Just got Minimoog #2634. Need PSU help.

[rjd2]

hi folks, so i just acquired my 1st mini! exciting times for me. it's in great shape, all in all.

so i am in the process of going over her and getting set up, calibrated, and restored, and have a few questions:

1-calibrating the PSU: the service manual says to calibrate + and -10VDC using test points 1a, 2a, and 3a on the OSC board( i have old osc board). it shows an image of R47, R48, and R135, but when i read across any of these pins, i dont get a reading for -10V. can someone point me to the best way to calibrate +/-10V on an original model? what are the best test points? (seems like TP's on the actual PSU board would be best....but what do i know?)

2-rectifier/board 5: the schems show a C1(1000mfd), C2(1000mfd), and C4(470mfd) on this board, however, this mini only had C2 and C4 stuffed, and both were 470mfd. i upgraded these to new sprague 500/50v caps, however, im wondering if i should stuff C1 while it's open? why wasnt this stuffed?

thanks for any help, can't wait to get up and running. -rj

[Kevin Lightner]

You can quickly measure the power by using the large, curved heatsinks on the power board.
They're likely close enough for your needs.
But ideally, check them at the rectifier board where the power wires are distributed.
The case can be used as ground, but again, the ground terminal on the rectifier board is best.
You can also check power at the accessory jacks on top or at the first three pins of any edge card visible.
Don't be surprised if the power droops slightly as you go away from the power supply (for example, checking the voltages at the VCO board.)
This is normal.
Also be aware that the psu board has a zener diode on it that's sort of like a tempco resistor.
That is, it's temperature sensitive. So allow the machine to warm up before fine tuning the power or anything else.
The most temperature sensitive things in a Mini are the VCOs and VCAs.

Moog changed what caps were used on the rectifier boards many times over the years of production.
They started with two 500uf and ended with three 1000ufs.
Most have two 1000uf's and a 470uf.
When I restore Minis, I replace every component on that board usually, including the 4 rectifiers and the small foil cap in addition to the electrolytics.
I usually upgrade them too, so if it came with two 500's, I'll tend to use two 1000's and 470.
If it's using a newer style temp-comp VCO board, I'll go for 3 1000's.
The high current draw of the later VCO board may cause problems if you keep 500ufs in there to power it, but 500s should be fine for the normal style VCO board.

Also, later units upped the fuse current rating a bit from 1/4 to 1/2 amp.
Either one is usually fine tho. If you blow a 1/4, try a 1/2.
If you blow a 1/2, there's likely something else wrong.
Use slo-blo or "time delay" fuses if possible.

I hope something here helps.

[thealien666]

Personally, when I got my Mini and wanted to calibrate the PSU, I connected my probes on C12 (+10 V on its positive side) and C11 (-10 V on its negative side) on the VCO board (old OSC board). That way, I wasn't too far away from the actual oscillator circuits. They're easily located near the right side of the left connector of the VCO board. On mine they look like black plastic cylinders with one rounded edge, that's their positive side. Their negative side is the flat edge. C12 is on the bottom and C11 is two resistors higher and they're both 5.6 mfd 35 V.

I also have a rectifier board with two 1000 mfd 35 V and one 470 mfd 50 V electrolytic caps, and one thin foil 0.01 mfd cap. I checked them and they showed respectively 1020 mfd and 1010 mfd, and 475 mfd. So I didn't see the need to replace them yet. The small one was showing up as 0.01 mfd also. Maybe they were already replaced with newer ones by someone else before I got my Mini ?

[Kevin Lightner]

I also have a rectifier board with two 1000 mfd 35 V and one 470 mfd 50 V electrolytic caps, and one thin foil 0.01 mfd cap. I checked them and they showed respectively 1020 mfd and 1010 mfd, and 475 mfd. So I didn't see the need to replace them yet. The small one was showing up as 0.01 mfd also. Maybe they were already replaced with newer ones by someone else before I got my Mini ?

Obviously you have a cap meter (or decent DMM)
Not everyone does. Cool that you do.
But unless it checks ESR also, it's a good idea to replace them anyway.
Even old, bulging caps can read fine for capacitance, but have increasing resistance that will eventually cause inrush current to cause problems.
Capacitance does tend to increase strangely enough as they age tho.
I replace them because I never want to see any particular Mini again and it's easy enough to afford them another 20-30 years or whatever.
But the inrush current thing can cause the rectifiers to blow sometimes, even before the fuses and I never know if someone might choose to use the accessory ports for a sample and hold, ribbon or whatever.
The later VCO boards take a LOT of current too, so if someone adds one, they can take an otherwise working Mini and blow something.

