Pulse Width Modulation

Posted on January 22, 2011 by John C.S. Keston

I setup and recorded this example of pulse width modulation using my Roland Juno-106. I started out playing an octave with the pulse width set to half way. Next I moved the pulse width up and down covering the full range possible. Eleven seconds into the recording I enabled the LFO on the pulse width and adjusted the speed of the modulation all the way up and then all the way down again. The picture shows the LFO in action. Notice that the LFO operates as a triangle, but with slight curves on each slope, which is probably a result of the analog circuitry. Here’s what it all sounded like.

Eerie Pseudo Oscillator Microtrack

Posted on January 15, 2011 by John C.S. Keston

Here’s another result of experimenting with the Roland Juno-106. I created this sound by using the VCF as a pseudo oscillator. I cranked up the resonance to full, and put the cutoff at about half way, then turned off both the square and sawtooth DCOs. You might assume that with the oscillators off, including the noise and the the suboscillator, that the synth would make no sound, well it doesn’t really until you almost max out the resonance. Maxing out the keyboard slider changes the pitch of the notes on a roughly twelve tone scale, but it’s not well tempered, so this creates an eerie detuning between the notes. See the photo above for the exact settings. In the recording you’ll notice an abrupt key change happening a few times where I turned on the sawtooth oscillator. I also nudged the LFO on the VCF here and there to add to the spooky factor.

I’ve done a similar thing with the Pro One before, but it sounds really nice in six voice polyphony on the 106. My guess at what’s happening is that a sort of controlled resonant feedback on the instrument’s noise floor is creating the pitches. This is something that doesn’t really work with digital filters on modeling synths, as far as I know. I know that I haven’t been able to reproduce this behavior on the Korg MS2000. I don’t fully understand what’s happening, but that’s my theory. If anyone has better insight, please share a comment on the article.

Eerie False Oscillator

Ableton Live Arpeggiating Analog Polysynth

Posted on January 14, 2011 by John C.S. Keston

Today’s One Synthesizer Sound Every Day involves using the arpeggiator built into Ableton Live. The arpeggiator, found under MIDI Effects, is a pretty simple tool, there are the usual up, down, up/down patterns as well as random, random once (repeats a random pattern), and random other (doesn’t play the same note twice).

This one minute and fifty second microtrack is composed of two layers of arpeggios created by routing Ableton’s arpeggiator to my newly restored Roland Juno-106. I added some filtered tempo delay and mixed in some reverb on the fade out to polish it off. I wish I had a piece of hardware that did exactly what Ableton’s Arpeggiator does (perhaps with the addition of a tap tempo button). In fact I started a discussion about this on the Electronic Musicians Network, a Facebook group started by my friend Robert Luna. Friend me, then message me there if you’d like to participate in the group. I’ll write an article soon compiling my research into dedicated hardware arpeggiators and hardware sequencers. Here are the arpeggios combined.

Two Layered Arp at 114 BPM

Juno-106 Glitchy Drone

Posted on January 13, 2011 by John C.S. Keston

Here’s another excerpt from the glitches that I recorded while the voice chip was failing on my Roland Juno-106. In this section I was just holding a note without touching any other controls, so all the variations in the sound were caused by the autonomous shorting that was going on inside the faulty 80017a voice chip.

Juno-106 Glitchy Drone

Juno-106 Bad Voice Chip Glitch

Posted on January 12, 2011 by John C.S. Keston

As I mentioned in my last article, I was quite disappointed that my Juno-106 arrived with a bad 80017a voice chip, not to mention a laundry list of other technical and cosmetic problems. However, in some ways I enjoyed the challenge of getting it back in working order. All the research, soldering, and fussing with the electronics was kind of rewarding in the long run.

Another thing I wanted to explore were the glitches I could get out of it while the voice chip was bad. So, I recorded 20 minutes of crackly, warbled, weirdness while adjusting sliders and pressing buttons. I’ll be sharing more bits and pieces of this in upcoming articles. For now, here’s an excerpt of glitches from the first few seconds of my experimentation.

Roland Juno-106 Glitch from Bad Voice Chip