Designing, and Building My Own Wind Turbine
Field Notes from Irricana, Alberta, Canada
by Steven Fahey

24 September
"Lights ON"
24 October
14 November
"Battery Test"
15 January
15 February
"Solar Panels"
5 June
"Water Pump"
31 December


Finally in the air! What a relief to have the windmill back together and running "on schedule". This wind turbine replaces the one destroyed in an accident in April. I'd already collected this 3-phase motor for another conversion, but it was supposed to be in addition to the old GE motor conversion, my first wind turbine. With the destruction of the GE, I had to get started on this motor conversion as a replacement .

Here's my data-logger screen (PICLOG: showing a 20kph wind giving me about 200 watts. The wind picked up Saturday afternoon and I had it logging for 2 hours when I took the photo. For the two hours you can seen 450 W-hr collected. Not bad but it could have been more. The tail was too light, and furling in the 30 kph gusts! I lowered the tower again on Sunday, added weight, and sent it back up. That was also an opportunity to add protective tape to the leading edges of the blades (oops forgot).

The conversion process required scavenging the magnets from the GE. There is enough room on the rotor of this one to use the same number of mags. That offered the promise that this conversion would produce roughly the same amount of power as the GE, too. Some simple bench tests showed that this was likely to be so.

The blades were damaged in the crash, too, and I decided to make new ones using the first as a template instead of trying to start a new design all over again. After laminating together a set of boards, I carved two down using the old third as a guide. There was one drawback to doing this. The old blades ended up weighing less than the 2 new ones, due to its age and dryness. Balancing the blades became a tricky question because I would either need a pound of steel counterbalance or have to "tweak" the heavy ones. In the end I discovered a way to adjust the heavy blades and just get them into balance. I did so by weighing each blade and finding its center of gravity. For the blades with more weight near the tip, I shaved them just a little at the tip to remove that weight. It took a lot of repeated shave-check-shave again to get it right. Now I have a well balanced set of blades with no counterweight to cause dynamic imbalance. Now that it's up and turning, I can see the benefit already. There is less vibration to feel when I hold on to the tower.

Fortunately, much of the tower and parts from the old wind turbine mount survived the accident. I didn't have to build a new generator mount or tail arm. To bring the tower back into service, I installed a winch for raising and lowering it. Using the winch makes the process of raising slower, but it affords more opportunity to stop, check on things, and to notice tangled cables. This happened during one raising with the winch already, so it's already proven its worth. The winch itself is a worm-gear winch, and it holds the tower in any position with no little pawls or free-spooling levers to get bumped. Instead of hand-cranking the winch, I have an eye-bolt that I chuck into the power drill, and use the drill to turn the winch much faster than I can by hand. One surprise with this worm-gear winch is how hot the gears get when I run it for 2 minutes! The drill isn't hot at all, but the friction makes the worm gear so hot I see smoke! I stopped 3 times while cranking the tower up, just to re-grease the winch.

Some additional notes, after reading comments about my diary entry:

October 25, 2009:

In the process of re-carving I noticed that the root was given an angle that would correspond to a TSR less than 4, while the tips are about 5. That's an oops on my part a few years ago when I made the first set of blades, but they actually work well on these motor conversions I make. There may be a little more "oomph" to get started turning with so much twist.

It's always spinning, even in the lightest breeze. You can feel the cogs when you turn it by hand, and the blades visibly "step-step-step" to start and stop, but that's no inhibition to turning in any wind that would produce power.

The winch is rated for 2000 pounds, and the manufacturer refers to a safety factor of 4, though I think that's a bit optimistic considering its size. It's the biggest worm-gear winch in the Acklands catalog. In my application, I'm putting about 1200 pounds on it just at the start of raising, so my operating safety margin is higher, but not large. No, I'm not completely fooling myself =)

I had to lower the tower to adjust the tail angle, then after Saturday's wind showed me that the tail was too light, I lowered the tower again to put weights on it. So in one weekend I raised the tower three times. The winch performed just fine, but one thing I did notice is how hot the worm gear gets. The grease would get very liquid at these temperatures and I stopped cranking half-way to re-grease. Not sure if that's a sign of doom or perfectly normal. Considering a bigger winch anyway.

October 26, 2009:

A storm front came in at about 10 PM, so I can report that all's well, with about 60kph up there (35mph or so). The tail is just about to furl in these gusts - a little fast but there's little I can do about it now.

I think the stator reactance is already limiting the output below 15 amps. I can't be sure until I look over the PICLOG data tomorrow, but the analog gauge isn't going much higher than 12A.

For the sake of knowing what happens, I hit the "kill" switch. Not a lot happened except it made less noise, indicating that it was running a little slower. This isn't a big surprise, because the stator resistance across phases is 3.5 ohms. Not much of a "short", is it? The GE generator, running in parallel-star had a stator phase resistance of about 1.2 ohm, so there was much more authority at the kill switch because the shorted current was proportionately higher.

Well, unless I resorted to re-winding the stator wires, I couldn't expect much better. This project wouldn't be finished this year if I'd tried doing that. My one remaining theory about motor conversions is the construction of the rotor and trying to re-use these squirrel cage rotors by turning them down on a lathe. Though the magnets stick frimly I'm thinking that the flux circuit isn't the best. Re-making the rotor, with thick cylinders of iron to plunk the magnets onto, is the way to go.

I have little to complain about because this old motor was (like the others) free, the magnets were scavenged from the old crashed genny, and the shop time to re-work the rotor was also free. It's running well in both light and moderate winds, which is where 90% of my wind is anyway.