by Doug Coulter » Sun Sep 12, 2010 5:00 pm
Depends on what you mean by accuracy -- long term, yes, it's probably the best any of us would do unless we built our own atomic clocks or bought them surplus (possible).
But the instant by instant accuracy (jitter) of these is pretty bad. I once faced that with synching things to WWV stations. Over long times, yes, it's as good as can possibly matter.
But over short intervals, atmospheric effects can be much worse than a wristwatch oscillator -- which is why even coming straight down from space (no "skip" that depends on ionosphere reflective layer heights that change path lengths) you need to get several GPS satellites to get a good fix (and of course there is deliberate jitter put into the sat signals, still).
So it depends on what you need.....a very slow adjusting TXCO based on a cycle counting error scheme with very long time constant using one of the above as a reference system for long term accuracy is needed to "have it all".
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.