High Pressure¶
RocketProps contains saturated liquid data and uses the corresponding states principle (CSP) to extrapolate that data to higher pressure.
Density¶
For those propellants in REFPROP or CoolProp, high pressure density data is curve fit to those high-quality data.
Those data are curve fit to a modified, polynomial version of the Tait equation
Gas&Liq 5th Ed defines the Tait equation with equation 4-12.2

When no data are available, there are a number of correlations available to approximate the density at high pressure.
COSTALD¶
The COSTALD model is a CSP model taken from the thermo project which is an open-source project on pypi.org developed by Caleb Bell and Contributors (2016-2020).
In the comparison plots below, it seems to be the most accurate model for all propellants, except for alcohols and parahydrogen.
Chang-Zhao¶
The Chang-Zhao is taken from Gas&Liq 5th Ed as defined by equation 4-12.3

There is a 2nd version of Chang-Zhao as modified in Journal of Molecular Liquids 160 (2011) 94-102 that is also made available in RocketProps.
Nasrfar¶
Journal of Molecular Liquids 160 (2011) 94-102 provides another model for compressed density as developed by Nasrfar Moshfeghian.
Model Comparisons¶
The following charts compare high accuracy density calculations (CoolProp) to the results of RocketProps curve fits and to each of the models mentioned above COSTALD, Chang-Zhao 1, Chang-Zhao 2 and Nasrfar.
Each chart shows the percent standard deviation (%SD) of each model as compared to CoolProp. The error of each model has a 95% chance of being within 2 times %SD, 99.7% chance of being within 3 times %SD.
The charts are shown in order from worst to best in terms of %SD.
Except for the three worst fluids, parahydrogen and the alcohols, the COSTALD, model matches high pressure propellant density better than Chang-Zhao 1, Chang-Zhao 2 or Nasrfar. Usually with less than 1% error; often much less.
For this reason, COSTALD is the default model for propellants without detailed high pressure density data within RocketProps.










