CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   FLUENT (https://www.cfd-online.com/Forums/fluent/)

 Anton December 7, 2006 13:22

Hello Everybody!

Do you have experience with DEFINE_UDS_UNSTEADY for _multiphase_ flows?

In the FLUENT UDF manual, paragraph 2.7.4,

we see the approximaton for the unsteady term for UDS:

[unsteady term] = = -( (rho*phi)^(n) - (rho*phi)^(n-1) )/delta_t * delta_V = = apu + su,

where

apu = -rho * phi^(n) * delta_V / delta_t, su = -rho * phi^(n-1) * delta_V / delta_t.

^ here means the time step, not the power, of course.

on the other hand, the natural idea to discretize the unsteady term is:

apu = -rho^(n) * phi^(n) * delta_V / delta_t, su = -rho^(n-1) * phi^(n-1) * delta_V / delta_t.

FLUENT's default is

apu = -rho^(n-1) * phi^(n) * delta_V / delta_t, su = -rho^(n-1) * phi^(n-1) * delta_V / delta_t,

though. The question is why does FLUENT take density values for calculating apu from the previous time step and not from the current time step?

Thank you very much in advance!!

 Anton December 11, 2006 11:14