HP 6289A power supply not outputting full current

[iroc86]

First time poster here but I've been following the forums and videos for many years. This is a great resource for learning electronics and I'm hoping to get some assistance repairing an '80s-vintage HP 6289A power supply. I think I've tracked down the problem but I have a few questions about the symptoms and diagnosis.

My 6289A will not output its rated maximum current of 1.5 A. If I connect a load resistor, max out the voltage knob, and try to dial up the current, the output peaks at around 950 mA. The same thing happens when I short out the + and - terminals as if I were setting the current limit. If the power supply sits for a while with a load connected, the output current will slowly increase beyond 950 mA, but when it cools down, it returns to the previous state.

First I checked all of the voltages and measured the ripple on the big filter caps. Everything seemed to be within spec.

The CC and CV regulation is spot on at lower current loads, so then I suspected that the regulator circuit was breaking down above ~950 mA output current. I swapped out many of the transistors and checked most of the diodes, but it didn't change anything. I also noticed that if I shunted the CC programming resistor R19 with a lower value, I could get the full 1.5 A output, but this didn't seem like an appropriate fix.

After some further diagnosis, I'm now thinking capacitor C16 may be bad. This is a 50 uF, 500 V ceramic on the pre-rectified 62 V rail presumably used for input suppression (there are no X/Y safety caps in this unit). It produces a fairly loud 60 Hz hum as the output current increases. Comparing the two output ripple measurements below, the waveform changes to a 60 Hz-modulated output when the supply hits a wall at 950 mA. Below this cutoff, the output seems fairly reasonable. If I manually raise the temperature of C16 with a heat gun, I can actually see the "bad" waveform change shape and become like the good one.

I haven't replaced C16 yet, but I wanted to get some feedback. Does this seem like a reasonable diagnosis? What failure mode in C16 is limiting the output current of the power supply? Or, am I off base and missing something else?

Here are a few pictures of the unit. I've also included a schematic for the 6284A, which is nearly identical except for the component values and voltage rails (I couldn't find a good schematic for the 6289A worth uploading). Thanks for the help.

[duak]

If I look at the ripple voltage waveform, I see the period is 16.7 ms corresponding to the line frequency of 60 Hz.  It ought to be 8.3 ms corresponding to 120 Hz.  The circuit should be balanced, but something on one side is not working well.  I would bet that either CR26 or CR27 or the transformer or one of the solder joints connecting these has gone blinky.  The buzzing noise tells me that the transformer is not happy and it could be caused by an imbalance.

I can't envisage any problem with C16 that could cause this problem.  If C16 had developed a partial short, it would get quite warm all by itself.  However, I've seen strange things that I wouldn't have thought would have caused trouble.

It's unlikely, but check the votages and waveforms on the other filter caps, C10 & C12.  These are floating power supplies not referenced to -OUT so take care.

If you don't find the problem, show us the waveform at TP 27 at full load.

I have a copy of the manual from the Keysight site, but the schematic isn't all too clear - not the best of scans.

[iroc86]

Thanks for the help, duak. You probably found the same (somewhat illegible) Keysight manual that I did. It's useful for comparing component values but I wish there was a better one floating around.

I agree that the 60 Hz waveform is a bit odd; the other filter caps all have a frequency of 120 Hz. Per the attached chart (from the 6284A manual), I checked all of the voltage test points a few days ago and everything was within spec. I also checked the voltages at various loads up to cutoff at 950 mA. The ripple across the capacitors remained the same except for the big C14 filter cap, which had almost 4 V of ripple at ~1 A output load. My other HP power supplies behave similarly, so I didn't think it was unusual.

If I have some time later I'll try to post a waveform of TP 27, which is essentially the voltage across C14. If I recall, it was a regular sawtooth wave (AC coupled) where Vpp was proportional to the output load. I don't remember seeing any distortion on C14 when the output went funky, so I originally assumed the problem was in the driver or regulator circuit.

From your thoughts about C16, I'll take another look at the transformer and CR26/CR27 rectifier diodes. It's possible that my heat gun test may have been hitting those components, although I tried to mask things as best as I could. I do believe that C16 is still making the noise because I can conduct the vibration through a screwdriver; it doesn't happen on the transformer... unless maybe the noise is resonating through C16 since it's mounted nearby.

[bob91343]

I would first suspect the 'current input circuit' and see if it's working properly.  I seriously doubt that the ceramic capacitor you are looking at is causing trouble.  Either the power supply isn't putting out enough voltage under load, or, more likely, the current control circuit isn't allowing sufficient current.  Look at the current sensing resistor(s) and the error amplifier.

[ArthurDent]

I agree with duak that either CR26, CR27, or a connection from the transformer through one of the diodes to C14 is open which accounts for the 60 hz instead of 120 hz ripple. With only one good diode or ½ of the transformer providing power, the D.C. voltage across C14 will droop too much at high loads.  Check the voltage on C14 compared to the output voltage meter at high load and see if the C14 voltage is too close to the output voltage causing that to limit the max current the supply can output.

I also see no way C16 could possibly hum and I’d wager that if you lift one leg of that capacitor and retest you will find that it’s the magnetic field of the transformer that is causing the hum. I have one supply that is quite loud at high loads. 

[iroc86]

Sorry for the delay in posting an update; I was out for a few days. I made some new observations based on all of your feedback:

• CR26 and CR27 turned out to be okay. For good measure, I swapped in a pair of 1N4004s, but the problem was still there.
• The voltage across C14 is 65 V at no load and 60 V at full load. The output is only around 10 V at full load with a 10 ohm resistor, so that rail is still 50 V higher than the output. The waveform at TP 27 is attached; there is no discernible change in the shape of the curve when the supply hits the wall. Only the ripple changes.
• I pulled one leg of C16 and the hum went away. It didn't affect the problem, though.
• I reflowed all of the solder joints coming from the transformer. No change, but I knew this was a shot in the dark.

Next I watched the output waveform while adjusting the current and voltage knobs. When the current knob is maxed out, the output is normal until the wall at 950 mA (adjusting voltage). When the voltage knob is maxed out, the output has the 60 Hz ripple at all loads (adjusting current). This seems to suggest that the current limiting circuit isn't operating properly.

From a diagnostic standpoint, I'm seeing two separate issues here, but they're probably related: 1) the current is limited too early, and 2) there is 60 Hz ripple on the output when the current is limited.

So, I thought that perhaps the output transistors Q6 and Q7 were being fed a wonky 60 Hz signal. This is true. TP 22 and TP 26, which are connected to the transistor bases, exhibit the 60 Hz waveform when the supply goes into current limiting mode. I think this makes sense as to why the output rails have the same kind of ripple.

Then, I checked the ripple on the regulator rails (referenced to +S) with varying output load. The -6.2 V, +6.2 V, and +12.4 V rails were all rock solid, which tells me that the 24 V between TP 34 and TP 37 is probably okay. However, the -4.3 V rail exhibits the 60 Hz ripple when the supply goes into current limiting mode. This rail is used in the Driver & Error Amplifier circuit. Might the -4.3 V rail be picking up interference through the transformer and passing it along to the output transistors? (The base of Q7, in fact, is wired into the same transformer tap as the -4.3 V rail.) I also swapped out VR4, the 4.3 V Zener diode, but it didn't make a difference.

I also tried to troubleshoot the Current Input Circuit, but I'm not sure how to interpret the output of the OR gate at CR3 and CR4. If I probe TP 17 relative to +S, I get a constant 2.9 V output regardless of the voltage and current settings on the front panel of the supply. How does this (and the Voltage Input Circuit, for that matter) drive Q3? Should I be referencing against a different rail?

I think we're slowly getting there. I appreciate the guidance very much.

[duak]

Iroc86 - sure, I could see that something in the -4.3 V supply would affect current limiting.  Most likely CR24 or CR25 or a solder joint in that circuit.  C12 could also be bad - if so, anything between 470u to 680u, 50 V or greater should be just fine.
 

[iroc86]

Well, I just swapped out CR24 and CR25, along with C12 from another "good" HP power supply that I have. Still a no go!

I also checked all of the test points that are connected to the -4.3 V rail. They all exhibit the weird behavior, especially TP 12 and TP 42 from the Voltage Input and Voltage Clamp circuits, respectively. At those locations, the output waveform hits almost 1 Vpp at the cut-off point, but then it quickly falls back down to ~60 mVpp with the 60 Hz modulation. Screenshot attached.

At this point, I'm kinda thinking that the -4.3 V rail is the culprit, but I'm not sure how else it could get 60 Hz interference except from the AC line. Since CR24, CR25, and C12 all seem to be okay, do you think the next logical step is to investigate the transformer? Is there some way I could test the windings to see if something is breaking down internally? I have a Variac--how about injecting a separate AC voltage to drive just that part of the circuit?

[duak]

Other than what we've discussed, nothing jumps up and says "i'm bad!".  It's kind of a complicated circuit with floating supplies and non-obvious interactions and now we're getting into subtle part failures.

You could check the windings for continuity - each winding should be, say, 10 ohms or less.  Shorted windings are harder to find - usually the transformer gets unusually warm with no load.

I've attached a multipage scan of the schematic, but it doesn't seem to be all that clear.  It has some voltages to check for.  The voltage across C12 is supposed to be 53 V so my advice above that it was OK to use a cap rated for less than 75 V is clearly wrong - I was reading the 6284A schematic.  A 63 V cap is OK but right on the 15% margin.

I'd check the parts around Q6 and Q7, in particular R36 and C6.  R81-R83 look like they should have a combined resistance of 40 ohms.

I'd like to see the waveform across VR4 (TP 41 to TP 23) as the load current is varied.  Note: TP 23 is pretty much +Vout so this measurement has to be taken carefully if the 'scope is grounded.

[xavier60]

The voltage control loop could be tested on its own by removing CR4 to disable the current control.

[duak]

xavier60 - that's an excellent idea to test the voltage loop.  Have to be careful with the load resistance though - without current limiting a 10 ohm load could easily draw too much current and kill a pass transistor.

