The HP5065A and its TED, some info!

[5065AGuru]

Hi,
After working for decades on the HP5065A I discovered an undocumented mod. that HP installed into it's later optical units.
If you inspect the twisted pair TED wires leaving the optical unit and find a heat shrink section that seems to have a component the size of a 1/4W resistor inside, that's the mod.
If you open the shrink up you will find a small RF choke, I have found various values.
In the past when I could get good performance with the choke installed I left it as is. Otherwise I cut the choke out.
Got one in last month with the choke and decided to make some definitive measurements.

With the choke installed the TED current is reduced to 70mA. Without the choke the current is around 230mA. The choke is just a cheap current shunt to drop the TED current!
Ran measurements at the following TED currents waiting a couple days each time for the Rb level to stabilize.

mA        A7TP2       SN ratio

70         2.2V          3194
100       2.4            3506
150       2.65          4027
200       2.85          4201
235       2.9            4305
250       2.95          4583
275       2.95          4583

As you can see the SN eventually leveled out at 250mA.

Just for fun I then installed the optical filter mod and the SN ratio increased some more to 7565.

At this point ADEV at 100 and 100 Sec. were 7.29X10-14th and 6.40X10-14th. 

So if your unit is performing well, leave the choke in. If you want a bit more margin just cut it out.
Once removed let the unit run for a week while monitoring the 2nd harmonic reading on the front panel meter. If it increases and pegs just turn the rear pot on A7 to give you a new reference level.

Cheers,

Corby

[KE5FX]

I'm confused.  What does the TED have to do with the ADEV?  I'd guess that changing the current through it will induce some linear drift in one direction or another until the cell temperature stabilizes, but does the final temperature matter that much?  If so, that does sound like some low-hanging fruit for optimization.

Also confused about what can be expected if you follow the advice in the manual for cell-flooding correction with that choke in place.  Assuming that it takes the bulk of the current, as your observation suggests, then forcing 1A into the TED+choke combination would smoke it to a crisp, I'd think.

[edpalmer42]

When I was restoring my 5065A, I pulled off the heatshrink tubing that was around the mod and found that the previous 'tech' had carefully soldered the red and black wires together and then applied the heatshrink.  No inductor.  The TED was completely shorted out.  Some people just shouldn't be allowed near a soldering iron! 

[chuckb]

Corby
A little off topic but my unit has a S/N ratio around 1500 with an A7TP2 voltage of 6.2V. This indicates a bright Rb lamp source. Do the better units (higher S/N Ratio) have a lower voltage on A7TP2?
 

[5065AGuru]

The TED is used to establish a particular level of Rb atoms in the cell. Too much and you get cell flooding where the amount of spectral line getting through the cell is reduced or eliminated completely! Too little and the signal degrades due to lack of atoms. Just right and you get the perfect Rb level that will promote the best SN ratio.
Years ago I one re-soldered the TED wires back into the circuit backwards. This kind of easy if not paying attention as the black TED wire goes to the post with the red wire on it and the red Ted wire to the other. Just remember black goes to red! Anyway after coming back next morning there was no discernible signal out of the cell. Restoring the proper polarity restored operation (after a few days of operation.) Now as to ADEV. The specified ADEV is reached if the SN ratio is above 250. However at 250 you will most likely be right at spec. for larger SN ratios the ADEV improves, typically by a factor of 4 or 5. Most users wont notice or cannot even verify that performance level. You are right that it is "low hanging fruit". Cutting out the choke is a trivial job and can increase the signal level appreciably.
You area also right about the cell-flooding correction procedure. It will work as written but will take a lot longer. The choke does survive a 1 Amp TED de-flooding as I have tried it before, but if I need to it I remove the choke so as to get the full 1Amp into the TED. One of the first things I do during a repair is to check the voltage drop across the TED. It's pretty close to a short so a couple tenths of a volt is all you should see. This of coarse if the choke is not in circuit! If the TED is open you will read the drop of the silicon diode that is located where the TED wires terminate at the 2 edge connector pins.

Cheers,

Corby

[5065AGuru]

Chuck,

A SN of 1500 with an A7TP2 of 6.2V P-P is fine. I kind of would expect a higher SN ration with that high of TP2, but you are definitely OK at those levels.
If you look at the chart you will see that the SN ratio increases as the TP2 signal increases.

