half porous model for distributor

[Bionico]

Hello.
I am working on an optimization project of a burner door for a boiler: it has the task to distribute the air+gas mixture from the fan to the heat exchanger as homogeneously as possible.
I have to check how to distribute the output flow from the distributor and I find it difficult to model it: it is a plate of 1 mm thick of micro perforated steel and the passage is further obstructed by a superimposed wire mesh. Moreover this plate is not flat but slightly arcuata
I think that the porous jump model is not good in my case because the flow that invests this element is not perpendicular, so I would be positive for the porous media; However, I need, in both cases, to define inertial and viscose resistance (in three directions if I do not use porous jump): without experiments is it possible?
I'm graduating so I'm not very expert on cfd, I'm learning, I hope you can help me :smile:

 

[stefano.garbin]

the thing is done experimentally, having in the middle of the pedestrians the wire mesh.
on a straight tube interpose the perforated sheet and the net. then go to "flux", i.e. impose the mass flow (or volumetric) of air and mix the delta pressure.
If you had only the perforated sheet you could try to simulate a small model and then use the results to define the coefficients of the porous medium.

 

[Bionico]

first of all thanks for the answer :smile:
Is there a way to know at least the order of magnitude of these parameters? I don't think that their limited variation is decisively affecting the result
the fact is that I am not sure I can carry out this experiment and I would like to avoid it possibly

 

[stefano.garbin]

It's been four years since I stopped dealing with premix burners in perforated sheet metal with external wire mesh. I had done a lot of simulations but I don't remember any value anymore unfortunately. . .
But the experiment is worth it if you can.

 

[Bionico]

I'm sorry
I managed to make the experiment! now I have two questions
the square interpolation of data results to me: dp/dm= 0.277v^2 + 0.136v
1) would result in 2 negative values per alpha and c2 (taking into account the darcy law): Why? Perhaps the experiment with reduced speed (of the order of 2 m/s) is not reliable but would it be better to use a more powerful fan?
2) alpha is very small, so viscose resistance is negligible (I had already guessed why it is a perforated plate), but in fluent can I set the permeability value equal to zero?

 

[Bionico]

Sorry, I correct the equation (I took the wrong one)
== sync, corrected by elderman ==
from here I get c2 = -450759,2 and alpha = -1,49e-10
c2 value (other than negativity) has a plausible size order considering that it is a 1mm perforated plate?