After stumbling across Chia-Ying Lee's Wind Band generator I was very impressed with the simplicity and usage when creating small scale power in low level residential applications. I scoured the internet looking for someone, anyone, who had taken the wind band idea to the next level. For example created an array of wind bands to try and harvest a usable slice of power to do something other then light a low power LED.
Anyway it was hunting around and absently staring out the window that helped me stumble across the idea. I saw the leaves on a nearby tree fluttering in the wind and that gave me the idea. What if you don't need a band, what if you could do the same thing more efficiently with a single blade.
My aim is to produce a device that can be made for a very low cost yet also generate enough power then was required to make the device.
So you know I am an electronics engineer, not a physicist or a mechanical engineer so there is likely a massive killer reason that I have overlooked why this design has not been implemented before. My prototype partially works but certainly needs further work.
Following is a step by step account of how to make one of my simple inventions which I am naming as a Wind Blade.
Step 1: Wind Generators
The wind powered generators that everyone is familiar with is are wind turbines. These are hugely power devices but have a number of key drawbacks. The propeller has to be up fairly high to access the stronger and more uniform wind currents and the costs to build and maintain a modern large wind turbine are enormous. Smaller turbines are better in terms of cost and maintenance but generally generate significantly less energy.
Wind bands are a fairly new alternative way of collecting energy from wind when the speed and current are far less stable for example in a residential estate, on a boat or on a remote sensor node.
Wind bands for anyone who hasn't heard of them consist of a tensioned band with magnets fixed to the center of the band. The band vibrates with passing air flow a bit like a guitar string being plucked. Coils positioned near to the magnets pick up the movement of the magnets and convert this movement to electricity.
The inventor of the Wind Bnd has a website here - Humdingerwind.com
The downside with wind bands is that the best power output can be achieved if the band's tension is correct for the wind speed. As the wind speed is constantly changing this can be near impossible to achieve. Auto tension devices have been mentioned but I have yet to find a public design that actually pulls this off.
Step 2: Bill of Materials
To create a wind blade generator you will need the following parts.
- 2 x Round Neodynium Magnets - 8mm Diameter x 3mm Thick - Farnell
- 2 x Plastic M3 Bolts16mm
- 4 x Steel M3 Bolts 20mm
- 6 x Steel M3 Nuts
- Insulated Coil Wire - Approx 0.4mm Thick
- Optional - 2 x M8 Steel washers
Tools you will need for the build
- 3D Printer with 1.75mm PLA Filament
- Super Glue or Glue Gun
- Soldering Iron
- Sharp Knife - Stanley / Exacto
Step 3: Printing the Parts
The majority of the parts for the wind blade can be 3D printed. With the exception of a few bolts, two magnets and some enameled copper wire.
The attached files contain the files to allow you to 3D print the Wind Blade design as a single print. Printing time for me was about an hour.
The main blade is printed as a single layer so you need to ensure your printer is setup correctly to allow you to achieve a good strong flat surface with no holes or cracks. I used a rectilinear fill method on this part to ensure that the plastic filaments ran parallel with the sides of the blade. This allows the blade to flex evenly without twisting. I tried a few different blade thicknesses with different results.
0.2mm resulted in a fairly poor single layer print.
0.35mm worked very well but is a little flimsy.
0.6mm is a bit too thick and doesn't vibrate as much.
In the end I used the 0.35mm blade.
The rest of the parts were printed with a layer height of 0.3mm and an infill of 30% which gave me plenty of strength without using much plastic.
Step 4: Assembling the Parts
Begin by using a sharp knife to clean up the coil holders, pay particular attention to part B for any imperfections on the inside of the ring. Once both parts have been cleaned up the coil former can be fit snugly onto the second coil piece and glued in position. If you don't want to use glue then I instead used a soldering iron to stitch together the parts as shown in the photo. This gives a very reliable bond but you need to be gentle with the soldering iron and remember to use your knife to clean up the side walls when your done.
Next take a 3mm drill bit and by hand clean up the 3mm holes in the spacers, the arms and the brackets. With all the holes cleaned out you can start to combine the parts together as shown.
Once the glue on the coil holders has set you can clean up the 3mm hole on this too.
Step 5: Winding the Coils
The coils need winding in a specific way to allow everything to work correctly. Begin by winding 80 turns onto your first coil and then use your glue to fix the coil wire in place around the coil former. If using a glue gun then take care not to heat the PLA plastic too much or it will start to deform. The end of wire you start with is end A and the end you finish with when the coil is wound is B.
When winding the coil keep the wire a little taught to ensure you get a good tight coil. Too tight and it will put too much pressure on your join and your coil former will pop open half way through winding. Very annoying when this happens and why I chose the soldering iron stitching technique in the end.
Once you have finished the first coil repeat for the second coil making sure you wind the same way as the first.
Once both coils are wound and the glue has dried then you can attach the plastic bolt and the spacer to the coil. Trap the coil wires underneath the spacer to provide a bit of strain relief. The plastic bolt can also be attached through the arm pieces and secured with a M3 nut.
Connect two of the wires together using soldering iron. The wire should be joined so that the wire A of the first coil should be connected to wire B of the second coil.
The other two wires should be left free to connect to your load.
Step 6: Calibrating
Next we have to calibrate the device. This is basically positioning the arms and the magnets so that everything runs as smooth and efficiently as possible.
Begin with the magnets, they should both be in the same position on either side of the blade and should slide nicely into the receptacle on the coil formers without touching or rubbing the sides.
Once the magnets are positioned correctly tighten the arm joints bolts so that the arms are close to the magnets but allowing a reasonable amount of movement for the blade to oscillate back and forth.
Blow on the blade to check that the oscillation is still smooth and your good to go.
Step 7: Measurements
To take measurements I connected both free ends of my coil wire to a resistor. By monitoring the voltage across the resistor I can determine the generated current and therefore the generated power. I used a 100 Ohm resistor and measured a voltage of +/- 10mV across the coils from a steady gentle breeze created by my desk fan. This implies a current of 0.1mA and a generated power of 1uW.
Current = Voltage / Resistance
Power = Voltage X Current
I guestimate that the power for the 3D printed parts alone was approx (1 hour x 12V x 2A) 24 Watts so it would take approximately 100000 hours to generate enough power to overcome the manufacturing process (at least the 3D printed stage). That's a bit too long in my opinion so at the moment the design is not viable.
I mentioned in the BOM that you could optionally add some steel washers. The washers go over the spacer piece and sit behind the back of the coil in an attempt to force the magnetic flux from the magnet to flow over the coil wire as much as possible. Finish by gluing in position.
I repeated my tests again using a 100 Ohm resistor and the same input wind speed but my measurements were largely the same.
The photo showing lots of noise but larger voltages is a test with the 100 Ohm load resistor removed. The voltages are a lot bigger now because there is no "load" or current flow on the coils.
The small voltages I am getting out of the current blade design don't allow me to rectify the AC signal to DC due to the 0.2V drop required by a Schottky diode. So the next step I want to do is to scale up the design and see I can can capture a bigger slice of the wind energy to provide a voltage I can rectify and allow me to start stacking multiple blade's together.
Bigger blade, bigger magnets, bigger coils.
If anyone wants to have a go at the project then I would very much like to know how you get on and if you have any ideas on how to improve the output voltage then please let me know.