CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   FLUENT (https://www.cfd-online.com/Forums/fluent/)
-   -   Radiation in Fluent (https://www.cfd-online.com/Forums/fluent/45054-radiation-fluent.html)

Arpit June 16, 2007 12:50

Radiation in Fluent
 
Hi,

I am solving a problem in fluent which has a rectangular geometry and a semicircular cavity on one of its edges...The cavity has air in it while the rest is solid aluminium. The diameter of the semicircle is at constant temperature (1200K) and there's radiation in the cavity and then the heat conducts into the solid. Can anyone tell me how to do this in fluent.

Do we really need a radiation model or boundary condition is suffice? Also what boundary type has to be given to the semicircular part (interface between air and aluminium)? Please let me know..

Thanks in anticipation, Arpit!

AJ June 16, 2007 14:32

Re: Radiation in Fluent
 
If you want to capture radiation heat transfer, you have to select a radiation model. You could switch ON simplest model (S2S or P1) that computes view factors and provide you heat transfer through radiation.

Fluent has higher level models (DO) as well may be better but computationally intensive.

Many of these models require you to input the surface emissivity etc which are important for radiation HT.

Hope this helps. AJ


AJ June 16, 2007 14:34

Re: Radiation in Fluent
 
Read this:

http://www.fluent.com/about/news/new...4v13i2/a21.htm

AJ June 16, 2007 14:38

Re: Radiation in Fluent
 
I think better will be to know the Adv and limitation of each model before you select one:

http://simba.ara.bme.hu/~cfd/fluent6...ug/node575.htm

Decide if this helps you. AJ

Arpit June 16, 2007 16:24

Re: Radiation in Fluent
 
Hey thanks for the reply!But the problem remains..

I tried using S2S model but it asked for a view factor file :(! What is that about? Also the most critical thing in my case is what boundary type i should give to the semicircle..should it be a wall/interface/interior or what :(?

Please help! Thanks. Arpit

Ahmad June 16, 2007 20:45

Re: Radiation in Fluent
 
It seems that you are dealing with a 2D problem, so the DO model won't be that computationally expensive. The model will ask you for emissivities of the materials you are using (Al, etc.), along with absorption and scatering coefficients, etc. (in the materials panel), which you can easily get from any heat transfer book. The DO model doesn't require view factors, and you can leave the default settings as they are. I understand that the air is trapped in the cavity with the base wall being fixed at 1200 K. If this is the case, care should be taken when modeling the air thermal expansion, and the density and other properties (in the materials panel) must not be left as constants ( try choosing the ideal gas law for density and kinetic theory for the rest). In Gambit, you should specify two regions (Continuums), one as a solid (Al block) and the other as a fluid (air in cavity), so the interface will automatically be the solid edge of the Al block, and you don't have to specify a BC in the interface.

Good Luck

Arpit June 17, 2007 12:21

Re: Radiation in Fluent
 
Hey Ahmad..you have got the problem very much...Continum definition is fine but what do we give in 'Specify Boundary Types' for the semicircular part? If we give it as wall...the heat does not conduct into the solid...I have tried interface and interior BC type but that also didn't help :( What to do for that? If one doesn't define it..fluent is automatically taking it as wall..

Tomasz Bednarz June 17, 2007 18:02

Re: Radiation in Fluent
 
Hey Ahmad you're a moron

Ahmad June 17, 2007 20:43

Re: Radiation in Fluent
 
Hey Tomasz

I totally agree with you. But could you please tell me why? I mean from a CFD point of you.

As for Arpit, I apologize for not helping.

Kind regards to all.

Arpit June 18, 2007 05:09

Re: Radiation in Fluent
 
hey..don'y say this..you guys are my last hope :(! I can't figure out what to do..please help me out!!!!!

Tomasz Bednarz June 18, 2007 08:16

Re: Radiation in Fluent
 
Ahmed,

No CFD reason, I just don't like you. You are a moron.

Tomasz.

Ahmad June 18, 2007 12:41

Re: Radiation in Fluent
 
Hey Arpit

You don't give any thing for the semicircular part in 'Specify Boundary Types'. By default, all external edges (faces) of the domain are considered as walls and all internal faces are considered as interiors. Hence the interface between the air and the aluminium block is an interior by default. Maybe you have another problem! Check if you have two edges/faces on the same semicircular interface. If so, you can fix the problem by either erasing one of them or, even better, by using the Connect command button in Gambit. Simply go to edges, click the connect command button and choose all the edges in the domain. The extra edges/faces will be erased leaving you with the necessary edges only (you'll get a messege if you have more than on edge/face in the same place).

I apologize if I haven't understood the question again.

Kind Regards

Arpit June 18, 2007 13:57

Re: Radiation in Fluent
 
Thank you very much Ahmad..I will try with the interior type..by default it doesn't take as interior though for me..Anyways I shall try and get back to you...

Thanks! Arpit


All times are GMT -4. The time now is 14:34.