Re: [time-nuts] HP 5061A problem

[Adrian]

Corby,

well, the OP Amp was shot.
When I pulled it out, I saw a marking 'BAD 7-8-97' on the side cover.
I was hoping to get it fixed because the 1uF ceramic cap at +15V was 
shorted and had burned the 75R resistor.
But, after some serious investigation, I'm realizing that it is 
incompatible with the unit I have.


The OP Amp module is marked 05061-6134 Ser. 2136. On the chassis under 
the module it says 05061-6092. The 5061A SN prefix is 1540A.
It has an Intech A-240 and 1826-0471 (OP07CJ). The inverting input of 
the OP07 is connected to J2/9 by a white wire, as on the schematic.
However, the plug has pins 8 and 9 bridged internally, so the inv. input 
is hard wired to -16V. J2/3 is n.c. inside the plug. And, the 8uF cap is 
connected between the amp input and output (J1 and J2/5).


I had the OP07 removed and the output of U1 bridged to J2/5 and it was 
locking. However, the response to fine adjust changes was very slow.
So, at least the basics are there and I should be able to get it running 
if I find the correct OPamp.


Adrian


cdel...@juno.com schrieb:

Adrian,

You say it does  not lock.

When you put in operate and push the logic reset button does the green
light come on?

What is the 2nd harmonic meter reading?

  If too low the logic will not turn on the green light. Raise the reading
via the A7 gain adjust thru the front panel hole.

Is the green light burnt out?

In operate and looking at the control voltage on the meter does the
control voltage "follow" a slow adjustment back and forth of the fine
Quartz adjust?

Corby

___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.



___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] HP 5061A problem

[Adrian]

Corby,

yes, the green light comes on, but when I turn the osc. fine control by 
more than +/- 30, it goes off.
The output frequency changes with the fine adjust, so that narrow range 
appears more like a pseudo lock indication. I can even unplug A8P6 and 
the green light is still on.
Output from the phase detector is there, but the oscillator control 
instrument doesn't show any significant move.


The 2nd harmonic meter reading is near full scale.

I'm suspecting a problem with the OP Amp. It does not work as described 
in the functional test.
And, when I apply DC to the input at A8P6, the control voltage shows 
only a small jump and goes slowly back to the initial value. That's not 
exactly what I expect from an integrator.


Adrian


cdel...@juno.com schrieb:

Adrian,

You say it does  not lock.

When you put in operate and push the logic reset button does the green
light come on?

What is the 2nd harmonic meter reading?

  If too low the logic will not turn on the green light. Raise the reading
via the A7 gain adjust thru the front panel hole.

Is the green light burnt out?

In operate and looking at the control voltage on the meter does the
control voltage "follow" a slow adjustment back and forth of the fine
Quartz adjust?

Corby

___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.



___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

[time-nuts] HP 5061A problem

[cdelect]

Adrian,

You say it does  not lock.

When you put in operate and push the logic reset button does the green
light come on?

What is the 2nd harmonic meter reading?

 If too low the logic will not turn on the green light. Raise the reading
via the A7 gain adjust thru the front panel hole.

Is the green light burnt out?

In operate and looking at the control voltage on the meter does the
control voltage "follow" a slow adjustment back and forth of the fine
Quartz adjust?

Corby

___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] HP 5061A Problem

[Adrian]

Corby,

the frequency test performed yesterday as follows:

- Peak Frequency 38.78 kHz at 1V eff
- EM Output connected directly to a 34401A
- measured -102 mV (10.2 nA)

- Peak Frequency 38.78 kHz at 2V eff
- EM Output connected directly to a 34401A
- measured -133 mV (13.3 nA)

- Audio Signal off
- EM Output connected directly to a 34401A
- measured -59 mV (5.9 nA)

Corresponding circuit check meter readings when EM output is connected 
to A7 J1:

- 36
- 46
- 24

That was measured with A11 mass spectrometer fully cw. There is no peak 
when adjusting the pot, the max. is always in the full cw position.


