CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   VOF model convergence (http://www.cfd-online.com/Forums/fluent/37981-vof-model-convergence.html)

lana September 22, 2005 20:26

VOF model convergence
 
Hi everybody,

I am trying to model a jet impinging on a molten pool of metal with the VOF model.

I am interested only in steady state solution, so I am using the implicit formulation for the interface, with PISO and PRESTO as numerical schemes, as suggested by the manual.

I am experiencing big difficulty in convergence: residuals oscillates of a couple of order of magnitude at each step and they remain almost horizontal... pressure along the interface becomes unstable after some hundreds time steps... i played with URF, but not sensible gains... Adopting the wall adhesion seems to give some improvement.

the manual anticipates this kind of troubles when differences between density and viscosity of the two fluids are great, in my case they are of 2 order of magnitude for viscosity and six for density.

Anyone has some advice for this kind of configuration? is it possible to get convergence or better quit?

Thank you very much.

edi September 23, 2005 04:38

Re: VOF model convergence
 
I think that the main trouble is that you chose the steady formulation. Consider to set up your case as an unsteady problem.

The manual is a bit too optimistic about VOF and steady calculations...

Hope this helps

Edi.

Eric September 23, 2005 04:56

Re: VOF model convergence
 
dear lana .. could you give some details about you physical model ? what is the impinging jet in your case ..is it a gas ?,

or is it a molten metal jet injected into different molten metal ..?

lana September 23, 2005 13:30

Re: VOF model convergence
 
hello, thank you for your interest.

Actually I am already solving it as unsteady. I am considering different VOF scheme (implicit and Geoconstructed so far) but I got pressure instabilities along the interface.

I think the main problem is at the wall boundary, since perturbation starts from there...

lana September 23, 2005 13:34

Re: VOF model convergence
 
Dear Eric,

I am simulating a typical welding configuration. My jet is hot gas (high velocity and very low density) impinging perpendicular on a molten copper pool.


MANOJ KUMAR September 24, 2005 02:14

Re: VOF model convergence
 
hi i doing also VOF in my problem a sheet of metal passing in a molten pool of zink and sheet is galvanized. but there is floating point error. Give me suggestion for doing my problem.i take unstedy


Eric September 24, 2005 06:16

Re: VOF model convergence
 
Hi Lana , I am modeling the same process, of impinging of inert gas on the free surface of a molten pool..I want to see the deformation of the free surafce ...my model is 2D VOF segregated unsteady ..I use velocity inlet at the top for the gas jet ..but how did you keplt the temperature of the molten pool constant or does it change during impinging ??

lana September 26, 2005 12:47

Re: VOF model convergence
 
I am not interested in the temperatures at this stage... I just set the viscosity and density constant in the molten pool.

I am starting to consider that the configuration could be instable itself, it happens that the surface oscillates when the jet momentum is too high. I will try to lower the jet momentum to see if it helps.

What about your solution? is it converging?

Eric September 27, 2005 04:13

Re: VOF model convergence
 
Hi Lana,

are you using Solidification/melting model simultaniously with VOF ?..I have had problem with convergence when using solidification model. you say that you are not interesting to temperature at this stage (temperature of metal molten pool should be higher that melting temperature.. how did you define the liquid state? ..in my case i use the same steps solution used by A.Nguyen and G.Evans in thier paper "Comput.Fluid Dynamics of gas jet impinging into liquid pool".. -if you don't have it I can send you copy ...

lana September 27, 2005 13:44

Re: VOF model convergence
 
Thank you Eric. Actually I used just that paper for validating my first VOF model. I used the same setup but there are not pretty much details on the convergence history. Once I got similar results I quickly switched to my case of interest, which happens to be yours also.

I tried SM model with VOF, but I was no able of running it properly... did you use it with VOF? 3 phases? how did you manage to let the VOF model to understand that two phases are decided by the SM?

For the stage I am I simply set a temperature distribution in the anode based on a previous model and wrote some UDF to simulate the solid state (high viscosity and sink momentum). However they have to be carefully tuned and I am still working on it...


MANOJ KUMAR September 28, 2005 05:43

Re: VOF model convergence
 
Hi Eric I doing VOF model for galvanizing sheet in that one is sheet which is liquid-solid and molten metal which is in container taking as a fluid. i define two phase and done same as fluent solution given in Bowl problem but there is floating point error. Please send me paper that u have my e-mail is ms_project05@rediffmail.com

Eric October 1, 2005 05:23

Re: VOF model convergence
 
Dear Lana .. In my case I use 2 phases (air and melt pool material) using SM model with VOF ..I read in the manual that "you can use VOF model with solidification/melting ..and you must take solidus, liquidus, and melting heat of phases that are not melting or solidifying (air) equal to zero..and you have only to set those values for phases that can melts or solidifies ...

Eric


All times are GMT -4. The time now is 03:44.