Hi,

I've just started working in CFD. Could anyone help me understand DNS ? I've few notions: 1. DNS is computationally time consuming 2. We need a very fine grid to carry out DNS

(how much fine my grid should be) 3. I can get 1/7th power law profile using DNS of pipe flow for Re>4000

Are my notions correct ?

Thanksa lot

Hi, DNS means directly solving the equations without any models, I mean if u consider RANS , here we are solving the averaged navier stokes equations , in LES, we are separating the flow in to tow parts one large eddy and another one is dissipated K.E , which occurs due to viscocity and this will be very fine scale. we are ressolving the large eddy and modeling small scales using smagorinsky or Germano dynamic model. where as in DNS, we are not manipulating NS equations , we are solving i,e we ressolving fine as well as large eddies,,if we want to ressolve both the scales(larger and smaller), our grid should be very fine to capture small dissipated scales(kolmogorov scales),,u can refer any CFD book, for the formulae to calculating kolomogorov scale..

all the best

Can anybody tell me where I can find some reviews or textbook on DNS? Thank you.

Hi, could you please post the reference to which you mentioned regarding the kolomogorov scale calculations Cheers Bob

You can get the calculation of Kolmogorov scale from the archive of the old messages on this site. Try search.

Cheers

I wonder most DNS codes using pseudospectral method use roughly 45-60 Chebyshev points in cross stream direction and around 32 Fourier points in stremwise and spanwise directions. Can such grid in wavenumber space capture the dissipative scales ?

Cheers Abhi

Hi,

A good review about DNS is the following paper and its references:

Moin, P., and Mahesh, K., DIRECT NUMERICAL SIMULATION: A Tool in Turbulence Research, Annual Review of Fluid Mechanics, Volume 30, Page 539, Jan 1998.

About Kolmogorov scales, you can find it in any textbook for Turbulence, for exapmle:

1) Wilcox, D.C., Turbulence modeling for CFD, DCW Ind., 1993.

2) Chen, C.J., and Jaw, S.Y., Fundamentals of Turbulence Modeling, Taylor & francis, 1998.

3) Pope, S.B., Turbulent flows, Cambridge University Press, 2000.

Thanks Ali.

I don't have access to the paper by Moin and Mahesh. Does anyone have the soft copy of this review article ? If so, I would greatly appreciate if you can send it by email.

thanks Regards Phillip

Thanks a lot Ali for sending me the paper.

Regards Phillip

direct navier stokes, usually

dns = direct numerical simulation