[iroc86]

These are all really good avenues for diagnosis, guys. I'll look into this again in more detail over the weekend and try out some of your suggestions. Good reminder on the current draw vs. load resistance--I have a 100 W adjustable power resistor in my Digi-Key shopping cart but I'm waiting on some other components before I place the order. I might want to hold off testing the control loops individually until I'm sure I won't blow up the supply.

Yesterday I began systematically lifting "non-essential" components from the board in an effort to rule out as many items as possible. It's not my preferred way of troubleshooting, but this one has me stumped. Of note was the removal of Q7, which is apparently only there to reduce the load on Q6 at higher output voltages. With Q7 gone, current still flows through R81-R83 (which tested good). I thought that maybe the 53 V rail was passing some interference through Q7, but it didn't make a difference. Aside from the -4.3 V reference voltage, there's really nothing else from the 53 V tap used here, so maybe I'm chasing a ghost. I'll post up a waveform of VR4 soon. FWIW, R36 tested good and C6 was one of the components I lifted.

Thanks for posting up the 6289A schematic. No worries on the cap voltage mixup; it's hard to keep things straight when referencing similar service manuals. I found another schematic that might be the best one yet. It's for the 6255A, which is the dual-output version of the 6289A. Near as I can tell, the schematics are identical because the '55A simply uses a separate PCB for each channel. I've attached the new schematic, which is all one page. It should be a lot easier to reference and has the correct component values.

[xavier60]

Much of the Base current for Q6 comes from the Emitter of Q7 via R35. VR5 sets the Base of Q7 at some voltage above the Q6's Emitter voltage.
This area would be worth while checking. I don't know what VR5's voltage is.

[xavier60]

It might be useful to know what voltage change there is on the Base of Q4 while going from the good state to bad with increasing current setting.

[iroc86]

I decided to put all of the components back in and start from square one again. I also changed my troubleshooting approach: instead of checking the transition between CV and CC modes by turning the control dials, which may affect other parts of the circuit, I am now changing the load to alternate between states. I only have 10 ohm power resistors on hand at the moment, so I set the CV/CC switch point to ~7.5 ohms using the chart below from another HP manual. With the panel dials set to 1 V and 130 mA, I can run one 10 ohm resistor for CV operation and add another resistor in parallel for CC operation. The bad waveform only appears in CC mode, regardless of output load, so I can troubleshoot at much lower power outputs that I originally thought.

By varying the load, I figured that I should be able to watch the output of the current and voltage comparators change when the supply switches modes ("error voltage" per the manual). This would occur just behind the OR gate on the anodes of CR3 and CR4. If I trigger off the output of the Voltage Input Circuit at CR3 (TP 12), I can detect a clear falling edge as the supply transitions from CV to CC mode. I presume that the fall time is dictated by the speed of Q6 and Q7 to adjust their output as necessary to accommodate the change in the load.

I expected to see a similar (if inverted) situation in the Current Input Circuit at the anode of CR4 (TP 16), but this was not the case. The output of Q2B is pegged at 3.6 V regardless of what the supply is doing. Consequently, the output of the OR gate (TP 17) always reads about 3 V. In all of my probing, I'm using +S as a reference per the manual.

I don't think the output of the current comparator should be a fixed value; frankly, I'm not sure how the supply is operating under these conditions. If I trigger off CR3 and watch TP 17 (output of OR gate), there is no change, AC or DC. Yet, if trigger the same way and watch the base of Q4, there is a small amount of ringing and then a definite change in DC voltage, suggesting a change in the output of the power supply. I'm not sure how the downstream amplifiers can respond to a change that doesn't exist. Maybe I'm probing incorrectly?

I also checked the voltage at VR4, the -4.3 V rail, to see if that was causing any of the output ripple. I'm not sure that's the case anymore. The DC amplitude is around -4.6 V and the ripple is less than 10 mV, which is within spec (the manual states normal ripple at 20 mV). It holds while varying the output, too.

[xavier60]

With the CC control at a high setting and no load, the Base of Q2a should be much more negative than the Base of Q2b. Q2b should be passing the full tail current pulling its Collector to something below 2V.
Low tail current maybe, R22.

[duak]

The way I think of this circuit is that the Voltage Input and Current Input circuits each have vetoes that override and reduce the voltage/current applied to the load.  Diodes CR4 and CR5 'OR' the veto signals from their respective comparison circuits.  I think this is a current operated circuit, so the voltage change at their cathodes may not be that large.   I can't think of any correct operating condition where at least one of the vetoes is not active, ie., in control.  There might be a short time when the voltage or current mode switches back and forth and neither is in complete control, but it takes a short time for one to take over and keep output within the set limits.

Does the anomalous waveform change much at lower current limit levels?

Are all the jumpers on the back terminal strip in place and tight?

This unit is in middle age.  What happens to the output if you tap on the variable resistors with the plastic end of the screwdriver?  Maybe one has gotten a bit blinky with age.

[iroc86]

Thanks for sticking with me, guys. I finally have some promising news below.

duak, if I remember correctly, the irregular waveform seems to lose some of its amplitude and shape at low current levels, but not much. If I dial it in "just right" at the CV/CC switching point, then the ripple goes through the roof at several volts peak-to-peak.

All of jumpers are good and tight. Tapping around on the board didn't seem to affect anything, either.

I took xavier60's advice and pulled CR4 to disable the Current Input Circuit, essentially turning the supply into a "CV only" unit. I made sure to put a high-value resistor across the output terminals to avoid any current surges. I was able to verify operation up to 40 V with no issues. I also tested the current output with a 10 ohm power resistor and was able to output the full rated current (!) by adjusting the voltage control. The output ripple was completely normal and did not exhibit any of the 60 Hz modulation as when in CC mode. So, the next step is to start troubleshooting the Current Input Circuit. xavier60, I'll check Q2 per your analysis and report back.

duak, I think your assessment of the CV/CC mode switching logic is correct, but I don't think the circuits are controlled by current. I tried lifting CR3 and putting an ammeter in series to measure the current, but it stayed fixed around 145 µA regardless of the output voltage and current. I read through the theory section in the manual again. HP describes a transient scenario to explain the way the feedback loop regulates the output. I highlighted the sentence that suggests it's a voltage-operated loop after all. What do you think?





To check this behavior, I set up the following test: CR4 removed (per above), 10k resistor across the output terminals, voltage control set to 10 V, and a 10 ohm resistor ready to short the 10k. I'm externally triggering my scope off the floating 10 ohm resistor lead and measuring the collector of Q1A, which is essentially the CV input to the OR gate. I'm using a second channel to observe the output voltage of the supply. When I connect the 10 ohm resistor, the scope triggers and captures the regulation happening in the Voltage Input Circuit to maintain 10 V despite the 1000X increase in output current (1 mA to 1 A). This all happens so fast that I couldn't see it before, and I assumed something was messed up because the OR gate was always at a stable value. As shown in the picture, the output voltage drops slightly due to the instantaneous load change, but Q1 alters its conduction to stabilize the voltage again... and then stabilizes itself.

I still think something is wrong with Q2 and/or its associated circuit based on the above, but I agree that perhaps it's normal for the base of Q3 to appear like it's sitting at some fixed value; if the regulation is working properly, I suspect that the supply will drive Q6 and Q7 at whatever level is necessary to maintain a zero change on the driver amplifier input.

[duak]

Iroc, you're right about this being a voltage operated circuit too.  You've clearly shown there are some significant voltage variation during CV/CC mode transistions and that should make fiinding the problem easier.  My experience with diode OR gates and transistors in general is that it doesn't take much of a change in voltage to cause a large change in current and I understood from your observations that the voltage variations were quite small.

I'd like to see what's happening on TP 14, 15 & 16 when you're getting the anomalous output during CC mode.  I would think that in CC mode, there should be almost no variation in any of these TPs.

The latest 'scope shot shows an interesting double pulse where the second pulse is of lower amplifude.  It'd be interesting to see where it's coming from.  I'd start with TP 11.  Also, the Voltage Clamp circuit (p 4.5) seems to operate in CC mode.  Perhaps there's something interesting on TP 42 and 43.

I have a Kikusui power supply that probably uses a similar design to hp's.  It would be clean at low and no output current but would put out a weird AC ripple at mid to high currents, but not at all voltages.  It turned out the capacitor on the output terminals was bad.  Your supply has a 490 uF cap across the output.  May I have you try putting 100 to 500 uF across the output, vary the current limit and see what happens?

[xavier60]

With the CC control at a high setting and no load, the Base of Q2a should be much more negative than the Base of Q2b. Q2b should be passing the full tail current pulling its Collector to something below 2V.
Low tail current maybe, R22.

The Base of Q2a should be close to -0.66V with the current control set to 1 amp and not loaded.

[iroc86]

It'll probably be a few days until I can get on this again, but I wanted to post back with some comments.

duak, I think I did check those TPs earlier on, but not with triggering the scope. Maybe we'll learn something if I set up a similar test as with the CV circuit. Also, regarding that second pulse, it may have just been a loose contact when shorting the output with the 10 ohm resistor leads, almost like switch bounce. I'll run through the triggering a few more times and see if it's repeatable.

That's an interesting story about your Kikusui supply. I don't have an LCR meter and can't really test any of these caps, but I do plan on replacing them down the road for good measure. I've been able to rule out the 490 µF output cap C20 by disconnecting jumpers A9 and A10, although I was hoping for an easy fix like that! In so doing, I saw the output ripple double to around 2 mVrms, so I guess the capacitor is still doing something useful. I have a spare 490 µF I can put across the back and see what happens when tweaking the current limiter, as you suggested.

xavier60, I'm impressed with your analysis of the circuit and being able to provide details about its operation. Can you elaborate on how you've come up with the numbers? The dynamics of the power supply operation are still a bit fuzzy for me, especially with the negative voltage rails and ground reference to +S. I'm also not entirely sure how the summing points at A5 and A6 are supposed to work.

