[time-nuts] Racal-Dana 1992 switches

Pete peterawson at earthlink.net
Sat Mar 29 11:05:57 EDT 2008


This topic has been addressed earlier; though with some
debate. I have proposed a simple heterodyne scheme for
beating 2 stable sources against each other & observing
a 1KHz difference frequency to resolve 1uHz deltas.
This is NOT a "state-of-the-art" scheme, but it will
provide better than 1E-12 resolution in less than 10s.

This scheme does require some non-standard items.

1. You need a stable synthesizer with external clock
    capability to yield 10.001 Mhz, phase locked to 
    one of your sources. The HP3336C or a PTS040
    work fine.

2. You need a 1KHz zero crossing detector to drive
    your counter input with low jitter. The ZCD requires
    2 opamps & a few passive parts, including 2
    inductors you'll need to wind by hand.

3. You need a level 7 double balanced mixer to
    heterodyne the second source & the 10.001Mhz
    signal. Mixers optimized as phase detectors, like
    the  mini-circuits SYPD-1 work well for this.
    You also need a diplexer on the mixer output
    to separate the 1KHz beat signal from the other
    mixer products. The diplexer is 6 passive parts.

The results are stable & provide counter readings of
9 significant digits down to 1uHz with the leading 4
digits of frequency assumed from the mixing process.
The counter gate time setting provides useful & often
necessary averaging of the readings; so a variable
gate time counter is handy. I've used a H-P 5335A,
& don't know much about the Racal 199x series,
but I suspect they would do just fine.

Pete Rawson



More information about the time-nuts mailing list