[Ahmed Khattab]

Originally Posted by g.hamel

Hello,

I reopen the post because the answers are bugging me.

if we have

and
then d = b and f = a (!?), could somebody tell me if it is the case? Then why define alpha and beta?

I have read a few topics about this question but the definition of D and F remains unclear.

Thanks in advance.

Hi,

i use porous media in my model also. i searched a lot for answers for your question. really i don't have solid data but if you studies the two definitions you will find that a, b are factors for the membrane itself whatever fluid flows across it. where alpha and beta are coefficients for specified fluid with specified membrane as they are factors of density and viscosity. when alpha and beta are implemented in equation delta p becomes function in alpha. beta, and velocity. but if you use a, and b delta P becomes function of a, b, v. density, and viscosity.

hope you will find it helpful.

good luck.

[ngj]

Good morning,

The confusion about the f and d coefficients are exactly the reason that we created a small calculator for waves2Foam, such that one can define physical meaningful parameters like porosity and grain size of the material:

https://openfoamwiki.net/index.php/Contrib/waves2Foam

Best of luck,

Niels

[g.hamel]

I made my owm research about D and F coefficients and now I am sure of their definition :

http://www.tfd.chalmers.se/~hani/kur...ukurReport.pdf

When dp = a*v² + b*v

Following the equation (2) page 2

a = 1/2 * rho * F => F = 2a/rho [no unit]
b = visc * D => D = b /visc [1/m]

In OpenFoam

d [0 -2 0 0 0 0 0] [1/m²]
f [0 -1 0 0 0 0 0] [1/m]

When we compare the units, the missing 1/m comes from dp/dx where dx is the size of the porous medium in the x direction (considering an anisotrop medium).

Finally :

F = 2 * a / (rho * dx)

D = b / (visc * dx)

I obtained my a and b coefficients by fitting a polynomial of the 2nd degree using matlab and the curve dp=f(v).

With the calculated coefficients, I obtained what I wanted.
I think it is correctly explained now.

[AJAY BHANDARI]

Hi all,

Can somebody please tell me what trTU() means and how it works in Ueqn.H file in porous simple Foam. I am very much confuse about its working. From the code I understand that Ueqn.A() gives the diagnol coeff of matrix U and tensor(I)*Ueqn.A() = tTU . then inverse of of the diagnol coeff of tTU are being stored in trTU(). as trTU() = inv(tTU).

But can someone tell me the maths behind it . How the matrix is being formed. How U is being calculated from that. Plz help. Any help will be appreciated.

[Outbound]

Greetings,

I'm having a bit of trouble adapting a study previously conducted in ANSYS CFX onto OpenFOAM.

I'm getting a bit confused as to how to properly calculate and insert the porosity values into the OpenFOAM porosity properties.

My flow is simulating a submerged dense vegetation layer, so I take it as isotropic flow.

The ANSYS CFX study which was published gave the values used for the porous media permeability K and the Forscheimer tensor T.

I'm just a bit confused as to what the D and F values will be then because I see some mathmatical formulas where determining them will involve a porosity value and others where it doesn't.

Screen Shot 2016-06-20 at 15.42.05.jpg
From POROUS MEDIA APPROACH FOR RANS SIMULATION OF COMPOUND OPEN-CHANNEL FLOWS WITH SUBMERGED VEGETATED FLOODPLAINS, M Brito, J. Fernandes, JB Leal, 2015

Can anybody help me clarify how do I obtain the correct values for D and F based on the available information?

Thank you for any help you might provide,

Edit:

Am I right in assuming that

D=1/K

and

F=(2*C_E)/SQRT(K)

Or is it just that F is just the already calculated T. Aren't they both the Forscheimer tensor?

[dewey]

Hi

I want to simulate A GAS FLUID FLOW inside a PIPE that has a POROUS obstacle in the middle and that pipe has a distribution of TEMPERATURE.

Do you know what solver i need to use? or some example that i can follow or something that i can use to help me?

thanks for your time.

[AJAY BHANDARI]

Hi alberto

Welcome to the CFD online community.

Yes the solver you can use to solve your problem is porousSimpleFoam.
It is present in incompressible folder of openFoam.

Also you can first practice the angledDuct case of it which will give you pretty much idea of how to solve your problem.

Hope this helps

Best
Ajay

[Saran16]

[QUOTE=Ger_US;269500]Hi all,

I would like to use Darcy-Forchheimer law for specifying Porous Zones in an application. how to calculate the constants in Darcy-Forchheimer law.....

