there is an error on wiki's page about 'Prandtl's one-equation model'

the Cd value should be 0.08 not 0.3

can anyone re-edit it?

You have done some impressive detective work to find that out! Reading in Wilcox's book I understand how this error occured. Wilcox talks about another constant which was 0.3, but I don't think that he even mentioned what the Cd constant was in Prandtl's one-equation model. I have not found the original reference to this model. I found some other references (Emmons 1954 and Glushko 1965) that used a Cd varying between 0.07 and 0.09. Your 0.08 value sounds okay then, but have you got a better reference for the actual Cd used in Prandtl's model?

Prandtl's one-equation model is rarely used now and perhaps we even should remove it from the global list of turbulence models and instead only link to it from a separate one-equation page. We can't have every model directly linked from the global list of turbulence models. What do you think?

log on http://www.simuserve.com/phoenics/d_...cs/lec7-22.htm then click on one equation models

Yep i we refer wilcox book, he mention that the Cd values varies from 0.07-0.09 (see page 77) and i also refer to the book from G. Biswas, V.Eswaran ( Turbulent flows. Fundamental, Experiments and modeling) they state that the Cd values should be 0.08. even though he did not mention why. i still dont get why the phoenics state Cd=0.1643.

this model is rarely used but i think it should be mentioned also, there is alot of other models refering to it. for example 'Baldwin-Barth model'. that's what i think.

I changed the constant and wrote a little footnote about this in CFD-Wiki. You can see what I wrote here. Please change it if you want it to look differently.

Dear Nurudin,

It has been years since I looked at Prandtl's 1 eqn model (I do not have Wilcox book handy).. However, be careful that not all L's (length scales) are the same in all references. Sometimes they refer to mixing length, or sometimes to characteristic scales.

For 1eqn models, you compute mu_t from

mu_t = C_m * k^0.5 * L_m

and

epsilon = C_d * k^1.5 / L_m

Substituting epsilon into mu_t (via L_m)

mu_t = C_m * C_d * k^2 / epsilon

which is the traditional 2eqn model for turbulent viscosity.. That means the product of C_m * C_d must gives close to C_mu ~ 0.09.. That means that the definition of L_m will change the value of C_?

Hope this helps,

Opaque.

im quite loss with phrase "definition of L_m will change the value of C_?" what that suppose too mean?

by the way..thanks to jonas larsson, u done a nice edit there. quite clear with all possibility of Cd values. thanks!

sorry, right now im understand what opaque mean by "definition of L_m will change the value of C_?"

so it seem we need to re-edit the Cd part. because according to CHAM, Cm=0.5478; Cd=0.1643, Cm x Cd =0.5478 x 0.1643 =0.09 thank to both of you...

Yes, that explains the difference. I have changed the text in CFD-Wiki a bit again to better reflect this. I hope that you like it. Feel free to add and change things in CFD-Wiki yourself. CFD-Wiki should be written by the visitors, not just me, and if we are going to succed with CFD-Wiki and keep it expanding it really needs additions from people like you guys.