Learning Forum

[charlie hurley]

Hello everyone,

I am currently working on a CFD model in ANSYS Fluent where I am simulating the condensation of moist air in a small duct. The setup involves a relatively warm moist air stream domain separated by a wall from a parallel cold air domain. I am using the Lee condensation model and am encountering some issues with pressure and temperature on the hot side. I have tried various different viscous models such as k-w and k-e with enhanced wall treatment, but this didnt seem to resolve the issue. I have also verified mesh independence. (I have tried the Eulerian Wall Film model aswell but switched back to Lee model due to some problems).

Any guidance would be greatly appreciated.

Here are the details of my setup:

• Duct Domain: Width of 1mm with symmetry boundary conditions.
• Turbulence Model: k-epsilon with standard wall functions.
• Condensation Model: Lee condensation model.
• Species Transport: Enabled to account for moisture.
• Boundary Conditions:
  • Inlet: Velocity-inlet with the initial inlet pressure set to 0 Pa gauge.
  • Outlet: Pressure-outlet with 0 Pa gauge pressure.
• Numerical Schemes: SIMPLE scheme with first-order upwind discretization.

Problems Encountered/Questions:

1. Pressure Issues: Despite setting the inlet to 0 Pa gauge pressure (atmospheric), the pressure at the inlet is showing as around 75 Pa gauge pressure for both duct (as shown below) with the cold domain pressure change being slightly higher than the hot domain. As my inlet pressure is atmospheric, shouldnt my inlet pressure be 0 Pa gauge, not the outlet? Or is this normal?
2. Flow Behavior: Experimentally i measured a pressure drop of 75 Pa for hot duct and 34 Pa for hot cold duct, due to condensation blocking the path of airflow. The model does not seem to account for condensation properly, so it doesnt give this larger pressure difference in the hot duct. Is there any way to resolve this?
3. Temperature Issues: The hot side temperature change across the duct is much larger than experimental results, this makes me think the latent heat of vaporization isn't being accounted for properly. Is there a way to fix this? Or could this be related to the above issues?

Any insights or suggestions on how to resolve these issues would be greatly appreciated!

Thank you in advance for your help.

Charlie

[Rob]

You've not set the inlet pressure as zero, read the panel carefully. So that's behaving as expected. 

Try just modelling the moist side and set a convection boundary for the cold side. It's fewer cells and probably easier to troubleshoot. Domains with separate fluid zones tend to be more difficult to work with. 

Latent heat is accounted for, but as I don't know what you've set relative to what you think you've set and whether the results are anywhere near converged it's somewhat difficult to comment. Condensation models are not simple, so chances are you've missed something or haven't correctly adjusted the model parameters. 

[charlie hurley]

I have tuned the condensation coefficients in the Lee model to match my experimental results for condensation rate, and i have also ensured that the results are converged via report definitions and running the model until all residuals sufficiently low, which still gives the same results.  

[Rob]

If you show more results I may be able to comment.