[time-nuts] Comparing Reference Accuracy
bruce.griffiths at xtra.co.nz
Fri May 23 01:52:25 EDT 2008
Stanley Reynolds wrote:
> This counter may be close : http://www.psplc.com/datasheets/vigo/VOITC-200.pdf
> or http://ztc.wel.wat.edu.pl/kalisz/T-2300R.pdf
Much better but still inadequate for characterising good oscillators, at
least for short averaging times.
Its actually not too difficult to achieve 25ps or better resolution with
a suitable interpolator.
You can actually achieve better resolution with an analog phase comparator.
The only constraints being that the frequency difference between the 2
sources being compared isnt too large and that such comparators have
deadbands towards the end of the range.
The latter constraint can be overcome by using a pair of linear phase
comparators driving the second with one of the signals in quadrature
with the corresponding input to the first comparator.
More information about the time-nuts