[time-nuts] Cs stability
Dr Bruce Griffiths
bruce.griffiths at xtra.co.nz
Tue Jul 17 05:23:34 EDT 2007
Tom Van Baak wrote:
>> Yes it can make sense.
>> Place one Cs clock in a chamber where the ambient temperature can be
>> adjusted to various fixed temperatures. Compare the phase of its
>> 5/10MHz and/or PPS outputs with respect to those of another Cs standard
>> held at constant temperature.
> I'm not sure I follow this. Yes, you will see a phase shift but
> how can you tell how much of said shift is due to a fixed
> phase shift (as if it were cable phase tempco) vs. how much
> is due to phase shift due to frequency offset (as if it were
> oscillator frequency tempco)? No change in temperature is
> instantaneous; during the (slow) change both phase and
> frequency (equals phase change over time) may change.
The cable phase shift can be compensated by having the interconnect
cable experience the same environment for equivalent lengths.
If the temperature is ramped slowly enough the uncorrected frequency
excursions will be relatively small, otherwise the phase shift due to
the integral of the frequency excursion over time may well be significant.
My point was that Cs standard only regulates (in the long term) the
frequency or rate of change of phase and not the initial phase which is
essentially arbitrary and may well have a significant tempco due to the
tempcos of the synthesiser delay and any buffer delay. The control loop
endeavours to control the frequency experienced by the Cs beam which is
subject to delays in the synthesis chain so that the phase at the
crystal oscillator output itself will be dependent on these temperature
More information about the time-nuts