CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   [Q] Convection term treatment in UDS (http://www.cfd-online.com/Forums/fluent/34789-q-convection-term-treatment-uds.html)

Ryan, Lee October 7, 2004 10:51

[Q] Convection term treatment in UDS
 
I have difficulty in dealing with convection term in UDS. I'm currently solving species transport equation using UDS in FLUENT.

Generally, as you know, phi-variable in general transport equation is "mass fraction" (or mole fraction). In this case, convection term is defined using "UDS_FLUX" and unit of the "UDS_FLUX" is kg/m^2.

However, when concentration is considered as a phi-variable, it's not easy to treat convection term in governing equation. Of course, there's a way to deal with this problem by multiplying density to all terms in governing equation but in my case, it's impossible to know the density value in calculation domain.

So, what I'd like to ask is "is there a way to deal with convection term properly without multiplying density when concentration is considered as a phi variable?"

I appreciate any comment.

fluidwork October 15, 2004 04:32

Re: [Q] Convection term treatment in UDS
 
Why did you say that you can't know the density value in calculation domain? FLUENT provides several options for definition of the fluid density: constant density and temperature- and/or composition-dependent density . So, I think you can obtain the density value through accessing the solver data by UDF every iteration. I don't know whether it's right or not. BTW. When you add the transport equation about concentration in FLUENT, do you need to define the concetration first ? Or, if you just need to define the convection term, the diffusion coefficient, source term and the boudary condition, then the UDS transport equation about concetration can work well?

Ryan, Lee October 15, 2004 05:04

Re: [Q] Convection term treatment in UDS
 
Thank you for your interest about my question. Fortunately, I've solved the problematic situation. Anyway, answer to your question is like below.

When you add the transport equation about concentration in FLUENT, do you need to define the concetration first ?

-> Yes, of course. In this case, concentration is not a calculated value using UDM but a "phi" value in general transport equation. So, proper modification of general transport equation is required. And then, as you mentioned, you only have to define convection term, diffusion coefficient, source term, boundary condition, etc.

fluidwork October 17, 2004 23:40

Re: [Q] Convection term treatment in UDS
 
Would you mind introduce more details about this question? How to define the concentration as the 'phi' variable and how to define the convection term in FLUENT.Because there's not enough introductions about the general transport equation in help document of FLUENT. I'm also interested in this kind of question. Thank you.

Ryan, Lee October 18, 2004 10:20

Re: [Q] Convection term treatment in UDS
 
My system is a multi-layered one. In some regions(computaional domain), there are convection and diffusion effect. In some other regions, there's only diffusion effect.(Or convection effect is negliglble).

Basically, when the convection effect is not negligible, concentration-based governing equation is impossible in FLUENT. (at least to me) In the previous comment, I told you "I fortunately resolved the problematic situation". That means I aprroached the governing equation with diffusion term and source term. As you can guess, in this case, concentration-based governing equation is possible and finally concentration value can be a "phi" variable.



All times are GMT -4. The time now is 13:14.