Minimoog rebuild, possibly PS issue (overload lamp)

[whinylittlerunt]

I’ve got a problem, and hoping someone can shed some light (pun intended).
Just finished servicing the boards on a 74 Mini. I screwed up, and put a 2.2 uF cap in backwards in the power supply. I didn’t notice this at first because I was checking the boards as I completed them, in series (so like I tested the power supply, made sure nothing blew up. Then tested with the key circuit, etc..). I went into this without being aware of the power sensing lines, and that all the boards needed to be present for the flow to work correctly.

Symptom is that the overload lamp comes on, dimly (not bright like when a fuse is blown), and the keys work but I’m getting just a noisy drone unaffected by controls.
I’ve since replaced the cap that was backwards even though it tested fine, and I went back through and did quick ohm checks on the semis just to make sure I didn’t short anything out. (I also test on a dim bulb but this has such low voltage it really doesn’t do anything to protect).
As far as restoration work, it was the standard rebuild - new electrolytics everywhere, matched resistor set on the oscillator board, all new LM741s, new sockets for the LM3046s. Also put in a new power cord.

I checked my voltages at the rectifier board once the supply was back in, and they looked good. Now I need to check at the power supply itself - I can’t seem to figure out where to measure the sense lines. I see it on the schematic; but I’m guessing just pick a component that connects to that point and just measure there, huh?

If a specific board is problematic, it should pull down the voltage at the supply right?

[analogmonster]

Not necessarily.

You should proceed systematically. Disconnect everything from the power supply, switch it on and check all supply voltages for correct voltages and for noise with a scope. Check ground for noise as well.

Connect a voice board and check whether waveforms appear on the scope and whether they look ok.

Check keyboard behavior for correct operation and do the scale and tune adjustment.

[whinylittlerunt]

I checked the voltages at the oscillator board. What three me off before was I was looking at the newer board layout and this is a version 2… oops.

So, +10v is good and rock solid. -10v is at about -9v and I can’t really adjust it. Trim pot doesn’t travel enough to bring it in. So I suspect either something is pulling down the -10 or the supply is still funny.

I tested the precision resistors and they all measure perfect.

I’ll check with the scope also, but as I mentioned, no oscillator functions work at the panel. The contour seems to work slightly.

[analogmonster]

Important is that you test the PSU only, with no other board(s) connected, neither any of the VCO boards nor another. Then you can see whehter the PSU itself carries the error or any of the (other) synthesizer boards.

[whinylittlerunt]

The lamp only comes on after everything is connected... if I have just the supply connected, the lamp stays off.

[analogmonster]

Check with a scope whether there is any signal or noise at the base of Q1204 when no external source is connected. That is the first tranistor of the overload lamp circuit. If there is no signal and the lamp flashes the lamp driver circuit is faulty.

If there is a signal without any external source connected my assumption is that the preamp for external sources is faulty. By the way: noise without any explaination is often a sign for unwanted junction break thrus within a faulty transistor. So what I've read so far my assumption is that there is a faulty transistor within the external input circuitry.

If I am right, you can find and identify this transistor by checking each emitter with a scope.

[whinylittlerunt]

Ok, well before I do that, let me just say I tried adjusting the -10v with only the supply connected, and I'm maxing out at -9.60v. There is barely any travel in the adjustment pot. I cleaned it just to be sure. Pulled and tested the transistors in the -10v circuit and everything checked out.

C20 on the power supply was what was in backwards, but that looks to be in the noise circuit.

What I did notice on the oscillator board, was after I put in the new chips, there was a TINY little solder bridge across 2 pins which I'm embarrassed to even admit because I'm super careful, but I guess it happens. Unfortunately, even swapping that chip out with a fresh one didn't change the symptom.

I'll populate everything and check the filter board again; you said Q1204... I'm presuming you mean Q12 on Board 4?

