CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   Reference Pressure (http://www.cfd-online.com/Forums/cfx/19290-reference-pressure.html)

Danard November 10, 2002 02:31

Reference Pressure
 
Hi all,

I notice that most of the tutorials in CFX5 set the reference pressure = 1.01e+5Pa and I can't find the explanation yet. How could you all set the reference pressure? You are based on what criteria? Thanks for any answer here.

Christian November 12, 2002 14:51

Re: Reference Pressure
 
Hi Danard,

1e+5Pa=1bar is atmospheric pressure.

1,013e+5Pa often is considered as "standard pressure conidtion".

The reference pressure is important for compressible similations.

Greetings, Christian

Neale November 12, 2002 15:11

Re: Reference Pressure
 
In CFX-5 the reference pressure is simply the "offset" pressure. All pressures for boundary conditions must be specified relative to this value, and the solver also solves for pressure relative to this value.

The main reason for this functionality is roundoff. For example, if you are solving a problem where the change in pressure across your device is small relative to the averaqe pressure level then it can be quite hard on the numerics. Say your average pressure level is 1 atm and the pressure change through your device is on the order of 10s of Pascals this might be a problem. So, instead, you can specify what you think the average level will be and the solver will solve relative to this value.

As was mentioned this can be more of a problem for compressible flows because the average pressure levels tend to be much higher than incompressible flows. In addition, for incompressible flows the magnitude of the pressure level is not really important, only the pressure gradient is. Also be careful if you are solving compressible flows that you do not use 0 refereece pressure and then specify 0 relative pressures for boundary conditions.

Neale


All times are GMT -4. The time now is 04:57.