Moog/Realistic MG-1 Mod Guide

[stiiiiiiive]

Hi Charlie, welcome to the forum

I have an answer to only one of your questions.
When one says a filter can self-oscillate, it means that the resonance, aka emphasis aka q-factor can go high enough for the filter to produce an oscillation pitched on the cut off frequency. And given the cut off frequency range, you guess the said oscillation, which can sound as a larsen, is audible, yep.

I'm sure some others Moogists will chime in to help you with the technical stuff

[charliekuldip]

Thanks for getting back. I’m no sure that I’m getting no self oscillation from the filter. I’m looking at swapping out U1009 IC for a new one. And updating the peak emphasis slider with a new one from syntaur.

[charliekuldip]

I meant to add that I bought a LM3046N for U1009 as a replacement for CA3046N
As I read they have same specs. Does anyone see any issues with that ?

[analogmonster]

...and I proceeded to step 5: adjust emphasis trim at pin6 of IC U1012, when I had my multi-meter to check the pins and adjust emphasis trim pot, at some point the oscillation I heard disappeared, and no matter what the trim pots are for emphasis, or filter scale, or filter cutoff, I can no longer hear any kind of self oscillation. Furthermore, the slider for 'Peak Emphasis' doesn't seem to have any effect on the sound.

According to the schema of the service manual the MG-1 uses a standard circuit to control emphasis and self oscillation. The resistor values of R93 and R100 are added. If both are turned into direction of zero ohm at some point the self oscillation must start. U12 adds the two phase halfs of the ladder output as a difference amplifier, and the amplification factor is determined by R99 (75k). If R93 and R100 are lowered to zero (which means maximum emphasis) and no self oscillation starts you can increase the value of R99 to lets say 100k or more. This increases the ladder output amplification accordingly. But you should check the output of U12 (pin 6) with a scope first, NOT WITH A MULTIMETER! Perhaps you killed U12 with that. The filter itself continues working, but the emphasis and self oscillation is dead of course. If no output can be seen WITH A SCOPE at pin 6 you should replace U12. I've seen such faults before.

Ive checked the IC at U1012, and I believe it is working correctly, but Im not 100% sure.
Does anyone have any tips on using multi-meters to probe ICs? ...or any other likely culprits that could cause the 'peak emphasis' to no longer have an effect?
...

My tip is DON'T USE A MULTIMETER! You can shorten IC pins with that. The neighbor of pin 6 (output) is pin 7 (+12V power supply). I don't know what happens if you shorten the output of U12 with its power supply, but that can't be good. Use an oscilloscope instead. If no emphasis happens any more, you probably killed the 741 opamp (U12), is my assumption. But this is a standard component, easy to obtain and easy to replace.

[analogmonster]

Thanks for getting back. I’m no sure that I’m getting no self oscillation from the filter.
...

?
You can hear it. And you can measure it with a scope at pin 6 of U11.

[charliekuldip]

Analogmaster-
Thanks for getting back to me. I had no idea about the risks of using a multi meter on the synth. I’ve got a scope I can use from here on out. Ur appraisal of shorting U12 while using the multimeter matches my experience. I’ve got a replacement chip I’ve ordered from Jameco and should be here on Wednesday. I also ordered a replacement for U9, and a new slider for the peak emphasis. But I’m gonna start by replacing U12 with the hope that solves things.

When using a scope on the synth are there any best practices for getting measurements?

Thanks for the quick reply and I’m feeling a lot more confident about the next efforts I’ll be making thanks to u.

I’ll post in the forum how it is goes.

[charliekuldip]

One other question I had was about the keyboard current source adjustment on page 16 of the service manual. I’m able to adjust hi end trim R2057 for 2.85 bolts with high c pressed, but when I hit the low F and adjust low end trim R2051 for 0 volts, I’m not able to get 0 volts. The float I can get is about 150 mV. The low end trim won’t go any further.

Is it possible a neighboring resistor or other component t near the low end trim is bad?