[xavier60]

This is the Harrison or floating type of design that you might have noticed others also mention.
Having the + output or +S be the ground reference for all of the functional blocks including for the control rails, allows the blocks to be connect directly together and control the Base of the pass transistor with respect to its Emitter.
Most linear lab and bench supplies are based on very similar principles.

Some voltages given are approximate.
The Base of Q2b is tied  to 0V so assume that the Emitters are -0.6V.  R22 goes to the -6.2V rail putting a drop of  5.6V across it resulting in 0.9mA of tail current for the differential pair.

The voltage on A5 doesn't change much, actually the bit it does change by doesn't matter at all.
The voltage across the Current Programming resistor and so its current remains constant.
This current causes a negative voltage on A5 in proportion to the resistance of the Current Control. -0.66V for 1 amp setting.
The -0.66V applied to the Base of Q2a causes it to be cut off causing all of the 0.9mA of tail current to flow through Q2b.
The voltage drop caused across R24 results in 1.9V at Q2b's Collector. It looks like 1.5V marked on the schematics?

The increasing voltage drop across the Current Sampling resistor that an increasing load current would cause, makes A5 change in the more positive direction.
As the voltage at Q2a's Base approaches 0V, it begins to conduct, leaving less and less current for Q2b causing its Collector voltage to rise, and eventually taking control of the output from the CV loop.

CR5 is a likely fault suspect.

Extra: Just occurred to me that Q2a's Base will not quite get to 0.66V because of CR5 being forward biased, and assuming it isn't faulty.
Anyways, a voltage reading of  Q2a's Base will be useful.

 

[duak]

Iroc, I probably can't improve on xavier's description of the Current Input circuit.  I'd probably overcomplexify and confusicate it just trying.  These days, I seem to confuse desktop stuffing and stovetop publishing all the time.

It's important to look at the waveforms on TP14 thru 16 while in CC mode.  I'd also suggest looking at +6.2 V and -6.2 V at high sensitivity with AC coupling.  When I think about it, any ripple on these supplies is coupled coupled directly into the veto signal and thus affect the output.  hp used a fairly elaborate regulator for the reference voltages to minimize noise because of this sensitivity.

If there is ripple on the reference voltages, I'd check CR22 & CR23 and C10.  A quick test is to connect your 490 uF cap across C10 (check polarity!) and see if there's any difference.  Since you're seeing 60 Hz, i'd suspect a diode or connection.

[xavier60]

I have been ignoring the ripple because I am hoping that it is the result of the power supply operating open loop because of a fault in the Q2 area.
The 60HZ component could be due to primary to secondary capacitive coupling although it does seem to be more than would be expected.

[iroc86]

Thanks for explaining the circuit theory, guys. I think this makes a little more sense now. Would I be correct in stating that the floating +S ground allows all of the functional blocks to shift their common references "automatically" as the output voltage changes? How does -S, the output "common," play into this? The operation is different than other control circuits I'm used to, where the reference points are fixed. What's the advantage of a floating design?

As for the troubleshooting, we are making some progress! I probed the test points in the Current Input Circuit while in CC mode and discovered that TP 14 (base of Q2A) was not behaving as described. It turned out that CR5 was indeed faulty (shorted) as xavier60 suggested. I presume that this short was causing Q2 to conduct constantly? I noticed that the OR gate now seems to work as expected, with the inputs at CR3 and CR4 switching according to CV or CC mode.

After fixing CR5, the current limit now reaches the full rated amount (1.8 A in over-range) . However, I am still experiencing the weird 60 Hz modulation in CC mode .

I took duak's advice and checked all of the voltage references again. I admit that I don't have the best probes for high sensitivity measurements, but I couldn't really detect any 60 Hz ripple on the rails. The 6.2, -6.2, and 12.4 V rails were all within 1 mVpp, too.

For confirmation, I also added the extra 490 µF cap across C10. There wasn't any change in the output ripple, although the ripple on the 12.4 V rail dropped from 4.5 mVpp to 1 mVpp. Either way, it's in spec, but at least I know my extra cap isn't toast. I also put the capacitor across the other filter caps, C12 and C14, and did not see any significant change on the output ripple. So, I don't think the ripple is a filtering issue.

So, thus far, I had only been able to observe 60 Hz ripple on the output. I couldn't find it anywhere else in the circuit. That may have changed now, but more probing is required--I found 60 Hz in the Voltage Clamp Circuit at TP 43. I think this circuit is only used in CC operation, so it makes sense that it would affect that operating state.

I have three waveforms below that show different states based on the position of the voltage control knob, relative to the CV/CC transition point, with the current limit set to 1 A. In full-on CV mode, the output ripple is normal and there's an obvious 60 Hz sine wave on TP43. At the transition point, the 60 Hz waveform changes, but it curiously corresponds to the 60 Hz output ripple (though inverted). In full-on CC mode, the 60 Hz waveform goes away entirely, and yet the output ripple remains.

I'm not sure what to make of this just yet, but I think I need to start probing some other areas to find where the 60 Hz interference is coming from. I actually disconnected the Voltage Clamp Circuit (via CR30 and CR32) in my testing a few days ago, just to rule out extra components, and the 60 Hz ripple was still there, so I suspect it might be coming from another place and just manifesting itself in the clamp circuit. I'm thinking the -4.3 V rail might be worth checking again now that the Current Input Circuit is fixed.

[xavier60]

 The 60HZ signal at TP43 should be the same as TP12. I would like to know what it looks like upto TP22.
In CV mode,

[duak]

The floating series regulator is a flexible design that allows for remote control and ganging of power supplies.  hp has a description of the design concept here: http://www.hpl.hp.com/hpjournal/pdfs/IssuePDFs/1962-07.pdf (hp Journal, July 1962 issue)    Just consider it to be positive ground with +S as the common reference and you're good to go.

An idea from way out in left field about the 60 Hz noise - perhaps the transformer's magnetic field is inducing some AC voltage in R54, the current sampling resistor.  R54 is a wirewound 0.66 ohm 20 W resistor and may have enough turns to develop enough voltage to affect the CC mode loop.  If R54 has a central hole, then putting a steel screwdriver in it should increase the 60 Hz noise.  If R54 is secured to the board with a steel screw, removing it might make a difference.

[xavier60]

I'm not sure what to make of this just yet, but I think I need to start probing some other areas to find where the 60 Hz interference is coming from. I actually disconnected the Voltage Clamp Circuit (via CR30 and CR32) in my testing a few days ago, just to rule out extra components, and the 60 Hz ripple was still there, so I suspect it might be coming from another place and just manifesting itself in the clamp circuit. I'm thinking the -4.3 V rail might be worth checking again now that the Current Input Circuit is fixed.

I suspect that the test points I mentioned earlier will have very little 60Hz on them or none. 
Because I don't know the value of CR3, I cant figure out the state of Q10.  So the 60HZ is from either the -4.3V, board leakage or just stray capacitive pickup and there is no real problem.
From my own design experience, some ripple in CC mode is expected because of the combination of BJT Early Effect and the lower loop gain  in CC mode.
Suspect areas of board leakage can be confirmed by directly probing the surface.

[iroc86]

Thanks for the PDF link, duak. That looks like an interesting read and should improve my understanding of these supplies. The ganging and remote control aspects make sense for the positive ground configuration.

Okay, so I made a silly mistake when I posted up those last scope captures--I had C20, the output filter cap, disconnected from the rear strapping points from prior debugging efforts. The presence of C20 doesn't seem to make a huge difference in CV mode, but it definitely affects CC operation (perhaps for the reasons xavier60 mentioned?). After installing C20, the output ripple went from ~6 mVpp to 660 µVpp .

That being said, I have a few more questions. I know that some output ripple is expected, but I want to be sure I understand things:

First off, the 60 Hz sine wave in CV mode at TP 43 was a result of poor probing and close proximity to the input transformer . However, when in CC mode, I could definitely trace the anomalous 60 Hz waveform from TP43 all the way to TP22. Again, I don't have the best probes for high-sensitivity measurements, but the averaging function on my scope got me close enough to where I could trigger on a waveform and be reasonably assured of a signal on those test points.

Out of curiosity, I compared the -4.3 V rail to the other references. It's definitely noisier and seems to correspond to the output ripple waveform. Even though the supply is working within its specifications now, could I assume that the output ripple is coming from the -4.3 V rail through Q3 and/or Q5? I would guess that some component of the waveform feeding the regulator (TP22) is also proportional to the sawtooth wave coming into the pass transistors, i.e. from CR26 and CR27.

Lastly, I have a question about probing VR4, the 4.3 V Zener diode. When I touched my scope ground lead to either side of VR4, the output of the supply went to zero, probably due to some kind of internal short. This seems strange to me, since I didn't have any other probes connected and the supply was essentially floating. I had to measure the -4.3 V by probing between TP41 and +S (not TP23). What would cause this? I didn't have any trouble probing directly across the 24 V, 53 V, and 62 V rails from the transformer.

[xavier60]

Lastly, I have a question about probing VR4, the 4.3 V Zener diode. When I touched my scope ground lead to either side of VR4, the output of the supply went to zero, probably due to some kind of internal short. This seems strange to me, since I didn't have any other probes connected and the supply was essentially floating. I had to measure the -4.3 V by probing between TP41 and +S (not TP23). What would cause this? I didn't have any trouble probing directly across the 24 V, 53 V, and 62 V rails from the transformer.

That's nuts!. All I can suggest for now is to make the probe's ground connection via a 100Ω resistor to TP23. Then probe TP23, looking for DC or RF.

[iroc86]

That's nuts!. All I can suggest for now is to make the probe's ground connection via a 100Ω resistor to TP23. Then probe TP23, looking for DC or RF.

I can give that a shot. I did notice that I was picking up a ~1.7 MHz AC signal when the supply was shorted out. It would even pick it up on the ground lead only, with the probe itself floating. Maybe some kind of interference in the line?

[duak]

Iroc, you could cobble up a 1X probe good enough for low frequencies from a coax cable with a BNC connector.  Use a 1K resistor on the probe end to help isolate the cable's capacitance from the circuit.

