Cricket NOR

Two CD4001 NOR chips: two tones, one slow chop, on when it is on · By Onion Madder · Mess O' Pedals

A CD4001 is four two-input NOR gates, and a NOR gate has one rule: its output is high only when both inputs are low. Everything this box does falls out of that. It has no Schmitt trigger, so an oscillator takes two gates, not one, and each oscillator keeps a spare input that stops it dead when you pull it high. Two chips is eight gates: a slow oscillator, two tones, and a two-gate latch. The slow one chops either tone or both, through a toggle each. The latch was drawn for a PLAY and a STOP button; it is built with a wire link holding PLAY instead and no buttons at all, so the box plays whenever it is powered. The two tones sum onto one volume knob. Named for the cricket, whose whole song is a tone chopped by a slow pulse, which is what this is.

The layout file is the guide

Like the 555 theremin and the Cicada NAND, the build document for this one is the file itself: ↓ Download Cricket_NOR.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.

The netlist

Quoted verbatim from the layout file. It was written from the board, so checking the board against it can only pass - what it is for is carrying the circuit somewhere else, and telling you what each strip is supposed to be doing while you stare at it.

The off-board wiring

Fifteen pads, and fourteen of them leave the board: four pots, three toggles, an LED, a jack and the supply. The fifteenth, +V, is a test point.

The board plays as soon as it is powered. J15 holds the latch's PLAY input on V+, so QBAR stays low and neither tone is ever silenced by it; the power toggle is the on and off. C4 still gives STOP a kick at power-on, and it loses to PLAY the moment it lets go. The latch, R6, R7, C4, D2 and D4 are all still on the board and doing nothing, which is harmless - see below for what a second board could drop.

Built, and heard

Built on the bench in September 2026, straight off the file above, and it works. It runs from chirpy and bright at one end to gritty and growly at the other. Everything is still on flying leads in the photo, so the drill template further down is a proposal rather than a record of a box that exists.

The latch does not latch on my board. As built, the PLAY button only plays while it is held down, and STOP does not silence anything - it gives a random dive-bombing noise instead. That is not what the circuit is supposed to do, and the layout above is not the reason: Q comes off pin 10 onto pin 13 and QBAR off pin 11 onto pin 9, cross-coupled the way a NOR latch has to be. So this is my board and not the drawing. The first thing I would put a meter on is the two links that carry that feedback, J12 and J14. Unproven - I have not gone back in yet.

Two bends of mine that are not on the file. One toggle between pot 3 and the positive leg of C5, which is the mix bus. A second between IC2 pin 7 and the wiper of pot 2 - pin 7 is the chip's ground pin and that wiper is the tone 1 timing node, so throwing it grounds the node the oscillator charges. I have not measured either of them.

So the buttons came off. A jumper from the PLAY pad to the V+ pad where the on button was holds pin 12 high, and that forces QBAR low and keeps it there, so the tones are never silenced and the board plays whenever it is powered. With the latch not latching anyway, that is less a simplification than a decision to stop asking it to. The layout file draws it as J15, in the same two strips.

Two things follow from that and neither is built yet. C4 stops earning its place - the 100n from V+ to STOP is there only to reset the latch at power-on so the box wakes up silent, and nothing wakes up silent any more. And IC2's C and D gates go idle, along with D2 and D4, the two diodes that carry QBAR into the tones. They are harmless left in place, and they are also half a chip doing nothing, which is worth knowing before you build a second one.

How it works, in one breath

A NOR with its two inputs tied together is an inverter. Take two of them: the first one's output into the second, a cap from the second one's output back to the first one's input, and a resistor from the first one's output to its own input. The cap charges through the resistor until the first gate flips, both outputs swap, and it charges back the other way. That is a square wave at about 1 / (2.2 R C), and there are three of them on this board. The rate pot with its 10K floor and the 1 µF give the LFO roughly 0.4 to 40 times a second; the pitch pots with their 1K floors and 47nF give the tones roughly 100 Hz to 10 kHz. Ballpark numbers, because a CMOS threshold is not exactly half the rail.

The trick is the spare input. Each tone's first gate uses one input as its timing node and the other as a control: hold that control low and the gate is an inverter and the tone runs; pull it high and the gate's output is forced low no matter what, and the tone stops. Two diodes bring two different highs to that input and a 100K holds it low when neither is there. One diode comes from the chop toggle, which hands it either ground or the LFO's square, so the tone runs on the LFO's low half and stops on its high half. The other comes from the latch.

The latch is the last two gates cross-coupled: each one's output feeds the other's spare input. A high on PLAY forces one output low, which lets the other go high, and the pair would hold that way after you let go; STOP does the same the other way round. The output that goes high when stopped is Qbar, and it runs along a bus strip to both tones' diodes, so stopped means both tones silenced regardless of the toggles. As built, J15 holds PLAY high for good, so Qbar never goes high and the latch is only ever in one state. The two tones each drop through a 10K onto the MIX strip, the sum leaves through a 10 µF into the volume pot, and the 2K2 and LED just say the power is on.

Unbuilt, and these are the bench numbers. The frequency ranges, which will land wherever the gates' thresholds put them. Whether the LFO's chop has an audible click at the edges when the toggle is thrown toward it, and whether that is a feature. And how loud two full-rail squares through 10K each are into whatever comes next: hot, is the guess. The floor resistors are the ones not to leave out, for the same reason as on the Hive and the Cicada - a pot at zero would tie a gate's output to its own input.

Things to try

• The chop is the LFO's low half. Swap one tone's CHOP wire from LFO to the LFO's other gate output and the two tones alternate instead of chopping together. Untested; it wants a pad the board does not have, so it is a wire to a chip pin
• A bigger LFO cap, 4.7 µF, for a chop slow enough to be a rhythm rather than a tremolo. Film or a bipolar electrolytic; the node swings both ways
• A third toggle that hands one chop input the other tone instead of the LFO, so tone 2 chops tone 1 at audio rate. That is the harsh gated sound the Cicada gets from a NAND, with the opposite polarity
• The Eurorack version, once this one is heard: the LFO and the latch are the rack-native parts of it, a clock and a gate

The two-gate CMOS oscillator is as old as CMOS and belongs to nobody. What is done with eight of the gates here - the chop, the diode-OR stop inputs, the latch with its power-on reset - and the layout and the box are Onion Madder's, drawn in Copper Bottom.