The Circon

LeafThe Circon

 The Story of The Circon

LeafThe Circon is a continuous controller, not an instrument. I built it originally in 1978, when Rachel and I had been asked to score Kubrick's The Shining. We were still assembling musical ideas, trying to come up with a novel, haunting melodic sound. The Moog Synth, with its keyboard controller, was poorly suited for the kind of color we were looking for.
Then I remembered nostalgically that I used to "play oscillator" (quite literally!) back as a student of Ussachevsky's at the Columbia Princeton Electronic Music Center. You see, they had no other controllers, certainly no keyboards, and that was as good as I could come up with for performing melodic ideas. Several years before that I'd seen an article in a magazine that showed a quartet of people playing electric instruments with an enlarged pitch pointer on a 90 degree curved scale, which showed where the notes could be found. So it wasn't as crazy an idea as you might think.
So for the Shining score Stanley had been listening to music like Sibelius' Valse Triste, and that cried out for such a device. We ended up building this instrument as an enlarged version of HP and Heathkit Oscillators, and Bob Moog was generous to design us a circuit that much improved the controller's stability and accuracy over my crude attempts at electronic engineering design!
At the time Rachel had found a couple of Italian cabinet makers, who were doing some woodworking projects for the brownstone. We asked them if they could build the little cabinet I had in mind. They said, "Sure, make us a sketch." So I gave them a prespective pencil drawing, with rough dimensions. But they came up with a perfect mousetrap = no openings of any kind!
Oops! How to install the circuit board and components? Yeah, we all had a good laugh, as I'd not really made that detail clear on the sketch, and they cut a neat rectangular opening in the bottom. Bob sent me some of the harder to find components, and I wired them onto a "vector board". We drove out to a Brooklyn plastics firm to get the white Lucite plexiglas (for the disk), and a smaller clear piece (for the two arms).
I cut these with a sabre saw, then sanded and polished the edges -- eventually working up to moistened toothpaste on a paper towel for a professional looking gloss (it really works!). The dial originally had a narrow paper strip afixed all around the arc which had the notes of the scale hand marked in small radial black lines and note names. A couple of years ago, while restoring the unit, I replaced that now worn and dusty guide with a new keyboard-like dial, designed in Illustrator in my Mac, then laser printed onto mylar film. Note how the spacing varies visibly all around the dial, as the pitches have to be found by trial and error, then carefully measured and marked. (Yes, it took a while.)
As I mention in the Open Letter, it would be neat to find a manufacturer to produce Circons for others who may wish to use one. Currently I have no plans to do this, and have no idea if it would be something many other musicians might want. Meanwhile, you can hear it in some of my more recent work on "Tales of Heaven and Hell", for exotic sounds and solo passages, most usefully for its human vibrato and phrasing in molto expressivo melodic performances.
To create those computer stored Circon tracks, please note that the still original controller has no MIDI. So, as cludgey as it sounds, an elaborate work-around was setup in the studio. Here's what happens: The Circon is connected to the original Moog synth with analog control voltage patch cables. Yup, the modular synth still works most of the time, an amazing track record. The output signal from the synthesizer is adjusted with a tuning meter for close to pure octaves. The Moog's output goes to the Fairlight Voice Tracker, which is set to track the simple high tone of the synth in pitch, volume, and brightness. The marvellous voice tracker does have MIDI out. So that output goes into a MIDI merger, and gets passed into the MIDI Timepieces, into the Mac, into Digital Performer, my sequencer program of choice.
Once in the computer, I store the performances at high res, for possible editing and tweaking. (That's really tough, though -- hand-editing these dense MIDI tracks, with all the pitch bend and continuous controller information going on in them!) At the same time, the MIDI patch-thru is fed out to one of the versatile, clean-sounding Kurzweil K2000, K2500 or K2600 units (it would also work fine with many other modern synths.) I've built several special Programs for the Kurzweils which are designed only to play the Circon signals. There's a good range of timbres available, and the K2000s or K2600s, like other modern MIDI devices, make it easy to map all sorts of useful controller interaction, much more flexible than, say, a Theremin or Ondes Martinot.
Are you still with me, not asleep yet? We're almost done. The Kurzweil audio output is the signal that gets recorded, usually after some ambience and reverb are added, which seem to go well with the intense vibrato of a Theremin-like paradigm. With a more modern Circon unit, there would be a direct MIDI out (see the note below), so all this "falderall" could be avoided. But now you know how all the nitty gritty of workaround routing is accomplished!
(To see more of the Circon in action, check out the music to "HeavenScent" (from TH&H) which can be downloaded in MIDI formats for your perusal. Informantion is on the HeavenScent download page... Before doing that, though, you might want to check out the images and descriptions just below, and Bob Moog's contributions, below that:)

