Result differences using SATSPLNdll

[hamschristoph]

Issue with Phase Calculation Using SATSPLNdll

Description

We have encountered an issue with the SATSPLNdll function in REFPROP. When using this function in the python wrapper, we receive different values compared to when the function is not used. Below is an example where the problem occurs:

• Gas components: CO2, Water
• Molar fractions (CO2, Water): 0.87784, 0.12216
• Temperature: 67.09 °C
• Pressure: 3.7 bara
• Property requests: CP (We used CP, but our main concern is the phase calculation. With SATSPLNdll execution, we get Q=998.0 & PHASE=Superheated gas; Without SATSPLNdll execution, we get Q=0.977797 & PHASE=Two-phase)
• Gas model: Peng-Robinson

Expected Behavior

For this example, a two-phase gas should be calculated.

Actual Behavior

• With SATSPLNdll: PHASE=Superheated gas
• Without SATSPLNdll: PHASE=Two-phase (as expected)

Question

Is there a problem in the calculation of SATSPLNdll?

Thank you for your assistance.

Versions

REFPROP Version: REFPROP 10
Operating System and Version: Windows 10
Access Method: Python wrapper

[ianhbell]

First point: don't use PR for such a non-ideal system.

It is a well-known problem that phase equilibrium calculations are very tricky for such asymmetric systems. Sometimes enabling the saturation splines with SATSPLNdll makes the calculations converge more reliably, sometimes not. Unfortunately you are in the case where it doesn't help. If you care only about binary mixtures, you could try to do tracing, but that is not supported natively in REFPROP.

[hamschristoph]

Thank you very much for your fast reply and for the hint regarding PR. We will no longer use PR for such non-ideal systems in the future.

Your suggestion regarding tracing is not feasible for us, as we have integrated the refprop wrapper into a simulation and therefore the gas composition as well as temperatures, pressures, etc. change dynamically. In addition, we often use more than two components. Since we want to avoid receiving output of superheated phase, even though the mixture is in two-phase state under the given conditions, we have temporarily deactivated the SATSPLNDLL in our implementation.

[ianhbell]

Unfortunately my suggestion to selectively disable the SATSPLN is about all that I can offer.