CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > OpenFOAM

[twoPhaseEulerFoam] Solver crashes with different RAS models

Register Blogs Members List Search Today's Posts Mark Forums Read

Reply
 
LinkBack Thread Tools Display Modes
Old   July 10, 2014, 09:31
Question [twoPhaseEulerFoam] Solver crashes with different RAS models
  #1
New Member
 
Kasper Linnestad
Join Date: Jun 2014
Posts: 8
Rep Power: 3
kasper1301 is on a distinguished road
Hi,

I have been running twoPhaseEulerFoam on a bubble column successfully with the mixtureKEpsilon turbulence models for both phases. However, if I switch to continousGasKEpsilon for the gas phase and LaheyKEpsilon for the liquid phase, the solver crashes after just a few timesteps. Does anyone know how to remedy this?

Additional info:
Initial turbulence is zero, hence set to 10^-8
Inlet turbulence is low, ~10^-3
Co < 0.25

Regards,
Kasper
kasper1301 is offline   Reply With Quote

Old   July 10, 2014, 10:02
Default
  #2
Senior Member
 
RodriguezFatz's Avatar
 
Philipp
Join Date: Jun 2011
Location: Germany
Posts: 1,097
Rep Power: 16
RodriguezFatz will become famous soon enough
How exactly does the solver crash?
__________________
The skeleton ran out of shampoo in the shower.
RodriguezFatz is offline   Reply With Quote

Old   July 10, 2014, 10:05
Default
  #3
New Member
 
Kasper Linnestad
Join Date: Jun 2014
Posts: 8
Rep Power: 3
kasper1301 is on a distinguished road
Floating point exception, the problem seems to be that both k and epsilon becomes unreasonably large very quickly.
kasper1301 is offline   Reply With Quote

Old   July 10, 2014, 10:10
Default
  #4
Senior Member
 
RodriguezFatz's Avatar
 
Philipp
Join Date: Jun 2011
Location: Germany
Posts: 1,097
Rep Power: 16
RodriguezFatz will become famous soon enough
1) I don't know twoPhaseEulerFoam, but can you set the number of iterations during each time step, like in pimpleFoam?
2) Are your initial conditions (zero turbulence, ...) physically meaningfull or just random? I had problems when I chose bad values of k and omega as initial conditions.
__________________
The skeleton ran out of shampoo in the shower.
RodriguezFatz is offline   Reply With Quote

Old   July 10, 2014, 10:16
Default
  #5
New Member
 
Kasper Linnestad
Join Date: Jun 2014
Posts: 8
Rep Power: 3
kasper1301 is on a distinguished road
1) Yes, that is possible. It just seems weird that it works without any problems with the mixtureKEpsilon model, and that it crashes with the other turbulence models.
2) The initial condition is a tank filled with water with no movement, so the turbulence should be zero (set to 10^-8 to avoid numerical issues). The inlet conditions are calculated according to [1, 2].

[1] http://aerojet.engr.ucdavis.edu/flue...ug/node217.htm
[2] http://support.esi-cfd.com/esi-users/turb_parameters/
kasper1301 is offline   Reply With Quote

Old   July 10, 2014, 10:20
Default
  #6
Senior Member
 
RodriguezFatz's Avatar
 
Philipp
Join Date: Jun 2011
Location: Germany
Posts: 1,097
Rep Power: 16
RodriguezFatz will become famous soon enough
to 1) -> I tryed to get a simple piece of pipe running with different turbulence models. Just 5 diameter long and periodic boundary conditions. There are some k-epsilon models I couldn't get running with all different schemes and initial settings. So don't worry...

Did you try to just increase the inner iterations? Did you try to have some more dissipative schemes, such as some upwind (how does your fvSchemes look like)?
__________________
The skeleton ran out of shampoo in the shower.
RodriguezFatz is offline   Reply With Quote

Old   July 10, 2014, 10:27
Default
  #7
New Member
 
Kasper Linnestad
Join Date: Jun 2014
Posts: 8
Rep Power: 3
kasper1301 is on a distinguished road
I just used the same schemes as in the RAS-tutorial with twoPhaseEulerFoam, haven't had time to experiment with different schemes yet.

Code:
divSchemes
{
    default                     none;

    div(phi,alpha.methane)          Gauss vanLeer;
    div(phir,alpha.methane)         Gauss vanLeer;

    div(phi.methane,kappai.methane)     Gauss vanLeer;

    "div\(alphaPhi.*,U.*\)"     Gauss limitedLinearV 1;
    "div\(phi.*,U.*\)"          Gauss limitedLinearV 1;
    "div\(alphaPhi.*,.*rho.*\)" Gauss linear;

    "div\(alphaPhi.*,(h|e).*\)" Gauss limitedLinear 1;
    "div\(alphaPhi.*,(K.*|p)\)" Gauss limitedLinear 1;

    "div\(alphaPhi.*,(k|epsilon).*\)"  Gauss limitedLinear 1;
    "div\(phim,(k|epsilon)m\)"    Gauss limitedLinear 1;
    
    "div\(\(\(alpha.*nuEff.*\)*dev2\(T\(grad\(U.*\)\)\)\)\)" Gauss linear;
}
Increasing the number of inner iterations may solve the problem, but I'm guessing the simulations will be slowed down (it currently takes ~40 hours), so I will probably just use the mixtureKEpsilon model
kasper1301 is offline   Reply With Quote

Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
different results between serial solver and parallel solver wlt_1985 FLUENT 10 April 11, 2012 15:25
MPI Crashes After FMG Solver - Parallel computing on Linux jonny_b FLUENT 0 April 11, 2011 12:36
CFX 5.5 Roued CFX 1 October 2, 2001 16:49
Setting a B.C using UserFortran in 4.3 tokai CFX 10 July 17, 2001 16:25
Error during Solver cfd guy CFX 4 May 8, 2001 06:04


All times are GMT -4. The time now is 21:08.