[maphd]

Hi,

Is subgrid filter length in large eddy simulation in fluent same as filter width? and is it also comparable to kolmogorov length scale? Can we use subgrid filter length instead of kolmogorov length scale?

I appreciate any help,
Thanks

[FMDenaro]

Originally Posted by maphd

Hi,

Is subgrid filter length in large eddy simulation in fluent same as filter width? and is it also comparable to kolmogorov length scale? Can we use subgrid filter length instead of kolmogorov length scale?

I appreciate any help,
Thanks

no at all...
Kolmogorov scale is dictated by the physics of the flow problem, the filter lenght is a computational parameter you decide.

[maphd]

FMDenaro, thank you for your reply. I am not very familiar with large eddy simulation. That's why my questions might be silly.
How I can read smallest length scale of large eddy simulation in fluent? If you know the answer, could you please tell me little more clear step by step directions?
I really appreciate it,
Thank you in advance...

[FMDenaro]

Originally Posted by maphd

FMDenaro, thank you for your reply. I am not very familiar with large eddy simulation. That's why my questions might be silly.
How I can read smallest length scale of large eddy simulation in fluent? If you know the answer, could you please tell me little more clear step by step directions?
I really appreciate it,
Thank you in advance...

the smallest flow structure you can see is of the order of the computational grid...
however, at this level of resolution Fluent has a relevant second order error to be considered.

[saeedi]

Flippo is right.

BTW, in case you want to figure out the Kolmogorov scale roughly, you need to calculate the approximate viscous dissipation rate from the simulation and then based on that you can calculate the Kolmogorov length scale roughly.

For viscous dissipation, you will need the strain tensor based on the fluctuating velocity.

The reason I said roughly, is that the viscous dissipation rate you get from the software may not be the exact value as you have some truncation error, etc. from your simulation as well.

[maphd]

Saeedi,
Thank you for your reply. Yes I need to calculate kolmogorov length scale roughly. if you know how to do it, could you please give me some step by step explanations on how to do it in fluent.
I really appreciate any help. Thank you in advance.

[saeedi]

Sure,

As I said, first you need to calculate the viscous dissipation rate.

For that, you need to have:

1-fluctuating velocity fiedl
2-calculate Sij' based on it
3-calculate the viscous dissipation rate (epsilon) as (0.5 *nu*Sij'*Sij
5-use the formulation for Kolmogorov length scale ( (nu^3/epsilon)^0.25)
6-you can then do time averaging for the obtained value (do it at multiple time and average it)

The way you extract the data on Fluent, I do not know. I usually do not use software.

[maphd]

Saeedi,
Thank you very much for your reply.
I relly appreciate your help

[FMDenaro]

I never directly tried a comparison but I suppose that a rough estimation could be done by considering that at the Kolmogorov scale, the Reynolds number is O(1).
Therefore delta = ni/ud being the velocity ud an intensity of the fluctuation at that scale

[maphd]

Dear Saeedi,
Do you know how to calculate Sij?
Thank you

[saeedi]

Sij is the strain rate tensor.

Assuming you are familiar with it:

Sij=0.5(duj/dxi +dui/dxj)

[maphd]

Saeedi,
Thank you for your reply.
On your step by step calculation on 3rd step it will be 2*nu*Sij*Sij instead of 0.5*nu*Sij*Sij , is that right?
Thanks,

[saeedi]

Yap, That's true.

Sorry about that. I wrote it in rush.

epsilon=2*nu*Sij'Sij'