Troubleshooting a 0-30V lineair power supply

[MichaelNL]

Hello everyone. I've been reading quite a bit on the forum but this is my first post!

A long time ago, way back in the early nineties I was in school and one of the projects was building a 0-30V 2A + 5V 2A lineair power supply. But live goes on, I ended up in IT and never really had anything to do with electronics anymore. So you can say my working knowledge of electronics is VERY rusty. :-)
A few weeks ago I came across the power supply again and thought I might as well try if it still works. I know it did in the nineties. They wouldn't let me leave school without a working power supply!

I turned the V-knob and the A-knob all the way to zero. When I switched it on the power indicator came on, just as 5V LED and the Amp LED. I measured a nice 5V too. So nothing wrong with the 5V part. I didn't test it under load though.
Turning the V-knob up makes the voltage on the output go from 0 to 30 Volts. So far so good. But when I turn up the A-knob only a little bit the output voltage increases quickly to about 44V. This also happens with the V-knob all the way down to zero. So clearly there is a problem and switched the thing off.

I opened the power supply and took out the boards to check them visually. I found nothing obvious. Nothing burned out, no leaky caps. I found out that the volt-LED didn't work anymore so I replaced it. I measured (in-circuit) all other diodes in forward and reverse with a multimeter on diode-test. All seemed to be fine (infinite in reverse, a normal PN voltage in forward). I desoldered all transistors and tested them with the hFE-function on my multimeter. All had values within the specs I found in datasheets. I'm not sure if that is the way to tell if they are faulty though.
I replaced the opamps with UA741's I had in my part-collection but that made no difference.
I replaced C8 and C9 because I happened to have those values but you guessed it, no joy.
Measuring the resistance of the A-potentiometer revealed that the values jumped around a bit here and there. Obviously this one was not too good anymore. I replaced both potentiometers (V and A) with new ones.

Now a schematic would be handy.

But finding the schematics of this power supply was another issue. It might be somewhere in a box but I couldn't find it. So I thought it would be a nice exercise to reverse engineer the schematics from the pcb (see schematic.pdf) and learn a bit about KiCad. There might be errors in the schematic. I tried my best to check everything but I just can't be 100% sure I've got it all right. If you spot something weird, tell me.

I tested a view voltages with both knobs to zero. The voltage across the zener diodes D3 and D5 are what they should be. The voltage across D4 is a little low, 11V. It becomes 12V when I turn the A-knob up a bit
The voltages coming out of the transformer are fine. The voltage at the rectifier D9 (measured across the caps) is about 44V. That's a familiar voltage: it almost matches the voltage I'm getting at the output when I turn the A-knob up a little.

It's noteworthy that this power supply has spent some time in a very damp cellar because of flooding. The power supply hasn't been in the water though. It's years ago and I'm sure there hasn't been any moisture near it in the last ten years because after the flooding I took it out and kept it in my house.

At this point I'm a bit at a loss. I don't know how to troubleshoot this further. It would be great if one of you could give me a view pointers so I can learn and brush up my electronics knowledge. I have a multimeter and a scope available to do more measurements. I would love to have this power supply working again!

Thank you in advance!

[HB9EVI]

I think you have to rework that schem; it's a complete pain to read how the components are ordered. Additionally several things are missing:

- I don't see a current sense resistor
- A-meter cannot work like shown in the schem
- voltage and current amp get the same feedback
- the current reference voltage is strange built up

no chance to say, where to search for the bug with that schematic

[coromonadalix]

the ammeter and voltmeter are not ok

the voltmeter is in parallel with the output connectors, not passing thru the 47k POT_U

The ammeter must be at least be in parallel with the current limitting detection resistor

The 2n 3055 collectors supply is connected where ?  this schematic is wrong and will not work, you're trying to output voltage thru the collector of the 2n3055, it should be the emitters.

The main output stage is wrong,  you should have some base resistors on each 2n3055, or at least an very low value ohm resistor on each 2n3055 emitter.

The 0.82 ohms resistors and 100ohms resistors are wrong too.

Where did you take this schematic???

see :  some similar designs

http://file.allpcb.com/bbs/p/wwwelectronicecircuitscom/wwwelectronicecircuitscomc20170227210203940.png

http://areksnotes.blogspot.com/2013/11/crude-power-supply-with-lm741-2n3055.html

http://www.electronicecircuits.com/electronic-circuits/regulated-power-supply-using-741-and-2n3055

http://www.electronics-lab.com/project/0-30-vdc-stabilized-power-supply-with-current-control-0-002-3-a/

[xavier60]

The original voltage LED being open would explain to original symptom. The CV op-amp U2 wouldn't have had any control of the Base of T3.
If the LEDs D8 and D10 are really wired the way shown in the schematic, it wont work well. The LEDs are supposed to allow either the CC OR the CV op-amps reduce the voltage of the Base of T3.

[xavier60]

It would make more sense if R14 and R15 are wired between B-E of the 2N3055's.

[MichaelNL]

Thanks for pointing out errors in the schematic. I had a hard time figuring out how it's all wired together because this thing consists of three pcb's and a lot of wiring going to the 2N3055's on the back of the case, the trafo and the front panel.

I'm going to do some more "homework" on how it's all connected and revise that schematic.

[xavier60]

Although it mainly looks like a good design, it has a short coming that's common in older designs. If the voltage Pot develops poor wiper to track contact, the output will suddenly increase voltage.
I would put some oil in the Pot. Preferably a high quality oil with low wax content. I use synthetic PAO compressor oil.
Sewing machine oil should be OK. 

[xavier60]

The replacement voltage LED has likely been fitted the wrong way around.

[MichaelNL]

@xavier60: You are right about the LED being in the wrong direction! It makes sense now. The broken LED was blocking, not a short. And a LED in the wrong direction is almost the same situation. I swapped the wires and it's working working now! Thank you very much for pointing that out to me.

But credit to the others too. My schematic is far from error-free. It will take some time to get it right but I will get there. The examples and suggestions help me a lot to get a clearer picture of how it should be. So thanks all for thinking here with me!