Xavier pointed out some limitations in the pass transistors that allow AC ripple to sneak by.  A consequence of the Early effect is basically that increasing the collector to emitter voltage will increase collector current.  The feedback circuitry tries to counteract this but it has limited gain and can't eliminate it all.

The intrinsic series impedances of the capacitors (Xc and ESR) and zeners (Rz) limit how well they can reduce ripple.  The electrolytic caps are getting old and could be replaced with new ones with larger values.

I wouldn't think that grounding the other side of the current sense resistor would shut the supply down - but who are you gonna believe, assumptions or your eyes?  I wonder if there's a connection between the circuit somewhere and the chassis - maybe the pass transistor heat sink insulators.  I'd measure with an ohmmeter and expect something north of 1 Mohm.  Anything less indicates a problem and the possibility of future problems.  It might explain why the diode in the CC circuit died - maybe one of the back panel terminals is grazing the chassis.  I'd be extra careful with putting the ground clip anywhere but +S - it'd be easy to frap one of the hard to get semiconductors.

If you see a 1.7 MHz signal then what could be happening is a circuit  is being formed by the  various capacitances to the chassis & safety ground  and the inductances of the line cords for the 'scope and power supply and either forming an oscillator or is acting as an antenna and is coupling in some nearby RF.

[iroc86]

This is super informative. I'm learning way more about analog circuit troubleshooting than I ever expected by posting this issue. I really can't thank you guys enough for sharing your knowledge .

I'll try the resistor trick on the probe that you fellows mentioned. Those unobtanium semiconductors have proven to be especially resilient with my amateur probing, but I don't want to push it. I did pull out the pass transistors at one point to check them, and I was careful to put the mica gaskets and pin standoffs back in, but it might be time for replacement insulators.

New caps are next on my list. It'll be interesting to profile the supply before and after to see if there were any changes. I'm also going to pick up a variable power resistor and some current shunts to adjust the supply per the manual. I have two other HP supplies that seem to work okay but could use a thorough evaluation, so it'll be a good learning experience.

Regarding the new capacitors, how critical is the ripple current rating and ESR for these old supplies? For example, the Sprague 5600 µF, 25 V main filter cap in the 6289A is physically very large, presumably for high current and low resistance. Short of buying computer-grade caps, I was going to put a few smaller electrolytics in parallel to get a similar advantage. Is there anything to watch for in doing this? As an example, this is the cap I was going to triple up to replace the 5600 µF: https://www.digikey.com/product-detail/en/UHE1H182MHD/493-1627-ND/. Unfortunately, I couldn't find any specs on the original caps, so I'm kinda shooting from the hip with respect to ripple current and ESR. I suppose I could calculate the ESR and in-circuit ripple current using an LCR meter and the waveform from my scope, but it may not be accurate if the capacitors have degraded.

[xavier60]

It's a massive topic, electrolytic capacitors. For many of the rules there are almost as many exceptions.
Low ESR capacitors generally use a water based electrolyte which tends to be less chemically stable than conventional electrolytics.
Best not to unnecessarily  use low ESR capacitors. I would rather not indiscriminately replace electrolytics, mainly those that show physical signs of problems. For the rest  I do in circuit ESR tests.

[bob91343]

While ESR can be an important parameter in circuits, it also can be an indicator of capacitor health.  Low ESR capacitors can cause regulator instability in some cases.

Experience is the best source of perspective regarding this.  I tend to discard old capacitors that have ESR much above 1 Ohm.  But smaller units often exhibit higher values.

Capacitance readings are important.  If a unit has less than, say, 75% of marked value, I toss it.  If its value is much above rating I won't use that as criterion for trash; tolerances are often sloppy.

Leakage can be important; if too much it will cause internal heating and even thermal runaway.  Especially when an increase in applied voltage causes more than proportional leakage current increase.

Old capacitors often recover after a period of use.  But they can fool you by suddenly deciding to fail.  There is no measurement of which I am aware that can predict this.

Certainly replacing all the electrolytic capacitors in a unit is a good idea sometimes but it sets you up for making errors such as bad solder joints, reverse polarity, and miswiring.  Plus it's time consuming and sometimes expensive.  The dimensions of modern capacitors prevent an easy replacement due to the mounting differences.  The modern cans are much smaller and have different mountings.

And I have mroe than once found a brand new capacitor defective.  I have one here rated for 100 Volts that had excessive leakage above about 75 Volts.  It was fine up to that stress but any more and it ran away thermally.

In summation, there is no simple answer and any course of action has its pitfalls.  Electrolytic capacitors in particular have many shortcomings and it's sometimes better to replace with some other type.  The most reliable are probably ceramic, in spite of their often sloppy tolerance.  Mica is good but even then, age can take its toll.

Since electrolytic capacitors and paper/plastic capacitors are usually wound in rolls, they can have annoying series inductance as well.  That's why you often see a small capacitor in parallel with a large one so that the combination has better characteristics than either one alone.

[xavier60]

Regarding the new capacitors, how critical is the ripple current rating and ESR for these old supplies? For example, the Sprague 5600 µF, 25 V main filter cap in the 6289A is physically very large, presumably for high current and low resistance. Short of buying computer-grade caps, I was going to put a few smaller electrolytics in parallel to get a similar advantage. Is there anything to watch for in doing this? As an example, this is the cap I was going to triple up to replace the 5600 µF: https://www.digikey.com/product-detail/en/UHE1H182MHD/493-1627-ND/. Unfortunately, I couldn't find any specs on the original caps, so I'm kinda shooting from the hip with respect to ripple current and ESR. I suppose I could calculate the ESR and in-circuit ripple current using an LCR meter and the waveform from my scope, but it may not be accurate if the capacitors have degraded.

With regard to large filter capacitors, there should be no need for anything out of the ordinary. So long as the capacitance value is scaled properly to the output current, any respectable brand of capacitor would naturally have an adequate ripple current rating also.
If space permits, upgrading the  voltage/size of the replacement capacitor usually improves durability.
I did a search to find how the DC load current relates to ripple current and got conflicting results.

[xavier60]

Needing some capacitors for myself, I found it difficult to find standard electrolytics.
Epcos/TDK, a trust worthy brand, has a series of standard type that claim to have an extremely long life span under low stress conditions,
http://www.farnell.com/datasheets/2367538.pdf?_ga=2.260007049.2004012069.1559655723-1727119234.1557827711&_gac=1.89793001.1558609709.CjwKCAjwiZnnBRBQEiwAcWKfYuGHUTqfk8jkJaF0-WFSzt7elG_M6tpqgaVBEIab-sNwByC66YENWBoCGCQQAvD_BwE

[iroc86]

I appreciate the feedback on replacing the caps! There are a lot of parameters to consider. I do remember reading that sometimes these new super-low ESR caps can spell bad news for old designs. It seems like there is no single answer to knowing when/if an old cap needs to be replaced. Perhaps the best course of action regarding my situation is to pick up an LCR meter and just see what I've got to establish a baseline. For all practical purposes, the power supply seems to be operating within its specifications, so maybe I shouldn't fix what's not broken.

I won't have a chance to get back on this project again until late next week, so I may have some delays in responding, but I'll continue to post updates for everyone.

[iroc86]

I wanted to post an update on my progress with this power supply. The unit is working pretty well these days. I conducted a full performance test per the service manual and the unit passed admirably. The output ripple is a little on the high side compared to my 6236B supply (which is nearly flat), but the 6289A is still within its 1 mVpp rating. Eventually I will get an LCR meter and test the caps individually, but for now, the performance is more than good enough for my use.

Out of curiosity, how much drift should I expect with the analog output meter? I've been chasing down an intermittent issue where I can't get the panel meter to calibrate properly. I think it might be related to thermal drift, but I haven't really been able to isolate the conditions. The meter will periodically read about 100 mV lower than the output voltage. It tends to drift around during use. The offset is consistent across the entire 0-40 V output range. At higher voltages, the meter is well within its +/- 2% accuracy rating, but at lower voltages it's obviously not. The output voltage is always rock-solid within 1 mV, either loaded or open circuit, regardless of warm-up time and temperature.

If I try to re-calibrate the supply after the meter has drifted, the adjustment that always goes out is the "meter zero" setting via R63. As soon as I reset that trimmer, the accuracy returns. I've attached the adjustment procedure from the manual and the meter circuit below. The full schematic is a few posts up if you want to see the whole thing.

I thought the drift might be related to the reference voltages, so I measured the +6.2, -6.2, and +12.4 V rails at initial turn-on and again after 30 minutes of warm-up time. The +/- 6.2 V rails didn't change at all, but the 12.4 V rail increased from about 11.98 V to 12.05 V. I can't imagine that 70 mV would make a huge difference, but I'm not sure. This rail has a +/- 1 V tolerance, so it's still within tolerance, but the manual doesn't provide stability specifications.

Any thoughts on this one?

[duak]

If there are reversible changes I would first suspect the parts with contacts in them.  If you twiddle S2's knob (not enough to take it out of the current range) or the variable resistors slightly, does the meter jump a bit?  If so, try cleaning its contacts. Make sure the soldered joints in this section as well as the mechanical connections to the meter are good.

If you can get some freeze spray, you might try components one after the other.  A heat gun can also work, but is less selective and it's easier to frap (or flambe) a part.

 

[iroc86]

If there are reversible changes I would first suspect the parts with contacts in them.  If you twiddle S2's knob (not enough to take it out of the current range) or the variable resistors slightly, does the meter jump a bit?  If so, try cleaning its contacts. Make sure the soldered joints in this section as well as the mechanical connections to the meter are good.

If you can get some freeze spray, you might try components one after the other.  A heat gun can also work, but is less selective and it's easier to frap (or flambe) a part.

The switches were all fairly sticky when I started repairing this unit, but I cleaned everything with contact cleaner a while back. Twiddling the knobs doesn't seem to make any appreciable difference in the meter reading. For good measure, I also pulled out all of the trimmers and thoroughly cleaned them, too. Unfortunately, it didn't have any effect. I also tried poking a few of the components and solder joints with a spudger, but didn't notice anything unusual.