There was still no 2nd harmonic though because the A7 input amp was shot.
I couldn't find the 1854-0023 in the xref, so I replaced them with BC546B's.
This worked until I reconnected the EM output to the amp.
Apparently the EM charges up to a high negative potential while 
unconnected. When it is reconnected to A7, the negative voltage peak 
instantly shoots the input transistor.
Touching the open connector with a scope probe shows a negative voltage 
transient of several hundred volts.
When it is connected to the amp, there are only -8mV across the 1M input 
resistor R10.
So there appears to be a very high resistance path between the EM and 
high voltage inside the tube.
This explains also why the meter is displaying beam current immediately 
after applying power to the 5061A.


Well, with a new transistor and avoiding the discharge scenario, I was 
finally able to get a 2nd harmonic that could be peaked with osc. 
coarse, A3 mod and A4.

The noisy 2nd harmonic signal is present at A8 J1.

So far it looks all good except for the EM problem that appears to be 
acceptable. At least the hope is still alive.


The unit didn't yet lock, so I will have to check the other parts of the 
control loop.


Adrian



cdel...@juno.com schrieb:

Adrian,

Most used High stability tubes are DOA but you might get lucky!

You say the low frequency test "worked".

What was the amplitude of the peak in nanoamps and was the peak well
above the noise or baseline level?

Once you were on the peak did you adjust the 41.6Khz amplitude to max the
signal?

Once on the peak you can adjust the mass spec pot on the top of A11 to
peak it.

If the peak was >8na and above noise then the tube appears to be working.

Anything below 4 or 5 na the tube is depleted, might be useable but STS
will be terrible.

If the peak is >8na then reconfigure for normal operation, in open loop
mode.

Turn the mod on and connect a scope to A8 J1 to observe the phase
detector output.

Adjust the fine and or coarse oscillator adjustments back and forth
across 5Mhz.

If you can see any output as you sweep through then set the osc.
frequency to peak the signal.

Then adjust the A3 mod and the A4 (top pot) to peak the signal.

A7 gain will adjust the amplitude.

Adj the osc freq. to null the signal (puts you "on" frequency) and see if
you have any 2nd harmonic now.

If so go ahead and see if it will lock.

Good Luck!

Corby

___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.



___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

[time-nuts] HP 5061A Problem

[cdelect]

Adrian,

You can use either Zeeman frequency as long as you use the corresponding
A1 frequency synthesizer setting.

See: http://www.febo.com/pipermail/time-nuts/2005-April/018171.html

This does not matter for troubleshooting but once you want to get "on
frequency" they need to match.

Corby

___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] HP 5061A Problem

[Adrian]

Corby,

many thanks for the comprehensive instructions.

After the harmonic generator alignment I measured 90mV into the 10Meg 
DMM input, so the beam current should be 9nA.

The floor with the harmonic generator pot fully ccw was 6nA.

I didn't change the generator output during the LF test. To get to the 
maximum, I just had to increase the frequency up to about twice the 
initial value of some 21kHz. The peak was a lot above the noise.


But, now I'm suspecting that the CBT has probably been exchanged by the 
previous owner and not been set up correctly.
Either he decided to dispose a dead CBT or he gave up half way... I hope 
the latter.
On the CBT label it says EM voltage -1615V. The applied voltage is 
actually -2500V.
I see in an old post you mentioned that the higher voltage should be ok 
for now.


Well, and something with the Zeeman frequency appears to be wrong, too. 
The measured (some) 41,6kHz from the LF test don't match the 53.53kHz 
sticker on the door. So I'm wondering what else needs to be checked.


The synthesizer output is 12.631,771,6MHz as indicated in the manual.

Btw. this 5061A has Opt. H96 which seems to indicate the added alarm and 
continuous operation outputs on the rear, and a A5 time delay divider. 
There is no clock.


Thanks,
Adrian

cdel...@juno.com schrieb:

Adrian,

Most used High stability tubes are DOA but you might get lucky!

You say the low frequency test "worked".

What was the amplitude of the peak in nanoamps and was the peak well
above the noise or baseline level?

Once you were on the peak did you adjust the 41.6Khz amplitude to max the
signal?

Once on the peak you can adjust the mass spec pot on the top of A11 to
peak it.

If the peak was >8na and above noise then the tube appears to be working.

Anything below 4 or 5 na the tube is depleted, might be useable but STS
will be terrible.

