Handy tables and charts.

[Harry's Lad]

The 232 bar point would indeed be near enough, an error of about 3 percent I think.
The error at 300 bar is still only around 8 percent.

Sorry Rich, but I don’t know where these percentages came from (or to what they refer). As I mentioned before, thermodynamics isn’t really my thing but, ‘Standard’ air behaves as the Ideal Gas that your model assumes at around 232 Bar. Above or below that pivotal pressure a mass-to-pressure error exists - below 232 Bar, the vessel’s pressure drop is slightly less than the Ideal Gas assumption anticipates; at higher pressures then the pressure drop is progressively much greater than your assumption would anticipate. Unfortunately, in the real world, the greater-than-anticipated pressure-drop means that at every iteration (fill) the pressure discrepancy increases and so, by cascading that error, there are considerably fewer fills to be had.

As I said, it hasn't led to complaints from users - only yours, Dave.

I haven’t complained. Far be it for me to complain. I’m just flagging an error.

Compressibility is actually also temperature dependent so unless yet another input parameter is called for, it's going to have to be an approximation.

Indeed it does but the temperature range we’re considering is comparatively small. Any temperature variation would apply equally to both models anyway.

And it's Rich, by the way, not Rob.

Noted and corrected. Sorry about that.

ATB
Dave