I'll pick up some freeze spray and see if I can isolate the issue, perhaps to the zeroing circuitry or the voltage references.

[duak]

Here's a really goofy thing: are you in a dry environtment?  I remember that one analog meter I had was sensitive to the static charge built up by dragging a finger across its face.  I found this out when I cleaned it with some IPA and the pointer followed the swab.  Totally weird and unexepected.  Also, many analog meters are sensitive to magnetic fields.  With the number of super magnets around these days, is there something nearby that might contain one?

There may be another explanation.  There two types of meter movement pointer suspensions - taut band and jewelled.  Jewelled movements can get a bit sticky over time that often manifests as a zero setting that doesn't stay put.  I don't know what type hp used.

Science news from the lab!:  I have an hp 6284A and the pointer is indeed sensitive to rubbing a finger across its window.  It takes a few passes all going the same way to move it.  A Lambda supply meter does the same but is even more sensitive.  Both are sensitive to a super magnet moved rapidly but seem to come to their original zero positions.  Hope I didn't demagnetize them.

[iroc86]

Here's a really goofy thing: are you in a dry environtment?

Actually... yes. I live in New Mexico.

That's really interesting about the static charge. It's not something I would have considered for this little HP supply, but your anecdotal evidence definitely provides some food for thought. I'll have to take a closer look. When I restored the supply, I polished the meter face with plastic compound and a special cleaner, so it's possible a charge may have built up from the rubbing. I also hadn't considered magnetic fields, but if that were the case, I'd expect my 6236B to act funny as well, since it's sitting right next to the 6289A on my bench. All good ideas, though.

It seems that the tolerance of my meter has actually gotten worse, or at least returned to its original state that I thought I had fixed: it's now inaccurate by about 1 V. When I was troubleshooting a few days ago, I may not have powered off the supply long enough to stabilize at a "cold" state, hence the 100 mV offset I was seeing. I don't think there's any significance to the magnitude of the offset, just the fact that it's worse than I thought.

I can try to measure the voltage at the meter terminals and see if it changes over time. I think that would at least indicate if the problem is with the calibration circuitry or the meter itself.

[duak]

Ayup, if you have a sensitive enough meter, look at the actual meter voltage - that'll tell lots.

I live in Vancouver, BC and with the humidity here, you couldn't build up a static charge even if you rubbed two nylon cats together.  Yet it seems I have a telekinetic abilty, at least on meter pointers.

I've travelled around NM a couple of times and I'll say its my favorite SW state.  My wife and I bumbered around Santa Fe the longest - a neat little city.  Away from the cities and towns the various parks, natural monuments and just things you find outside are quite unlike what we have here.

[iroc86]

That's cool! I'm actually up in the mountains, so it's not quite as desert-y as SF or even Albuquerque... more like what you'd expect in Colorado. Lots of tall trees and snow-capped peaks. The isolation can be pretty intense, but I agree that the parks and natural sights are quite impressive. I visited Vancouver a few years ago on my way to a cruise in Alaska, and I was super impressed by the city. It was clean, refreshing, and the people were very welcoming. I have a few friends who are infatuated with Victoria, so that'll probably be a stop in the near future.

I'll report back on the meter measurement soon. This is shaping up to be a busy week, but we'll see how things go!

[iroc86]

Quick update. I had a chance to set up some repeatable conditions to better define the meter drift in this power supply. Throughout all of the characterization, I left the supply adjusted to 20 V output voltage. Regardless of temperature, output current, or meter switch position, it always regulated to within +/- 10 mV of the setpoint, suggesting that the regulation circuitry is working great. I'm actually surprised that it's so stable considering the analog potentiometer controls.

My test involved a load resistor set to 26 ohms, per the service manual, which loads the supply to 0.77 A at 20 V. This is about half of its maximum rated output, so this seemed like a good test point. I then connected a DMM to the terminals on the panel meter and measured the voltage as the unit warmed up. After about 10 minutes, the meter input voltage increased by about 1 mV, from 36.8 to 37.7 mV. The meter seems to have a sensitivity of about 1.8 mV/V on the 50 V range, so 1 mV of drift would be quite perceptible--about the width of the needle itself.

Interestingly, I noticed a fair bit of fluctuation at the 10-minute mark, jumping around +/- 0.5 mV. The output voltage was solid at 20.00 V throughout. After about 30 minutes, the meter voltage seemed to stabilize again at the 36.8 mV mark.

I don't currently have a means of datalogging with a DMM to take a bunch of points over a long period of time, but it appears that the meter circuitry is going through some kind of warm-up jitter before it fully stabilizes back at the original level. Admittedly, the HP manual does suggest performing all calibration steps after the supply has been turned on for at least 30 minutes, but I would have expected a little better stability on the metering, especially for a lab supply that may be frequently turned off and on.

My next step is to get some freeze spray and start isolating components in the metering circuit...

[cbutlera]

Many of the symptoms described in this thread, including the buzzing, are consistent with a leakage path on the PCB between the mains voltage area and the current input circuit.  I would focus particularly on the track between the current control potentiometer R16B on the front panel, and the A3 terminal on the rear.  On my somewhat similar HP Harrison 6111A, that track (green arrow in the photo) runs frighteningly close to the tracks connected to either side of the dropper resistor for the neon indicator (red arrows in the photo).  At the closest point next to the resistor pad, the creepage distance is just 0.6mm!  On a 50+ year old PCB with no solder mask, it would be a surprise if there wasn't significant leakage between these tracks.

This is clearly unsafe by modern standards, and I'm amazed that it was ever regarded as acceptable.  The troublesome live tracks carry the connections between the transformer primary, the power switch, the neon indicator and its dropper resistor.  On my unit, I made a number of 10mm long cuts to isolate these tracks, and reconnected the switch, neon indicator and resistor to the transformer primary with flying leads.  I then connected two of the now floating tracks (indicated by the red arrows in the photo) to ground, to act as a guard.  The transformer primary terminals are still connected to the PCB, but there are now no live tracks going anywhere else on the PCB, except to the immediately adjacent pads, which I used to terminate the flying leads.  As can be seen in the photo, I made a small transparent shield to cover the remaining live area of the PCB and I also put clear heat-shrink on all of the exposed live terminals.

As a side note, if you are trying to achieve high stability, make sure that C1 has a very low leakage current. Even a few nanoamps is too much, and will give rise to tens of microvolts of drift.  My 6111A used a Sprague 1 µF film capacitor in this position, and replacing it with a new polypropylene capacitor improved the stability by about a factor of 50.   It can now hold 10 volts to within about +/- 20 µV for an hour or more, once warmed up.  Although bear in mind that the 6111A has an ovenized Zener reference and a switched precision resistor network to set the voltage.

[iroc86]

Well, it's taken me a bit longer than I expected to go back and try to tune up this old power supply!

Thanks for your insight, cbutlera--apologies on the very late reply. I see what you mean about the possibility of leakage with the unmasked PCB traces. My 6284A doesn't seem quite as tight as your 6111A. The trace from R16B to the A3 terminal actually jumps between top and bottom a few times and has at least 1 mm of clearance to the adjacent traces. Compared to the photo you posted, my supply also has a bit more clearance between parallel traces when they neck down to squeeze between pads, etc. Even so, this is good knowledge for repairing other vintage gear. I like your solution with the flying leads and improvised guard traces. Also, thanks for the tip about the leakage current of C1 (is this the same as C4 on your 6111A's schematic?). My 6284A will drift about 10-20 mV at 20-30 V... not a deal breaker for my use, but a little better performance is always nice.

Now, about the meter drift I've been seeing: I applied some freeze spray to the various components in the meter circuit (the schematic is a few posts up) while measuring the meter voltage across TP48 and TP49. Everything was thermally stable except for Q12 and Q14. When Q14 got cold, the voltage went down; when Q12 got cold, the voltage went up. This makes sense since these BJTs are in a differential amplifier arrangement and are just jellybean PN2907s per the service manual.

In an earlier post, I noted that the meter voltage fluctuated up and down until the supply reached a stable temperature. I wonder if I'm seeing the effects of these two transistors coming up to temperature at different rates and "pushing/pulling" each other until they stabilize. In the photo below, you can see that these are two discrete components with no thermal interface. Interestingly, HP chose to use a matched pair for Q11, the other differential amplifier in the meter circuit (TO-39 can, bottom right of photo).

Just for kicks, I soldered in two new PN2907s to see if they track a little better. I don't have any data yet, but I'll report back. This may not even be the issue I'm chasing, but it's interesting nonetheless.

[duak]

Iroc, could you make up an aluminum bracket that thermally ties the two transistors together?  I envisage something that looks like a staple for paper after it's been crimped.  hp often uses an 'S' shaped bracket to tie matched devices together.

If you have two new devices with long enough leads, maybe stick two together head to head inside a metal tube and then bend the leads to fit the holes.

I see that there are what look like 1/2 W resistors below the transistors.  Do they get warm during operation?

[iroc86]

Hey duak, I haven't had a chance to try the thermal bracket yet. The 1/2 W resistors nearby don't get appreciably hot, maybe 5 degC over ambient. There is a rather toasty (100+ degC) wirewound resistor nearby, but I don't think it's causing any thermal gradients. I tried putting some FR4 board in between the warmer components to act as a barrier, but it didn't make any difference.

I think I still need to run some more tests and determine if the drift is being caused by the amplifier circuit or the meter itself. The two transistors seemed promising, but I'm not so sure anymore...

[iroc86]

I ran a few new tests. I removed the panel meter from the circuit and applied an external voltage to see if the drift was coming from the meter movement. It wasn't. The needle remained dead-even for the duration of my test (about 30 min, plenty long enough to see the drift on the power supply meter circuit).

I then measured the voltages of TP46 and TP47, referenced to +S, on a cold unit over the course of 60 minutes. The panel meter is connected across these test points, so the difference between TP46 and TP47 is essentially the same voltage that the meter sees (through some dropping resistors, of course). To eliminate any interference from the mechanical meter movement, I installed a 100 ohm resistor in place of the meter. I also measured the meter voltage across the 100 ohm resistor for comparison to the test points.

