TRITON

Since we launched Rossum Electro and produced our first module (Evolution), folks have been bugging me to build an analog VCO. In their day, the E-mu Modular VCOs were the most stable and had widest range of any of that era. But I waited for inspiration to strike, and it did, twice!

So what is “Zing Modulation” anyway? In thinking about what makes an interesting timbre, I have always felt ring modulation (actually, “four quadrant multiplication”) held a prime spot, but it’s limited in usefulness because it produces frequencies that are not harmonically related to its inputs. This is due to the trigonometric identity:

2 x sin(A) x cos(B) = sin(A + B) + sin(A – B)

where A and B are frequency components of the ring modulator inputs.

But I realized that if the two inputs could be forced to be periodic with the same frequency, this limitation would disappear. To understand why, just realize that there is no “memory” in a ring modulator – if you start over from the same spot and give it the same waveforms, it will produce the same output every time. So the output of two inputs that have the same period must itself be periodic at that same frequency, which means all its frequency components must be harmonics of that frequency. And it’s fairly easy to arrange two oscillators to be precisely periodic – it’s called hard sync*. Bingo!

But because of that crazy trigonometric identity, there are going to be zillions of harmonic components, all adding together in weird combinations of phases, and these will vary in interesting ways depending on the exact details of the incoming waveshapes. Below are some examples of the bizarre waveshapes coming out of TRITON. The yellow waveshape is the carrier oscillator with Zing modulation; the green is the original carrier waveshape. The blue and red traces are the modulation oscillators.

The second inspiration was to build a VCO that would provide interesting yet precise control of the waveshapes. The circuitry underlying TRITON’s symmetry control was remarkably challenging. Varying the symmetry of a triangle wave from sawtooth, through triangle, to inverted sawtooth has been done many times before, but doing so in a fully analog, voltage controlled manner that has absolutely no effect on the underlying frequency is extremely challenging. Add to that challenge the requirement for the waveforms to be visually perfect up to 20kHz – now that would be fun (i.e. really hard)! I had to throw away an entire prototyped design because it wasn’t quite accurate enough at the highest frequencies. The final circuit is the only time I’ve ever had to specify a capacitor as small as 0.5pF to get the precise results I wanted. Here are some of the pure waveshape ‘scope photos. The top photo is the shapes at about 250 Hz, below it are the shapes at about 20kHz!

Finally, I realized that I’ve never seen anybody play with the phase relationship between hard synced oscillators. Combining that with Zing modulation gives even more voltage controlled timbral variation. These waveforms differ only by the phase of the MOD 2 oscillator!

* I’m pretty sure I was the person who first coined the term “hard sync.” I didn’t invent hard sync; I think that honor goes to Serge Tcherepnin. When I heard about it and how great it sounded, I immediately implemented it. But E-mu modular VCOs already had one form of sync, so we needed another name to distinguish the two. Hence “hard sync” and “soft sync”.

View or Download TRITON Documentation

TRITON Quick Start Guide (PDF)

TRITON Manual (PDF)