led coil generator problems

[luke.w.pc]

i am trying to make a simple generator which i have seen which basicly has a small cylinder with a magnet in and coils around the outside which generate electricity which lights an led.
i've got a pen casing and coated it with sevral levers of magnet wire (enameled copper) and put a magnet in and shook it around, my multimeter showed up to 5 millivolts(no way near enough to light an led), so i put toghether a full wave rectifier with 4 diodes and checked it and i still only got 5 millivolts.
i know an led takes a few volts to light and ive acualy seen one working just connected dierectly to the coil.
i would like to know if anyone has tried to make one of these before or has any ideas of why its not working.
thanks in advance for any help
luke

 

[KrisBlueNZ]

The coil may be generating much more than a few millivolts but the multimeter may not be able to respond quickly enough, especially if it's digital (these have averaging circuitry).
The obvious way to increase the output is to increase the number and lengthwise density of the turns, by using thinner wire or winding several layers or both, and reducing the thickness of the coil former, i.e. using a thin tube instead of a ball point pen casing. You can even use a non-magnetic metal such as copper; thin but fairly strong copper tubes are readily available.
When driving an LED from a coil in this way, you should put a reverse-connected diode across the LED, because LEDs can be damaged by reverse voltages above around 5V and the coil will generate both polarities depending on the direction of movement of the magnet.
Also you can get low-current LEDs that will light quite brightly with only one or two milliamps.

 

[CDRIVE]

Your plastic pen isn't going to provide much flux density and you'd be far better of charging a large cap by rectifying the AC. When it's charged use a switch to turn on the Led. The cap will have far more current capacity than your coil will.

 

[CDRIVE]

I forgot to mention that Schottky Diodes are best for this application because they have a very low forward voltage drop.

 

[Wabajig]

Or germanium diodes may be cheaper. Make sure your PRV is at least 100V for any diode you use, due to the counter EMF of the coil. Hope it helps.

 

[JimW]

I teach a class where we make these all the time. Here are a couple of suggestions:
- The pen casing is not big enough to allow a strong magnet to be inside. We use a half inch diameter plastic tube. 3/8 inch would work as well.
- Use a neodymium magnet. Stronger the magnet, the more voltage you will generate. If you don't have a Neodymium, use the strongest you can get in the tube.
- Use two LEDs, one in each direction. Then you will light one in each direction of the magnets travel.
- You can't use a metal tube of any kind. It will resist the travel of the magnet as you shake it.
- Lots of windings with thin magnet wire. We fill a 1 inch wide area to a height of 1/4 inch. This and a 1/2 inch Neodymium magnet will get you enough voltage to light 4 LEDs arranged in a full wave bridge configuration (two at a time).

JimW