CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Main CFD Forum (
-   -   k-w turbulence model (

Aaron January 6, 2003 13:59

k-w turbulence model
Recently I have been investigating the Wilcox k-w turbulence model, and I have a question that hopefully someone(s) can help me with.

Wilcox cites the A.A. Townsend approximation for k as k~t^(-n). I am unclear about what time span the "t" is referring to. Looks like the units don't match, though I'm not sure if they have to in this case.

An explanation of the equation and the "t" would be greatly appreciated.

Thanks in advance.

Aaron January 7, 2003 15:21

k-w turbulence model: update
After reviewing the equation, my colleagues and I can only conclude that the "t" is actually a "Tau" or a stress tensor without subscripts, and not a time scale or time span. So at a minimum, its a confusing simplification, and possibly a typo. Any comments?

Alec Browne January 8, 2003 06:33

Re: k-w turbulence model: update
I'm assuming you're referring to Wilcox's turbulence and CFD book published by DCW industries.

They have a web page devoted to known typos, and reporting newly-discovered ones:

I suggest you check it out! Good luck

Aaron January 8, 2003 21:59

Re: k-w turbulence model: final word
After sharing e-mails with the author, I've learned that the equation is correct as written k~t^(-n) and t is indeed time. The time scale for this t is infinity and leads to the asymptotic solution for k.

Bredberg January 20, 2003 10:58

Re: k-w turbulence model: final word
Townsend gives the relation k~t^-n for the (initial) decaying (grid) turbulence and give n the value = 1.25+-0.06. Note that n rises to 2.5 in the final decay. This physical behaviour is one (good) reason for introducing turbulence govern (NOT y+) damping functions. Wilcox and others uses this relation to set the value of the coefficient for the destruction term (C_eps2, beta) in the epsilon/omega-equation.


All times are GMT -4. The time now is 20:39.