If the peak is >8na then reconfigure for normal operation, in open loop
mode.

Turn the mod on and connect a scope to A8 J1 to observe the phase
detector output.

Adjust the fine and or coarse oscillator adjustments back and forth
across 5Mhz.

If you can see any output as you sweep through then set the osc.
frequency to peak the signal.

Then adjust the A3 mod and the A4 (top pot) to peak the signal.

A7 gain will adjust the amplitude.

Adj the osc freq. to null the signal (puts you "on" frequency) and see if
you have any 2nd harmonic now.

If so go ahead and see if it will lock.

Good Luck!

Corby

___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.



___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] HP 5061A Problem

[Tom Knox]

If you have another reference like a GPS or Rubidium it makes set up much 
easier since you simply tune to 5MHz and then adjust current.
The High Perf Tubes often needed to be degaussed, I have the 10638A but no 
cable if that is any help contact me off list. 
Also placing an external power supply on the ion pump is often needed if the 
clock has been stored for a long period. 
In the past I could talk you through set up on the phone but now I would need 
it in front of me, sorry.
Thomas Knox



> Date: Sat, 21 Sep 2013 14:27:38 +0200
> From: rfn...@arcor.de
> To: time-nuts@febo.com
> Subject: [time-nuts]  HP 5061A Problem
> 
> I got a 5061A with high performance tube that I'm trying to get running.
> 
> The Cs oven works and the -3500 and -2500V are there.
> The beam current has its maximum at the correct frequency.
> 90MHz from the multiplier are there and they are phase modulated with 
> 137Hz at about 0.01 rad.
> Measured on a spectrum analyzer, the 137Hz sidebands are some -47dBc.
> I'm not sure if that low modulaion index is correct though.
> 
> Also, the harmonic generator alignment worked as described in the manual.
> Measured with the EM output into a 10M input DMM, the floor was around 
> -60mV, finally set to some -90mV after alignment.
> The low frequency test worked as well and delivered a maximum at some 
> 41.6kHz.
> 
> An unusual observation is that the meter indicates beam current 
> immediately after switching the CBT on.
> Nevertheless, it could be tuned to the correct maximum after warming up.
> However, there is still no 2nd harmonic and it doesn't lock.
> 
> Measured with a scope directly at the EM output, there is mainly noise 
> with just a faint signal when maxed with the A3 mod level potentiometer.
> 
> Unfortunately, I couldn't find reference data in the manual, neither for 
> the beam current nor the multiplier modulation index.
> 
> So far, everything seems to work properly except for the missing 2nd 
> harmonic and the unusual beam curren behavior.
> 
> Any help would be much appreciated.
> 
> Adrian
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
  
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

[time-nuts] HP 5061A Problem

[cdelect]

Adrian,

Most used High stability tubes are DOA but you might get lucky!

You say the low frequency test "worked".

What was the amplitude of the peak in nanoamps and was the peak well
above the noise or baseline level?

Once you were on the peak did you adjust the 41.6Khz amplitude to max the
signal?

Once on the peak you can adjust the mass spec pot on the top of A11 to
peak it.

If the peak was >8na and above noise then the tube appears to be working.

Anything below 4 or 5 na the tube is depleted, might be useable but STS
will be terrible.

If the peak is >8na then reconfigure for normal operation, in open loop
mode.

Turn the mod on and connect a scope to A8 J1 to observe the phase
detector output.

Adjust the fine and or coarse oscillator adjustments back and forth
across 5Mhz.

If you can see any output as you sweep through then set the osc.
frequency to peak the signal.

Then adjust the A3 mod and the A4 (top pot) to peak the signal.

A7 gain will adjust the amplitude.

Adj the osc freq. to null the signal (puts you "on" frequency) and see if
you have any 2nd harmonic now. 

If so go ahead and see if it will lock.

Good Luck!

