XOR Ring Mod

Two inputs, one chip, one knob · By Onion Madder · Mess O' Pedals

Plug two oscillators into it and it smashes them together. An XOR gate is a ring modulator for square waves: its output is high only when its two inputs disagree, and that is the same thing as multiplying two square waves, so out comes the sum and the difference of the two pitches. This is that gate with a front end on each input that squares up whatever you feed it, so it is drawn to take the outputs of the other builds on this site straight from their jacks. One CD4070 does all of it, on a 12 by 22 board.

The XOR Ring Mod layout drawn as a stripboard diagram, titled XOR Ring Mod, 22 by 12 stripboard. Twelve orange copper strips run left to right, lettered a to l, with 22 numbered columns. The CD4070 chip, U1, stands in the middle on strips e to k, with a short diagonal link drawn across its body. Left of the chip: C1 and the resistors R2, R1 and R3 standing across the strips, C4 lying along strip h over a cut, and two long blue links running down to the bottom strips. Right of the chip: R7, R8, R4, R5 and R6 standing across the strips, C2, C3, C5, and D1 lying along strip e near the right edge. Gray X marks show nine cut holes. Off-board parts sit in the margins with a wire to each pad: input jacks JK2 and JK3 on the left wired to IN A, IN B and GND3; the power toggle SW1 and the DC jack DC1 at the top right on +9V and GND; the output jack JK1 and the A100K volume pot POT1 at the bottom right on GND2 and OUT. Below, the same board drawn from the solder side, mirrored, every solder point a dark dot. A parts list and an OFF THE BOARD block run underneath. — The board as Copper Bottom draws it: the component side on top, then the same board from the solder side with the columns running the other way, so a cut is the same hole in both pictures. The panel parts sit around the board with a wire to every pad, and the **OFF THE BOARD** block at the bottom says where each leg goes.

What you'll need

tap to expand / collapse

Fourteen parts on the board plus the chip and six links, and six off it.

1× CD4070 quad XOR, DIP-14 (U1) - spot it: fourteen legs, the number printed on top, a notch at one end. A CD4030 is the same thing under an older number. Three gates are used and the fourth is tied off. It solders straight to the board
1× 1N5817 Schottky (D1) - spot it: fat black body, a gray band at one end. Reverse-polarity protection on the supply; the band goes toward the chip
4× 100K resistor (R1, R4, R7, R8) - spot it: bands brown-black-yellow. R7 and R8 make half the supply; R1 and R4 hang each input on it
2× 10K resistor (R2, R5) - spot it: bands brown-black-orange. One in front of each input gate
2× 1M resistor (R3, R6) - spot it: bands brown-black-green. The feedback that gives each input its snap. Do not swap them with the 100Ks
3× 100nF ceramic (C1, C2, C5) - spot it: printed 104. C1 and C2 couple the two inputs; C5 sits across the supply
2× 10 µF electrolytic (C3, C4) - spot it: a small can with a stripe down one side; the stripe leg is the negative one. C3 holds the half supply still, stripe to ground. C4 is the output cap, stripe away from the chip
6× wire links - offcuts of component leg. One of them, J5, runs diagonally under the chip's body, so it goes in before the chip does

Off the board

1× A100K log pot - volume. Lug 3 to OUT, lug 1 to GND2, wiper to the output jack tip. The output is a full-swing square wave, so you will want it
3× 1/4 inch jacks - in A and in B on the left, out on the right. Input tips to IN A and IN B, both input sleeves to GND3; output tip to the volume wiper, sleeve to GND2
1× DC jack, 9 V center negative
An SPST toggle for power, in the positive supply lead between the DC jack and the board's +9V pad. Every build on this shelf gets one, and the drawing above shows it

The layout file is the guide

Like every layout on this shelf, the build document for this one is the file itself: ↓ Download XOR_Ring_Mod.json - import it into Copper Bottom and the editor gives you the board picture, the placement walkthrough, the electrical checks and a printable build sheet, all from the same file. No coordinates are written out on this page on purpose: the editor is the viewer, and a number typed twice is a number that can drift.