Cheers,

Corby

[KE5FX]

The TED is used to establish a particular level of Rb atoms in the cell.

Hmm, interesting.  I was always under the imrpession that the TED's job was solely to maintain the Rb reservoir in the tip.  Is it reasonable to speculate that if increasing the TED current results in an SNR improvement, the real problem is that the cell oven temperature is hotter than it ideally should be for that RVFR assembly?

[5065AGuru]

The TEDs job is to maintain the proper amount of Rb in the tip while allowing the correct amount into the cell.
My best guess is that each cell has a slightly different amount of Rb when fabricated.
Also they probably try to put in a little too much.
That way the TED can be used to hold just the right amount in the tip while allowing the proper amount out into the cell.
The Varian R20 and V4700 which preceded the 5065A also used this scheme but instead of a TED they had a metal braid sleeve that went over the cells tip and had a route outside of the oven to allow relative cooling via ambient temperature. That worked but you had to be very careful when doing maintenance as if you torqued the metal sleeve you could snap the cells tip off!
Trying to juggle the cell temp. would be problematic as you want a low cell temp (around 65 C) for best performance and the match between the main and filter cells will degrade at differing temperatures.

Cheers,

Corby

[edpalmer42]

To add a little more TED info .....

Depending on the age of your 5065A, the TED may or may not have a plug/socket in the twisted pair that connects it to the rest of the circuit.  So measuring the current might require desoldering.  As a quick check, without desoldering, measure the voltage drop across the TED.  If you're pumping 1A through the TED, expect to see somewhere in the range of 0V2.  I measured this and Fig. 3-1 in the manual hints at it by specifying a power supply that puts out 1A at ~0V2.  I've also seen a value of 0V18 mentioned in a posting on Time-Nuts.  In normal operation, I see ~0V06 across the TED.

As for why the inductor was added in the first place, the TED is wired in series with the power supply that feeds the 100 KHz frequency divider.  So if the TED were to fail open, that would kill the divider and the optional clock.  Could the inductor be intended to provide an alternate path to keep the divider powered?     That seems like a bad idea to me because then you'd have no indication that the TED had failed.

[KE5FX]

Could the inductor be intended to provide an alternate path to keep the divider powered?     That seems like a bad idea to me because then you'd have no indication that the TED had failed.

That was the idea behind the diode.  The wires to the TED could be disconnected entirely (and sometimes were, in the case of one of my units), and it would still work.

Seems like the inductor would have been soldered directly across the diode if they were trying to limit the voltage drop.

[edpalmer42]

Could the inductor be intended to provide an alternate path to keep the divider powered?     That seems like a bad idea to me because then you'd have no indication that the TED had failed.

That was the idea behind the diode.  The wires to the TED could be disconnected entirely (and sometimes were, in the case of one of my units), and it would still work.

Seems like the inductor would have been soldered directly across the diode if they were trying to limit the voltage drop.

Thanks, I missed that diode.  The TED appears on a few sheets in the schematic.  For those playing along at home, the diode is CR3 in Fig. 8-6.  So it looks like I got it backwards.  The power comes from the +20V pin on the 100 KHz divider (module A4) and feeds through the TED to the Phase Detector Assembly (module A8) and both ends of the TED are designated as +20V.  Yeah, not confusing at all.   

[5065AGuru]

And just to keep you guessing the +20V current through the TED to A8 continues on and also supplies A10,A11,A13, and A1.    The total of those 5 modules currents comes to around 235mA and is the current through the TED.
I pointed this out a while ago to someone planning to replace the A1 synthesizer with a modern DDS based unit. I have done this before and my design pretty much matched the current draw of the original A1. New designs must take this into consideration!

Cheers,

Corby

[Uky]

Since the 5065A that I has been repaired som many times that I think I can do it in my sleep, I would be intersted in a modernized A1 design since the one in my instrument is full of unobtanium logics. I have the schematic drawing for the TTL-version and it could be one option to recreate that design but a DDS design is more appetizing. If made public, I am interested.

