Now that the design is complete, let’s build this thing!
I printed all the parts on my 3D printer, and ordered 4 mm rods from Amazon. The rods are really hard to cut, so I had to use a Dremel. It’s good they’re very strong, but it makes manufacturing a pain in the butt. Anyway, here’s the finished product:
As you can see, the handle, pillow blocks, gears and motor are all mounted onto the wooden platform. Let’s turn the handle and see what comes out of the generator!
The oscilloscope shows three phase power! I don’t know exactly how fast I turned the handle, but it was around 1 or 2 revolutions per second. What I do know is that it was a comfortable pace; not too fast or too slow. Sinusoids have a peak to peak voltage of about 10 volts. Definitely usable!
Next, let’s hook up the three phase power to a hex bridge and bulk capacitor:
When I turn the handle at a comfortable pace, I get about 11 volts. What’s cool is that the voltage across the capacitor depends on the maximum speed I turn the handle; if I turn it slowly, the voltage is pretty small regardless of how long I turn the handle for. Since there’s no load, if I turn the handle quickly for even a small amount of time, the voltage jumps up and remains large. The voltage will eventually drop due to parasitic resistance across the capacitor.
Now, let’s put a load across the pin headers (and across the capacitor) and see what the waveform looks like. I found a 100Ω resistor laying around.
Let’s break down the waveform. First, the voltage across the resistor jumps up to about 8 volts when I start turning the handle. Then, since the speed I turn the handle isn’t consistent, the voltage jumps around between 7.44 volts and 9.68 volts. When I stop turning the handle, the capacitor discharges through the load.
Let’s say the average voltage across the load is 8.5 volts. Then, using P = V^2/R, the resistor is dissipating almost 3/4 of a watt. Sure enough, when you touch the resistor, it feels warm.
In order to regulate the output voltage better, since the last test showed the power is all over the place, let’s put a buck-boost regulator on the output:
Now, when I turn the handle, the output of the buck-boost regulator is unstable for a second, then it snaps to 5 volts. I used my variable load to see how much power the regulator can output, and it was around 2 watts. The power output is throttled by the mechanical component of the system. When I try to output more than that, I can feel the plastic crank and gears struggle under the load. My biggest concern is the plastic that mates the crank to the shaft of the first gear will strip, so I’ll try rebuilding the system with that in mind. But not bad for a proof of concept!
EE Diary 