[time-nuts] Any Spectracom hackers out there?
John Ackermann N8UR
jra at febo.com
Sun Oct 24 15:51:40 EDT 2004
Bill Hawkins wrote:
> Just finished tracing the schematic of a Spectracom 8163 WWVB
> receiver and standard comparator. The lock light on the 8163
> runs off the receiver AGC line. There are two MC1496 phase
> detector chips. One detects the AGC level and the other, 90
> degrees over, detects the phase difference between WWVB and
> the local 10 MHz VXO. This is done using 20 KHz reference
> signals against 60 KHz (crystal filtered) WWVB. All dividing
> is done with 74LS90 chips.
> I'd expect the amplitude shifts for decoding time to come from
> the AGC amp, leaving you free to disconnect the VXO and replace
> it with precision 10 MHz. But there's a microprocessor in there
> somewhere, so that's just a guess.
> Be glad to look at a copy of the schematic and render an opinion
> just for the chance to look at Spectracom documentation. Reply
> and I'll send my address.
> Bill Hawkins
Hi Bill --
I have an 8164 which is very similar to the 8163 but has a built-in
oscillator that is frequency locked to the WWVB signal (it's stable to
about 1x10e9; interesting, but not up to snuff for a real standard :-)
). I also have an 8161-16 thats' nothing but a comparator. The RF in
all the 81xx units appears to be pretty much identical, but I'll be
happy to copy the relevant stuff for you -- let me know where to send it.
I don't actually have to disable the internal 10MHz VXO; that signal is
divided down to 1MHz and then sent to the CPU board which divides
further to 1pps and steers that 1pps to align with the WWVB second
marker derived from the time decoder. If I break the 1MHz line and
insert my own signal, that should do the trick without the rest of the
unit being any the wiser.
More information about the time-nuts