 

[KE5FX]

I'd been planning to add thermal cutouts to the RVFRs in my 5065As for a while, and decided to do it with thermistors and an MCU rather than simple thermal fuses in order to support data logging as well as overtemp protection.   The thermistors are placed at more-or-less arbitrary locations under the shield covers at both ends of the physics package, so they don't return very accurate readings, but they are OK for the original intended purpose of detecting thermal runaway and can give a reasonable view of relative temperature changes for logging purposes.




In the plot above, the controller is monitoring the second harmonic and photo readings as well as the thermistors at both ends.  As it turns out, the cell temperature has a massive influence on the photocell current and 2nd harmonic levels as well as the frequency. 

Normally this unit runs at an (indicated) cell oven temperature near 43.5C, photo I near 10, and 2nd harmonic near 30, but at the time I started the plot above, the cell oven had been disabled for a half-hour or so and allowed to cool down.  Notice that the 2nd harmonic level starts out near 70(!) at left, corresponding to a reported cell temperature of about 40.5C. 

As the oven cools further the 2nd harmonic level starts to drop; by 39.0C it is back to the ~30 vicinity.  At that point I turned the cell oven back on, and you can see it warming up in the second half of the plot.  The second harmonic reaches a clear peak just below 40C -- unfortunately at a point where the df/dt is as high as it ever gets -- and then falls back down to 30 as the oven approaches its final temperature.

As an experiment, I changed A11R5 from 430 ohms to 560 ohms to bring the cell oven temperature down to a level near the 2nd harmonic peak, and performed a basic realignment.  A7TP2 is capable of about 4V peak-to-peak on this unit (and always has been), but I'm not sure where it ended up.  The C field obviously had to be readjusted to take out the frequency error.  There seems to have been little or no improvement in short-term performance, though:



Note that these two traces were taken over five years apart.  The apparent improvement below t=0.2s is likely due to a better loop gain setting, and isn't relevant here.  Same for the apparent degradation beyond t=40s, as the DOCXO is what is being measured beyond t=~5s.

It's hard to judge long-term performance immediately after power-cycling and realignment, but I don't see any evidence that it has gotten worse.  I wouldn't be surprised if lowering the cell oven temperature actually does make it a little worse, though, given the increased frequency tempco at the new setting.  After it settles down for the next few days I'll do an overnight run against the other 5065A and a GPSDO. 

So, Corby -- is this effect the same as what's being observed when the TED current is adjusted?  Or would you expect the cell temperature and TED temperature to have independent effects? 

[5065AGuru]

My thoughts on changing cell temperatures:
The nominal temperature (65-66 degrees C) sets the temperature of the reference cell and the filter cell.
The filter cells temperature is critical as to allow as little of the interfering line through as possible.
If you wanted to test the performance at say 63 to 69 degrees you could see if it effects performance. What you were seeing is not the same as the TED info. Your data is taken in a short enough interval that any change in cell temp. you make would not effect the Rb level in the cell as it takes a couple days for that to occur.
Cheers, 

Corby

[KE5FX]

After a couple more days to let things settle, I don't see a medium/long-term difference at the new temperature setting:



It's possible that Rb2 is much better than Rb1, of course.  If so, I wouldn't be able to tell what's really going on.

[5065AGuru]

Yes, Those plots look great! Thanks for sharing.

Cheers,

Corby

[notfaded1]

Maybe dumb question but Since the 5065 has the rotary dial meter and 5061B a similar setup.  Would it be possible to tap all of the levels on the dial at the same time and data log them?  I don't know enough about the circuitry to know if measuring more than one tap on the rotary switch at the same time is possible without interfering with each other?  Can you measure them all at the same time?

Bill

[KE5FX]

Maybe dumb question but Since the 5065 has the rotary dial meter and 5061B a similar setup.  Would it be possible to tap all of the levels on the dial at the same time and data log them?  I don't know enough about the circuitry to know if measuring more than one tap on the rotary switch at the same time is possible without interfering with each other?  Can you measure them all at the same time?

Bill

Yes, because the tap points are upstream of the meter switch.  All of the parameters could be monitored given enough single-ended DAC channels and appropriate voltage divider resistors.  All of the meter switch positions are ground-referenced except the (old) oscillator oven supply.

That said, the extra load on the points being monitored will have a small effect on the meter reading for the selected switch position, and vice versa.  That could be avoided with some opamp voltage follower stages, but it probably isn't worth the trouble.