CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   Inertial and viscous coefficient for porous media (http://www.cfd-online.com/Forums/main/65364-inertial-viscous-coefficient-porous-media.html)

 franzdrs June 12, 2009 11:54

Inertial and viscous coefficient for porous media

Hi,
Does somebody know if the C2 and 1/alpha coefficients for the porous media model MUST be positive values? I have found these coefficients using one of the methods described in the Users Guide (Deriving the porous coefficients based on experimental pressure and velocity data) and the C2 coefficient turns to be negative because the quadratic term of the regression equation is negative. In the Users Guide itself, in the example of this method, they get 1/alpha to be negative (-242282). So, does anybody knows if that actually matters? Since FLUENT does not know that my quadratic term was negative so that with the negative sign of my C2 the value turns positive, I dont know what FLUENT will make with that.
Some help anybody?
Franz

 mettler June 12, 2009 14:16

my permeability and inertia coefficients were both positive I found mine from actual laboratory data tho. I am pretty sure I followed a method described in a paper by Boosma to get the coefficients. Also, there are a lot of papers dealing with research of heat transfer in porous media, and most have values for the permeability and inertia. K, permeability, cannot be negative due to the square root of the value. I can't remember which one is which in Fluent - sorry

 vishyaroon June 12, 2009 15:36

I have faced similar problems in the past. However as Mettler mentions these values have to be positive (physically negative values do not make sense). So the common approach we follow to get positive co-efficients is to modify the data points obtained from the lab data that represent the low velocities, within the experimental tolerence limit.

We choose the low velocity points because generally these are the ones that result in the negative values (numerically they affect the x-intercept).

 franzdrs June 13, 2009 06:27

yes, I understand that these coefficients should be theoretically positive. However, since these values are empirical, would not be possible that when the regression gives a negative coefficient, then when FLUENT uses them the positive coefficient compensates for the negative one and at the end one has a good result? I guess if I knew how FLUENT works with these values then I could tell, but I dont know that much. Somebody?
Thanks

 subha_meter August 12, 2011 22:07

porous media model

FLUENT uses a Darcy-Forchhimer type equation which has a viscous term and an inertial term. The equation looks like

Delta P = E1*Re/Ga + E2*Re&\^2/Ga

where E1 & E2 are Ergun constants - 150 & 1.75 or 180 & 1.8 and many other different combinations are possible depending on system and porous medium.
Re = Reynolds number, Ga = Galileo number

In simpler form, the equation can be written as

or Delta P = v/alpha + 0.5*C*v*|v| (v = velocity, 1/alpha = viscous resistance, C= inertial resistance)

Hope it helps

 dreamlifter747 September 14, 2013 06:13

inertial and viscous co-efficients for porous media

hello guys ,

I have trouble finding the alpha and beta values for modelling plasma (blood ) flow through a porous media which acts as the filter. All i have is the porosoity report and material properties.
Is it possible to calculate the inertial and viscous co-efficients ?

 mettler September 15, 2013 23:33

The only was I was able to get the two coefficients was run a physical experiment and measure the pressure drop thru the various porous media I was studying. After knowing the pressure drop you can back calculate the two coefficients.

good luck

 All times are GMT -4. The time now is 15:01.