Thursday, February 21, 2013

Zero-friction bicycle dynamo


I wanted this non-friction dynamo for my bike that you can buy, but then I thought, why don't I just make it myself?

I've always preferred dynamos above battery powered lights. There are several advantages:
- No need to change battery
- No concern with turning it off when not cycling
- Most of the time you can turn it on with your foot while cycling
- The superior feeling of satisfaction, using your muscles to produce light

My experience is, however, that normal dynamos often fail. Sometimes they simply stop working without any hope of repairing them, which has happened to me a couple of times. Less serious, but still not acceptable, is that they rarely work when it's raining, or even less likely, when it's snowing. Sometimes the position of the dynamo changes slightly and has to be readjusted. Another annoying fact is of course that it becomes significantly heavier to cycle.

My idea was to instead use magnets, coils and LEDs to make a dynamo. In that way you should be able to make it more robust, and hopefully solve all the problems I mentioned above.

As always, I searched the Internet to find out what other people have done before (see links at the end of the post). I saw one solution where one magnet was put to motion by other magnets on the wheel, but I preferred another one, where the magnets on the wheel simply pass close by a coil with an iron core. No moving parts at all in other words, besides the wheel.

This is the circuit I used:



I got a hold of two good coils, so I decided to create one complete system each, as seen above, for front and back lights. Very simple, as you can see. I chose to add a rectifier bridge to make use of both the positive and negative voltage peaks. Each diode has a forward voltage, which will increase the amount of voltage needed before current is flowing, so I used schottky diodes (they have a lower forward voltage).

Most people seem to use voltage protection on their circuit (with a zener diode). I'm not sure it's needed. If you measure the voltage of the open circuit you will probably see a higher value than the LED data sheet recommends, but the coil will not act as a perfect voltage source. The voltage will drop to the forward voltage of the LED, and the current will be saturated to a maximum level. If you don't have an oscilloscope it will of course be difficult to know what this maximum level is. In my case, since I use three LEDs in parallel I figured the current wouldn't be too high for each diode to damage them. They have been working fine for over a year now.

To find a good coil, look for an old relay (I found a couple in a heap of electronic trash at my fathers place). Some work better than others, so you may want to temporarily connect it to the LED and pass a magnet over it to test it, before attaching everything to the bike.


I used shrink tube to protect the windings, as seen below.


I bought a rectifier bridge from the local electronics shop, and placed it straight on top of the coil. To know what wire is what, I used red shrink tube on the positive, and black on the negative. To keep it all in place, I again used shrink tube, covering both the rectifier bridge and the coil (not seen in the picture).


As a final step for the coil, I used even bigger shrink tube encapsulate the thing (I know, it's starting to sound like an obsession). I made holes for the iron core (which conveniently was threaded at the end) and the two wires, and sealed the ends of the shrink tube with electrical tape. To prevent water from getting in through the hole for wire, I covered it with silicon.

There was already a metal clip on my bike, probably for use with bicycle bags, so I used that to mount the coil. As with most other dynamos, I connected ground to the whole body of the bike (by simply sticking it in between the nut and the metal when mounting). Since I had two coils, I placed an extra metal piece and mounted the second coil at the end of it.


My first plan was to use magnets from an old hard drive to put on the spokes, but by coincidence I found two reelight magnets on a trashed bike. Besides the extra effort needed to mount hard drive magnets to the spokes, they should be equally good, or better. The hard drive magnets are without doubt stronger. If interested, see the links at the end of the post. I placed my magnets pretty close to the nave and for me that gave enough brightness. The further away from the nave they are, higher currents and therefore higher brightness you will get.



I got the LEDs from old battery lights. I had to scratch some PCB traces on the backside, and solder a bit together.



As a final step, I used the cover of an old 2.5 inch usb hard drive to protect the coils, and to mount a switch (the switch finally broke after turning it on with my feet too many times, so now the lights are always on)



And the result!


(It looks like it's blinking slower than it actually is because many blinks come between the frames of the video)
Possible improvements:
  • I'm using two magnets. I could add at least two more to increase the blinking frequency.
  • The magnets I use are not very strong. If I use some hard-drive magnets, the intensity should go up.
  • The intensity of the light can simply be increased by placing the coil and magnets further away from the nave. At the same time, you might be need to add some protection in your circuit.
  • If you don't want the light to blink (in some countries it's illegal), a capacitor can be placed in parallel with the LEDs. I've even read of examples where a capacitor big enough was used, to keep the light on when not biking for a minute or so. In that case it will, however, take a while to charge it up.

Solar panel for heating


My old man is planning to make a solar panel, after seeing this project:
http://24volt.eu/solfangare.php

Every winter he needs to heat up his workshop to avoid water pipes freezing, and occasionally make it warm enough to work in. Hopefully that can be avoided by placing a solar panel on the wall.

The plan is to replace a door which is anyway almost never use, with a glass and a dark background. Air holes above and below should provide enough natural air flow so that no fan is needed. Another difference, with respect to the link above, is that the air is not taken from outside. The idea is that this will increase the efficiency even more when it's cold outside. You can see an initial drawing below.


I'll keep you updated about the progress!

Mobile phone as oscilloscope

A friend from the Repair Café where I sometimes help out, showed me something really cool. He had adapted an old oscilloscope probe to be connected to the 3.5 mm mic/headphone jack on his android phone. Apparently there are several apps to be downloaded, that plots the microphone signal. Imagine having an oscilloscope in your pocket wherever you go!

Found a link about it:
http://www.nicomania.de/en/ipod/oszilloskop-tastkopf-fuer-ipod-iphone-und-android/

I want one...