Steam Ejector 2D simulation
 

I need to study pressure and temperature for large scale ejector, with following boundary :
Primery inlet pressure 6.7 bar and 453 K
Sec Inlet pressure .094 bar and 318 K
Out let 0.3 bar and 343 K
I used K-e realizable model
first primary steam expand and be in wet region then entrained Secondary steam and mix at wet region then flow shocked and out saturation steam.
solution always diverge before starting calculation due to many reasons:
# Divergence detected in AMG solver: k -> Increasing relaxation sweeps!
# Divergence detected in AMG solver: Coupled -> Decreasing coarsening group size!
Divergence detected in AMG solver: WSB-eq
Divergence detected in AMG solver: WSN-eq

I used Density based and wet steam as multiphase and tried to change it to pressure based and mixing but without any results

can anybody help me to solve may case as i'm beginner in Fluent and CFD
I tried to Decreasing coarsening group size and Increasing relaxation sweeps but solution still diverge and floating

Hi
did you find the solution for this issue . i need help as well

AMG solver problems
 

Hi, i have same problem and need help.
in my case i am modeling a fermentor sterilization by 2 bar steam, and the model that i chose is density based and inviscid flow. like ahmedwaheed i have convergence problem in amg solver and wsb equaion.
anybody help me?

Hi,
from the pressures and temperatures of you reported is assumed a ratio of flow rates of about four times, please do you confirm?
The compression ratio is low (this is easier) and the energy of the motive steam with about 17 degrees centigrade of overheating is great.
To lead to the convergence calculation, however, you should first use a UDF because you cannot set in "boundary condition" final pressures hoping that does not diverge, the Fluent calculation has to do it in small pieces, then I suggest 2000 iterations for motive steam ramp, 1000 for suction and 1000 for discharge and subsequent 1000 for closing the calculation. The number of iterations is set in the "calculation activities".
With this you may already have solved most of the problems.
Another tip, you can set an under-relaxations to 0.01, and pay attention to the courant number.
An ejector has very violent turbulence because there are very high speed (4 or 5 times the speed of sound), so in the "boundary condition" occurs that is set "Intensity and Hydraulic Diameter", do initialization hybrid set a courant number minimum of 10 and maximum of 30 and an "under-relaxations" of 0.4
Surely with these settings the calculation will tend to converge, it is obvious that to everything upstream there must be a nearly perfect mesh!
hi and good work