[time-nuts] Caveats on Allan Deviation with ultra stable oscillators

[Magnus Danielson]

Hi,

On 05/29/2014 06:47 PM, Richard (Rick) Karlquist wrote:
> On 5/29/2014 9:15 AM, cdelect at juno.com wrote:
>
>> On these latest oscillators at the longer Tau (>100sec.)
>> the Quartz versus Quartz data shows much better performance
>> than the Quartz versus Maser!
>>
>> What is happening (I think) in this case is that both Quartz units
>> have exceptionally low and similar ageing.
>
> Or alternately, the quartz units are injection locking to
> each other.

When they do, they drift together, so the delta-drift is zero and thus 
do not polute the measurement.

Injection locking would also cancel some of the noise naturally, depends 
if one oscillator locks to the other or if both locks to each other.

Running one of them off frequency would reduce the noise cancellation 
due to injection locking.

Cheers,
Magnus