The data is plotted below. The ~1 mV rise in meter voltage at startup is the same as before, but note the unusual drops around 24 minutes and 49 minutes. This behavior correlates directly with two little humps on the TP46 graph. The power supply sat on my bench, undisturbed and with no load, for the duration of the test.

I then swapped out Q11 (the TO-39 dual BJT) and replaced it with two 2N2222s, just to see if there'd be any difference in response. The absolute numbers have changed a bit, but the weird fluctuations remain... and only after the unit has been powered on for upwards of an hour. The power supply output voltage (30 V in my tests) remained rock solid the whole time, so I'm pretty certain this is drift from the meter circuit only.

Any ideas on what might be causing this? Could it be from a passive component (e.g., a bad resistor), or would this be more likely a semiconductor issue? The only original transistor left is Q15, which functions as a current source according to the service manual. It's a 2N2907.

[xavier60]

What is Q15 doing? Is it operating as a constant current source?
I can't figure out how the whole circuit can work properly with constant current fed to the Emitters of the output transistors.
Is VR6 fitted as shown?
What voltages are across VR6 and R62?
What is the Colector voltage of Q15?

[iroc86]

Per the manual excerpt below, Section 4-38 states that Q15 is a current source to set the bias current for the amplifier. I'm not sure how it works.

Yep, VR6 is there. Earlier revisions of this supply had a slightly different arrangement with resistor R40 in its place. Interestingly, the schematic has the polarity of the diode incorrect. My PCB matches the proper orientation, though--cathode to +12.4 V. It's supposed to be a 4.22 V Zener.

The voltage across VR6 is 4.00 V. Across R62, it's 3.30 V. The collector of Q15 to ground (+S) is about 8.65 V. Vce measures 0.024 V.

[xavier60]

With Q15 saturated, it isn't operating as a current source which I don't think it should be for now. Check for changes in operating state of Q15.

[iroc86]

With Q15 saturated, it isn't operating as a current source which I don't think it should be for now. Check for changes in operating state of Q15.

Maybe not... but an LTspice simulation seems to show that Q15 is saturated with Vce right around where I measured at ~20 mV. Hmm.

[duak]

Here's some thoughts:

As per section 4-39 in the clip from the manual above, the purpose of the circuit is to prevent damage to the meter movement during overload by limiting the current to it.  Q15 is a current source that can provide a certain maximum current of about 5.5 mA.  Assuming D1, R62 & Q15 are good, then when Q15 is saturated, it's an indication that it is not limiting current.  Is the meter pointer significantly above maximum?  If not, what happens to Q15's Vce when the meter reads above maximum?

When I look at the 'meter' curve in the above graph I think that a fan is coming on at 30 minute intervals.  Could the HVAC be starting and stirring the air up?  Less likely is that the HVAC or some other repetitive load such as a pump is affecting line voltage.

When I think about it, Q12 & Q14 are within a closed feedback loop so their temperature differences should be insignificant.  OTOH, the VBEs of the transistors in the first pair are terrifically important and should track for minimal drift.   My copy of the manual shows the first pair is in separate packages and are designated Q11 & Q13 whereas IROC's schematic shows a dual part Q11 and the above picture shows they are in a common package.  hp probably made a production change to improve the circuit by going to a dual part in one package.

I wonder if the transistors in Q11 are actually matched.  What is the part number on the package and in the manual?

[iroc86]

Interesting thoughts, duak. (FYI, I forgot to label D1 correctly in my LTspice schematic; it's representing VR6 with a ~4.7 V Zener, which is probably close enough to the real 4.2 V part for simulation purposes.)

No, the meter pointer is in the middle of its range. I'm doing all my testing at 30 V. The meter graduations go up to 50 V. When S2 is in the lower range, it's divide-by-ten for 0-5 V.

I think I see how Q15 is supposed to work based on your description of the current limiter. But what does the manual mean about Q15 setting the bias current for the amplifier per section 4-38? If you disconnect the Q15 collector in the LTspice simulation, the meter voltage drops to essentially zero, so Q15 is still doing something even when in saturation.

I measured Vce of Q15 and selected the 5 V range while driving the output voltage higher. At about 10 V, Vce starts to increase, but even when the supply is set to 50 V (ten times the meter range), it only reaches about 70 mV... so it's still basically in saturation. Now, if I try the same thing on the lowest amperage range (180 mA), Q15 will come out of saturation when the output current reaches 0.5 A or so. Is this behavior inherent to the difference between sampling the output voltage vs. the current sense resistor voltage?

Regarding the dips on the meter curve, I'll pay attention to the airflow in my room. We're using heat now, so there wouldn't be a heavy electrical load from AC, but temperature could play a difference. I can try putting a thermocouple in there and running the test again. Even so, I'd be surprised if it's the furnace, as my lab bench is about 15 feet away from the nearest vent and I had the lid installed on the power supply case. This thing would be super sensitive if a 1-2 degC ambient temperature change caused the panel meter to deflect so significantly.

That being said: for Q12 and Q14, component temperature really does seem to make a difference. I can squeeze one of the BJTs with my finger and the meter input will move by about 200 µV. If I squeeze the other one, it'll shift by the same amount in the other direction. This isn't really enough to see on the panel meter, but when I hit the transistors with freeze spray, the meter input will move by 2-3 mV, which is upwards of 2 V on the meter on the 50 V range... quite visible.

I took a closer look at the 6289A errata (attached) and HP does actually call out replacing Q11 with a dual part in later revisions. Under Change 4, they're saying it's a 2N4045, which appears to be a matched pair with "tight Vbe tracking" per most of the data sheets I've found. That's probably what's in my unit, or at least something close.

While playing around with a few things as I was writing my reply above, I noticed that I can get the get the panel meter to move significantly by touching the components in the meter amplifier circuit. I know the human body is going to have some conductivity, but I didn't expect to visibly see such a change. This leads back to cbutlera's post in Reply #46 about PCB leakage. Might this be what we're seeing with the meter amplifier circuit and any adjacent components?

[duak]

I goofed in my calculation of the current in my previous reply; it should be 5.5 mA.  (I corrected the reply).  A more common name for the circuit Q15 is a part of is a current source.  A current source can always provide less current than it's designed for - it's just never supposed to provide more, hence it also limits current.  An analogy is a valve that's fully open ie., saturated - it can't do anything more to get what's downstream to accept more flow.

It's not incorrect to say Q15 provides a bias current.  IMHO it would be more clear to say that Q15 provides the emitter current for the output stage of a differential amplifier.  This is the lion's share of the current provided to the circuit; R64 and R65 provide a smaller amount.  Most flows to -6.2V through R71 & R70 and the rest through R74.

Q15 could be bad in that its gain has fallen off or it developed big leakage but it's really unlikely.

Curiouser & curiouser...

I wouldn't expect operation to change between Voltage and Current modes, unless there was something wrong.

When touching the parts with a finger, are you making electrical contact with any of the actual circuitry?  If not, and the circuit responds when touched and then recovers instantly when not, the circuit could be oscillating.  Maybe look at some of the signals with a scope to see if you get something weird to happen.

It could be leakage from something or other on the PCB but the picture doesn't indicate any contamination.  OTOH, there could be a bad part somewhere - capacitors get leaky and so can transistors.  The meter circuit doesn't have any capacitors though.

I can't remember if you've looked at the 12.4, 6.2 and -6.2 V right on the parts in this circuit while the event is happening.

I have an hp 6284A.  If the meter circuit is similar enough and if I can scare up some space & time I'll see how it responds.

[xavier60]

If Q15 is supposed to operate in an unsaturated state, it would be odd that the designers would have added this extra source of possible variance to the meter circuit.
Try adding an extra 0.6V of diode drop to the zener.

[duak]

I pulled the hp 6284A out of retirement and managed to clear a space for it on the bench.  It was made in '79, has a dual transistor for the meter amp and has a zener diode VR6 instead of a resistor.

Using a Fluke 8062A DMM, I measure 3.804 V across VR6 with the meter reading anywhere between zero and full output voltage.  I see that Q15 is saturated with a Vce of about 50 mV.  I expect that the voltage across VR6 is reduced by Q15 drawing more base current because it is in forced Beta mode, ie., the collector current is limited by the rest of the meter circuit so the voltage across R62 is reduced and the base current must rise to maintain equilibrium.  If I select low voltage range and pin the meter, Q15 comes out of saturation with a Vce of about 500 mV and I measure 3.822 V across VR6.   At this time Q15 draws less base current because it is working as a transistor and its collector current is contributing to the voltage drop across R62.  Repeating what I said above I think Q15 is part of the circuit to protect the meter and it works by limiting the current for out of range values.

Assuming VR6 is a 4.22 V zener as per the update instructions, this tells me that VR6 might be running at a low current.  BTW, hp's original design used a resistor.  I suspect the resistor was replaced by a zener not necessarily for stability but rather if the +12.4 V supply failed with an overly high voltage, the meter movement would be damaged because the meter current limit would also be increased.

I don't have a logging meter so I have not tried to replicate IROC's test of meter circuit stability over time. I did not find that touching Q12 or Q14 cases made a difference in the meter reading or in the voltage between TP45 & TP47.

[xavier60]

I pulled the hp 6284A out of retirement and managed to clear a space for it on the bench.  It was made in '79 and has a dual transistor for the meter amp and has a zener diode VR6 instead of a resistor.

Using a Fluke 8062A DMM, I measure 3.804 V across VR6 with the meter reading anywhere between zero and full output voltage.  Q15 is saturated with a Vceo of about 50 mV.  I expect then that the voltage across VR6 is dragged down by Q15 drawing more base current because it is in forced Beta mode, ie., the collector current is limited by the rest of the meter circuit so the voltage across R62 is reduced and the base current must rise to maintain equilibrium.  If I select low voltage range and pin the meter, Q15 comes out of saturation with a Vce of about 500 mV and I measure 3.822 V across VR6.   At this time Q15 draws less base current because it is working as a transistor and its collector current is contributing to the voltage drop across R62.

