Can CFX calculate the properties of liquid carbon dioxide?

[snolkeve]

I am using CFX to simulate some properties of a liquid CO₂ pump. When I was reading the Modeling documentation, I found a paragraph that says:

For each fluid associated with an access key, several blocks of information are available, including
parameter definition, superheat table, and saturation table information. However, because many calculations
do not require all of this information, a less complete set of property information can be included
in the .rgp file. In this release of CFX, only the dry superheated vapor model is supported, corresponding
to MODEL=3 in the .rgp file. Other settings of this parameter are ignored by the CFX-Solver, so all the
saturation data that would normally be read when using the non-equilibrium (MODEL=1) or equilibrium
(MODEL=2) models will be ignored. This means that the $$SAT_TABLE section does not have to exist
under a $$$Database access key, that the SUPERCOOLING parameter can be zero and that only the
$$SUPER_TABLE section is necessary.

Does this mean that CFX does not support the use of RGP tables for liquid CO₂ in calculations?
If it does not support this, what would be the recommended method for performing such calculations?

[Mark O]

Both liquid and vapour values are stored in the RGP superheat tables when you create an RGP file.  For example, there will be a density table for the liquid RGP material and if also created there will be separate density table for the vapour RGP material. But to avoid any table interpolation clipping for the liquid density at the saturation boundary being clipped to the saturated value you need to turn on a beta feature. See

https://customer.ansys.com/s/article/How-to-fix-wrong-Liquid-RGP-density-values-for-subcooled-liquid

[snolkeve]

I just tried to access this platform, but it said I don’t have permission. Is there any other way to access this content? Thank you.

[Mark O]

I have pasted the content here

Q:  I am running a CFX case where the liquid is being modelled with an RGP file.  The liquid density is too low when compared with REFPROP.  The REFPROP state shows subcooled liquid.  Is there a way to fix this?

A: The liquid densities are too low because the default behavior is for CFX to use the saturation values for properties. The adjustment for subcooled liquid is a Beta feature in CFX 24.1.    This can be activated by turning on CFX Beta features. 

Once the Beta Features are active, edit your RGP liquid material in CFX Pre.  Go to the Material Properties tab and check off "RGP Liquid Properties (Beta)". Make sure that the RGP Liquid Properties value is set to Subcooled and not Saturated.  CFX will now access the subcooled values for density and you should see higher (correct) density values in your simulation results.

[snolkeve]

I also wanted to ask — after checking some references, I found that model-3 corresponds to the Dry Superheated Vapor Model. Does this mean that the property lookup is only performed in the region where T > T_sat(P)?

[Mark O]

No. The tables cover the entire pressure and temperature range reported by the table generator when the table is created.  However, by default, the CFX solver will clip the values for the liquid property to the saturation values at Tsat,Psat. The beta feature to set the RGP Liquid Properties value to Subcooled is required to have the table interpolate into the T< Tsat, P < Psat region.

[snolkeve]

For my model, the inlet is located in the liquid region where T < T_sat and P > P_sat, while the outlet pressure exceeds the critical pressure. As I understand it, by default, the properties should be clipped to the saturation values before reaching the critical pressure — is that correct? When I enabled this (beta) feature, I found that the calculation became very difficult to converge. I’m not sure what’s happening — is my understanding incorrect, or could this be a two-phase issue?

[Mark O]

My bad. Subcooled interpolates into  T< Tsat, P > Psat 😀

[Mark O]

Without the beta feature the liquid properties in the region T< Tsat, P > Psat are clipped to the values at Tsat,Psat and so the liquid is incompressible. With the beta feature the liquid becomes compressible and pressure waves can propagate through the liquid at the speed of sound for the liquid. You may need to run with a smaller timescale.