[Chris Ron]

I have a problem in which a pressure has been given to me in torr which is 10^-07 and the values I need to obtain should be absolute but whenever I set reference pressure to zero the solution shows an error "Overflow". Moreover what should I specify in Reference state in material properties I am using air ideal gas.
Regards

[ghorrocks]

The solver can handle zero reference pressure just fine. This means there is something about your simulation which is either not physically possible in CFX (such as a negative absolute pressure) or your simulation is numerically unstable.

The numerically unstable issue has a FAQ: https://www.cfd-online.com/Wiki/Ansy...do_about_it.3F

You appear to be doing a simulation at extremely low pressure. This is likely to be difficult as any overshoot in pressure will result in negative absolute pressures and crash the solver. Are you sure that the Knudsen number of your simulation is OK so that you can simulate a flow like this in a Navier Stokes solver like CFX?

[Chris Ron]

Originally Posted by ghorrocks

The solver can handle zero reference pressure just fine. This means there is something about your simulation which is either not physically possible in CFX (such as a negative absolute pressure) or your simulation is numerically unstable.

The numerically unstable issue has a FAQ: https://www.cfd-online.com/Wiki/Ansy...do_about_it.3F

You appear to be doing a simulation at extremely low pressure. This is likely to be difficult as any overshoot in pressure will result in negative absolute pressures and crash the solver. Are you sure that the Knudsen number of your simulation is OK so that you can simulate a flow like this in a Navier Stokes solver like CFX?

I am new to CFD and CFX in particular. I think you are right, the density at this pressure is about 1.5*(10^-10) and as a result the knudsen number is high. Can you guide me about this?
Thanks for your earlier response.

[ghorrocks]

Wikipedia has a good page on Knudsen number: https://en.wikipedia.org/wiki/Knudsen_number

If Kn<0.01 then continuum flow (and NS equations are applicable)
If 0.01<Kn<10 then transitional flow (and NS equations may be applicable, depends)
If Kn>10 then free molecular flow (and NS equations are definitely not applicable)

If your Knudsen number is above 10 then the Navier Stokes equations are not applicable, so you cannot use CFX. For these Knudsen numbers you need to look at molecular dynamics simulation software.

[Chris Ron]

Originally Posted by ghorrocks

Wikipedia has a good page on Knudsen number: https://en.wikipedia.org/wiki/Knudsen_number

If Kn<0.01 then continuum flow (and NS equations are applicable)
If 0.01<Kn<10 then transitional flow (and NS equations may be applicable, depends)
If Kn>10 then free molecular flow (and NS equations are definitely not applicable)

If your Knudsen number is above 10 then the Navier Stokes equations are not applicable, so you cannot use CFX. For these Knudsen numbers you need to look at molecular dynamics simulation software.

Thanx for your reply. I will definitely, further look into the information you have provided.