The refrigerant data exposed in the response object was erroneously subject to a unit conversion in our egress function. This has been fixed.
No change in calculation results
Since the data was correct in the calculation itself and the bug was only in how the data was presented in the response object, there are no changes to any calculation results.