why different results on same model with cosmosflo 07 and flow simulation09?

qihongming · May 24, 2009, 22:32

Same model on cosmosfloworks 2007 and flow simulation 2009 produce different results? please see attachment, this is simple model.

Results:
cosmosfloworks 2007:
Results---surface parameter(Select elements)---click Evaluate-Tab local----Temperature (Fluid) maximum is 59.0421C

However the results of flow simulation 2009 is 77.7825C

I'd appreciated if you can tell me the reasons?

Attachment have exceed 100K, could you tell me howto attach my files?

ghorrocks · May 25, 2009, 20:24

Hi,

There is no attachment so I can't be specific but if any CFD analysis has not been adequately show to have mesh independance, adequate convergence and small enough timestep (if transient) then the result of the analysis is nothing but a random number. Comparisons between different random numbers are pretty pointless.

Glenn Horrocks

qihongming · May 26, 2009, 01:14

Thank you very much.
Please see following attachment.

ghorrocks · May 26, 2009, 01:30

Hi,

This is a CFX forum. You won't find many people with Cosmosworks here. I cannot view the file and would not have the time to do a detailed analysis of your simulation if I could open it.

Glenn Horrocks

May 24, 2009, 22:32	why different results on same model with cosmosflo 07 and flow simulation09?	#1
qihongming New Member qihongming Join Date: Mar 2009 Posts: 14 Rep Power: 17	Same model on cosmosfloworks 2007 and flow simulation 2009 produce different results? please see attachment, this is simple model. Results: cosmosfloworks 2007: Results---surface parameter(Select elements)---click Evaluate-Tab local----Temperature (Fluid) maximum is 59.0421C However the results of flow simulation 2009 is 77.7825C I'd appreciated if you can tell me the reasons? Attachment have exceed 100K, could you tell me howto attach my files?

May 25, 2009, 20:24		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,701 Rep Power: 143	Hi, There is no attachment so I can't be specific but if any CFD analysis has not been adequately show to have mesh independance, adequate convergence and small enough timestep (if transient) then the result of the analysis is nothing but a random number. Comparisons between different random numbers are pretty pointless. Glenn Horrocks

May 26, 2009, 01:30		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,701 Rep Power: 143	Hi, This is a CFX forum. You won't find many people with Cosmosworks here. I cannot view the file and would not have the time to do a detailed analysis of your simulation if I could open it. Glenn Horrocks