CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Static pressure-adaptation gradient (http://www.cfd-online.com/Forums/fluent/28874-static-pressure-adaptation-gradient.html)

Andrew Parker August 30, 2001 06:45

Static pressure-adaptation gradient
 
Hi

Does any body know when you should stop adapting the static pressure gradient in your model? I know that the refine value that should be entered should typically be about 1/10th of the computed value, but when should you stop this process. For example if I am using the two-layer zonal method I know that I have to continue adapting until the y+ value is below 4 or 5-what is the similar value of the static pressure gradient.

Thanks

Andy

Philipp August 30, 2001 09:28

Re: Static pressure-adaptation gradient
 
this might sound a bit stupid, but I think you should stop when you reached a result that is in regard on accuracy versus CPU time OK for you. If you cannot say what is a good result cause you have nothing to compare with, than I would suggest to go forward and forward until the changes of the result are getting smaller and smaller. Of course it's you again who has to decide when the changes are small enough... Do you already tried different mesh types (e.g. compare structured and unstructured models with each other) or different models (laminar, k-e, ... whatever fits more or less to your problem)?

best regards philipp

Nurul Murad September 3, 2001 00:04

Re: Static pressure-adaptation gradient
 
hi..from my limited experience...usually you will do grid independency when you refine your grid...ie. in pressure gradient. if you refine double the original grid size everytime and the solution doesn't change much (usually 5% and below is good enough)...then you can confidently say that it is time to stop refining. from my experience...you will need a large grid size by doing grid independcy using k-e turb model as oppose tu LES. LES will require about half of what k-e usually acquire. please check on this..thanks..


All times are GMT -4. The time now is 10:46.