CFD Online Discussion Forums

CFD Online Discussion Forums (
-   FLOW-3D (
-   -   Restart for solidification (

atulverma May 12, 2009 11:18

Restart for solidification
hi everybody!
I am using a shot sleeve for filling the die and I am stopping the simulation after the die is filled completely and the piston at that time is still pushing the fluid therefore I am having pressure iteration failure which continues in the restart also and this have to be avoided to get correct velocity field to get correct segregation pattern .
I cannot use fill fraction as stopping point for filling because I have multiple mesh block and I don't think we can specify the stopping to be the point where the fill fraction in some mesh block is 1.0(if it is possible please tell me how to do it).
Now I want to give restart for solidification the possible ways I think that will lead to converged solution:

1.Stop the simulation little earlier,but that leaves voids which cannot be filled(only partially filled cells can be filled).

2.Give limited compressibility option from the beginning of the restart. Please tell me how this can be done...because generally the solver invokes this after few unconverged time steps.If this can be done from the beginning of the restart i think we can get correct velocity field and thus correct segregation patterns.

This is the limited options I can think about I you have any suggestions "please please " reply.I am not aware if i can play with boundary conditions in this case...

melissa May 14, 2009 13:13

How do you have the motion of the piston defined? You could remove the piston for the restart.

atulverma May 15, 2009 06:25

hi melissa!
I have defined prescribed motion for the piston with velocity as the function of time.I would like to remove the piston but the boundary conditions and the boundary itself have to be redefined.I have never tried that but it seems to be a possible option.Using symmetry boundary condition just in front of the piston in restart will be o.k?
My aim is to pressurize the fluid during casting so that I could see the segregation due to shrinkage effects, for that I require Coupled Motion but the problem is that the solution does not converge and compressibility is invoked thus the incompressible fluid is treated as compressible,which is wrong.

All times are GMT -4. The time now is 06:40.