DIY 1000 watt wind turbine

We built a 1000 watt wind turbine to help charge the battery bank that powers our offgrid home. It's a permanent magnet alternator, generating 3 phase ac, rectified to dc, and fed to a charge controller. The magnets spin with the wind, the coils are fixed, so no brushes or slip rings necessary.

Update: DIY Amp Hour meter for monitoring charge!

All about homebrew wind and offgrid power systems, the complete reference!
Remove these adsRemove these ads by Signing Up

Step 1: Build the magnet disks

We had 12" steel disks hydro cut. We cut a template for mounting the magnets. Then we mounted 12 grade n50 magnets around the outside edge. We then built a form, and poured the resin with hardner.

Step 2: Build the coil disk

We wound the nine individual coils, soldered them in a 3 phase wye configuration, and encased them in resin. We used 35 turns of 2 parallel strands of 14 gauge enameled (magnet) wire for 12 volts. Use 70 turns of single strand for 24 volts. # 3 phase diagram shown here shows 3 stator coils. each of those coils is actually 3 coils in series. coils 1,4, and 7 are series together, 2,5, and 8 are series together, and 3,6, and 9 are series together.

more details, see the following page 15 for the series star, 1-y diagram.


Step 3: Build the bearing assembly

Picture of Build the bearing assembly
Two Harley Davidson wheel bearings are inserted into the pipe, with a smaller pipe locked between them to keep them in place.

Step 4: Construct the blades

The blades are 2" x 6" pine, cut at 10 degrees on a table saw, and sanded into a rough airfoil. Not perfect, but close enough.

More can be found at

Step 5: Further Reading

Hi, How did you determine which wire gauge to use?
1000W / 12V = 83.3A
from any of the ampacity tables i could find online 83 Amps would need arround AWG 4 for enclosed wire and AWG 7 for open air application. Am I missing something? Please explain.
sspence (author)  laiskasmariui1 year ago
Ampacity tables do not take into affect length of the wire. Each coil , plus each phase carries 1/3 of the current. 35 turns of 2-in-hand 14awg equals 11awg. Each turn is about 6" of wire, so it's about 17' of wire. This equates to a 8% voltage drop at maximum power. If you want to drop to 3%, you could try dual 9awg, but that would be hard to wrap.
Thanks for the response :) I didn't realize that the three phases share the current.
Sorry in advance if these are silly questions!

You say here that doubling the number of turns of wire doubles to voltage output, right? Do you know what effect more magnets would have? Would doubling the number of magnets (and the number of coils), but keeping it at 35 turns double the number of watts output without doubling the volts?

Also, do you know how many rpm you'd need to get out of this thing for it to actually reach 1000W?
sspence (author)  katherine13311 year ago
the number of magnets has to do with the number of coils. It's a 9 coil to 12 magnet ratio. 650 rpm is the designed rpm. as you add more load, it takes more wind to maintain rpm.
Mrdman24422 years ago
By wiring the alternator in star you sacrifice some current output. My thought would be wire each phase separately to a bridge rectifier then combine the DC output from the 3 rectifiers in parallel. Is there something wrong with this idea?
I am building a vertical turbine with a savonius (starter) inner rotor and a larger H type main rotor on the same shaft. With this setup in low wind, below startup, the turbine will still be turning and by connecting the rectifiers in series I could get usefull voltage at low current. When the winds pick up the rectifiers will be switched back to parallel so the main rotor can kick in.
Is this possible, or is there some reason 3 phase is only wired in star or delta?
sspence (author)  Mrdman24421 year ago
each phase is being rectified separately and then paralleled. you can switch from star to delta with some additional effort (electronic or mechanical speed monitor) but the benefits are outweighed by the additional complexity in my opinion.
kalo022 years ago
Hi there, i'm not sure if this has already been asked but I was wondering if there was a reason why you chose to do it in three phase AC and then rectify it instead of generating it as DC in the first place? Is it just so there's no need for a commutator?
how would you get a DC voltage form an alternator if you don't rectify it?
sspence (author)  kilofeenix1 year ago
We do rectify it. See the picture above with the 6 diodes?
sspence (author)  kalo022 years ago
That's one reason. With no brushes or slip rings, there's less maintenance.
Looks very professional, but achieve AC voltage before the rectifier is how much? and the treatment you how the absence of wind (low voltage). Please reply to the email
sspence (author)  lecaonguyen1 year ago
It's not professional, it's a backyard hack ;-)

The ac voltage is determined by the number of turns in the coils, and the wind speed vs. load. If the voltage is too low, the batteries do not charge.
Mrdman24422 years ago
Do you have an idea of how feet of magnet wire you used? Thinking about useing aluminum wire as its cheaper but I need an idea of how many feet to buy as the weights are different.
sspence (author)  Mrdman24422 years ago
aluminum doesn't handle current as well, so you'll need to derate accordingly. I used about 200' of wire, but that's dependent upon voltage and current needs.
mcsk8rg2 years ago
why do you need to rectify teh current comming from the coils ??? you can just connect all the +ve n -ve terminals of the coils n you are done !

you can get current directly from ALL the coils all at once !
sspence (author)  mcsk8rg2 years ago
There are no + and - terminals on the coils. It's 3 phase AC. Diode bridge is necessary to convert to dc.
ironman12342 years ago
if i were to do this could i use 19 gauge wire? would i use the same amount of coils, turns, and magnets? and would i use the same n50 magnet or would i use a different size?
sspence (author)  ironman12342 years ago
with lighter wire, there is an opportunity to over heat the coils and cause a fire if you pull too much current. Off the top of my head, you'll probably be limited to 30 amps or so.
valiox2 years ago
Because I don't have enough money to buy big Neodymium magnets,I have a big question> What if I am using N50 type, but wich will be 9mm in diameter, and 5 mm height, and I will reduce the number of turns for every coil, so in that case I will increase the number of the magnets, and the number of the coils?

sspence (author)  valiox2 years ago
With that many changes, all I can say is have fun experimenting.
bchurchman2 years ago
Gotta love a fellow tinkerer! Was looking at your other projects. What's a good weekend project without the Fire Department involved? LOL
sspence (author)  bchurchman2 years ago
The only time I had to call the fire department was when I was burning grass, and it got away from me. Embarrassing, as I was a member .....
danlong2 years ago
One question.