Circon, front view

LeafGot a cute little digital camera early this year (this was in 2001). It's perfect for taking those off-the-cuff shots and odd closeups of things you are trying to describe on the web. So I thought you might enjoy seeing some new views of the Circon as it appeared at the start of 2001. First, here it is, sitting on the oak kitchen stool that I usually place it on when I'm recording with it (or even just when practicing). This image is much better illuminated than the original flash snapshot, so you can study the device better here than the photo at the top of this page. (As usual, click each of these images for a large, much clearer view.)

LH side
Circon, side view

Let's move in closer to the left side of the Circon. This is the domain of the left hand, which "plays" a pivoted short arm for control of volume and expression (your fingers press it down and let it spring back up quickly for each detached note, and/or ease it up and down musically for a more singing passage). You can clearly see that I used the very expedient device of a rubber band (you may know it by it's more technical name: "elastic polymorphic potential energy temporary storage device"). A simple push-pin from an art shop (now long gone, alas) holds the upper end, while the lower end slips around the arm, about a third of the way up, as shown.
Above that are two knobs, from left to right: "Range" and "Scale", which resemble the similar controls on an analog synthesizer. The first knob's control shifts the whole voltage higher or lower, adding a constant additional +/- voltage to whatever the main pitch arm is set to generate. The second control is simply in series with the arm's voltage source, so it varies how much a voltage change occurs with a constant angular motion of the main pitch arm (the "slope" of the curve). Together you carefully adjust to obtain accurate octaves, and assure that the A's are near the usual concert pitch of A = 440 Hz.

Circon, interior view

If you were to invert the Circon and look at the bottom, you'd discover this opening access to the internal wiring. To the right top is the precision potentiometer for the main pitch control, a short gray shielded wire coming from it. Most of the circuitry itself is contained on this simple "vector board", which I wired to resemble a printed circuit board, following the schematic Bob Moog designed. It's not an elaborate circuit, but is efficient and has proven itself with trouble free operation for nearly two dozen years, as this is written. I sometimes worry if our latest digital wonders will reach such a longevity. I guess the better ones will. My earliest GDS and Synergies are still fully functional, for example. Then again, so is the Circon, thanks so much, Bob! (And see more about Bob's contribution to the Circon recently added just below.)

© 1996-2012 Wendy Carlos -- All Rights Reserved

Up(Top of the Page)
Behind the Scenes, with Bob

LeafDuring my search for some other early technical papers recently, I came upon the original docs from Bob Moog, a key part of the story behind the Circon, some of which I've attempted to describe above. Without further ado, here are the two important pages for your perusal. (As usual, click each of these images for a large, much clearer view.) 

Letter from Bob Moog

Here's the gracious letter of April 10th 1978, which Bob sent me along with his schematic (below) for the Circon. We spoke only a once on the phone about it, yet he grasped at once what was needed. You can read here how open and straightforward is his description of the inner workings of his proposed circuit. I recall he kindly laughed at the much more primitive circuit I had originally proposed, which contained none of the voltage stabilizing and regulating niceties. He was very quick in coming up with this, then mailing it to me -- it arrived less than a week after we'd spoken about it!
I called him back right away, excited and thankful, and also delighted to have the feedback by the master of analog music instrument design. You'll read that he offered his time and expertise to me again for the future. As it turned out, two film scores would come up first. And soon after completing "The Shining" I prepared for moving to the new loft space, constructing the new studio, and working on the score to TRON. So I ended up having much more on my mind than new control tools for several years afterward. Then came digital synths, time to start to move into the newer technologies (with some reluctance for the tradeoffs -- MIDI presents some compromises for continuous controllers, at least). So I never did take him up on his kindness. Can't help but to wince a bit now, knowing that this offer has, well, expired.

