CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Fluent discretization (http://www.cfd-online.com/Forums/fluent/30402-fluent-discretization.html)

Chris Michaels October 17, 2002 10:58

Fluent discretization
 
Hi.

I was hoping someone could shed some light on exactly what some of the discretization items are:

1) For pressure-velocity coupling, how do you decide on SIMPLE vs. PISO?

2) What pressure discretization is best to use for agitated mixing tanks? (standard, linear, 2nd order, body force weighted)

3) Same for momentum and energy.... how do I decide on what is best to use?

In general, are there any that are gridding-scheme specific (meaning something I should use for HEXs vs. TETs)? I know some of them make calcs shorter, but will the "better ones" have significantly longer iterating time for a 600,000 to 1,000,000 cell geomtery?

Thank you all in advance. If you can, please share any knowledge you may have on this, for I have all but mastered Gambit and now work more with Fluent. Take care.

CM

Jimmy October 21, 2002 07:45

Re: Fluent discretization
 
3) Same for momentum and energy.... how do I decide on what is best to use?

I work on free jets and the parametric studies i've done show the the most important parameter is discretization order of the momentum equation. The use of 1st order completely underpredict shear layers developement and turbulence production. At least 2nd order is desirable, even QUICK if tou can.

About number of cells, i've seen in my calculations that increasing their numbers logically increase the time needed for 1 iteration but, as a counterpart, the solution converges in a lesser number of iterations.

As far as i can remember PISO scheme is specially built for transcient problems but never used it. For reynolds averaged calculation you should use SIMPLE.


All times are GMT -4. The time now is 19:08.