CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   FLUENT (https://www.cfd-online.com/Forums/fluent/)
-   -   root mean square (https://www.cfd-online.com/Forums/fluent/29836-root-mean-square.html)

 John May 29, 2002 13:48

root mean square

Dear ALL

I am currently trying to model water flowing through a pipe in Fluent 5.5. To validate the model I have found a number of published papers on the subject. Many of the papers use a parameter called the 'root mean square velocity'. What does this mean and how can I get it from Fluent?

 Steve May 29, 2002 17:03

Re: root mean square

Urms is the root mean square of the turbulent fluctuations in the x direction. It can be recovered from the turbulent kinetic energy if the flow is isotropic (i.e. u'=v'=w') using k=1/2(u'^2+v'^2+w'^2). However the flow is unlikely to be isotropic and so if you are using an eddy viscosity model you will need to know the ratio of u':v':w'(which for example is 4:3:2 for equillibrium flat plate flow).

For eddy viscosity models the flow is assumed to be isotropic so it will be difficult. For second-order closures you can recover Urms from the Reynold stresses.

Steve

 Steve May 29, 2002 17:10

Re: root mean square

Sorry 4:2:3 not 4:3:2

 John May 30, 2002 09:55

Re: root mean square

Hi Steve,

Thank you very much for your response. I am using the k-eps 2 equation model which if I understand correctly assumes the eddy viscosity is the same in all directions. Can I therefore assume that it will be impossible to get the rms of the turbulent fluctuations from Fluent directly (unless I change to the Reynolds stress model) and the only way I can get them indirectly is to know the ratio of u'v'w' for a pipe?

Thanks once again, John

 Steve May 30, 2002 16:14

Re: root mean square

Thats correct, however if the data you are comparing to has all three components of the turbulent fluctuations than you could combine them to give k and compare to k from FLUENT.

 All times are GMT -4. The time now is 19:43.