PropSi returns erroneous values for NH3 in Excel

[bwiencke]

At certain pressures (below 79000 Pa) erroneous temperature values are returned or no values are returned. This issue is not present when using RefProp.
CoolProp Version: 6.6.0 and 6.6.1
Operating system: Microsoft® Excel® 2021 MSO (Version 2401 Build 16.0.17231.20236) 64-bit

For example:

1. =PropsSI("T","P",57565.979,"Q",1,"AMMONIA")
   Expected value: 229.15 K
   Returned value: 337.45 K

2. =PropsSI("T","P",30086.278,"Q",1,"AMMONIA")
   Expected value: 218.15 K
   Returned value: Inputs in Brent [219.766178,223.766178] do not bracket the root. Function values are [180.302048,624.573688] : PropsSI("T","P",30086.27794,"Q",1,"AMMONIA")

[ibell]

There was a problem with the saturated vapor density ancillary function in 6.6.0. It was fixed in 443a2fd . Building from source should fix this problem.

[ibell]

I had to make another fix to the ancillary (sigh...) but it should be working now. The consistency plots look much better after this fix:

Ammonia.pdf