Learning Forum

[Ali Qasim]

Hi everyone,

I am modeling a cross-flow energy exchanger for heat transfer from humid air to a cold fluid separated by the membrane. I have performed the simulation and for validation purposes compared the results with experimental results and analytical calculations. I used the effectiveness-NTU method for the analytical calculations of sensible effectiveness.

Surprisingly, there is a huge difference between analytical and numerical simulation values.
I have used symmetry boundary conditions in numerical analysis (half domain is modeled with half flow rate).

In the case of analytical calculations, do I need to use “half domain with half flow rate” or “full domain with full flow rate”??? In both cases, the results are different.

Regards,

[Kishan Konannavar]

Hello

How does the numerical solution compare with experimental results?

It is possible to model a smaller domain if the actual geometry and physics allow. If the reduction in domain is appropriately accounted for in the form of geometry and boundary conditions in Fluent, you should get accurate results. 

Given that different analytical results are obtained for "full" and "half" domains, I would suggest going with full domain, which would be simpler. "Halving" a domain needs to be appropriately accounted for in the analytical treatment as well in the effectiveness relations that may have been used. 

If not checked out already, I would suggest visiting the following materials on AIS which could be of some help

How Heat Exchangers Work | Ansys Innovation Courses

Forced Convective Heat Transfer in a 2D Staggered Tube Bundle — Simulation Example - ANSYS Innovation Courses

 

Thank you

Regards

Kishan