Corby 

___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] HP 5061A Problem

[paul swed]

Adrian
It looks like you are gathering a lot of good detail. I have a couple of
thoughts. But will say there are many on this list that are far far more
knowledgeable then me. My experience was in building a hp 5060/61 combo. A
mixed marriage that worked.
That said. You don't mention the current reading. I understand thats really
relative but would be curious.
So possibilities. The photo multiplier in the tube is actually a tube. They
can go noisey with age. That would be the masking effect that would prevent
2nd harmonic.
To much current may be an additional issue. Lastly the frontend transistors
of the photo mult preamp may be going bad. Just some thoughts.
Regards
Paul.
WB8TSL


On Sat, Sep 21, 2013 at 8:27 AM, Adrian  wrote:

> I got a 5061A with high performance tube that I'm trying to get running.
>
> The Cs oven works and the -3500 and -2500V are there.
> The beam current has its maximum at the correct frequency.
> 90MHz from the multiplier are there and they are phase modulated with
> 137Hz at about 0.01 rad.
> Measured on a spectrum analyzer, the 137Hz sidebands are some -47dBc.
> I'm not sure if that low modulaion index is correct though.
>
> Also, the harmonic generator alignment worked as described in the manual.
> Measured with the EM output into a 10M input DMM, the floor was around
> -60mV, finally set to some -90mV after alignment.
> The low frequency test worked as well and delivered a maximum at some
> 41.6kHz.
>
> An unusual observation is that the meter indicates beam current
> immediately after switching the CBT on.
> Nevertheless, it could be tuned to the correct maximum after warming up.
> However, there is still no 2nd harmonic and it doesn't lock.
>
> Measured with a scope directly at the EM output, there is mainly noise
> with just a faint signal when maxed with the A3 mod level potentiometer.
>
> Unfortunately, I couldn't find reference data in the manual, neither for
> the beam current nor the multiplier modulation index.
>
> So far, everything seems to work properly except for the missing 2nd
> harmonic and the unusual beam curren behavior.
>
> Any help would be much appreciated.
>
> Adrian
> __**_
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/**
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

[time-nuts] HP 5061A Problem

[Adrian]

I got a 5061A with high performance tube that I'm trying to get running.

The Cs oven works and the -3500 and -2500V are there.
The beam current has its maximum at the correct frequency.
90MHz from the multiplier are there and they are phase modulated with 
137Hz at about 0.01 rad.

Measured on a spectrum analyzer, the 137Hz sidebands are some -47dBc.
I'm not sure if that low modulaion index is correct though.

Also, the harmonic generator alignment worked as described in the manual.
Measured with the EM output into a 10M input DMM, the floor was around 
-60mV, finally set to some -90mV after alignment.
The low frequency test worked as well and delivered a maximum at some 
41.6kHz.


An unusual observation is that the meter indicates beam current 
immediately after switching the CBT on.

Nevertheless, it could be tuned to the correct maximum after warming up.
However, there is still no 2nd harmonic and it doesn't lock.

Measured with a scope directly at the EM output, there is mainly noise 
with just a faint signal when maxed with the A3 mod level potentiometer.


Unfortunately, I couldn't find reference data in the manual, neither for 
the beam current nor the multiplier modulation index.


So far, everything seems to work properly except for the missing 2nd 
harmonic and the unusual beam curren behavior.


Any help would be much appreciated.

Adrian
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] HP 5061A problem

[wje]

Possibly, but from what I remember, the real issue is erosion of the EM 
dynode surfaces from the high level of ion bombarment, resulting in 
reduced output and lower S/N, and eventually such low multiplication 
that the tube is considered 'failed'.

Bill Ezell
--
They said 'Windows or better'
so I used Linux.



[EMAIL PROTECTED] wrote:
> Per the last statement in the message below:
>
> "The danger running without a working ion pump, though, is that at 
> eventually you'll damage the electron multiplier."
>
> Is it correct to presume then that the EM becomes covered in
> "cesium splatter" and thus the S/N output gets poor?
>
> -Brian, WA1ZMS
>
>
> -Original Message-
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
> Behalf Of wje
> Sent: Monday, August 18, 2008 1:43 PM
> To: Discussion of precise time and frequency measurement
> Subject: Re: [time-nuts] HP 5061A problem
>
>
> If you install the second unit, your 5061A still won't work. +20v at the 
> sensor output will signal the ion current monitor to shut down the 
> electron multiplier supply.
>
> If you have some high-value resistors around, can you make up a test 
> 200M resistor string? If so, see what happens when that's connected 
> across 4 & 5 of the first 3500 v supply.
>
> However, even with the failure case you describe with the first unit, 
> I'd expect your 5061A to still work at least for a short time. Do you 
> have beam current? If not, you might have other problems as well. Check 
> the voltage at A15 pin 4. You should have +18 v or so. If not, more 
> troubleshooting is in order.
>
>
>
> Bill Ezell
> --
> They said 'Windows or better'
> so I used Linux.
>
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
>
>
>   

