Hi..there,

I need your opinion about what I'm thinking

The turbulent viscosity can be written using the low k-e model like this nu_t=cmu*fmu*k**2/e fmu : dampling function cmu : constant value from standard ke model.

Generally, the dissipation term is calculated from the trasport equation.

But, I'd like to use only the one-equation to make the turbulence model.

if we adopted the mixing length for e, we can write e=k**(1.5)/lm lm: the mixing length of Van Driest

Finally, the turbulent viscosity is nu_t=cmu*fmu*(te**0.5)*lm..

Is this resonable to compute the turbulence values with Navier Stokes equation? Is there any mathematically discrepancies in this one-equation?

Please, tell me anything about this question Have a nice day

It's OK, you are allowed to do that. The question is where you are going to get the distribution of the mixing length. If you can guess it correctly, there's no need to solve an equation for it. However, guessing the mixing length is not easy and will depend on the case you are trying to simulate.

Hrv

Thank you for your help,

What do you think if I use lm as Cebeci-Smith length model or lm(l_e) from Norris and Reynolds's damping function under a near wall region (z+=around 100)?

lm=cappa*z*(1.-exp(-z+/a) from Cebeci-Smith model

or

lm=Const*z/(1.+5.3/Re_z) from two-layer model

which is more reasonable to compute the turbulent viscosity and turbulent kinetic energy?

Jay