CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   CFX (https://www.cfd-online.com/Forums/cfx/)
-   -   pressure losses modelisation (https://www.cfd-online.com/Forums/cfx/22304-pressure-losses-modelisation.html)

 Alex March 16, 2006 05:52

pressure losses modelisation

Hi! I am studying a system connected to a fan which permit to pump air into the system. To make optimisation of this centrifugal fan, i want to model the sytem by a domain (or subdomain) whit the same pressure looses and i will connect it with the fan domain. I would want to model pressure losses throught a subdomain. The deltaP=1500 Pa.This subdomain is at the entrance of the fan domain. I believe that i have to use a subdomain with a momentum source with a directional loss model. How can i define the permeability and the loss coefficient (darcy law) to obtain a delta P=1500 Pa between the inlet and the outlet.

Thanks!

 Robin March 16, 2006 14:08

Re: pressure losses modelisation

Hi Alex,

You can't. If you want a specific pressure drop, you can simply specify a constant momentum source within the domain. A general momentum source in the x direction is defineds as:

-dP/dx = S_x

If you want a pressure drop of 1500 Pa in the x direction, just divide by the distance in the x direction and you have your source.

That said, it's generally not a good idea to do it this way. One reason is that the pressure drop will not vary with velocity. If you happen to have flow moving in the opposite direction, it will cause a pressure rise.

The truth is you are missing some information. You may measure a pressure drop of 1500 Pa, but this will vary with velocity. What you really need is to know is how your loss varies with velocity. Then you can implement your pressure drop as a proper loss model.

You could reverse engineer your drop. If you know the flow rate, just devise a linear loss coefficient that will give you approximately the pressure drop you want.

Refer to the documentation for the equations for all of these.

Regards, Robin

 Alex March 17, 2006 06:08

Re: pressure losses modelisation

Thank u Robin.

 All times are GMT -4. The time now is 17:44.