HP 5061A - bad cesium beam tube?
I received a 5061A that I think probably has a dead tube, but I'd like
to be sure before I give up hope.
I expect the unit has been off for quite a few years, so I first powered
it on with CS OFF. The ion pump reading pretty quickly dropped below 25
and stayed there. I left it like that for about 12h before switching it
to LOOP OPEN. The ionizer, cesium oven, etc cycled on and off in that
state for roughly another 12h before the ion pump reading was low and
stable enough for these circuits to remain on (below a reading of about
25). Once it got to that point the ion pump dropped down to about 20
over a few minutes, where it stayed. I then coarse-tuned the OCXO to
match the frequency of my HP 5061A. Unfortunately, the beam current
indication was only about 2. I waited another 12h. The ion pump reading
had decreased to about 18 and the beam current was still around 2.
Here are all the indicator readings, and the specified normal range in
the manual:
| indicator | value | normal |
|--------------+-------+--------|
| battery | NA | NA |
| supply | 40 | 35-45 |
| ion pump I | 18 | 0-15 |
| osc oven | 39 | 20-45 |
| cs oven | 14 | 5-35 |
| 5 MHz | 37 | 35-45 |
| mult | 42 | 35-45 |
| beam I | 2 | 15-30 |
| control | -2 | |
| 2nd harmonic | 15 | 20-45 |
| 1 MHz | 37 | 35-45 |
| 100 kHz | 37 | 25-45 |
I measured the +3500V supply as +3.2kV with it still powering the
beam. That's within the >+3kV spec.
The ion pump reading is still above the 15 spec. Should I leave it on
for a couple days to see if that comes down? The manual does say this
could take a few days to be in line. The extremely low beam current
reading concerns me, however. I haven't touched the BEAM I METER pot
yet, so I don't know how that's set.
Anything else I should do?
Thanks
Matt
Matt,
If you're seeing a second harmonic reading of 15 it's very likely that
there is still some cesium left in your tube. No telling how long it
would last though. If you haven't yet I'd recommend downloading a copy of
the service manual as there is a lot of good information there for
adjustment.
You should be able to get it to lock with a beam current of only 2, but it
may be difficult to see the peaks if you follow the normal startup
procedure. Do you have another reference you can use to get the oscillator
close to on-frequency and try closing the loop?
Depending on the measurement equipment you have available I'd suggest
checking the following:
- Measure the electron multiplier voltage (-2500V supply A19).
Depending on the age of your unit this could be adjustable on the A15
assembly either with a potentiometer or by changing out a resistor. The
"Beam I Meter" adjustment on the front panel only changes the meter
indication, but by changing the electron multiplier voltage you can try to
increase the beam current (should be measured at the output of the tube
where it plugs into A7-J1, there is a procedure in the manual)- If the tube is running low on cesium (likely) you can improve the
SNR for a while by boosting this voltage
- If the tube is running low on cesium (likely) you can improve the
- Perform the loop gain adjustment procedure on A7. By increasing the
gain of the AC amplifier you can boost the amount of 2nd harmonic signal
available to help with lock
If you can get it to lock with the above then there are further adjustments
you can make. Mine is in a similar boat and I just keep it in CS Off most
of the time and power up occasionally for a test.
Chris
On Sat, Feb 25, 2023 at 10:25 AM Matt Huszagh via time-nuts <
time-nuts@lists.febo.com> wrote:
I received a 5061A that I think probably has a dead tube, but I'd like
to be sure before I give up hope.
I expect the unit has been off for quite a few years, so I first powered
it on with CS OFF. The ion pump reading pretty quickly dropped below 25
and stayed there. I left it like that for about 12h before switching it
to LOOP OPEN. The ionizer, cesium oven, etc cycled on and off in that
state for roughly another 12h before the ion pump reading was low and
stable enough for these circuits to remain on (below a reading of about
25). Once it got to that point the ion pump dropped down to about 20
over a few minutes, where it stayed. I then coarse-tuned the OCXO to
match the frequency of my HP 5061A. Unfortunately, the beam current
indication was only about 2. I waited another 12h. The ion pump reading
had decreased to about 18 and the beam current was still around 2.
