This electronic instrument project began with an experiment of walking ring counters. The basis of the design is that feeding outputs of a shift register back to the input through a logical inverter creates this rippling bit effect where, all ones then all zeros are passed down the line. When mixed by a resistor ladder this feedback renders "sinewaves" of different pitches, or noisewaves of different lengths if fed back thru exor. This utone circuit uses CMOS logic, a resistor ladder, and a few other simple pieces to create audio forms. The scale inherent in this instrument is the undertone series, giving divisions of the main clock frequency. I built this keyboard to tap the different outputs of the shift register and there were these pitches, divisions of the clock by 2 through 16. I consulted the finely tuned ears of the Owen Gardner (of Horse Lords ) in order to identify the scale. He responded over sms message using only cowboy language and western slang to describe the series.

How's my driving? There is a Micro-power PLL chip, the CD4046BE, which is used for its oscillator and exor gate. There are two CD4015BE shift registers shifting 16bits, a TL072 for buffering and such. There is an exponential current source for controlling the oscillator and there is a logic inverter for cycling the shift register. From each output of the shift register chips there are two resistors one 102 going to the touch-pad associated with the put, and another of a higher value to mix with the rest of the stages in the resistor ladder.

All keys sequentially all modes

One Expressive

Utone Filter Mess

If an oscillator signal is fed to the data input of the shift register, the whole circuit acts somewhat like a digital low-pass filter. In the noise modes there is something like resonance.

DIY-ing this design

I love diy-ing things! Here is a schematic painting of the utone design and some notes toward construction. In the picture below you can see yellow shapes along the bottom of the frame. These shapes denote octaves, evens, and odds. The squares are octaves, circles are odds, diamonds even. In the noise modes the odds tend to give longer noise patterns and evens will give more pitched, shorter noise patterns. In the painting it says 102 for the 4046 cap, in the right angle it says 470... it's up to you. Also you could use different values for the resistor mixer in order to mix less of a sine wave and more of a special wave.

To be clear, some of the pins which need to be connected for these Integrated circuits to function (enable pins, reset pins) I have omitted. If you need to see what to do with those pins try referencing my other schematic further down the page. This is an art schematic, generalized to show form and motion. no kidding, also I bet if you have some context using these devices (CD4046, CD4015, 2n3904, 2n3906) like from image search and a breadboard, reading the data sheet, you'll know what to do. Or you could very easily reference the F Mims style schema below that uses right angle wires on grid paper to describe the circuit. The toggle switch that sets noise modes or sine modes should be a SPDT center off. Center off is sine mode.

Below is an image of the first prototype I made of this design, I cut the pcb with the method outlined here. This is a single sided board so you can see all of the circuitry that is doing the work. I cut the spots for the ICs and soldered those in, then cut the traces for the touch keys. Then soldered the rest of the parts and finally the resistor ladder tornado.

Making right angle sine waves with CMOS may bring the questions for some "what is the difference between analog and digital?" or "what makes a system digital?" possibly useless questions, but could the answer be right angles or no right angles? -- nearly right angles, because there are never true right angles, always rise time, fall time and ringing from fast flipping bits in the tube of the copper trace. Thats the point, even the one or zero has a place where there is slew that must be squared up. But if we are still to entertain the dumb question, would it be sharp angles? maybe its memory storing bytes.. continuously variable simultaneous calculation as opposed to sequential calculations? CPU... The dichotomy of these two is itself a digital thought, either or.