[time-nuts] Where does the Z3801A 1 PPS come from?
Tom Van Baak
tvb at leapsecond.com
Mon Mar 14 14:58:27 EST 2005
> I wonder if the Z3801A has the same PPS stability as the 58503B. You
> mentioned the 58503B 1 PPS jitter was spec'd at some number of
> picoseconds. When I've looked at both my Z3801As, I see about 30-50ns
> of peak-to-peak noise (with 100 second averages and logging times of up
> to several days; one unit is a bit quieter than the other).
Depending in how deep we want to take this
mystery I / we could check several models of
Z38* and 585*.
But the external 1 PPS divider is also suspect
so here are two easy tests to run.
1) Try the old hair-dryer test and see how much
the TI readings change as you gently warm the
external 1 PPS divider box. A string of 'HC390's,
especially in ripple carry mode, should go wild
and confirm the tempco hunch. This is a test
you can do in a matter of minutes.
2) Eliminate the Z3801A 1 PPS architecture from
the equation and just test the latency and stability
of your external 1 PPS divider. Split any 10 MHz
sinewave into two, run one side through the external
divider to TI start (1.0 V DC) and the other side into
TI stop (0.0 V AC). Look for similar 2.5 ns variations
as your plot. You may get a hint within a few hours;
and know for sure in 12 or 24.
More information about the time-nuts