[time-nuts] Anyone want an engineering job making a GPSDO comparator for T-Bolts?

Charles Steinmetz csteinmetz at yandex.com
Sat Nov 9 00:07:26 EST 2013


Bob wrote:

>We successfully "replaced" an internal low-quality TCXO with an external
>T-Bolt signal with much help from Time Nuts; but one guy called me up and
>suggested we discipline the INTERNAL oscillator instead of replacing it with
>an external signal.

Whether or not this is a good idea depends entirely on your 
particular needs.  A GPSDO works by relying on the quartz oscillator 
for stability at low tau, and the GPS for stability at higher 
tau.  In the case of a quartz OCXO, the crossover generally occurs 
between perhaps fifty and several hundred seconds.  If you need good 
stability at tau < 100 seconds, where the GPS is not very stable, the 
disciplined TCXO may be much worse than the disciplined OCXO (both 
because the oven stabilizes the crystal better, and because OCXOs 
generally use better crystals in the first place).  When excellent 
stability at tau < 100 seconds is required, you would likely find the 
disciplined TCXO inferior to a disciplined OCXO.

This is not to say that disciplined TCXOs are necessarily junk -- 
Jackson Labs makes some that reportedly perform fairly well -- but I 
suspect JL uses better TCXOs than the ones you usually find.  Even 
so, their disciplined TCXOs are not as good as their disciplined OCXOs.

Best regards,

Charles





More information about the time-nuts mailing list