Well, that's difficult to say, having no "standard" crystal. With my limited recent experience, I think it compares well with other methods. I'm accustomed to seeing differences of 2% or sometimes more between different methods and different people measuring the same crystals at their own workbenches.
PHSNA is definitely not absolute. But when evaluating its performance I saw that its results fell in alongside those from other methods.
In a presentation on crystal measurements and filter design, I compared several methods and/or instruments and plotted their results for four crystals which were tried on different equipment. See page 21 of this PDF:
For the "actual" value of the crystals, I took the average of two measurements I thought were most trustworthy. On in which and employee at International Crystals used their commercial equipment. I think s.s. might mean Squires-Sanders. The other was measurements taken by Bill Carver using a VNA. The other measurements I did myself using various methods. PHSNA isn't listed since this presentation predates PHSNA.
Mr. Hayward said "... where Lm is the motional inductance and the reactance slope is approximated by ¦¤X/¦¤F". In his simulation, the simulated 100 mH was calculated to be 99.93 mH using the method. If real life measurements agreed as closely as that approximation, we're in good shape.
73-
Nick, WA5BDU
Nick Kennedy