I did peek inside the E not long after I got it and I do remember that
the OXCO looks very different, it is much smaller than the one on my
Nortel unit. If it is the OXCO I can always source a potentially better
OXCO and then I can have more fun playing with it. Being a technician I
really enjoy fixing things and seeing them work again. One other big
difference in the E model is it takes a single 24VDC power input and has
a little switcher module in it to generate the voltage the GPSDO
actually uses, this is similar to the setup on the Nortel unit, however
it did not look to be built nearly as good. I will give it a little
more time and if it does not improve I will pull it out of its box and
start to poke around inside.
Paul.
On 2/27/13 8:23 PM, John Miles wrote:
Sounds like a bad OCXO. The crystal frequency jumps slightly, and the
disciplining loop then has to compensate by steering the DAC voltage
through the loop filter. If that's the problem it may get better if you
leave it running for a few more weeks.
The smoking gun is the attack/decay characteristic of the DAC adjustment.
At least in theory, no issues on the GPS side of the loop should result in a
fast leading edge at the DAC.
I haven't seen any Thunderbolts with genuinely bad OCXOs yet, although some
of them are definitely better than others. The Thunderbolt-E, however, is a
very different model than the ones that are normally found on eBay. Almost
everything you hear about a "Thunderbolt" refers to the older non-E version.
I've heard (but can't exactly cite) that the Thunderbolt E models were a
reduced-cost, higher-volume product. If that's true, you can assume that
they don't use the same OCXOs, because those OCXOs were much better -- and
would have been more expensive -- than they had to be.
(The Thunderbolt E should not be confused with the more common Thunderbolts
running 'E'-level firmware.)
-- john, KE5FX
Hi, Paul. Glad to see you made it here. For the time-nuts; Paul and I
have spoken about these issues, and I have had similar experience. The
only thing I would add is the the spikes to seem to occur less
frequently with a better antenna placement, but when they start to
cluster, they fire off one every two or three seconds, sometimes
faster.
JimT
On Wed, Feb 27, 2013 at 2:08 PM, Paul Berger <[email protected]> wrote:
Hello all:
I am an electronic technician with 30+ years experience fixing
computers. In my spare time I like to play with old computers and
electronics. Recently I got bit by the precision timing bug, partially
after running across the wonderful leapsecond.com site, which lead me
here. First I got a FE-5680A, then after reading some of the archives
here I bought a Nortel / Trimble 45000 GPSDO and still later a used
Thunderbolt-E...
_______________________________________________
time-nuts mailing list -- [email protected]
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
_______________________________________________
time-nuts mailing list -- [email protected]
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
