[Giorgitseli]

Hello everybody.

I am new user regarding the transient calculations and I would appreciate any useful information about how I can select an appropriate time step for my simulation.

Moreover does anybody know if I can "insert" in CFX a steady state flow field and then run a model for the solids that exist in the fluid domain???I hope that if this is an available option in CFX I could reduce the required total computational time. Regarding the particles I use the algebraic slip model (no deposition).


THANX

[ghorrocks]

By solids I assume you mean ASM of small solid particles.

Yes, you can do this, but the whole idea of ASM is it couples back to the flow. You can turn bits of the solver off with expert parameters, so you can turn off the pressure/momentum equation but solve the ASM equations. Alternately using particle tracking you could do this as a one-way coupled simulation.

[Giorgitseli]

"You can turn bits of the solver off with expert parameters, so you can turn off the pressure/momentum equation but solve the ASM equations. Alternately using particle tracking you could do this as a one-way coupled simulation."

Thanx for the rapid answer!
1.)How can I turn off the pressure/momentum equation and only solve the ASM equations??I ll check the particle tracking method because I have not yet worked with this!
2.)Any idea for the time step selection?

[CFD_UJI]

hi! from what i know, you have to select a time step so that the Courant number of your problem is approximately 1. thanx

Daniel

[Zigainer]

Originally Posted by Giorgitseli

1.)How can I turn off the pressure/momentum equation and only solve the ASM equations??I ll check the particle tracking method because I have not yet worked with this!

ghorrocks already answered your first question:

Originally Posted by ghorrocks

You can turn bits of the solver off with expert parameters

[ghorrocks]

hi! from what i know, you have to select a time step so that the Courant number of your problem is approximately 1.

This one keeps coming up, it is wrong.

CFX is an implicit solver and does not have the CFL<1 (compressible) or Courant No<1 restriction. You should determine time step size by a sensitivity analysis, and you will probably end up with an acceptible time step size larger than 1.

This restriction applies to explicit solvers.