Learning Forum

[Canh Doan]

Hi everyone,

I have a simple 2D mesh as the figure below.

I want to directly assign the laser intensity (W/m2) to a particular cell.

Is it possible in fluent?

If possible, how should I do it?

Thank you so much.

 

[Canh Doan]

[Canh Doan]

This is my figgure:

[Rob]

Have a look at source terms. 

[Canh Doan]

Hi Rob, thank you for your reply.

 

Can you explain more about source terms? I am a fresh student in this field so I might misunderstood your words.

 

 

[Canh Doan]

Hi Rob, Do you mean the DEFINE_SOURCE term is UDF code?

 

As the UDF manual, it is possible except for the discrete ordinates radiation model. However, as far as I know, the radiation heat transfer and the interaction between laser radiation and gas should use the DO model.

[Rob]

Yes, but you can assign a source term directly without needing a UDF or expression. For energy absorbion by the media you want DO or Monte Carlo radiation models. 

If you just want to add energy into those cells (heat source) I'd split them out (register and separate option) and then use a fixed source term. 

[Canh Doan]

I have tried to add directly the radiation source to a gas medium as you said. I set the divergence angle is 0 degree (collimated beam). But the result is a diverged beam as in the picture below. I don't understand why that happens because as theory, the radiation source go into a gas with a refractive index is 1 will not change direction. Can you help me correct it?

 

[Rob]

In 2d don't select any surfaces to display on, you'll get the contours then. 

I suspect you're getting the beam (left) and that's warming the glass up (right). What conductivity did you set in the glass? Refine the mesh around the beam too - adaption may be a good choice. 

[Canh Doan]

Hi, thank you for the advice. After re-mesh, I found the result is the same as the previous case. Can you help me explain why the beam is diverging when I set "theta" and "phi" is ~0 and the beam direction? This is my result and setting:

         

   

[Rob]

What did you set the divisions as in the main DO panel? You may find you need more mesh too - look at the resolution as the beam reaches the glass.

[Canh Doan]

This is my DO model setting. I set only one band because I assume that my laser has only 1 wavelength is 1.07 um.

I just want to find the trend of the solution so I think a coarse mesh is better to save time.

I am curious that

1. Could I set a collimating beam in my model?
2. Could I assign the intensity to a particular cell inside the domain via UDF?

Please tell me is it possible or not?

 

Thank you so much

 

 

[Canh Doan]

I am worry that I cannot assign the irradiance into Ansys Fluent via UDF code because I cannot find the keyword "irradiance" in the UDF guideline documents.

[Rob]

Increase the Discretisation.  The DO model uses an angular discretisation and the default is a little low. Don't go over 3-4 or the cpu overhead gets excessive. 

[Canh Doan]

It doesn't work. I don't think it is a problem.

[Rob]

Why do you need a wavelength band for one wavelength? What absorbtion coefficient is set?

[Canh Doan]

It is only for a test case. My final target is a gas domain that absorbs laser irradiance and generates heat. Each gas has an absorption coefficient depending on wavelength so I need to indicate the wavelength band here.

The absorption coefficient was set in the air is 0 and glass is 0.05.

My target in the DO model is to make a collimating beam or a focusing beam into a gas domain. But up to now I cannot control the beam direction and divergence angle so I ask here.

Maybe you also don't know how to set a collimated beam inside a gas domain.

[Canh Doan]

Hi Rob, is it possible by using the UDF code?

[Rob]

Have you tried refining the mesh down the beam as well as increasing the discretisation? 

[Canh Doan]

I refined the mesh and increased the discretisation but the light beam is still wrong.

[Rob]

Is that with the angle on the beam as before? 

[Canh Doan]

Yes. I set 2 cases with the same angle.

[Rob]

Rotate the domain by a degree or so. Then read the DO documentation to explain why.