[youtube:http://youtu.be/I8W20uwtJ3Y]

A “Joule Thief” is a simple voltage booster circuit. It can increase the voltage of a power source by changing the constant low voltage signal into a series of rapid pulses at a higher voltage. You most commonly see this kind of circuit used to power LEDs with a “dead” battery, but there are many more potential applications for a circuit like this.

In this project, I am going to show you how you can use a Joule Thief to charge batteries with low-voltage power sources. Because the Joule Thief is able to boost the voltage of a signal, you are able to charge a battery with a power source whose output voltage is actually lower than the battery itself. This lets you take advantage of low voltage power sources such as thermoelectric generators, small turbines, and individual solar cells.

## Project Steps

### Background: How Does a Joule Thief Work?

This circuit used in this project is a modified “Joule Thief.” A Joule Thief is a self-oscillating voltage booster. It takes a steady low voltage signal and converts it into a series of high frequency pulses at a higher voltage.

Here is how a basic Joule Thief works, step by step:

1. Initially the transistor is off.

2. A small amount of electricity goes through the resistor and the first coil to the base of the transistor. This partially opens up the collector-emitter channel. Electricity is now able to travel through the second coil and through the collector-emitter channel of the transistor.

3. The increasing amount of electricity through the second coil generates a magnetic field that induces a greater amount of electricity in the first coil.

4. The induced electricity in the first coil goes into the base of the transistor and opens up the collector-emitter channel even more. This lets even more electricity travel through the second coil and through the collector-emitter channel of the transistor.

5. Steps 3 and 4 repeat in a feedback loop until the base of the transistor is saturated and the collector-emitter channel is fully open. The electricity traveling through the second coil and through the transistor are now at a maximum. There is a lot of energy built up in the magnetic field of the second coil.

6. Since the electricity in the second coil is no longer increasing, it stops inducing electricity in the first coil. This causes less electricity to go into the base of the transistor.

7. With less electricity going into the base of the transistor, the collector-emitter channel begins to close. This allows less electricity to travel through the second coil.

8. A drop in the amount of electricity in the second coil induces a negative amount of electricity in the first coil. This causes even less electricity to go into the base of the transistor.

9. Steps 7 and 8 repeat in a feedback loop until there is almost no electricity going through the transistor.

10. Part of the energy that was stored in the magnetic field of the second coil has drained out. However there is still a lot of energy stored up. This energy needs to go somewhere. This causes the voltage at the output of the coil to spike.

11. The built up electricity can’t go through the transistor, so it has to go through the load (usually an LED). The voltage at the output of the coil builds up until it reaches a voltage where is can go through the load and be dissipated.

12. The built up energy goes through the load in a big spike. Once the energy is dissipated, the circuit is effectively reset and starts the whole process all over again. In a typical Joule Thief circuit this process happens 50,000 times per second.

### Modified Joule Thief That Can Act as a Battery Charger

In order to make a battery charger, I made a few changes to the standard Joule Thief Design.

First, I added a capacitor to the node between the resistor and the first coil. This helps to stabilize the output a little.

Then I added a 6-volt zener diode to the base of the transistor. This helps to protect the transistor from being damaged by voltage spikes. The base-emitter junction is the weakest point of the transistor. Most small NPN transistors will have a maximum allowable base-emitter voltage of 6 volts or less. So I added a 6-volt zener diode between the base and the collector of the transistor. This will limit how high the output voltage can get but it will protect the transistor and make it last much longer.

At the output of the second coil, I added a diode. This allows the output voltage to pass through but it prevents electricity from the battery draining back through the transistor.

The capacitors and the zener diode also help protect the transistor from high voltage spikes that can occur if the circuit is turned on without a load. The voltage of the second coil will jump up as much as it needs to in order to be discharged. If there is no load attached, the coil voltage can reach over 60 volts. This could quickly damage the transistor. The zener diode and the capacitors help to limit these voltage spike.

### Wind the Toroid

The transformer in the circuit is made by winding wire around a ferrite toroid. These toroids can be purchased from electronics suppliers or they can be salvaged from old electronic equipment such as power supplies.

Take two pieces of thin insulated wire and wrap them around the toroid ten times. Be careful not to overlap any of the wires. Make the wires as evenly spaced as possible. To hold the wires in place while I was prototyping, I wrapped the toroid in tape.

### Prototype the the Circuit on a Breadboard

It is always a good idea to prototype any circuit on a breadboard before soldering it together. This lets you check to make sure that all of your components are working. It makes it easy to test the performance of the circuit and make any necessary changes.

### Solder the Circuit Together on a Circuit Board

In order to make the finished circuit as small as possible, I soldered the pieces onto a perf board. This kind of circuit board lets you position the components anywhere on the board that you want and make custom connections between them on the bottom side. When I was done soldering, I trimmed the board around the components.

To make connections at the input and output of the circuit, I used PC board screw connectors. This lets you easily connect a variety of components to the input or the output. You can use whatever power source or battery you want to work with. To avoid mixing up the terminals, I recommend labeling the positive and negative terminals of each one as either input or output.

### Select a Power Source

You can use any use any low voltage DC source to power the Joule Thief charger. You can use solar cells, hydrogen fuel cells, thermoelectric generators, or small DC generators. You could hook it up to a wind turbine. You could even set up a small generator connected to a hamster wheel. Use your imagination.

The source of the electricity doesn’t matter. However, the input voltage will affect how high the output voltage can get. With the components that I used, you will get the best performance from power sources that are between 0.9 volts and 2.0 volts (with a maximum at 1.50 volts). Below 0.9 volts, the circuit will have difficulty boosting the voltage high enough to effectively charge a battery. Above 2.0 volts the output voltage can start to get too high and it will be limited by the zener diode that is protecting the transistor.

### Use You Jewel Thief Charger to Charge Batteries

This charger can easily charge battery packs up to 10 volts. However, keep in mind that in order to boost the voltage, the current drops significantly — so it will take a while to charge a battery. Because it charges at such a low current, it can’t effectively charge devices like cell phones because they need a certain amount of current in order to switch to charging mode. The best kind of battery to charge is a simple battery pack made of either NiMH or NiCd batteries.