Here are all the indicator readings, and the specified normal range in
the manual:
| indicator | value | normal |
|--------------+-------+--------|
| battery | NA | NA |
| supply | 40 | 35-45 |
| ion pump I | 18 | 0-15 |
| osc oven | 39 | 20-45 |
| cs oven | 14 | 5-35 |
| 5 MHz | 37 | 35-45 |
| mult | 42 | 35-45 |
| beam I | 2 | 15-30 |
| control | -2 | |
| 2nd harmonic | 15 | 20-45 |
| 1 MHz | 37 | 35-45 |
| 100 kHz | 37 | 25-45 |
I measured the +3500V supply as +3.2kV with it still powering the
beam. That's within the >+3kV spec.
The ion pump reading is still above the 15 spec. Should I leave it on
for a couple days to see if that comes down? The manual does say this
could take a few days to be in line. The extremely low beam current
reading concerns me, however. I haven't touched the BEAM I METER pot
yet, so I don't know how that's set.
Anything else I should do?
Thanks
Matt
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe send an email to time-nuts-leave@lists.febo.com
Matt well written up comments on what you are seeing.
All is not lost. Let it run. Frankenstein takes some 48 hours to lock when
I first was able to make it work.
Now over several years it locks in under 2 hours.
Beam = 0 and its not broken.
Its interesting that your offset is -2 that may be a good sign that its
actually trying to lock or is locked.
Especially if its staying at that offset.
The fact that the multiplier is 42 really appears a good sign.
OK fact. Those lamps go flat on the center bottom contact over years. Are
you sure the green lamp is good. Swap it with the yellow if you know the
yellow is good. If thats the issue a drop of solder fixes the center
contact.
If all is working you have a low beam tube. So its going to have a lower
signal to noise. But it will hang around the right frequency. This is
getting picky because if I didn't have other CS to check I wouldn't know
honestly.
Regards
Paul
WB8TSL
On Sat, Feb 25, 2023 at 11:25 AM Matt Huszagh via time-nuts <
time-nuts@lists.febo.com> wrote:
I received a 5061A that I think probably has a dead tube, but I'd like
to be sure before I give up hope.
I expect the unit has been off for quite a few years, so I first powered
it on with CS OFF. The ion pump reading pretty quickly dropped below 25
and stayed there. I left it like that for about 12h before switching it
to LOOP OPEN. The ionizer, cesium oven, etc cycled on and off in that
state for roughly another 12h before the ion pump reading was low and
stable enough for these circuits to remain on (below a reading of about
25). Once it got to that point the ion pump dropped down to about 20
over a few minutes, where it stayed. I then coarse-tuned the OCXO to
match the frequency of my HP 5061A. Unfortunately, the beam current
indication was only about 2. I waited another 12h. The ion pump reading
had decreased to about 18 and the beam current was still around 2.
Here are all the indicator readings, and the specified normal range in
the manual:
| indicator | value | normal |
|--------------+-------+--------|
| battery | NA | NA |
| supply | 40 | 35-45 |
| ion pump I | 18 | 0-15 |
| osc oven | 39 | 20-45 |
| cs oven | 14 | 5-35 |
| 5 MHz | 37 | 35-45 |
| mult | 42 | 35-45 |
| beam I | 2 | 15-30 |
| control | -2 | |
| 2nd harmonic | 15 | 20-45 |
| 1 MHz | 37 | 35-45 |
| 100 kHz | 37 | 25-45 |
I measured the +3500V supply as +3.2kV with it still powering the
beam. That's within the >+3kV spec.
The ion pump reading is still above the 15 spec. Should I leave it on
for a couple days to see if that comes down? The manual does say this
could take a few days to be in line. The extremely low beam current
reading concerns me, however. I haven't touched the BEAM I METER pot
yet, so I don't know how that's set.
Anything else I should do?
Thanks
Matt
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe send an email to time-nuts-leave@lists.febo.com
Chris and Paul, thanks for the notes.
It now locks! It wouldn't last night, but I just tried again and the
continuous operation lamp illuminated. Unfortunately, the beam current
is now close to 0 and the 2nd harmonic is at about 10.
