[time-nuts] LPFRS phase ripple?
csteinmetz at yandex.com
Tue May 23 06:53:38 EDT 2017
> one shows quite some phase ripple (appx. 1 ns RMS with a frequency around 0.007 Hz) which spoils ADEV in the 60 - 200s range, very similiar to the typical GPSDO bump.
> My first guess would be a problem in the control loop locking the VCXO to the Rb
Good guess. The most likely explanation is that the loop is unstable or
marginally stable over a range of frequencies. This could be due to
wrong component values in the loop filter due, for example, to component
value drift (probably a capacitor, maybe a resistor) or installation of
wrongly valued parts at the time of assembly, OR unexpectedly high f/V
sensitivity of the EFC characteristic of the VCXO.
It could also be a problem with the power supply to the loop filter, or
frequency modulation in the Rb physics package, but both of those are
much less likely than instability of the VCXO locking loop, IMO.
More information about the time-nuts