[time-nuts] Measuring ADEV using TBolt-Tic tester

[ws at Yahoo]

You missed something somewhere
The Tbolt's RMS noise is 100 ps for the phase and 10 ps noise on the PPT 
using unfiltered one second data.
These include the RMS sum of several noise sources.
The resolution is sub ps.
The noise is much greater than the resolution so averageing works fine

ws
************
Azelio Boriani azelio.boriani at screen.it

OK, so the Tbolt hardware resolution is 100pS. If you have a hardware
resolution of 100pS and do an average over the data, yes, you can obtain
greater resolution but your data has to cross the 100pS boundary to have any
variation. If your phase moves under the 100pS window your averaging can
only say that the phase has not crossed the 100pS boundary by an arbitrary
lower resolution obtained with averaging. Of course all must be stable.
Maybe the trick is to use a sampling oscillator that is not so stable so
that it can move the 100pS window to reveal if the phase is near one edge or
the other and "feed the average" with numbers that are not all the same.