The 5 MHz signal seems pretty stable relative to the one from my
5065A. It drifts a little back and forth, but not much. I expect this is
a symptom of having a low SNR.
Chris, I do have the service manual; I'll take a look at the loop gain
adjustment and the -2500V supply.
Still, I'm not thrilled about having a unit with a weak tube. The seller
claimed this functioned correctly (from the pictures though I don't
think they knew how to test it). If I can get a good discount I'd keep
it, but otherwise I may send it back.
Matt
The shipping is going to be nasty but then you have dealt with that already.
Frankenstein's beam is 0 and it was a bad tube given to me. The original
was a option 004 and it was totally dead.
It gets tricky if you have a RB and the CS. Who is right? If you have a
gpsdo thats very helpful in understanding which one may be drifting.
Frankenstein's beam has always been 0. But the 2nd harmonic is 32. I
actually inserted an amplifier in the beam measuring circuit and could read
the beam current or leakage. Who knows it was very small and did seem to
peak. But it does lock and doesn't seem to drift with respect to the other
Cs. I use a GPSDO to set the 5 MHz. Eliminates the whole double peak
possible issue.
Depending on how much you paid. It could be a keeper. Frankenstein was $125
at a flea market. But the lessons I learned from it have been far more
valuable/fun.
Regards
Paul
WB8TSL
On Sat, Feb 25, 2023 at 2:56 PM Matt Huszagh huszaghmatt@gmail.com wrote:
Chris and Paul, thanks for the notes.
It now locks! It wouldn't last night, but I just tried again and the
continuous operation lamp illuminated. Unfortunately, the beam current
is now close to 0 and the 2nd harmonic is at about 10.
The 5 MHz signal seems pretty stable relative to the one from my
5065A. It drifts a little back and forth, but not much. I expect this is
a symptom of having a low SNR.
Chris, I do have the service manual; I'll take a look at the loop gain
adjustment and the -2500V supply.
Still, I'm not thrilled about having a unit with a weak tube. The seller
claimed this functioned correctly (from the pictures though I don't
think they knew how to test it). If I can get a good discount I'd keep
it, but otherwise I may send it back.
Matt
Hi
One way to “decide” the who’s right argument is to get a reasonably good
GPS (Trimble NetRS from eBay maybe) that will run off of 10 MHz (yes this
means a doubler if you have a 5061a). Feed the atomic clock into the input
on the GPS.Then log the RINEX files for a while (= a month maybe). Send
them off to NRCan (free) for analysis.
If you send it in about 3 weeks after the end of the time period, you will get back
a pretty good clock estimate. It’s certainly good enough to tell you how well
your Cs or Rb is doing. With an Rb, you might do pretty well with a week’s
worth of data.
Bob
On Feb 25, 2023, at 5:06 PM, paul swed via time-nuts time-nuts@lists.febo.com wrote:
The shipping is going to be nasty but then you have dealt with that already.
Frankenstein's beam is 0 and it was a bad tube given to me. The original
was a option 004 and it was totally dead.
It gets tricky if you have a RB and the CS. Who is right? If you have a
gpsdo thats very helpful in understanding which one may be drifting.
Frankenstein's beam has always been 0. But the 2nd harmonic is 32. I
actually inserted an amplifier in the beam measuring circuit and could read
the beam current or leakage. Who knows it was very small and did seem to
peak. But it does lock and doesn't seem to drift with respect to the other
Cs. I use a GPSDO to set the 5 MHz. Eliminates the whole double peak
possible issue.
Depending on how much you paid. It could be a keeper. Frankenstein was $125
at a flea market. But the lessons I learned from it have been far more
valuable/fun.
Regards
Paul
WB8TSL
On Sat, Feb 25, 2023 at 2:56 PM Matt Huszagh huszaghmatt@gmail.com wrote:
Chris and Paul, thanks for the notes.
It now locks! It wouldn't last night, but I just tried again and the
continuous operation lamp illuminated. Unfortunately, the beam current
is now close to 0 and the 2nd harmonic is at about 10.
The 5 MHz signal seems pretty stable relative to the one from my
5065A. It drifts a little back and forth, but not much. I expect this is
a symptom of having a low SNR.