The only way I can get it to 0 volts is if the hi-end trim For high c is less than 2.85v. More like 2.65v in this scenario.

[analogmonster]

Analogmaster-

Thanks for your compliment but I am just a _monster_ not a _master_

Thanks for getting back to me. I had no idea about the risks of using a multi meter on the synth. I’ve got a scope I can use from here on out. Ur appraisal of shorting U12 while using the multimeter matches my experience. I’ve got a replacement chip I’ve ordered from Jameco and should be here on Wednesday. I also ordered a replacement for U9

I did not get the point. Why do you want to replace the transistor array (U9)?

, and a new slider for the peak emphasis.

...and why this? Does it have another value?

But I’m gonna start by replacing U12 with the hope that solves things.

You can check with a scope first whether a replacement is necessary or not. Measure the amplitude / signal of the filter output signal at pin 6 of U12 first. According service manual this should be a +14mV/-12mV signal.

When using a scope on the synth are there any best practices for getting measurements?

- Don't touch any components with your hands while the synth is powered on / power cable is plugged in
- Find a proper place to connect the ground clamp of the measuring head
- Be very careful not to short IC pins when touch a pin with the measuring head
- Pre-select 10mV/grid unit vertical maximum (5mV would be better) and about 10msec/grid unit horizontal on the scope to see a proper result when measuring pin 6 of U12
- If nothing appears on pin 6: Pre-select the scope for 5V/grid unit and measure pin 7 for +12V to be sure that U12 is power supplied, meaning if the power supply is present and no output at pin 6 then U12 is definitely dead

Thanks for the quick reply and I’m feeling a lot more confident about the next efforts I’ll be making thanks to u.
I’ll post in the forum how it is goes.

You are welcome. If there are more questions / points to discuss don't hesitate to ask. When I started in engineering synthesizers people helped me as well.

Best

Carsten

[charliekuldip]

analogmonster -
Replacing U12 with a new chip worked! The peak emphasis is now working again! Thanks for your feedback.

One thing I have on my 'mod wishlist' is adding an external input for VC to control the LFO. I see some mentions of that on the forum but no real specific instructions.

Does anyone have any schematics for adding a VC input for the LFO?

[analogmonster]

analogmonster -
Replacing U12 with a new chip worked! The peak emphasis is now working again! Thanks for your feedback.

That's cool So you repaired an original Moog synth, which I consider as great success, as these analog veterans deserve to be kept and "cherished".

One thing I have on my 'mod wishlist' is adding an external input for VC to control the LFO. I see some mentions of that on the forum but no real specific instructions.

I agree. CV controlled LFOs are magic. And if two VC-LFOs control each other, it is even cooler LOL. I emulated ratcheting with an VC-LFO controlled by an EG. Listen to my track You from time point 5:13, there you can hear it.

Does anyone have any schematics for adding a VC input for the LFO?

It depends on the LFO circuit of course, _where_ you have to feed the CV in and _how_ you feed it in. I did this for my Formant LFOs, see also here: VC-LFO
For the MG-1 LFO you must feed in the voltage at the connection between R142 (9.1K) and the MOD RATE slider R141 via a 1K resistor, I assume, but I will check it with LTSpice and communicate the result here.

[analogmonster]

...
For the MG-1 LFO you must feed in the voltage at the connection between R142 (9.1K) and the MOD RATE slider R141 via a 1K resistor, I assume, but I will check it with LTSpice and communicate the result here.

I was wrong. The MG-1 - LFO is current driven, not voltage driven like other LFOs (like my Formant LFOs I modded with voltage control). But you can add a CV between R144 and R146 via 470R resistor in your MG-1. See the LTSpice results of the original MG-1 LFO and a VC variant at MG1_LFO and MG_VC_LFO. The triangle and pulse LFO outputs change in frequency when an e. g. 5V sine is added to the point I mentioned above, but as side effect the LFO amplitude decreases a bit. To compensate that a more complex circuit would be needed or better a voltage driven LFO, no current driven LFO.