Breadboard sawtooth oscillator core from scratch

Nov 01, 2026 Last reply: 1 day ago 4 Replies
Jump to replies (4)

Most oscillator projects you come across start at the finished end: a designed PCB, a panel, a calibration ritual. This one walks the opposite road. A sawtooth core is laid down on a single breadboard with a short list of cheap parts, and every component gets explained as it enters the circuit. For anyone who has wanted to get into DIY synths but felt locked out by jargon, this seems like a genuinely good entry point.



The core itself is built around a 40106 working as the Schmitt trigger inverter, with an RC network charging and discharging against ground to make a repeating sawtooth. A TL074 covers the op amp duty, and a TL072 is listed as a swap since only one op amp is actually needed. Power comes from either two 9V batteries with clips or a dual supply in the 9 to 15V range. The rest of the bill of materials stays delightfully short: four 100k resistors, one 1N4148 diode, a 1 uF and a 2.2 nF foil capacitor, a single jack socket and a pile of jumper wires.



What stands out is the teaching. Circuits are drawn by hand on dot-grid paper, and oscillation is explained through a water analogy: voltage as pressure, current as flow, resistors as narrow pipes, capacitors as balloons and diodes as one-way valves. Basic waveforms like square, triangle, sine, sawtooth, shark fin and reverse saw all get defined along the way. The prototype tracks volt per octave and produces proper synth tones when fed CV from a Korg SQ-1 step sequencer.



Not everything is settled yet. Sine wave generation is still an open problem on the research list, and the volt-per-octave conversion is deliberately saved for a later part of the series. That leaves room to grow, which for a part one feels about right.



Project by Moritz Klein. Watch it on their YouTube channel:

formatting link


The pick that impresses me most is the 40106 as the Schmitt trigger inverter. It is a cheap logic chip rather than a dedicated oscillator part, yet with the RC network it still yields a clean repeating sawtooth. The trade-off, as I see it, is precision: there is less to fine tune than with a purpose-built oscillator IC. For a first core on a breadboard, that seems like exactly the right compromise.

Does anyone know how the output stage is arranged on the breadboard? A jack socket is in the parts list, but I am curious whether the buffers and the AC coupling sit between the core and that socket, or whether the raw sawtooth from the RC network goes straight to the output.

As far as I know the maker has not spelled out the exact signal order. What is clear is that this first part covers the oscillator core, buffers, AC coupling and voltage dividers, so those stages are at least all explained within the episode. The precise wiring between them is not something that has been stated.

If anyone builds this and wants to trim the parts count, the TL072 swap looks like the obvious move, since only one op amp is needed anyway. A second variation worth trying: bump the capacitor value and run the same core as an LFO. The maker has said an LFO differs from this core perhaps only in capacitor size, so one breadboard could host both experiments.

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required