The caps you mentioned above on the VCO board are solid tantalum.
They're usually ok on Moogs and don't need replacement, but some Minis have silver ones fwiw.
(The ones I use for replacement are yellow.)
It actually doesn't matter too much really about the voltages being absolutely perfect unless one is really picky and also expects to drive it via CV externally.
More important is them being equally the same, + and - .
But of course, it's proper and desired to get them exact as per spec.

The biggest problem experienced by 1st time restorers is that they don't know what to look for or everything to do or how.
I've received in instruments where the owner replaced certain parts fine, but completely ignored others.
But how can anyone learn unless they do?
It's good to know about one's instrument and money can be saved if all goes well.
If someone really screws things up bad, there's always professional techs.

Btw, Arps use different types of tantalum caps and they can often die.
The failure modes for most T-caps is a direct short too, so it's a good idea to replace them all on an Arp.
But they're expensive and doing this on an Omni or Quadra can really add up.
I've only seen a couple die on Moogs, so while I sometimes replace them, I don't do it on every one.
It's always a good idea to replace any that go across power rails in Arps tho.
I can't tell you how many 2600's I've seen that died from bad T-caps and it's never fun seeing smoke come out of your instrument.

[thealien666]

A very informative and detailed explanation. A big thank you Kevin!
I'm mostly a self taught electronics hobbyist with about 30 years of experience fixing and building things.
I hadn't thought about ESR. Good point.

Since it's pretty inexpensive to do, I'll go ahead and replace all those caps and diodes on that rectifier board as a preventive measure, just to be on the safe side. I have no way of knowing how old these caps are. If they're the originals, they would be from 1975 which is only 10 years younger than me, and I'm also drying up with age as they do!

[dougt]

Btw, Arps use different types of tantalum caps and they can often die.
The failure modes for most T-caps is a direct short too, so it's a good idea to replace them all on an Arp.
But they're expensive and doing this on an Omni or Quadra can really add up.
I've only seen a couple die on Moogs, so while I sometimes replace them, I don't do it on every one.
It's always a good idea to replace any that go across power rails in Arps tho.
I can't tell you how many 2600's I've seen that died from bad T-caps and it's never fun seeing smoke come out of your instrument.

Yep and not just ARPs. I've replaced tons of them in Oberheims, Sequentials, and Yamahas from the late 70s-early 80s where they're sitting across the power rails shorting them out... Kevin, do you know why the ones in Moogs have been more reliable?

[MC]

it's never fun seeing smoke come out of your instrument.

And that smoke smells like h3ll too.

[Kevin Lightner]

Kevin, do you know why the ones in Moogs have been more reliable?

No, but I have a couple guesses...

1. Moogs tend to run on lower voltages than those other synths.
2. Moogs seem to do more filtering at the power supply. The caps down the line may not see as much ripple.
3. The brands and style of t-caps used by Moog are different than the types used by those other synths.

..and yeah, when they blow, they really stink.
But normal electrolytic caps and resistors can reek too.

When I was a tech at Roland US, we used to get in a lot of blown CR-5000 and CR-8000 drum machines.
They had a regulator that would die and take almost every IC along with them.
We'd just have to sniff a machine to know what happened sometimes.

[thealien666]

...When I was a tech at Roland US, we used to get in a lot of blown CR-5000 and CR-8000 drum machines.
They had a regulator that would die and take almost every IC along with them....

That reminded me of those dreaded early Commodore 64 PSU which had a 7805 regulator inside that would overheat (due to an engineer that decided to encapsulate everything inside in opaque epoxy, à la ARP modules), and which its overtemp protection would fail and pass the input voltage to the output (25 V instead of 5), frying everything it powered! The smell of burned chips and transistors is choking, rivaling that of t-caps !

[Kevin Lightner]

I'd never heard of that C-64 problem before, but I don't doubt it a bit.
A 7805... properly heatsinked.... should be able to handle up to 1.5A, but most designers build in some tolerance and don't let them handle more than 700-1000ma.
But encapsulated, all bets are off even if heatsinked.
The heatsink itself needs to dissipate its own heat somehow.

That's kind of amazing too- a C64 ran on 1.5amps or less of 5v?
Forgive my ignorance- I'm an Apple II guy, but even Apple II's came with a large multivoltage switcher with a lot more current.
I guess not having a bunch of peripheral slots in a C-64 brought the draw down compared to the Apple, but it's still hard to believe they didn't go to a TO-3 version (7805K) or other regulator with more umpfh.
Personally I prefer LM317/LM337 regulators when I can use them.
They require more external "glue" components (a 240 ohm resistor and a trimmer, if I recall) but they're better at regulation and adjustable to boot.
It always amazed me that CAT SRMs used plain old 78XX style regulators.
THEN, they mounted them high on their leads only with a metal heatsink on top.
Great way to have one break off or short after a few weeks on the road.

