[evrikon]

Hello together,

I simulate a simple axial fan with MRFSimpleFoam. Everithing works fine, however, I have got very low pressure jump predicted. I' looking now where mistake in my settings is.

The question is what the type of boundary condition I should use for the rotor. Rotor in my simulation is a rotating fan wall inside of the MRF zone and rotates of course with the same angular velocity as the whole MRF zone. If I would use fixedvalue; uniform (0 0 0) like in the tutorial mixerVessel2D, does it stay at rest? Should I take movingWallVelocity instead (s. below)?

From my ../0/U;
.....
boundaryField
{
STATOR
{
type fixedValue;
value uniform (0 0 0);
}

ROTOR
{
type movingWallVelocity;
value uniform (0 0 0);
...........

Other question for my full understanding: does MRF model in OF use an absolute velocity formulation or relative velocity formulation? "absolute" means it solves the equations still in non-rotating coordinate system, relative I mean as rotating coordinate system. This question refers to the problem I encountered, namely with absolute velocity formulation the wall should rotate, with relative should stay at rest relatively to the rotating zone where it is located in.

Alex

[evrikon]

In the meantime I tried itself, read through the forum etc. If the case set up correctly fixed value condition for a wall inside of a rotating zone leads to correct radial velocity distribution on this wall. One can see it in paraview.

[nash]

hi evrikon,

sorry for bringing this old thread up.
do you manage to get the right pressure rise?
i simulate the radial fan now, and i got lower value of pressure rise. and the residuals of velocity fall only to 3rd order of magnitude.

for this simulation, i use approach of mrf and turbulence model laundersharma and komegasst (low re model implementation).

any suggestion for better result?

[be_inspired]

Quick question.
Did you make any faceZone? It is to know if boundary faces are selected or not ( including wall faces)