[nymand94]

Hello

I am simulating the flow through a hvac system.
To start off I am using simpleFoam, laminar for simplicity. I intend to add turbulence later on, when things are running.

The problem:
In the HVAC system, there is a filter, which I know the pressure drop across, as a function of massflow.

I want to simplify the simulation as much as possible. I am considering 3 different approaches.

Option 1:
Remove filter from the simulation entirely, and link the the patches of the inlet and outlet, using cyclic, cyclicAMI or fan constraints

Option 2:
Pretty much the same as option 1, though using the more recently released NCC - Any advice of pro/cons for either is appriciated.

Option 3:
Simulate the filter as a porouse region. converting the whole simulation to porouseSimpleFoam, ballparking the values of the DarcyForchheimer coeffs.

I have a attached a picture of the situation. please let me know if i need to clarify anything.

Any advice is appreciated. - Do I need to consider any 4th option? NCC vs AMI

[Yann]

Hi,

I would go for the 3rd option. No need to switch solver, you can use porous zones in simpleFoam thanks to the explicitPorositySource in fvOptions (or fvModels)

If you have the pressure drop of your filter function of velocity you can compute the Darcy-Forchheimer coefficients.

Regards,
Yann

[nymand94]

Thank you for the advise.

Do you know if it is possible to have a region with a specified pressure loss? I found the fanPressure patch, but I think a pressure loss/gain region would be better in my instance.
Else I would use the patch internally at the outlet of the filter

[Yann]

For any source term applied to a zone inside the mesh, I would say fvOptions/fvModels is the way to go. You can have a look at the different sources available but explicitPorositySource seems to be appropriate.

If you prefer to set the pressure drop on a patch you might find jump conditions such as fixedJump.

Yann