I agree too: a bunch of burned semiconductors definitely smells worse than a single, small capacitor.
But what IC's don't tend to do is blow up like firecrackers with a loud boom when hooked up backwards.
I've screwed up before and installed an electrolytic backwards.
It wasn't pretty and I learned QUICKLY to double-check my work.
Not all caps vent elegantly (as with Macintosh Imacs) where one can just see the vent broken apart.
Some truly explode.

[thealien666]

Not surprisingly, later C-64 PSU used the TO-3 casing and a proper heat sink inside a ventilated box. I guess hundreds of repairs and exchange under warranty had Commodore learn their lesson.
And yep, that little C-64, which had better graphics, better sound, and a better keyboard feel than a basic Apple II and cost a whole lot less, ran all that on 1.2 amps (very close to the 7805 current limit), thanks to dedicated custom chips.

To get back on the subject of PSUs, call me old fashioned, but I really don't like new switching type that are used more and more in electronics. Sure they're more precise, physically smaller, are usually low cost, and offer better efficiency than linear types. But they're also much more sensitive to bad design and cheap components, or failing capacitors due to heat. Most of them have some type of protection against out of tolerance behavior due to failing components. But that's precisely their Achille's heel. I stopped counting the number of DVD players and computer motherboards I've seen, switching themselves off, of behaving erratically due to a failing switching type PSU. Not to mention that they're operating internally with dangerously high voltages, making them hazardous to diagnose and fix. They also require much more specialized diagnostic tools to calibrate and restore. Finally, if badly designed or out of specs, they can be a source of massive RF noise than can even bleed into audio circuits.

I'd rather have a good old big transfo (almost never fails) with an oversize rectifier bridge or diodes (very seldom fails), big aluminum electrolytic caps (can fail, but can be easily replaced before they do), and voltages regulators on proper heat sinks (rarely fail unless too close to operating specs for too long, and/or insufficiently cooled).

But I sound like my 72 years old father, who still thinks vacuum tubes sound better than transistors... I guess he likes even order harmonics distortion...

[Kevin Lightner]

Switching power supplies have come a long way though.
When built well, they can be safer, smaller, lighter, auto-detect incoming line voltages, produce many different voltages and produce less heat than a linear.
Poorly made or overtaxed in whatever application they're being used in, I completely agree with you.
For example,I have a Club Of The Knobs switcher powering the heating elements on one of my downstairs toilet seats.
It wasn't all that good for a synth, but heats a bottom nicely in the winter.

I don't like tubes more than transistors, but they definitely have their own unique sound and response.
It's not just the harmonic content when distorting, they can have different transient responses and coloration that's desirable.
One of the most beautiful sounds ever is a Minimoog plugged straight into an old 40w Fender with reverb or similar amp.

As for C-64s vs Apple IIs in the graphics dept, the specs would have one believe they are better because they have greater resolution and more colors (which is true), but the graphics are simply sharper on an Apple. Mono-font text also.
They have a kind of fuzzy look on the Commodores, especially around the edges.
I wouldn't want to stare at C-64 graphics or text all day or use their keyboard for lengths of time.
No slam to you or others that love 'em, I simply don't. I've programmed both (and Vic 20s, Data Generals, Amdahls, etc.)
I will definitely agree that a stock C-64 has greater sound capabilities than a stock Apple II tho.
Great sound chip!
The Apple CAN do sampling and playback with no extra hardware. While not very high resolution, it's possible to do.
I admit that I'd probably rather play games on a well equipped C-64 than an Apple II.
The graphics and better sound of the Commodore can make for some very cool games.
They're just two very different beasts.
I prefer the Apple II platform mostly because of all the peripheral slots they provide.

I have a small page up on main site about Apple II stuff I use: http://www.synthfool.com/apple2
It's come in handy for all sorts of things in the world of synth repair- EPROM copying and burning, calibration, detecting intermittent problems, etc.
Mine usually runs about 8 times faster than a stock Apple or Commodore too.
This due to one slot having an after-market accelerator.

Awhile back I was able to very easily make an analog style sequencer on an Apple equipped with a multichannel ADC/ DAC board.
Has real knobs (Moogs!) that are read in by the ADC.
1-16 CV input and outputs, gates, trigs, S-trigs... whatever.
I put a video on YouTube of it playing a Moog 55: http://www.youtube.com/watch?v=mRFRFnrNni0

I love programming things like that in AppleSoft BASIC.
Add one line of code to get, say, inverted sequences.
Or add two or three more lines and have it quantize the ins, outs or both.
Run forward, backward, random.... whatever you'd like. Easy to program.
One could even do special coding like "if an F is played twice in a row, wait 3 seconds, transpose everything, change all C notes to A, then run backwards."
This could drive an unknowing operator quite mad trying to figure out what some weirdo programmer built in for it's "rules."