CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   Help in porosity for knitted fabric??? Linear or Quadratic resistance coeff (http://www.cfd-online.com/Forums/cfx/92495-help-porosity-knitted-fabric-linear-quadratic-resistance-coeff.html)

 SS460 September 15, 2011 15:50

Help in porosity for knitted fabric??? Linear or Quadratic resistance coeff

Hi,
For knitted fabric having porosity of 30 and 50%
similar to these photos
http://www.arrigoni.it/img/agr/vento/vento8.jpg
http://www.arrigoni.it/img/agr/vento/vento7.jpg

Any suggesions of resonable values for
or permeability
I need these values to define the porous material.
I have spent considerable time searching for any useful literature for this issue.

What is the range of values you expect would be appropriate for such fabric?
For example: Quadratic resistance coefficients = 300 kg/m^4?

Is it possible to calculate these values?
I have gone through almost all threads regarding above parameters in this forum,
I have been in this link too:
http://www.cfd-online.com/Wiki/Fluen...e_drop_data.3F

but the problem is that I don't have and can't perform experiment to obtain the required data.

Could obtaining the loss coefficient K_loss and head loss be of any help?
http://www.thermal-wizard.com/tmwiz/

Regards

 ghorrocks September 15, 2011 18:57

You can estimate the resistance of a single isolated cylinder quite accurately, the drag of a cylinder is well known. If you search the literature you can probably find the drag/resistance of a bank of cylinders. You should be able to define a resistance coefficient from this.

 SS460 September 16, 2011 01:29

Is this what you are meaning by isolated cylinder (second and last objects)
http://www.centennialofflight.gov/es...nts/TH12G8.htm
found through this:
http://www.centennialofflight.gov/es...ients/TH12.htm

From Wikipedia I found this
http://en.wikipedia.org/wiki/Drag_coefficient

But which resistance is defined through the drag coefficient and how?
Also, the knitted fabric is very thin, would that effect the cylinder assumption?

Regards

 ghorrocks September 16, 2011 06:36

Quote:
 Is this what you are meaning by isolated cylinder?
Yes, that is correct. The drag coefficient is actually a function of Re, but if you are lucky enough to be in the constant drag coefficient regime then a constant is fine.

Quote:
 But which resistance is defined through the drag coefficient and how?
Convert the drag coefficient into a pressure drop versus flow velocity curve. Then this can be implemented as a porous resistance or source term.

Quote:
 Also, the knitted fabric is very thin, would that effect the cylinder assumption?
I do not think you understand what I am getting at - the fabric is a whole bunch of little cylinders next to each other. The total drag is the sum of the drag of each little thread, and each thread can be assumed to be a cylinder.

But the drag of an array of cylinders next to each other is different to an isolated cylinder. You will have to take this into account.

 SS460 September 16, 2011 09:17

Thanks again dear ghorrocks :)

What i'm trying to do is to examine the effect of the fabric porosity on blowing air wind (So it is an open area not pipe for instance)

I found the following link that provides a way to determine the resistance loss coefficient in (m^-1)
(Is it enough to define the porosity of the fabric accurate enough?)
http://jullio.pe.kr/fluent6.1/help/html/ug/node236.htm
It is located under this title:
Quote:
 Deriving Porous Media Inputs Based on Superficial Velocity, Using a Known Pressure Loss
So how about determining the head loss coefficient for a screen with required porosity:
http://www.thermal-wizard.com/tmwiz/...ns/screens.htm
Then continue with above approach?

Quote:
 Convert the drag coefficient into a pressure drop versus flow velocity curve. Then this can be implemented as a porous resistance or source term.
Any source that provides the needed correlations of drag coefficient? From my search I didn't reach any helpful thing.

Regards

 SS460 September 16, 2011 09:28

Quote:
 So how about determining the head loss coefficient for a screen with required porosity: http://www.thermal-wizard.com/tmwiz/...ns/screens.htm Then continue with above approach?
By taking another thought, the above link for head loss in screen is for pipe, so it won't be accurate for the case of open blowing air?

Regarding the drag force, how can I use it to obtain the pressure drop, or head loss coefficient? any helpful equation?

Sorry for the "shallow" questions but i'm structural engineer and lots of what I learned in fluid has vanished :)

 ghorrocks September 17, 2011 07:15

Quote:
 Deriving Porous Media Inputs Based on Superficial Velocity, Using a Known Pressure Loss
This approach looks good. I don't have time to read it in detail but it looks promising.

Quote:
 drag force, how can I use it to obtain the pressure drop
If you have the drag force of the membrane then you can divide it by the membrane area to get the pressure drop across the membrane.

 SS460 September 17, 2011 09:32

Many thanks dear ghorrocks

Just want to make sure about one correlation I found during my search:

loss coefficient = (1-porosity)*(drag coefficient)

Best regards :)

 ghorrocks September 18, 2011 07:42

I have no idea about that. It does not seem to be applicable to the CFX approach as CFX does not use a loss coefficient of that form.

 SS460 September 18, 2011 15:27

I see,
I found it in this publication
http://journals.tdl.org/ICCE/article/viewFile/4173/3854
equation 8 in 3rd page.

When I used it with the assumption of cylinder having Cd=0.8-1.2 it gave reasonable results, although it may not be accurate or even correct!
Best regards

 ghorrocks September 18, 2011 18:26

But it appears that they are simple grossing up the individual cylinder drag to the local porosity. This sounds like a very approximate (ie not very acurate) approach to me as it does not take into account proximity effects. But if you have nothing better and do not need very accurate results then you could use it anyway.

There is an empirical formula for meshes, which can also be used for fabric, I guess:

delta_P/(0,5*rho*V^2)=1,62*(1,3*(1-f)+(1/f-1)^2),

where f is porosity

 SS460 September 21, 2011 14:55

Thanks A_Lyaskin for the formula, do you have a reference?

Best regards