Newton Hunt suggests that larger quartz crystals are more stable than small ones, which explains the stability of the SAT pitch reference. The size of the crystal has little to do with its stability. Much more important is the design of the electronic circuit that makes the crystal oscillate. The frequency of a quartz crystal oscillator does vary with temperature, but not nearly so much as a tuning fork. There are things that a designer can do to stabilize the frequency, such as include compensating elements (usually capacitors) that have opposite temperature characteristics from the quartz crystal, and so cancel out most of the frequency variations. The SAT is, no doubt, temperature-compensated like this. But crystal stability is not really an issue for piano tuning because the piano strings will vary with temperature at least 10 times as much as even the poorest quartz crystal. The biggest errors in quartz frequency sources come from initial calibration. That is why even cheap sound cards in computers, once they have been calibrated, become excellent secondary frequency standards (at least for musical applications). One well-known application for cheap quartz crystals is wrist watches. These have very tiny 32.768 kHz crystals that cost the watch maker about $.20 in quantity. Yet these same crystals are able to maintain time to within 2 minutes per month, which is, in piano tuning terms, 0.08 cents. So you don't need a big expensive crystal to make a stable frequency source. Also, someone mentioned the use of the dial tone as a pitch reference. It seems that in many areas of the USA, the dial tone is a major third (F-A). Some years ago I tried to get our local phone company to commit to what the accuracy of that A-440 was, but they could give me no assurances that any particular accuracy would be maintained. Robert Scott Ann Arbor, Michigan
This PTG archive page provided courtesy of Moy Piano Service, LLC