I am currently simulating an entry hall of a large Building to find out the heat built up inside this hall. The medium is of course air (thickness <1) and I have some semi transparent faces (glass). So I used the solar calculator to generate the heat quantity in conjunction with the DO radiation model to simulate the radiation heat. But I always get negative temperatures on the floor and side walls, even when I set an adiabate temperature for them. The temperature values of the glass ceiling seems ok (about 90°c) I tried standard values and advanced settings as described in the fluent manual, tried the DO model without the solar module and even defiend all glass walls as opaque. But its always the same. The solution becomes converent but the temperatures are totally unrealistic. Could please anybody help me ????

Best regards

What did u put for the absorption and scattering coefficients of air? It should be set to a small values other than zero (e.g. I put 1e-09).

Also, try increase the angular discretization i reckon?

Hope this helps.

Hi, I put 0 for absorption( constant), 0 for scattering (isotropic),and refractive 1 (constant ) under materials.

For concrete ( the floor) I put the same.

Does it make any difference to put 1e-09 or zero ?

I increased the angula dis. for divisions up to 10 and the pixels to 4 and even added 4 bands ( under non gray ) ( I did several simulations with increasing difficulties )

But the result was alway the same, so I dont think is has something to to with the angular dis. or gray bands.

I will try this evening another simulation with very low absorpion factors. Maybe something will change.

What values would you set for Air and the floor and glass materials ??

Thank you very much for your help !!!! I will write the result tomorrow.

I tried it out but sadly it did not work with this parameter or other ( tried 0,1 0.5 and 1 as coefficient). Any other ideas ???

Even i hv a similar kinda problem ... i hv a reactor which is nearly 1 mt by 1 mt n the walls ar heated at 1160 C the gas in it is h2 ( later i will introduce SiCL4 and CH4 too, so it will involve reactions too) .. i used the rng k e model and used the DO model as suggested by fluent but for that i need to input the scatering coeffients and the absorption coefficient which i m not able to get even on the internet .. m really stuck .. help me

I am also experiencing a weird problem with the radiation model. I am solving a coaxial conterflow heat exchanger: there is hot gas on the outer shell, and cold gas in the central tube. I need radiation in the outer shell (T>1000C), but when I turn on the radiation module, the separation wall between the two gases becomes extremely hot, even hotter than the hot fluid. I use DO and I specify the WSSGM, wall emissivity 0.6. Any experience with that?! luca

no dude/dudette

Dude. That is really weird. I am trying everything comes into my mind, but there is really no reason why the wall temperature should rise. After a while the wall is radiating more than the hot gas, and the gas temperature rises instead of decreasing, really weird! luca

Hi, for mehane you can take an abs. coef. of 0 cause it is a non visible gas. The scattering coef. therefor should also be 0 ( or very very small) and Siliciumtetraclorid is also invisible so I assume you can take the same values...

so u mean to say that ..all the gases behave same ... as far as i know CH4 is a greenhouse gas so it should absord more heat ... so should behave differntly towards radiation

Of course they behave different but as far a I understand the fluent manual, the absorption and scattering coef.are dependent on the visibility of the gas. This means that all invisible gases have the same values ! the other properties as densety cp value etc. are different.

If I get it wrong please tell me !!

no even i was also trying to contradict my own self ... i also agree with u ..but i m really confused ... i think radiation must has something to do with absorbtion coefficient ...

moreover in the same kinda problem .. fluent itself gave a value of 0.01 for absrobtion coefficent.. for the mixture of the gases( SiCl4,H2 and CH4)... but that was i think only for showing ..n were not accurate ..but still they gave something

Did you call them for the coefficient or where did you get it from ?

And your thinking is right. Radiation has someting to do with this coef. the greater the coef. the more rad. is absorbed by the medium and the smaller e.g. 0 in air the more rad. gets through.

But my problem still exists, do you have any clue why I get temperatures belw 0°c on a hot summer day ? ( see my first post)

the company for which i m working now ... they had a differnt model made for them from fluent itself i saw it ... n i really dont hv ne idea... abt the deatils of DO model..i can give a email id from fluent ..may b he can help u hes is fluent suport helmi@fluent.be

hii have u ever implemented a udf as a boundary condition ?? cos i hv know temp at certain positions of my wall and i want to put them as a boundary condition , so in other words its like :

x,y = 370,0 temp = 600 C x,y = 370,50 temp = 650 C . . . n so on

i m not able to write the C code .. i hv no idea what will be the inputs and how the fluent will understand the output .... can u please help me .. i thhink its a very short code ...

plzz i m stuck badly .. help

I have written UDF's but never for boundary conditions. It takes a while to understand how to do it, and I think You should just read the manual and practice. There are many examples similar to the one You are trying to implement. luca

The absorption coefficient for methane-air reaction can take as 0.5 and scattering COEFFICENT AS 1E-09. i hope this will help you.

Hey, i accidentally saw this! Most probably the reason for ur abnormal temperature. Taken from Fluent Manual--user guide (pg. 12-47): 'THe thin wall treatment should be used for semi-transparent walls only where abs or emission by the walls is not significant. In cases with significant emission or abs of radiation in a participating solid material, such as the absorption of long wavelength radiation in a glass window, the use of semi-tranparent thin walls can result in the unphysical temperatures in the numerical solution. To avoid potentially erroneous results, one or more solid cell zones should be used to represent the solid material in such a case.' ...

Hopefully this really helps~!