Confused by the defintions of Pabs and Ptot for quite a long time. For example, a pipe flow, many books only talk about the pressure drop along the pipe and there are some pretty mature equations for calculating the pressure drop. But CFX created several definitions for pressure, for example, Pabs, Pstat, Pref, and Ptot.

In almost all tutorials, Pref is 1atm. This is arbitrary and not difficult to specify. However, for outlet, CFX let Pstat be zero. This is to say, Pabs=Pref at outlet. In other words, Pref is not arbitrary, it is determined by the Pabs at outlet. My understanding is because outlet is connected to the atmosphere, Pabs is 1atm at outlet, then let Pref=Pabs, thus Pstat=0. This tells us if we want to specify Pref, we should know Pabs at outlet beforehand.

Here there is a question for us: How can we know the Pabs at outlet? If the pipe is running under high pressure (for example, 30atm), what's the Pabs at outlet?

Looking forward to your answers. Many many thanks!

David

Hi David,

The answer is in your question. You know that you pipe is at a high pressure, 30 atm, thus you must know the pressure somewhere. Use this to determine the appropriate pressure boundary condition.

Robin

Thanks Robin. I strongly believe you can help me to solve this problem out.

In fact I have got a converged solution with Pref=32atm and Pstat=0 for my two-phase flow with reactions. You know, Pref is arbitrary, and I can give it another value, that doesn't matter. But I am not sure the Pstat=0 is right or not, I can not find the reason why I specify it like this.

CFX requires to specify a relative Pstat at outlet. According to Pabs=Pstat+Pref, I should know Pabs at outlet. But how can I know? If I know the pressure drop along the pipe, only after I know pressure at inlet then I can know the pressure at outlet. That's impossbile because I only know the inlet velocity. In CFD books, a zero gradient of variables are adopted. And I also tried Pref=32atm and Pabs=1atm at outlet, the solution is not converged. This means the value of Pstat at outlet is quite important for the solution.

I really need your help, because I can not find the answer from the books. I am looking forward to your answer.

David

Hi David,

You are on the right track. Regarding Pref and Prel, just make sure they add up to the correct Pabs. As for what boundary condition to use, you basically have three choices:

1. Total pressure inlet -> static pressure outlet 2. Total pressure inlet -> velocity or mass flow outlet 3. Velociy or Mass Flow inlet -> static pressure outlet.

If the inlet is at the entrance from a resevoir or plenum, then your Total Pressure at the inlet is roughly equal to the static pressure of the plenum (since the velocity in the plenum is nearly zero). If you can make an assumption about the entrance losses, you can adjust the Total Pressure appropriately.

What you do at the outlet will depend on what you know.

If your pipe is ejecting into a resevoir or plenum, then the Static Pressure felt at the end of the pipe is roughly equal to the static pressure of the plenum. This differs from the inlet condition because there is little recovery of the dynamic pressure. For this case you should specify the Average static pressure to be equal to that of your plenum. The average static pressure boundary condition will allow the local static pressure distribution to reach equilibrium while maintaining the average over the outlet. The mass flow rate will then be determined in the solution.

If you know what your flow rate is, then you can specify the flow rate at your outlet by either velocity or mass flow, preferably mass flow. If you specify a mass flow at the outlet, the velocity and pressure distribution at the outlet will come from the solution.

If you specify a velocity or a mass flow at the inlet, you MUST specify a static pressure at at least one outlet or opening. If you don't, the solver will pick the first control volume to set the pressure level (because pressure must be specified somewhere) and you may find that it will become unstable (if the first control volume is not in a stable region of flow). In any case, it is generally undesirable to specify the flow in AND the flow out since you are effectively eliminating the conservation equation from your calculations.

I hope this helps.

Best regards, Robin

David,

- Set Pref to a representative pressure that your device operates at.

- Set the outlet relative pressure to zero (either uniform or average).

- Set your inlet to velocity specified (as you have done).

Remember that "Relative Pressure" means relative to your reference value (P_abs=P_rel+P_ref). So if you set P_ref=30atm and you know that P_abs at the outlet is 32 atm then you need to set the P_rel at the outlet to 2 atm.

The pressure drop will be computed as part of the solution. i.e., the inlet static pressure will adjust itself to the velocity and outlet pressure that you have specified.

Neale.

Thanks Robin and Neale, I am much clear now.

David