Hi:

I have a homework to simulate the turbulent flow in a channel, my teachear saied there are 8 vortices in the channel, so which turbulent model should i use! thank you

Dear ztdep,

The number of vortices has no relation whatsover with the trubulence model you employ. One of the good models that have been found succesful for internal and external flows even with moderate adverse pressure gradients is Menter's k-w SST model. Wilcox k-w model would also be a good choice, though it suffers from the freestream dependency problem.

Hope this helps Regards,

Ganesh

ztep wrote:

I have a homework to simulate the turbulent flow in a channel, my teachear saied there are 8 vortices in the channel, so which turbulent model should i use!

diaw writes:

Have you tried simulating with transient DNS, with no convection stabilisation?

Hi:

ganesh, thank you for your response!

But some studuents can not predict the vortices with a standard k-e model,

I will try the models you have suggested! Thank you again!

Hi:

diaw, would you like to give some information about some free DNS code!

Thank you !

Hi ztep,

If you email me the model geometry (channel shape) & flow parameters, I'll have an initial go at it using a rather unconventional approach.

I can then send you the results & let you try & have a go on your own. If it works out, then I'll let you have a copy of the model code-file & you can try for yourself.

It would be based on open-source, downloadable software.

diaw...

Hi:

diaw, it is only a 3 dimensinoal straightn square channel, the renolds number based on the Uin and the weight of the channel is 10000!

ztdep@yahoo.com.cn

Thank you very much!