[time-nuts] GPSDO control system

[Charles Steinmetz]

I've lost track now of who it was, but someone asked if there was a 
better strategy than a GPS disciplined quartz oscillator or a GPS 
disciplined rubidium.  My best standard uses an SRS PRS10 rubidium 
disciplined by GPS PPS at tau = VERY long (> ~12 hours), using the 
PRS10's internal PPS locking.  The rubidium, in turn, disciplines my 
best OCXO at tau >300 seconds.  So, for tau less than ~300 seconds 
the stability is determined by the OCXO; for tau greater than ~300 
seconds and less than ~12 hours, the stability is determined by the 
rubidium; and for tau > ~12 hours, the stability is determined by 
GPS.  I have found that this works very well.

Note that the PRS10's digital PPS PLL has a bewildering array of 
adjustable parameters.  One can easily get lost exploring different 
combinations.  My advice is to keep it as simple as possible.

Best regards,

Charles