___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] HP 5061A problem

[wa1zms]

Per the last statement in the message below:

"The danger running without a working ion pump, though, is that at 
eventually you'll damage the electron multiplier."

Is it correct to presume then that the EM becomes covered in
"cesium splatter" and thus the S/N output gets poor?

-Brian, WA1ZMS


-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
Behalf Of wje
Sent: Monday, August 18, 2008 1:43 PM
To: Discussion of precise time and frequency measurement
Subject: Re: [time-nuts] HP 5061A problem


If you install the second unit, your 5061A still won't work. +20v at the 
sensor output will signal the ion current monitor to shut down the 
electron multiplier supply.

If you have some high-value resistors around, can you make up a test 
200M resistor string? If so, see what happens when that's connected 
across 4 & 5 of the first 3500 v supply.

However, even with the failure case you describe with the first unit, 
I'd expect your 5061A to still work at least for a short time. Do you 
have beam current? If not, you might have other problems as well. Check 
the voltage at A15 pin 4. You should have +18 v or so. If not, more 
troubleshooting is in order.



Bill Ezell
--
They said 'Windows or better'
so I used Linux.


___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] HP 5061A problem

[wje]

If you install the second unit, your 5061A still won't work. +20v at the 
sensor output will signal the ion current monitor to shut down the 
electron multiplier supply.

If you have some high-value resistors around, can you make up a test 
200M resistor string? If so, see what happens when that's connected 
across 4 & 5 of the first 3500 v supply.

However, even with the failure case you describe with the first unit, 
I'd expect your 5061A to still work at least for a short time. Do you 
have beam current? If not, you might have other problems as well. Check 
the voltage at A15 pin 4. You should have +18 v or so. If not, more 
troubleshooting is in order.

The danger running without a working ion pump, though, is that at 
eventually you'll damage the electron multiplier.

Bill Ezell
--
They said 'Windows or better'
so I used Linux.



Morris Odell wrote:
> Hi all,
>
> My 5061A has failed and I've isolated the problem to a faulty A18, the +3500 
> volt ion pump power supply.
>
> The existing one has almost certainly has had the internal 200M resistor go 
> open as the output voltage is about 5 kV and it does not decay quickly when 
> the power is off.  The output voltage to the current sensing circuit is 0.
>
> I have a replacement which I suspect is faulty - when energised from a +18 
> volt source it only produces 2 kV which decays quickly and the sensor output 
> is +20 when the thing is unloaded. That 200M resistor may have gone low! Is 
> this one worth installing ie: will +2 kV pump enough ions?
>
> So I'm looking for a replacement or advice from anyone who may have 
> attempted to repair one.
>
> If anyone has a parts donor of course I'd be happy to pay appropriately 
> including shipping down here to Australia.
>
> Morris 
>
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
>
>
>   

___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

[time-nuts] HP 5061A problem

[Morris Odell]

Hi all,

My 5061A has failed and I've isolated the problem to a faulty A18, the +3500 
volt ion pump power supply.

The existing one has almost certainly has had the internal 200M resistor go 
open as the output voltage is about 5 kV and it does not decay quickly when 
the power is off.  The output voltage to the current sensing circuit is 0.

I have a replacement which I suspect is faulty - when energised from a +18 
volt source it only produces 2 kV which decays quickly and the sensor output 
is +20 when the thing is unloaded. That 200M resistor may have gone low! Is 
this one worth installing ie: will +2 kV pump enough ions?

So I'm looking for a replacement or advice from anyone who may have 
attempted to repair one.

If anyone has a parts donor of course I'd be happy to pay appropriately 
including shipping down here to Australia.

Morris 


___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.