Learning Forum

[hurch346]

Hello,

I am currently trying to model a segment of a plate heat exchanger, i.e. two air streams separated by a plate (one hot and one cold). The hot air is humid and is meant to condense on the plate, which should be shown by the Eulerian wall film model. 

 

I am using species transport, coupled with EWF to do this. However, I have run into an issue. The temperature fields, and species mass fraction fields should look have a triangular gradient, due to the cross flow configuration. When i turn EWF model off, the pattern looks correct, as below: 

However, when i use EWF model, this pattern becomes uniform across the y axis for the film thickness, and temperature, as shown below:

 

Here are my settings for EWF:

For my boundary conditions, i have set an inital condition for the film thickness. And a phase change driven by the wall-boundary-layer model. 

To check convergence, i have been using report definitions for film height and film mass transfer. I dont believe this is a mesh issue either as i have tried various different meshes. Any help would be greatly appreciated, thank you.

 

[Rob]

If you set an initial condition for the film how long have you run the film (solver time steps) relative to the domain residence time?

[hurch346]

Thanks for the reply, I think maybe i didn't run the solver for long enough. There is 10 sub iterations per iteration, and initally i was running around 300 iterations. By running for more iterations i've noticed that my report def plots begin to increase quite late, this is making me unsure about when to stop the simulation. Would you recommend stopping when these report plots flatten out?  P.S. this is in steady-state

Thank you 

[Rob]

Yes, you need to let the model stabilise. Check the film updates in the Console window for film time. Then compare that with the domain plug flow residence time.