Chris, I do have the service manual; I'll take a look at the loop gain
adjustment and the -2500V supply.
Still, I'm not thrilled about having a unit with a weak tube. The seller
claimed this functioned correctly (from the pictures though I don't
think they knew how to test it). If I can get a good discount I'd keep
it, but otherwise I may send it back.
Matt
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe send an email to time-nuts-leave@lists.febo.com
paul swed paulswedb@gmail.com writes:
The shipping is going to be nasty but then you have dealt with that already.
Frankenstein's beam is 0 and it was a bad tube given to me. The original
was a option 004 and it was totally dead.
It gets tricky if you have a RB and the CS. Who is right? If you have a
gpsdo thats very helpful in understanding which one may be drifting.
Frankenstein's beam has always been 0. But the 2nd harmonic is 32. I
actually inserted an amplifier in the beam measuring circuit and could read
the beam current or leakage. Who knows it was very small and did seem to
peak. But it does lock and doesn't seem to drift with respect to the other
Cs. I use a GPSDO to set the 5 MHz. Eliminates the whole double peak
possible issue.
Depending on how much you paid. It could be a keeper. Frankenstein was $125
at a flea market. But the lessons I learned from it have been far more
valuable/fun.
Yeah I'd very much like to keep it. I've already learned a lot and
enjoyed trying to get this to lock. But, the price I paid is too much
for something that's EOL. If the price comes way down (which I think it
should) then I'll hold onto it and since it works to an extent, I should
be able to get some use out of it.
I have a trimble thunderbolt GPSDO. Unfortunately, I'm living in an
apartment for at least the next couple months and can't get a good GPS
signal here, which is why I used the rubidium.
Matt
Bob Camp kb8tq@n1k.org writes:
One way to “decide” the who’s right argument is to get a reasonably good
GPS (Trimble NetRS from eBay maybe) that will run off of 10 MHz (yes this
means a doubler if you have a 5061a). Feed the atomic clock into the input
on the GPS.Then log the RINEX files for a while (= a month maybe). Send
them off to NRCan (free) for analysis.
If you send it in about 3 weeks after the end of the time period, you will get back
a pretty good clock estimate. It’s certainly good enough to tell you how well
your Cs or Rb is doing. With an Rb, you might do pretty well with a week’s
worth of data.
Thanks Bob, that seems like a useful technique I wasn't aware of. I
actually have a NetRS too. Unfortunately, I can't get a good GPS signal
in my apartment. I may move around the middle of the year and hopefully
the next place will have a suitable location for my GPS antenna.
Matt
Have you tried putting your GPS antenna outside the window? Today's
energy-efficient glass has a metallic layer that does not pass RF.
David N1HAC
On 2/25/23 9:04 PM, Matt Huszagh via time-nuts wrote:
Bob Camp kb8tq@n1k.org writes:
One way to “decide” the who’s right argument is to get a reasonably good
GPS (Trimble NetRS from eBay maybe) that will run off of 10 MHz (yes this
means a doubler if you have a 5061a). Feed the atomic clock into the input
on the GPS.Then log the RINEX files for a while (= a month maybe). Send
them off to NRCan (free) for analysis.
If you send it in about 3 weeks after the end of the time period, you will get back
a pretty good clock estimate. It’s certainly good enough to tell you how well
your Cs or Rb is doing. With an Rb, you might do pretty well with a week’s
worth of data.
Thanks Bob, that seems like a useful technique I wasn't aware of. I
actually have a NetRS too. Unfortunately, I can't get a good GPS signal
in my apartment. I may move around the middle of the year and hopefully
the next place will have a suitable location for my GPS antenna.
Matt
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe send an email to time-nuts-leave@lists.febo.com
"David G. McGaw via time-nuts" time-nuts@lists.febo.com writes:
Have you tried putting your GPS antenna outside the window? Today's
energy-efficient glass has a metallic layer that does not pass RF.
Yeah, that does work. However, I haven't really found a good way to
mount it outside the window and it requires that my window be at least
somewhat open whenever I'm using my GPSDO. I suppose I could use this
for occasional checks.
Matt
