Hi Bernd,
Den 2021-08-10 kl. 03:23, skrev Bernd Neubig:
Hi Magnus,
Magnus Danielson wrote:
I have one OSA 5515, at it fails to lock up. The 8655S OCXO heats up nicely
from
5,000166 MHz, with a bit of an undershot and stabilizes at about
4,999999140 MHz. That for sure looks like the tell-tale of SC-cut crystal in
that oven. I've been able to persuade it up to 4,999999225 Hz as closest
using the oscillator trimmer, but that's it.
Sorry to correct you: This warm-up behavior indicates that an AT-cut crystal
is used. For an SC-cut the frequency during warm-up will rise.
Yeah, I realized that later, especially as Nigel sent me the datasheet
for the 5655S. It's a 5th overtone AT-cut. It was too long since I
refreshed those particular memory cells. So do not feel sorry for
correcting me, I am thankful you took the time to correclty correct me.
My facepalm moment was so short that I failed to actually facepalm.
Unfortunately I cannot help you with the described problem, as I do not have
any details about that device.
Thanks to the datasheet from Nigel, I see that the trim-pot allows me a
3 Hz correction and the EFC range is more than +/- 0.5 Hz. With more
than 0.8 Hz error, I see that I will need to trim the OCXO manuall more
first, or else it will not be able to pull in more. It could be that the
oscillator drifted too much to be able to reach 5 MHz sharp.
It is interesting that I can clearly measure the beat-note on the
frequency, but the pull-in power, which should be strong for these
offsets, is very weak but there. I can measure the very weak climbing
slope and the integrator voltage increase ever so slighly.
The 5155 steering has a fairly direct counter out of three 74LS193 (4
bit up/down counters) that feed a Harris DAC. So if the analog
integrator crosses a limit, the counter increases or decreases and the
counter/DAC takes over the state. Thus acting as a "digital capacitor"
of sort. Kind of neat in all it's simplicity.
PS. I brought forward your name as balloter for 1139, John Vig was quick
to support it.
Cheers,
Magnus
_______________________________________________
time-nuts mailing list -- [email protected] -- To unsubscribe send an
email to [email protected]
To unsubscribe, go to and follow the instructions there.