Assuming VR6 is a 4.22 V zener as per the update instructions, this tells me that VR6 might be running at a low current.  OTOH, hp's original design used a resistor.  I suspect the resistor was replaced by a zener not necessarily for stability but rather if the +12.4 V supply failed with an overly high voltage, the meter movement would be damaged because the meter current limit would also be increased.

I have not tried to replicate IROC's test of meter circuit stability over time.  I did not find that touching Q12 or Q14 made a difference in the meter reading or in the voltage between TP45 & TP47.

Is it possible for you to decrease the temperature of Q15 to see if it drops out of saturation?

[xavier60]

Anyway, I'm hoping that iroc86's  meter drift problem is being caused by Q15 drifting in and out of saturation because the it's operating too close to the edge.
Another way of covering up this possible cause is by decreasing R62 by a small amount.

[duak]

xavier, I don't have any cool spray to try on Q15.  I had some dry ice packs here the other day but they're long gone now.

I also edited my previous reply to hopefully clarify it but it was about the same time you submitted your reply.  As I said, I think it's to protect the meter.

One way to increase the current limit is to put a resistor in parallel with R62, say 1K0 and see what happens to the drift.

[iroc86]

Thanks for testing out your 6284A, duak. I'm really glad we have something to compare to. I have a 6236B triple output that I thought was similar, but the meter circuit is much simpler... no amplifiers, just a resistor network. I'm not sure how/if HP is limiting the meter current on that model (FYI, it has zero drift from what I can tell!).

Out of curiosity, did you let your supply warm up before taking your measurements, or were you going from a cold unit?

That's interesting about Q12 and Q14 not responding to human touch. I'd also like to extend xavier's request about Q15's temperature--I'd be curious if either Q12 or Q14 would pull one way or the other when doused with freeze spray (on my unit, if I spray them at the same time, the voltage across the panel meter more or less stays the same). I saw that you don't have any spray, but would you be willing to try heat gun?

It looks like you're getting Q15 to drop out of saturation when switched to voltage display, unlike my unit. Can you also try it on the current range?

To your earlier questions about the voltage rails, I did do some cursory measurements on the 12.4, 6.2, and -6.2 V rails. They're all rock solid throughout the entire warmup cycle. The 12.4 V rail is a little low at 12.0 V, but it's still within HP's +/- 1.0 V tolerance.

I took a few more measurements over the weekend and will try to post up some new graphs tomorrow. FYI, I'm not doing this with a logging meter, either... just a pair of Fluke 27s, my HP 3466A, and a stopwatch. I'll put on one of Dave's videos and just burn through the data points while I watch on the side.

More to come. Thanks, guys!

[xavier60]

The input selection possibly alters the amplifier's input Common Mode voltage which will also alter the current draw through Q15.

[duak]

I suppose I was a professional logger back when I was paid to write down numbers.

The final numbers were after a 30 minute warm up.  They were a few mV lower when it was cold.

When the meter reads full scale on either voltage or current, the voltage between TP46 & TP47 is almost exactly 1.000 V.  Because hp used wirewound pots even the fine adjustment is somewhat granular and is obvious with a DMM and damn near impossible to set to an exact XX0.0 V.  By heating Q12's case with a soldering iron I can reduce the above voltage by about 10 mV and by heating Q14's case I can increase the above voltage by about 10 mV.  It's hard to say how much their temperatures changed but I'd guess it's more than 10 deg C.

Pinning the meter in both voltage and current modes cause Q15 to come out of saturation.  For all in-scale readings Q15 is saturated.

The diff amp has a gain of X10.  With an output voltage of 1 V for full scale, the input voltage is 100 mV.   R74 is the common emitter resistor that long-tails the input stage of the diff-amp and It'll have about 5.5 V across it.  Since both the current sense resistors and the output voltage connect to +R, the + output terminal, there isn't really a common-mode voltage to consider.  Anyway, the input to the diff amp will cause a variation of +/-100 mV across R74 which is probably not significant vis-a-vis Q15.  However, if something is on the edge this variation may be significant.  Note that the sensed voltages into the diff-amp are of opposite polarities for the voltage and current modes so there is a distinction between them.  hp goes over this in the manual.

I'm going throw out something here - the current calibration adjustment R56 is wirewound.  This is where the resistance element is a thin wire wrapped around an insulator in the shape of an arc.  A moveable contact on the wiper touches one or more turns of the wire and makes a variable resistance.  I mentioned above that the voltage and current pots are wirewound and that the adjustments are granular.  Suppose R56 is just on the edge of contacting the next turn on the resistance wire.  Slight temperature changes and/or extraterrestrial ethereal forces may be just enough to tip it over and make the meter jump.  Maybe a cleaning and slight back and forth working of the wiper will make a difference.

[iroc86]

I too am seeing about ~1.000 V between TP46 and TP47 when the meter is at full scale. The linearity seems to change, too. I don't have any data to back up that assertion, but it seems like the meter reading will fluctuate as I adjust the output voltage. It usually reads within a "needle's width" when going to/from specific output voltages, but occasionally it'll end up a half-graduation higher or lower. That might have something to do with the wirewound pots, but I'd expect consistency because I can always get the output voltage to return to its original value. The meter circuit is just sampling off of that, so I figure the issue is somewhere in the amplifier.

Regarding the wirewound adjustment pots, that's a very good thought about the adjustment being on the hairy edge of the contact. Early on in my troubleshooting, I removed R56 and R72, ran them through their full ranges, and cleaned the contacts with CAIG DeoxIT F5 Fader lubricant. I've also turned the pots to a different setting just to move the wiper to another area, but it didn't make any difference. I can try tossing in some modern trimmers to see if that changes anything.

Thanks for heating up Q12 and Q14 with your soldering iron. Those fluctuations are about the same as I'm seeing.

Here's a new set of charts. This time I also logged the temperature in the vicinity around Q12 and Q14, just in the open air. The lid of the case was closed as best as possible with all of the test leads coming out. I let the unit warm up for 35 minutes before switching the furnace on (just to rule out that possibility). As expected, the panel meter responded slightly, but nowhere near as significant as the first set of data I posted. I shut off the furnace at 52 minutes in, and the temperature measurement stabilized. Interestingly, the meter voltage was on a rising trend the whole way, even after the temperature mostly stabilized around the 50 minute mark. The meter just started to level off at the very end, but I had to shut things down and go do something else.

After a a few hours, I tried again on a cold supply (though only over 20 minutes and no furnace cycling). The response was almost identical, including the odd blip at the ~12 minute mark. I have no idea what caused the two big dips in my data from a few days ago.

duak and xavier, I'm going to probe around some more based on your responses and will report back...

[iroc86]

I did some more testing with Q15 and meter current limiting. The maximum current through the meter is about 1.7 mA when pegged in either voltage or current modes. (FYI, when I'm referring to voltage and current "modes," I mean the setting of the panel meter switch, not the output voltage or current limiting modes of the supply itself.)

duak, can you confirm that you saw Q15 drop out of saturation when your meter was reading voltage only? If so, this is different behavior than what I observe. My supply will only limit meter current via Q15 when the meter is in current mode. This seems odd to me, since the resistor networks in front of switch S2 provide the same input voltage to Q11 regardless of mode--the only difference is the polarity, as you mentioned.

Here's what I mean: In the high range modes (50 V and 1.8 A), the maximum input voltage to Q11 is 100 mV based on the resistor networks, which corresponds to a full-scale reading on the meter. This is based on 50 V across R59-R61 for voltage mode and 1.2 V across R56-58 in current mode (i.e., a 1.2 V drop across 0.66 ohm current sense resistor R54 with a 50 V input and 27 ohm load = ~1.8 A). Note that in this scenario, R56 is adjusted to 100 ohms, which correlates to the actual pot adjustment on my unit.

In the low range modes (5 V and 0.18 A), the 100 mV input again corresponds to full-scale on the meter, except now it's at the 5 V (voltage) and 120 mV (current) input levels due to the divide-by-ten resistor network. Anything exceeding a 100 mV input to Q11 should presumably cause Q15 to start limiting the meter current. At full output voltage or current in the low range mode, the input to Q11 will be about 1 V. So, anything from 200 mV to 1 V ought to provoke some response from Q15.

I'm not sure why Q15 is dropping out of saturation in only the current mode. That's when the base of Q11B is grounded and Q11A is positive. Might this have something to do with the nature of the differential amplifier and the way the voltage references are set up? Or perhaps Q15 doesn't need to fully drop out of saturation to limit the current in voltage mode? (In voltage mode, Vce of Q15 will increase slightly, to maybe 60 mV or so, but it's vastly different than the ~500 mV in current mode.)

I know this isn't directly attacking the meter drift problem, but if this part of the circuit isn't working properly, it might be a clue.

[xavier60]

RF could explain the odd behavior like being over sensitive to being touched by hand.

[xavier60]

It is actually normal for the amplifier to draw a little more current when one input has a + input and the other is zero,
 compared to when one input has a - input while the other is zero.

[duak]

Iroc, I measure about 1.56 V between TP46 and TP47 when pinned in voltage mode and about 1.58 V when pinned in current mode.  Likewise Q15's Vce is 485 mV and 5.10 V respectively.  Xavier points out that because the two modes are slightly different the common mode voltage into the diff amp  will affect the current drawn from Q15.

I'll bet if you reduce the voltage across VR6 to about half of what it is now, it should limit maximum meter readings to the upper third or so of its range.  You should also definitely see Q15 come out of saturation.

These things were introduced in 1967 when it was cheaper to use discrete parts than an op-amp.  I'd bet the original circuit didn't have Q15 but during a design review or testing an internal supply failure pointed out how the meter could be damaged.  VR6 replaced the resistor as a running change after some field failures surfaced.  These days, I'd use an op-amp and a band-gap reference diode and be able to limit the meter current to 125% of full scale +/- a couple of percent.

I've encountered one discrete circuit where one of the transisors developed a strange leakage current that manifested as an intermittant drift and replacement solved the problem.

