“Build a small circuit from salvaged components, to make an LED shine brightly from low-voltage sources.
I feel a special connection to this circuit – whether you call it the “One-Volt LED” (named by Z. Kaparnik in 1999), or the “Vampire Micro Torch” or “Joule Thief” (designed in 2002 by Big Clive). The essence of this popular circuit comes from Kaparnik’s article in Everyday Practical Electronics, which included schematics for a “Bright Light” that will power LEDs for ages, pulling the leftover energy out of a nearly-dead 1.2V or 1.5V battery. It’s a truly magical little circuit, and versatile enough to work with a broad range of odd NPN transistors, salvaged ferrite cores, and “unusable” batteries. Sustainability heaven!
OK, OK… it’s really not ALL that mysterious. But I do think this circuit tickles our minds because it has a certain mysterious beauty, with its cycles of feedback, collapse, and decay, the interplay of electricity and magnetism, and so on. Despite its relatively simple construction, its behavior is complex and powerful.
As you can see, you have a battery hooked up to an NPN transistor through a couple of oppositely-wound wire coils on a ferrite toroid (AKA ferrite core or bead). One of the windings goes from the battery’s positive terminal (+) to the transistor’s base, through a resistor. The other winding goes from (+) to the collector. But, note how the wires cross: the battery’s (+) is connected to opposite ends of the two windings. This is important because, when power flows through the wires, it induces a magnetic field in the toroid. But it’s going in opposite directions through each wire.
Our LED is connected between the transistor’s collector and emitter legs. As yet, there isn’t enough voltage to make it light up. Power is, however, flowing through the resistor, then through the base to the emitter of the transistor, just enough to start pushing open the pathway for power to flow from collector to emitter.
As this happens, a positive feedback loop begins, wherein the electricity flowing through the wires induces a magnetic field in the ferrite bead, as mentioned before. The magnetic field from this new pathway pushes more power through the base-connected coil, and the flow of electricity keeps feeding the magnetic field, until the toroid becomes saturated – meaning it essentially can’t hold any more and the field starts to shut down – things are a little fuzzy here for me, but bear with me. 😅
So, when this happens, it stops the flow through the transistor and the magnetic field starts dumping its energy back into the electric coils, which shoves current through the oppositely-wound coil. This power is added in series to the trickle coming from the battery, it’s sent to the collector and the LED’s positive leg, and with the transistor closed up, all that sweet voltage courses through the LED at last, lighting it up brightly. (Big Clive’s video shows the circuit dumping around 22.5V into a 9-chip series LED bar!!)
But the effect is short-lived, and soon it also decays, turning the lamp off and beginning the cycle anew. According to Clive (again), a normal frequency is around 80-100kHz, depending on the ferrite core you use and when it reaches saturation.”