CFD Online Discussion Forums

CFD Online Discussion Forums (
-   FLUENT (
-   -   Anyone ever use third order? thoughts? (

Paul June 20, 2008 10:09

Anyone ever use third order? thoughts?
I read an article published by someone who said he got good results for efficiency curves of cyclones using third order MUSCL equations. Anybody else use them? Any thoughts/ideas on how to make them converge?

Also, there is no third order option for pressure. Should I just leave it on second order?


red lemon June 20, 2008 10:18

Re: Anyone ever use third order? thoughts?
yes leave pressure as second order unless PRESTO or QUICK is better depends on your application and mesh. Use MUSCL for all grid types so you can to reduce diffusion but solution may be more difficult to converge and less tolerant of poor cell quality.

Paul June 20, 2008 12:34

Re: Anyone ever use third order? thoughts?
Yeah I'm having difficulties getting this to converge. Pressure as second order has worked fine for me before so I've left it as is.

The residuals are bouncing around a little but staying somewhat constant. I can tell from the static pressure contour plots however that this thing is nowhere near convergence. I mean I can obviously screw around with the relaxation factors until it finally converges but its not going to give me good data. So far its run for about 15000 iterations on RSM steady.

What I take to be a bad sign is the lack of stabilization of the static pressure of the outlet (which I have plotting while iterating), which tells me that this isn't going anywhere useful.

Any ideas?

red lemon June 20, 2008 14:06

Re: Anyone ever use third order? thoughts?
15000 iterations is a lot! Sounds like the overall case setup needs checking. Also what is grid quality like ? Ensure first and second order are stable first.

All times are GMT -4. The time now is 00:13.