how to reduce correctors for transient flow
 

Hi, everyone:

In a intake manifold, one pressure inlet, four pressure outlet, for each iteraction, the correctors are over 20, I have chang the pressure underrelaxtion factor,but it does not act active, I have checked the .info file, NSP is always over than 1000, how to solve the problem?

Thanks

Re: how to reduce correctors for transient flow
 

In transient run, Uder-relaxation factors are not used. You can probably try to set under-relaxation factor for pressure correction by

NavCenter=>Analysis Controls=>Solution Controls=>Solution Method, in Solution setup, the 3rd item is URF, give a value of 0.7 or so.

Re: how to reduce correctors for transient flow
 

hi Chinayu,
the other way to get the pressure correctors below the limit of 20 is decrease the time step size. What is the value of courant number for your simulation?

Re: how to reduce correctors for transient flow
 

To be honest it could be any number of things, although the obvious ones to look at are courant number, grid quality and turbulence model. Also look at boundary region suitability and ensure that you are not getting any whacky gradients from poor simulation setup. Which equations are not converging and are you getting any other warnings?

Re: how to reduce correctors for transient flow
 

hi chinayu

in the command window

type RELAX and enter

it will ask for which parameter to be changed .....

example

RELAX 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 it will change all param to 0.5

i hope you got required solution ..wish u all the best

anand

Re: how to reduce correctors for transient flow
 

hi anand,
as john mentioned, you cannot do any kind of relaxation that is based on transfer of information from previous to the actual solution step, which is the case you suggested. You would change the physics of your problem in this way. Look at any CFD book (e.g. Malalasekera). The only way you can relax with transient starCD simulation and PISO is decrease time step size, or change the relaxation for pressure correction, relates to the PISO algorithm, or just go out to a pub.

Re: how to reduce correctors for transient flow
 

I have decreased time step to 0.1CA and the courant number is far below 1, so I think the time step is short enough. I have also checked the mesh quality, the mesh quality is not bad, but the presssure boundary are changed, I get the boundary pressure from GT-Power.