[maphd]

Hello All,

I want to calculate turbulent dissipation rate in Large eddy simulation in fluent 14.

Could anyone please give me some step by step calculations on how to do it in fluent?

Thanks,

[LuckyTran]

consult either of these two:

http://www.cfd-online.com/Wiki/Turbu...ssipation_rate


http://www.cfd-online.com/Wiki/Intro...kinetic_energy



what you need are the spatial derivatives of the fluctuating velocities (so you can then assemble the symmetric part of the fluctuating strain rate tensor). This is somewhat challenging to do since you don't know your fluctuating velocities until you have converged mean statistics, which I think is the hardest part of this problem. Once you get the fluctuating velocities, you're nearly there.

[maphd]

LuckyTran,

I calculated fluctuating velocities. But do you know how to calculate the first derivatives (partial derivatives) of the fluctuating velocities? I need those to calculate the strain rate.

Thank you,

[LuckyTran]

Wow this is an old post!

I think it would have been much easier to just work in instantaneous strain rates and then determine the fluctuations. Not sure why I didn't think of this back then.

[CeesH]

Hi Maphd;

what works for me is to set up a custom field function:

[eff-viscosity]*[strain-rate-mag]*[strain-rate-mag].

Export the values over a number of timesteps and average. Of course, check if the statistics converge.

Cheers!
C

[asmita_iitb]

hello,
Q.1
q(heat flux)=h(Tamb-Tw)

q- surface heat flux
Tw - wall temp
q,Tw unknown


how to update Tw and q through udf in fluent?

[maphd]

Hi CeesH,
Thank you for your reply. Actually, I am trying to calculate strain rate then from there I will calculate turbulent dissipation rate.
Is the formula you wrote giving me the strain rate? I need to calculate fluctuating strain rates, then from 2*NU*sij*sij I will calculate turbulent dissipation rate (Nu: kinematic viscosity, sij: flctuating strain rates).

Hi LuckTran,
Thank you for your reply. Do you know how I can calculate instantaneous strain rates? is the above formula gives that to me? Can I calculate strain rate and then calculate turbulent dissipation rate as I explained above?

Thank you,
I appreciate your help.

[CeesH]

Why do you want to calculate the instantaneous strain rate?
FLUENT gives that under derivatives - strain rate. You are indeed right it should be nu, not mu.

[maphd]

CeesH,

I think I got confused on this. Yes you are right, fluent gives me instantaneous strain rate.

the formula you wrote in your first post, what does it equal to? Is it strain rate?
Thanks,

[LuckyTran]

Velocity gradients and strain rate is already available as a Fluent variable (which is why it would be easier to do it this way). You don't need to calculate the derivative yourself. I actually don't recommend to calculate gradients from velocity, use the Fluent one. If you calculate it yourself then you need to discretize it yourself.

The reason for the nu and mu confusion is the definition of turbulent kinetic energy. Usually k is defined as turbulent kinetic energy per unit mass (turbulent kinetic energy is most often a misnomer). Whether k is the actual kinetic energy or kinetic energy per unit mass will determine whether your dissipation rate will be by mu or nu.

[CeesH]

yes, what I did was implement the custom field function where Sij is the strain rate magnitude found under derivatives, and the effective viscosity is the kinematic one.

[maphd]

CeesH,

So, you wrote the formula for turbulent dissipation then?
Also there is no strain rate magnitude in fluent, I can choose strain rate under derivatives. do you mean that one?
Thanks,

[CeesH]

that is the strain rate magnitude yes defined as sqrt(2S_ij S_ij)

[marialhm]

hi, maphd, have you succeeded in calculating turbulent dissipation rate in LES, and how? I also want to do so.

Many thanks.

Maria