SynthGuy's Synthworld: Gettin' Some Serious Control Of The Moog 55 Modular!

The next round of modifications For the Moog 55 came about to increase the system's usefulness in studio and live performance.

Although the 55 looks like a very large and powerful system, there are some things about the way it's implemented that limit its abilities, and not always in ways that are immediately apparent.

Well, some are obvious. Seven VCOs and ONE lowpass filter? I mean, C'MON.

But hey, that's a pretty obvious one, and you knew the job was dangerous when you took it.

Sure there's also a highpass filter, but on its own, it doesn't give a lot of flexibility for voice shaping. There's a set of 6dB/octave lowpass and highpass rotary switches that I still haven't figured out much use for even after all these years (and I'm certainly open to suggestions here).

There's other things that are less obvious, but pop up as annoying limitations as you use the system more and more.

For instance...

Hey, there's no real dedicated LFO. Sure, there's the big standalone 921 oscillator, but using it as an LFO seems like a criminal waste of its many capabilities. It's the only oscillator in the system that can be hard synced via the Clamp input, and who wants to give up an incredible audio oscillator as a lowly LFO anyway?

On the other hand, it is syncable and resettable when used as a clock, and the adjustable differential auxiliary output makes for great stereo panning effects when patched to a couple of VCAs, it has a reverse sawtooth output...maybe it IS just a very capable (and expensive) LFO. It's puzzling. At any rate, where are the simple LFOs? Not here...

The 921B oscillators can be set low enough to be an LFO, but each of the two banks of 921B oscillators are tied to a single 921A controller. You can't use the other two oscillators as audio oscillators tracking the keyboard without the third (LFO) oscillator tracking the keyboard as well. What the what??!??!

Yes, that's right...there is NO way to control these oscillators individually! There's no separate CV tracking inputs, and no way to separate them from the 921A controller. So in reality, instead of having 7 oscillators in the system, you have 2 groups of 3 oscillators that must track the same pitch CV (so basically you have two fat oscillators) and one super oscillator that will, in most instances, end up being an LFO.

Erg.

This was a sobering realization as I started trying to create larger independent patches.

Another hidden limitation is that all three 921B oscillators in a group share the same pulse width, and the same amount of PWM if applied. If you set the 921A controller so that one oscillator is a square wave, all three are square waves. There are no individual CV inputs for pulse width on the 921B modules. When you consider just how useful poly PWM is for creating lush animated leads and pads, this becomes a baffling omission for a top of the line system.

Even the MiniMoog lets you set separate pulse widths for each oscillator, and even though they don't have PWM, there are certainly some classic sounds to be had there.

Now, to be completely fair, all of these things I've listed here as limitations actually tend to underscore what was still the main purpose for this system back in 1974.

It was designed as a studio instrument, where you would use recording techniques like overdubbing to build up a complex sound, instead of having lots of individual control of all the basic elements. While I could see and understand this approach to using the instrument, I was increasingly missing the touches that would further unlock the many capabilities that lurked beneath the surface, tantalizingly out of reach. And for live performance, having this kind of flexibilty greatly increases the usefulness of the system, since it can be patched to do more at once.

So, here's what I did about it (deep breath).

The schematics show that each 921B already has the necessary circuitry to be controlled by multiple CVs for both the pitch and the pulse width. The only thing required would be a summing opamp circuit for each, an easily implemented circuit. This would duplicate the actions of the summing circuits in the 921A...in essence giving you two (or more) separate control inputs to these oscillators. This seemed like a promising place to start...

I also needed to create some jacks as inputs for these new CVs, but the oscillator panels are already filled with jacks, and I didn't want to lose any of the current sync or FM capabilities.

My solution was to replace the mono jacks used here with TRS stereo jacks. Now, if I plugged a cable in halfway (first click) then the incoming signal would go to the original sync or FM destination. If I plugged the cable all the way in, then the signal would go to the new pitch or PW destinations. I could use a TRS splitter to drive both destinations at once if desired, and best of all, the mods would be invisible from the front panel.

I designed and built the additional summer boards using surface mount technology. This gave me a very small footprint, as the circuitry is fairly simple, and allowed me to use some very beneficial components.

With pitch inputs to summers, the input resistances must be very closely matched to tolerances of .1% or better to get proper tracking from multiple CV sources (like if you planned to drive an oscillator from a sequencer and also a keyboard to make transposition changes). Companies back in the day like Emu Systems or Sequential Circuits often had someone dedicated to measuring and binning vast quantities of 100k ohm metal film resistors all day long to feed production needs, using accurate test meters or a specialized test setup.

For my summers, I opted to use precision surface mount thin film resistors from Susumu that are already factory matched to .01%. These excellent resistors cost a little more, but the end results are worth it. I also chose a surface mount precision low drift dual opamp for my design that served as both the pitch and PW summers. All of this was built onto a Surfboard 9161 prototype board including a fine tuning trimmer for the pitch CVs.

Now, I know that there are some people who are wary of their abilities to solder surface mount components, but with good equipment, practice and patience, it's not really that difficult, and the end results can be quite good.

There was even enough room at the bottom to drill a small hole to allow for mounting the boards on the oscillator module via a stick-on nylon standoff.

As I thought about how I might use this, it occurred to me that it would also be useful to be able to disconnect the CVs coming from the 921A to increase the amount of independent control for each oscillator. For this, I decided to add a set of small switches to each 921B for the pitch and PW controls coming from the 921A. This way, each oscillator can be controlled from an external CV, the 921A, or both!

Hey, now we're gettin' there!

After carefully selecting and drilling the locations on the oscillator front panel, I installed a pair of short bat toggle switches. Using a nice dress nut instead of the standard supplied hex ones made them look more like original equipment on the panel.