, dualized using polyDualMesh (90deg)


hope some one can help...

if anyone gets some results, or finds any errors, on the ftp site there is an incoming folder where the corrected file can be dropped.

Thanks to all