[time-nuts] 10811-60120 (serial prefix 2023) Manual/Information
Matt Huszagh
huszaghmatt at gmail.com
Sat Jul 31 19:28:48 UTC 2021
Hello,
I posted this on the HP-Agilent group
https://groups.io/g/HP-Agilent-Keysight-equipment/topic/10811_serial_prefix_2023/84530740?p=,,,20,0,0,0::recentpostdate%2Fsticky,,,20,0,0,84530740
but was redirected here.
I'm debugging a faulty HP 8663A and am currently investigating the
10811A reference oscillator (the 8663A auxiliary fuse is blowing among
other problems, so the oven could be at fault). There's a good manual
for this online. However, both ko4bb and keysight have the same 2028+
manual. My version has prefix 2023. Additionally, I wasn't able to find
much information on how the 60120 differs from other versions.
It would great to get some advice on the symptoms I'm seeing. And, if
anyone has a manual pertaining to this iteration of the oscillator, that
would be incredible.
There appear to be a number of differences between the 10811A in my
instrument and the one described in the manual. Several examples (I
haven't done a thorough search; this is just what I've noticed and there
is probably more):
- Q6 does not appear to be present in the 2023.
- No thermal fuse (F1) in the 2023.
- U1 is powered directly from the oven supply (not a 10 V regulator; I
haven't yet figured out what U2 does in the 2023).
- U1 directly drives Q7, rather than through a series resistor.
My 10811 is not obviously faulty. In fact, I believe it's probably
ok. But, the oven supply current draw is a little high (during warm-up
and after settling) in comparison to what's stated in the manual. And,
before I write it off as definitely not the problem, I'd like to get
some more informed opinions. I measured the current draw as a function
of time with the oscillator starting at room temperature (it was left
unpowered for a day prior to this). I supplied +20VDC to the oven. Here
are the results:
| time (min) | current (mA) |
|------------+--------------|
| 0 | 510.6 |
| 1 | 510.6 |
| 2 | 511.8 |
| 3 | 512.4 |
| 4 | 513.0 |
| 5 | 513.5 |
| 6 | 514.1 |
| 7 | 514.7 |
| 8 | 233.0 |
| 9 | 199.8 |
| 10 | 181.2 |
| 11 | 171.9 |
| 12 | 166.1 |
| 13 | 162.0 |
| 14 | 158.5 |
| 15 | 157.3 |
| 16 | 156.2 |
| 17 | 155.0 |
| 18 | 154.4 |
| 19 | 153.3 |
| 20 | 152.7 |
| 21 | 152.1 |
| 22 | 152.1 |
| 23 | 152.1 |
| 24 | 150.9 |
| 25 | 150.9 |
| 26 | 150.4 |
The precision I've given from about 8 minutes on is a bit
misleading. The actual value oscillated about +/- 1 mA (the values
before this appeared very stable). The manual stated that the oven
current should not oscillate, but given the small amplitude of
oscillation I figure this isn't something to worry about. Also, I assume
the transistor heaters will need to turn on and off to maintain the
temperature setpoint. I would expect this to create some oscillation in
the current.
The manual says that the warm up current should be between 380 and 490
mA for 5 to 10 min, followed by a decline and with the oscillator
settling to a current between 60 and 150 mA after an additional 10 to 15
min. So, the currents are a little high. But, given this appears to be a
different design and I can't actually find anything functionally wrong
with the oscillator (the 10 MHz signal also appears fine), it doesn't
seem to me that this is actually a problem.
I also checked voltages against those specified in the manual when the
oscillator was at room temperature. I've provided the results in the
table below (as I mentioned earlier, Q6 doesn't appear to be present in
this oscillator, so I checked other parts of the circuit to would
normally be shorted to these Q6 nodes).
| voltage point | oven cold (measured) | oven cold (target) | conclusion |
|---------------+-------------------------------------+--------------------+---------------------------|
| Q6B | starts just above 2, drops steadily | 2 | ✓ |
| Q6C | 11.5177 | 11.4 | ✓ |
| Q6E | starts at 1.5, drops steadily | 1.3 | ✓ |
| Q7B | 12.7 and dropping | 12.7 | ✓ |
| Q7C | 19.76 | 20 | ✓ |
| Q7E | 11.5293 | 11.4 | ✓ |
| Q8B | same as Q6E | 1.3 | ✓ |
| Q8C | same as Q6C | 11.4 | ✓ |
| Q8E | 280 mV | 0.23 | slightly high (TBD) |
| U1 pin 1 | 1.63, dropping | 1.8 | slightly low (TBD) |
| U1 pin 2 | 0.368 | 0.23 | slightly high (TBD) |
| U1 pin 3 | 0.368 | 0.23 | slightly high (TBD) |
| U1 pin 5 | 4.1697 | 4.1 | ✓ |
| U1 pin 6 | 4.1725 | 4.1 | ✓ |
| U1 pin 7 | 12.7, dropping | 4.8 | This is way off. But, |
| | | | I think it may be ok, |
| | | | since it appears U1 p7 |
| | | | is shorted to Q7B. That's |
| | | | probably just a version |
| | | | change. |
| U2 pin 2 | couldn't find this | 10 | |
| U3 pin 6 | 18.646 | 19 | ✓ |
Nothing appears obviously wrong.
I think it's also worth mentioning a few other things that surprised
me. The oscillator is about 10 years older than the 8663A. The 8663A has
a serial prefix of 3043, whereas the oscillator (as already mentioned)
has a serial prefix of 2023. It surprised me somewhat that HP would use
a 10 year old reference oscillator in new synthesizer. But, I don't have
enough experience to know whether this is normal or not.
Also, the oscillator is labelled as an Agilent part. Apparently, Agilent
wasn't formed until 1999, so presumably Agilent took an old oscillator
and rebranded it. Is that correct? In any event, it sounds to me like
this isn't the oscillator that was originally shipped with this
frequency synthesizer.
Best
Matt
More information about the Time-nuts_lists.febo.com
mailing list