This next part is clutching at straws:  Since this is a precision circuit it might benefit from using transistors that weren't optimized for switching.  I used 2N5087 & 2N5089 low noise transistors years ago when I designed some low noise stuff but there are others that may be more available.  I believe the ubiquitous 2N2222, 2N2907, 2N3904 & 2N3906 were all gold doped to reduce switching times.  This makes them slightly leakier and may contribute to popcorn noise.  That being said, I've used these transistors for various low level iinear circuits and they've been just fine.  I see in the manual that hp specified a Sprague 2N2907A but if there was a problem with any of the above, I'm sure they would have changed parts.

[iroc86]

All very interesting stuff, guys. xavier, thanks for pointing out the current draw difference vs. polarity on the amplifier circuit. I think I can see how that might affect Q15 downstream.

I didn't know about the doping on the jellybean switching transistors--those things are so ubiquitous that I never really considered how a switching transistor might behave differently than an amplifying transistor. It's hard to say what's actually in these units, too, with HP's internal numbering scheme. Over the course of refurbishing a variety of HP equipment, it's amazing to see how often their parts lists have conflicting cross-reference numbers. I certainly give them credit for improving the designs over the years, and it's testament that a design from fifty years ago is still useful and usable, despite the issues that crop up with age. (My 6289A appears to be from around 1986.)

Re: VR6 and its effect on Q15 saturation, I did notice some of that in one of the LTspice simulations I ran. I reverted back to the original design with a resistor and it changed where the "knee" of Vce appeared in relation to the amplifier input voltage at Q11. I think I will try adding another diode in parallel with VR6, as xavier suggested, and see how that affects the saturation point. I might have a few other Zeners I could swap in, too.

So, where are things right now? I replaced R63 and R72 with regular resistors to eliminate any potential temperature effects with the wirewound pots. From the little bit of preliminary testing I did, it doesn't seem like that's the issue. Even over just 10 minutes of warm-up, the meter drifts by about three widths of the needle. It's tough to systematically diagnose this issue because I want to start from "dead cold" every time to ensure I'm not chasing my tail. That, and the intermittent times when the needle is a half-tick low for no apparent reason at all. It just takes time...

Next, I'll probably swap out Q15 and see what that does. I don't have any spare 2N4045 TO-78 packages for Q11, so that'd be the only original semiconductor left.

At some point, I probably need to ask myself how much of this is "good enough" and within my expectations for a 30-year-old piece of hardware based on a 50-year-old design. As a side experiment, I'm thinking of spinning up a PCB for the meter circuit with modern hardware, in the same topology, to see if the drift is any better. There are a few matched pair SOT-23 transistors out there that might be interesting to experiment with.

[xavier60]

The easiest way of causing Q15 to stay in saturation for slightly higher current is to  put a resistor across R62 to reduce its value by about 10%.
Also check for possible RF interference or oscillations. Although unlikely in this case, that configuration of transistors can oscillate.

[duak]

Well, the US put men on the moon (and returned them safely to the Earth) using technology contemporaneous with these supplies.  The initial temperature drift problem is probably inherent in the design.  I have an hp 6002A and it uses op-amps in the same basic topology but it was far more difficult to get working right and it's got a noisy fan that runs all the time.

The instability you see is another thing.  There's got to be something blinky in there.  Keep puttering along with it and it'll be found.  BTW, If you bypass Q15 to 12.4 V with 470R, it may give more clues.

My supply was missing a cover over the power resistors in the back so I tried to find what it looked like.  In the process I found this link showing someone replacing the analog meter with a digital one:
 

The digital panel meters I worked with would accept either +/- 200 mV or +/-2 V so at first glance it wouldn't need a meter amplifier/limiter circuit.  The meter accuracy could be improved - just sayin'

I think I saw a posting on EEVBLOG recently where someone had replaced the LED/LCD with Nixies in an hp something or other.  Personally, I like the look of LEDs and they're from about the same time.

[iroc86]

Yeah, the random instability is really the kicker here. After you all helped me diagnose the output current issue (back in May, gosh!), I started using the power supply semi-regularly. It became really annoying when the needle would sometimes read a half-tick lower at turn on. The supply held its output voltage just fine of course, but I'd have to pull out a DMM and check it just to confirm. Kinda frustrating on a benchtop lab supply that's inherently going to be powered off and on over the course of a day.

I'll add both of your suggestions to my list and report back. Should have some time this weekend to play around and let the unit go through a few warm-up cycles.

The digital panel meter is a neat idea. I had actually considered this a while ago and picked up a bunch of original HP 5082-7611 7-segment displays, the classic red digits they used in all of their older test gear. I'm a bit of a purist when it comes to refurbs and restorations, so I thought this'd be a cool way to upgrade the old 62xx supplies without resorting to modern parts that ruin the vintage look. My plan was to use an ICL7107 and eliminate the original meter amplifier altogether. (FYI, this is essentially what HP does in their E36xx supplies. I took apart an E3630 triple output supply we had at work and found a TC14433 ADC in quite nearly the exact datasheet application circuit for a 3.5-digit voltmeter. )

This was the post that gave me the idea, though he was swapping LCD to LED in an HP system multimeter: LED display for HP 3457A multimeter - I did it :-). Perhaps it's the one you were thinking of? Not Nixies, though.

[duak]

Forward to the past!: https://www.eevblog.com/forum/testgear/nixie-display-for-hp3478a/msg2744792/#msg2744792

An idle thought - use a single chip instrumentation amplifier to drive the meter, like the http://www.ti.com/product/INA128   One resistor sets the gain.  Add a few diodes or LEDs across the output (after a series resistor) to clamp the voltage to protect the meter and you've solved the drift problem.

[iroc86]

Ha, very cool! +1 for that fellow's ingenuity with the Nixies. I have to say, I agree with him about the LCD display on those later model HP meters... probably why I'm still hanging on to my 3466A even though it's only 4.5 digits.

Nice idea about the instrumentation amp. I came across those when I was researching the amplifier used in these power supplies. I suppose HP's design is something like a discrete BJT version of the same thing? The resistor network across the input amplifiers is very similar, as well as the downstream amplifier (Q12/Q14?).

[iroc86]

The current limiting circuit with Q15 is continuing to catch my interest. Possibly not related to the glitchy amplifier, but who knows.

duak, can you check to see the orientation of VR6 on your 6284A? Was it per the schematic or the errata in my Reply #52 above? I've looked through several service manuals for HP supplies of a similar design and they show both arrangements (even with schematics that are clearly drawn by different illustrators/engineers). I suspect that the cathode to +12.4 V is the correct orientation since it's functioning as a clamp per the overvoltage condition you mentioned earlier. I'm fairly certain this circuit wouldn't work at all if VR6 was flipped around, but I'm just curious.

I tried a few other Zener diodes in place of VR6. I measured the Vce of Q15 and the meter current at maximum output voltage (50 V) in the low range meter mode (5 V). The higher values reduced Vce as well as the maximum current through the meter.

- 4.2 V VR6 (stock part), Vce=75 mV, 1.64 mA limit
- 5.1 V VR6, Vce=38 mV, 1.59 mA limit
- 6.8 V VR6, Vce=27 mV, 1.50 mA limit

This seemed strange to me. I could see how increasing VR6 would extend the saturation point of Q15, but with that, I figured the current limit would increase, too.

So, I ran some simulations in LTspice and found an interesting relationship between VR6 and Q15. At values of VR6 below a certain "knee point," the current through the meter is limited by Q15 acting like a transistor and throttling back. At values above the knee, Q15 is fully saturated; interestingly, raising the Zener voltage further will reduce the current even though Q15 is still in saturation. This relationship is illustrated in the plot below (output voltage = 50 V, meter on 5 V range). I've also attached a new LTspice model in case anyone wants to play around.

Based on swapping out VR6 with various values, I surmise that my circuit is behaving on the "right half" of the plot, beyond the knee: VR6 voltage goes up, maximum current goes down. This would also explain why I'm not seeing Q15 drop out of saturation. (I don't have any Zeners lower than 4.2 V to see if I can find the knee, but I can try to get some.)

This is still strange to me, since the transition point occurs around 4.5 V @ 50 V output. With a 4.2 V value for VR6, I should be seeing Vce rise to at least several hundred millivolts, as duak observed in his 6284A. All I can figure is that maybe some of the resistors in this circuit are out of tolerance and affecting the response in unusual ways. At the very least, I know that R62 is about 2.7% out (770 ohm vs. 750)... not enough to matter, but if enough of those variances stack up, who knows.

That all being said, I buttoned up the unit "as-is" and I'm moving on for now. I need the bench space back and this'll give me an opportunity to use the power supply for some other projects, so we'll see if the stability issue rears its head again. At the moment, I'm only experiencing slight drift (~0.25 V on the panel meter) over a 20-minute warm-up period, which is probably within the design spec and not much of an issue on the larger range. To recap, I replaced Q12, Q14, and Q15 with new 2N2907s and swapped the adjustment pots with regular resistors.

I may see about improving the meter circuit with either a modern matched transistor pair, the instrumentation amp duak pointed out, or perhaps a digital conversion with an ICL7107... all with increasing levels of time and effort.

[xavier60]

The 4.2V zener would be shunting some of R75's current to the 12V rail while still leaving just enough current for Q15's Base.
Higher voltage zeners allow more of R75's current to flow out of Q15's Base and through R62, increasing its voltage drop and reducing the current available to the amplifier.
It would have been better to have reduced R62 to rule out possible problems being caused by Q15.

[duak]

G'day y'all,

Xavier's pointed out what's happening with Q15 when VR6's voltage is increased.  In other words, higher VR6 voltages lead to the voltage at Q15's emitter dropping relative to the meter circuit's reference point  (+ Vout) and so it can't source as much current to the diff amp as before.

VR6's cathode is indeed connected to +12.4 V in the 6284A.  If VR6 were reversed it would act as an ordinary silicon diode and conduct with a forward drop of about a 0.6 V.  Q15 would turn on slightly and pass a far lower collector current, probably << 1 mA.