[time-nuts] DMTD phase shifter

[Bruce Griffiths]

Stephan

For ultra fine phase shifts one can always use NIST's approach of using
a splitter (or directional coupler) to sample the OCXO signal then use a
mixer as a dc controlled attenuator, attenuate the output with a fixed
attenuator and recombine it in quadrature with the original signal. Very
small phase shifts at the reference frequency can be achieved in this way.

Stephan Sandenbergh wrote:
> Corby,
> Thanks - it makes sense. I was suspecting that such a low-frequency phase
> shifter would be done using cable lengths. However, I can't say that I've
> got much experience with this kind of thing.
>
> I have another question:
>
> The rule of thumb would be that the reference oscillator must be 3x more
> stable than the DUT. I haven't got much to measure my standards against, but
> I have got 3 identical units. According to literature the separation of
> variances method provide a reasonable estimate.
>
> How well does a measurement, based on separation of variances, compare to
> one that was done using a sufficiently stable reference?
>
>   

It works best when all 3 oscillators are compared (pairwise)
simultaneously using 3 DMTD systems and 3 time interval counters.
It tends to be less useful for longer tau as the 3 oscillators share the
same external environment.

> Regards,
>
> Stephan.
>
>
>   
Bruce
> 2009/7/24 Corby Dawson <cdelect at juno.com>
>
>   
>> Stephan,
>>
>> I have a DMTD system that was built by NBS (now NIST) in the early 80s.
>>
>> The coarse phase shifter is made up of ten lengths of miniature coax
>> cable bundled up in a shielded box with toggle switches on the front
>> panel to select the delay used.
>>
>> The LSB is about 2us.
>>
>> It also has a fine adjust that consists of a couple varactors in a phase
>> delay circuit.
>>
>> I have found that with a good quality offset L.O. I can get the precision
>> I need (<1X10-13th at a 1 second interval) without using the fine adjust
>> circuit.
>>
>> As far as the small time interval value I usually start my measurements
>> at around 2 to 4 us delay with the DUT set for a slowly increasing phase
>> shift,
>>
>> The accuracy required depends on what your measuring but with a 10Mhz
>> input and a 1hz beat note +-1us is equal to +-1X10-13th so most counters
>> will work just fine. Just be advised that the us and below digits will be
>> jumping around quite a bit!
>>
>> Also make sure that the delay is adjusted so the counter updates every
>> second, you can be updating every two seconds which nullifies the
>> cancellation of the noise in the L.O.
>>
>> Corby Dawson
>> ____________________________________________________________
>> Always a good call. Click now to establish your local phone service!
>>
>> http://thirdpartyoffers.juno.com/TGL2141/fc/BLSrjpTHbDiki6pTvGTKKCArqOqjNQCdDWO6HMJXAtNrfARisO7WahgNOso/
>>
>> _______________________________________________
>> time-nuts mailing list -- time-nuts at febo.com
>> To unsubscribe, go to
>> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>> and follow the instructions there.
>>
>>     
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
>
>