I am new to openfoam




how to find the a and b values ????

[AJAY BHANDARI]

Hi Saran

As you are new to open Foam I would advice you to first go through the angled duct case in porousSimpleFoam tutorial. Details of which can be found out in the following link.

http://www.tfd.chalmers.se/~hani/kur...ukurReport.pdf

You will be able to understand what DarcyForchiemmer law means and how d and f values are given.

Hope this helps

Best
Ajay

[Saran16]

Thank You for Your time ajay

[RygeltheXVI]

Hello All,

This Tech Report may prove helpful for those having difficulty understanding Darcy-Forchheimer settings in OpenFOAM.

Kind Regards,
RygeltheXVI

[Ahmed Khattab]

Originally Posted by Ger_US

Hi all,

I would like to use Darcy-Forchheimer law for specifying Porous Zones in Exhaust System application. But I am confused with coordinate system specification. Can anyone please explain what is e1 and e2?

and I have the values

a = 9.367
b = 1.029E7
alpha = 0.5 * a * density [kg/m^4]
beta = viscocity * b [kg/m^3s]

How can I calculate d and f parameters from the above data?

My guess: d= beta/viscocity [1/m^2]
f=alpha/density [1/m]

Am I correct?

if not, please tell me how to calculate d and f parameters.

thank you in advance

Hi Ger_US

According to d and f, you can say that d refers to Darian part of the Darcy-Forchheimer equation while f refers to Forchheimer part. So by comparing Darcy-Forchheimer equation by the source term added to Navier-stokes equation in OpenFOAM with Darcy-Forchheimer equation you will find that

D = 1/k while k is the permeability.
F = Cf/(k)^0.5

There us another way to express permeability by using power law.

Best Regards,

Ahmed

Sent from my SM-N9005 using CFD Online Forum mobile app

[shaiashe]

Hi all,

Is there a way to define the A and B coefficients such that they vary in space within the porous zone?

thanks,

[Outbound]

The quickest solution I can think of is by defining different porous zones, but if you want to have a smooth transition of porous coef values I don't know if it's easy to define the coefs by the use of functions, although it should be possible.

[deepbandivadekar]

Originally Posted by olesen

Note there is also useful little trick in darcy/forchheimer specification: since negative coefficients are physically meaningless, they can (mis)used to specify a multiplication factor for strongly anisotropic porous media. This can be quite convenient.

This is an old thread but what does this mean exactly? Can someone illustrate?

[fidu]

Originally Posted by Outbound

Greetings,

I'm having a bit of trouble adapting a study previously conducted in ANSYS CFX onto OpenFOAM.

I'm getting a bit confused as to how to properly calculate and insert the porosity values into the OpenFOAM porosity properties.

My flow is simulating a submerged dense vegetation layer, so I take it as isotropic flow.

The ANSYS CFX study which was published gave the values used for the porous media permeability K and the Forscheimer tensor T.

I'm just a bit confused as to what the D and F values will be then because I see some mathmatical formulas where determining them will involve a porosity value and others where it doesn't.

Attachment 48453
From POROUS MEDIA APPROACH FOR RANS SIMULATION OF COMPOUND OPEN-CHANNEL FLOWS WITH SUBMERGED VEGETATED FLOODPLAINS, M Brito, J. Fernandes, JB Leal, 2015

Can anybody help me clarify how do I obtain the correct values for D and F based on the available information?

Thank you for any help you might provide,

Edit:

Am I right in assuming that

D=1/K

and

F=(2*C_E)/SQRT(K)

Or is it just that F is just the already calculated T. Aren't they both the Forscheimer tensor?

Hi am want to simulate as well some vegetation with the leaf area density (LAD) [m^2/m^3] where I calculate the frochheimer coefficient as f= 2 c_d LAD [1/m] where c_d [. ] is my form drag coefficient. As I want to do a comparable study to a windtunnel experiment I have to scale the entire case by 1:160. Now my question is, how do I ensure a good scaling when f has the dimension [1/m] quantity or better when LAD has the dimension [m^2/m^3]? How did you solve that?

Thanks in advance

Best
fidu

[Jingxue Wang]

I also have the same question. Where can I get the values of a and b for porous medium? For my case，I want to simulate the flow around trees where the tree canopy is treated as porous media.

[shaiashe]

Hi Wang,

This is a complicated issue. There are no two single values. You should do some reading; Start with Finnigan (2000), and work your way to the present. Otherwise, calibrate against field measurements.

Shai