[wind81]

Hi, I would like to model upper room uvc system. I found out that most of the work done is using fluent DO radiation model. It is possible to model it using cfx monte carlo model ? Anyone know how to set the radiation model using cfx? Thanks

[ghorrocks]

I have no idea what an "upper room uvc" system is.

CFX has several radiation models, including a Monte Carlo model. Have a look in the documentation and tutorials for how to use it.

[wind81]

Originally Posted by ghorrocks

I have no idea what an "upper room uvc" system is.

CFX has several radiation models, including a Monte Carlo model. Have a look in the documentation and tutorials for how to use it.

Hi,ghorrocks thank you for your reply. Upper room uvgi, the the uv light device mounted above the head level in the room for disinfection. From the literature, I have saw that people are using the fluent DO radiation model but cfx doesn’t have the DO radiation model and there is lacking of documents for UV modeling using CFX. May I know anyone have experience model uv using cfx?

[ghorrocks]

The Discrete Ordinates radiation model in Fluent looks like the Discrete Transfer model in CFX. Have a look to see if that does what you want.

UV light is just a band of the electromagnetic spectrum. The radiation models in CFX are designed to handle the entire electromagnetic spectrum. So the models in CFX should be able to handle it, and experience in modelling any other part of the EM spectrum will be useful as you model in the UV part of the spectrum.

[wind81]

Originally Posted by ghorrocks

The Discrete Ordinates radiation model in Fluent looks like the Discrete Transfer model in CFX. Have a look to see if that does what you want.

UV light is just a band of the electromagnetic spectrum. The radiation models in CFX are designed to handle the entire electromagnetic spectrum. So the models in CFX should be able to handle it, and experience in modelling any other part of the EM spectrum will be useful as you model in the UV part of the spectrum.

Thanks ghorrocks, let me try on this.

[Gert-Jan]

If you find a decent and validated way to calculate the UV intensity in 3D using the Discrete Transfer method, I would also be interested. I never managed to obtain reliable results.

[wind81]

Originally Posted by Gert-Jan

If you find a decent and validated way to calculate the UV intensity in 3D using the Discrete Transfer method, I would also be interested. I never managed to obtain reliable results.

Hi, Gert-Jan, yes it happened to me also. I just found that CFX Discrete Transfer method can not model directional radiation flux.

[ghorrocks]

I thought the entire point of the discrete transfer model was to model directional radiation flux. I do not use radiation modelling much, but every time I did was with the discrete transfer model and I did a validation and it gave me good results.

Don't forget it is a tunable model, like mesh size, convergence tolerance and time step size. You need to adjust it to get it accurate enough but not too big as to make the simulation too big. The main parameter to tune was the number of rays to trace or something like that (it has been a while).

[Opaque]

Originally Posted by wind81

Hi, Gert-Jan, yes it happened to me also. I just found that CFX Discrete Transfer method can not model directional radiation flux.

You are modeling a diffuse UV lamp, or a collimated UV beam (say laser)?

The directional Flux feature is for collimated beams, not for isotropic lamps.

[Gert-Jan]

My customer has quartz tubes with a plasma inside. So no laser. Probably a diffuse lamp. For the geometry of interest, they provide an intensity in x,y,z-table format that I import. Works pretty well.