Both inputs come in on the left and the output leaves on the right, the way the jacks go on my boxes. Input B reaches the left edge along strip d, and the output reaches the right edge along strip l. J5 is the one link under the chip: it carries input B's squared signal from pin 11 across to pin 6, diagonally between the pin rows, and it has to be soldered in before the chip.

The netlist

My script wrote this from the design and the board was checked against it in Copper Bottom: every connection it names is on the board and nothing on the board is missing from it. GA and GB are the two input gates, QA and QB their squared outputs, and MIX is the XOR of the two. Pin 10 is the spare gate's output and is left empty, which is why it is not listed.

# XOR Ring Mod - the intended circuit, written from the design # One CD4070. Gates 1 and 4 square the two inputs with hysteresis, # gate 2 multiplies them, gate 3 is spare (pin 8 to ground, pin 9 on QB). VIN: @+9V D1.A VCC: U1.14 D1.B C5.A R7.A GND: @GND @GND2 @GND3 U1.2 U1.7 U1.8 U1.13 R8.B C3.B C5.B VREF: R7.B R8.A C3.A R1.B R4.B IN A: @IN A C1.A XA: C1.B R2.A GA: U1.1 R2.B R1.A R3.A QA: U1.3 U1.5 R3.B IN B: @IN B C2.A XB: C2.B R5.A GB: U1.12 R5.B R4.A R6.A QB: U1.11 U1.6 U1.9 R6.B MIX: U1.4 C4.A OUT: @OUT C4.B

The off-board wiring

Six pads leave the board: two inputs and their ground, the output and its ground, and the supply.

Pad	Goes to
IN A / IN B	The two input jack tips. Anything with a wave on it: the output of another build, a synth, a drum machine. Turn the source up; see below for how loud
GND3	Both input jack sleeves
OUT	Volume pot lug 3. The wiper goes to the output jack tip, and lug 1 to GND2
GND2	Volume pot lug 1 and the output jack sleeve. Same net as GND
+9V / GND	The supply, through the power toggle. DC jack positive to the toggle and the toggle to +9V; negative to GND

How it works, in one breath

An XOR gate only understands high and low, so each input first goes through a gate of its own that decides which one it is. C1 blocks whatever DC the source carries, R2 feeds pin 1, and R1 hangs pin 1 on half the supply, which R7 and R8 make and C3 holds still. R3 runs from that gate's output back to its input, and because the gate does not invert, the feedback pushes the way the output already went: by the arithmetic the input has to move about 0.4 V past the middle before the gate flips, and the same back the other way. That is what makes a ragged or slow wave come out as a clean square, and it is why an unplugged input sits still instead of buzzing. Input B gets the same treatment from R5, R4 and R6 on pins 13 and 11.

The two clean squares meet in the third gate, pins 5 and 6, and its output on pin 4 is high only when they disagree. Flipping one square wave in step with another is multiplying them, so what comes out is the sum and the difference of the two pitches and their harmonics, and neither original on its own. C4 takes the DC off, and the volume pot sets how much leaves. The fourth gate does nothing: pin 8 is tied to ground and pin 9 rides along on input B's signal, so neither of its inputs is left floating.

The front end wants a swing of roughly 1 V each way from the middle, about 2 V peak to peak, to switch reliably. That is arithmetic, and the CD4070's own switching point is only promised to land somewhere in the middle third of the supply, so a particular chip may want more. The builds on this site put out a square wave as big as their supply, so with their volume up there is plenty.

Things to try

None of these have been tried, because none of this has. They are the obvious knobs.

Two of the other builds into it. A Mosquito 556 on one input and a Dragonfly VCO on the other, and sweep one while the other holds still
One input only. The empty side sits still, so you hear the other input squared up, which is a fuzz of sorts on its own
Close pitches. Two oscillators nearly in tune give a difference tone that is slow enough to hear as a beat
Change the snap. R3 and R6 at 470K make the inputs harder to flip, so a quiet source drops out; at 2.2M they flip more easily and a slow wave may chatter at the edges

An XOR as a ring modulator for square waves is the standard Lunetta trick and belongs to everyone who has ever built one, and a gate with feedback as a Schmitt trigger is out of every CMOS cookbook. Putting both on one chip, and the layout, are mine and so is anything wrong with them.