Bob's Circon circuit, annotated

Here's the circuit design Bob came up with for this custom controller. I've added my commentary, annotations, wherever you see the lighter pencil additions (plus the bottom view of the transistor), to his original hand-drawn ink schematic. It's pretty much what I followed when I built the actual device. Note the two independent sections, one fairly simple for volume and expression, the other fairly subtle, for pitch. I note that in March of 1995 I had to replace the op amp chip with a newer, and probably more stable, substitute. I wish I could have afforded one of the highest quality plastic-film potentiometers made by companies like Penny and Giles. They have a much more consistent taper, so the spacing of the intervals would remain more constant. You can see that from the front photographs, the "keyboard" keys are not all the exact same width. But it's still close enough. The real problem would be if that Allen-Bradley sealed carbon pot would fail, as any replacement would present its own curve and semitone spacing. Fortunately, it gets little use, so I doubt I'll ever need to face that eventuality. (And anyway, there are perhaps a couple of hundred of the same vintage AB controls throughout the studio and the original Moog synth...!)
For awhile Bob and I discussed perhaps making this into a commercial product. But the idea seemed impractical, and sort of faded away. After his excellent new designs for MIDI Theremins, it seemed like a similar final output stage might allow the Circon to speak MIDI directly, and we again considered manufacturing a few for the market. At least I expected to convert my own custom unit to MIDI. Too many other projects came up, for both of us, so it never happened, nutz. And now I don't see how it ever will. But the concept it still sound, and provides a somewhat more tractable way to obtain Theremin-like music passages. Please feel free to adapt the idea to your own electroacoustic music making tools, if you'd like. It's quite a lovely addition, if I do say so myself...! ;^)

Oh -- there's also something else you may want to check out. On the Resources page, in the new top section of PDF files which I've just added to our site, there's a pdf I've made for the more technically curious among you, from scans of an article in the January 1952 Popular Mechanics (I had spotted it in an uncle's dusty basement, when I was a kid, and he let me take it). It describes a built-it-yourself project (very common then, not so much now, alas!) of four "Wobble Organs", comprising an Electronic SATB "barbershop quartet!" Fascinating. I was lucky to find a clean used copy of this issue on eBay, blindly trying to figure out which one it had appeared in (found it in two tries!).
In grad school I recalled the article only vaguely (the magazine copy had been lost years earlier), but remembered that one could learn to play melodies from a curved rheostat/potentiometer dial. See -- we all build on the past, nothing is ever really "original!" That encouraged me to work out the means to try the same idea, but this time using standard audio test oscillators, first in my parent's home, then later at the Columbia-Princeton Electronic Music Center (with Vladimir Ussachevsky and Otto Luening). A little clumsy, but it worked! And it sure sounded different than the usual technique we used in the CPEMC: bits of recorded magnetic tape pieces, carefully cut and spliced together by hand along a grooved aluminum editing block, to form "melodies and phrases" -- ick! (You all think you have it so hard now, DO you...? Ha! ;^/ )
"Tomorrow's Electronic Barbershop Quartet", by L. A. Meacham
Popular Mechanics Magazine, January 1952
(Original article Copyright 1952 Pop Mech)

Download the file to read and/or print HERE.
If your browser experiences problems download a zip version HERE.

Vladimir encouraged the musicians who worked there to try to be innovative, expand the horizons, and found my idea for real-time playing of a calibrated HP oscillator most amusing and useful. So he allowed me to adapt, with greater precision, one of the studio's newer, better units (I recall a gray panel Heathkit sine/square oscillator). It was my memories of that graduate student experience which much later inspired creating the Circon Controller for the Moog, when the Kubrick film project, "The Shining", came up. At first Stanley seemed to be gravitating towards at least a few spooky quiet melody-driven things like haunted waltz music (for example, Sibelius's "Valse Triste"). Later he abandoned that idea completely and stayed with more textural elements. I, for one, was sorry to see it go, if only for the contrast and variations it would have allowed us.
You can read about and listen to some of my Circon performances on volume one of "Rediscovering Lost Scores". Track four of that album, "Nocturnal Valse Triste", is my actual final Sibelius performance, which wound up not in the film (a long story, read the notes on both volumes) but instead at the end of Vivian Kubrick's excellent documentary on the making of "The Shining." And you already have probably seen up above on this page that there are several appearances of the Circon on "Tales of Heaven and Hell." One of the tracks there, "Heavenscent", is described in detail on the Resources Page.

Up(Top of the Page)

Back Back to the Wendy Carlos Home Page

Wendy Carlos, The Circon
© 1996-2012 Serendip LLC. No images, text, graphics or design
may be reproduced without permission. All Rights Reserved.