[analogmonster]

...
What I did notice on the oscillator board, was after I put in the new chips, there was a TINY little solder bridge across 2 pins which I'm embarrassed to even admit because I'm super careful, but I guess it happens. Unfortunately, even swapping that chip out with a fresh one didn't change the symptom.

Nothing to be embarrassed about. It happened to me as well.

I'll populate everything and check the filter board again; you said Q1204... I'm presuming you mean Q12 on Board 4?

Well, it is definitely Board 4. My circuit drawing from 1970 says "Q1204", a newer MINI service manual says "Q25 on Board 4"; both mean the same transistor.

Take a look at the schematics. What I mean is: The "External PreAmp And Overload Display" circuit (Board 4). The (external) input is limited and AC coupled pre-amplified by a 2 transistor differential amp and a filter stage. The result is limited again and hits the 3 transistor lamp driver.
The question is: Is there any "ghost" noise created by a defect transistor? This would let the lamp shine and would create noise in the synthesiszer, as if you would add an external noise signal to the MINI by the "External Input".
So if there is no external signal connected and the preamp generates noise there is a defect transistor, what you can check the way I said.

[whinylittlerunt]

I’ve not gotten a chance to go in and work on this issue yet but wanted to at least post a small video demonstrating what was wrong, and plan to get back to it this coming week.

https://youtu.be/zhl42h8d7kU?si=nnQdaxjBqB20LYIz

[analogmonster]

According to the video the noise vanishes when all oscillators are leveled down in the mixer. But there is a tone playing on any keystroke with a constant frequency independent from the key which is pressed. Where does it come from? Filter self oscillation? Or is that part of the problem?

[whinylittlerunt]

According to the video the noise vanishes when all oscillators are leveled down in the mixer. But there is a tone playing on any keystroke with a constant frequency independent from the key which is pressed. Where does it come from? Filter self oscillation? Or is that part of the problem?

I don't know. This is what I was presented after doing all the work. Overload lamp on constantly but dim, and the symptom shown in the video. I didn't power up and check each board for functionality between parts swaps, and I already mentioned that I put a power supply 2.2uf cap in backwards and subsequently found a solder bridge on one of the 741's on the OSC board. That's all on me. Learned my lesson to not rush. Those have since been corrected and I did do a basic continuity/short check on the semiconductors after fixing my screw up.

I'm going to go back through and check voltages again. I did originally swap out the damaged power cord with a new one and in the process had to re-do the wiring to the 115/230 switch, so I'm going to go back through my steps from the start and check everything. The instrument was working fine before this.

[whinylittlerunt]

Ok so right off the cuff, I'm missing -5v somewhere. I went back and took measurements, starting right at the regulator board (was wondering why I wasn't getting any readings and then realized I needed the supply board plugged in...). I pulled out the 1N4004 diodes I installed and measured them at about .513 voltage drop, and then checked some brand new stock ones I just got and they were more around .530 voltage drop, so I installed new diodes.

I'm unable to get the -10v any better than -9.6v, and then on the -unregulated line I'm supposed to be seeing -20v according to what is printed on the headphone amp section, pin 21, and I'm only getting -14v. I didn't include any other screenshots of the circuits on that board because the readings are the same at the corresponding pins. )Pin 1 on the noise generator circuit is getting +10v but for some reason it was not measuring from pin 1 on the socket. It measured out at R24 and R3.

So that is where I am at so far.

[analogmonster]

I will never understand why the Moog engineers defined the PSU concept that way in that times. PSUs are that damned easy to construct, but not that way.

Enough of ranting. First of all you need clear precise voltages from the PSU. But this can be measured ONLY when really nothing else is connected, no keyboard, no VCOs, VCA, VCF and so on. A naked PSU. Before you ensured this no other measurements and debug activities make sense.

[whinylittlerunt]

I have been testing this with only the supply connected.

[analogmonster]

If you remove the regulator board that only the rectifier board is left, what are the outputs of the rectifier board? They should be about + / - 20V DC. Please check whether these voltages are achieved and check with a scope that these voltages are without any noise, especially the ground line.