What is the separation between the coil and the magnet face? I didn't see a stat on the field spread. I did note that the magnet has a 175 F max temp before damage and I have visions of a 100F plus day with the turbine cranking. How do the magnets stay "cooler" along with the coils themselves during power production?
yummy692 years ago
hello, how are you all guys ?
Can i make thins thing with hard disk megnets because im unable to buy this shape and neodymium magnet. from my place. so i think hard disk magnets are neodymium magnets and they can be use for this procedure ? Can it be ?
i have 200 pieces so i can use them together/
sweenig yummy692 years ago
I don't think you can use hard drive magnets. These magnets are polarized one way on one end and the other on the other end. So, the flux lines would go in the upper face on one side, travel through the structural steel (and the magnet itself) and exit the upper face on the other side. Hard drive magnets are not typical magnets. I suppose, if you were able to arrange them properly, you would have alternating poles, but that seems really difficult.
jmart902 years ago
i definately want to built 12 of these, each with 3 blades 8 feet in length! can anyone give me tips for building these cheaply?? (under $300 american??)
sspence (author)  jmart902 years ago
It's more like $500 plus, with all new materials, not counting the tower. Scrounging wire and magnets, you could do it for less than $100.
ashbranmeg3 years ago
Did you alter the polarity of your magnets? If not, wouldnt the output of your generator be DC with no need for a rectifier? If there is no alternating of the field, there will be no reversing of the flow? That means DC not AC?
sspence (author)  ashbranmeg3 years ago
 Yes, the magnets alternate polarity. Trying to make a dc gen directly is an interesting concept. let me know how it works.
You are a very patient man, and you seem to have been put in good company...T.A. Edison, one of the countries, or maybe the worlds greatest inventors. I , and probably many other readers, appreciate your hard work, and you have impressive credentials. Thank You.
sspence (author)  jimramos2 years ago
Thank you, but I'm just standing on the shoulders of others.
dpaiva3 years ago
Hi everyone. Please, I need some advise. I have these small 20mm rounded magnets (100pcs). They seam very strong. The thing is that they are so small, they get to far from one to other. Needless to say, I did not get any good voltage.

I am planning to use only 6 coils with about 300 turns and rectifying every single coil and then put then in series or parallel depending on the wind. 6 Magnet pieces together X 8 around each disk.

Does it make sense? Is there any problem rectifying each coil?

Also, I used some resin they use on cars that seams to interact with the magnet. I will find where to get this epox instead.
Thanks in advance
sspence (author)  dpaiva3 years ago
So, you are making a 6 phase machine, that would be 12 diodes, two per phase. you can't series alternate phases, but you could series two coils and make a 3 phase (1,2,1,2,1,2) then you would only need 6 diodes. resin is not magnetis. what do you mean "interact"?
dpaiva sspence3 years ago
Hi, thanks for the quick reply.

- Interact meaning it pulls or push the magnet, acting like one. I will look for other resin kind. When I say putting them into series, it would be after rectifying them (2 diodes per coil + capacitor).

- My concern on making it into phase is that since I have different magnets sizes I would not know how to perfectly mach the disposition of the coils X magnets. I did see on your pdf the ratio would be 1/1.333 though, but I just wanted to keep it simple to avoid more frustration. :)

Extra info: The magnets I got are 20mm diameter x 12mm height. 8 per disk. (Total 16 magnets blocks)
Each coil: 300 turns x 22awg = 40mm diameter. (Total 6 coils)

Hey, I really appreciate all your attention and help.
sspence (author)  dpaiva3 years ago
fiberglass resin is non magnetic. capacitor?
schel3 years ago
All I gotta say is AWESOME!! well done!

the applications of this design are wide open!
2 of these -sans the blades...could easily gen a bunch -o-power on say,...a stationary bike...or heck a small lawnmower engine could spin a row of these ...make for Da' BomB camping gen

Thanks for sharin' yer info bro!
can you help me how can i built 10 mw wind power plant. you have any idea. plese help me i need you help.
sspence (author)  hemang charaniya3 years ago
mw, as in milliwatt? or MW, as in megawatt? No, I only build units in the 500 to 1000 watt range.
menahunie3 years ago
Interesting post; but is it just to show everyone what you did? or show them how to duplicate what you did?
I have to go with the latter since you left out allot of build information.
The cost from doing this "project" cost vs. power output is far more than what I did.
I went to a junkyard found a 260 amp alternator ( $100)off an ambulance and mounted and attached a wind prop to it. I pulled the voltage regulator and used a Sevcon unit for control. I used #4 welding cable with soldiered terminals to take the alt. output directly from the rectifiers. For charging the batteries using a HF ( high Freq. battery charger is best ). There is more electronics for the out put to the house etc..
You can also use the motor from a treadmill; you see them all the time on someone's curb they are throwing out...
I included some math about windturbines:
Power AVAILABLE in the wind = .5 x air density x swept area x (wind velocity cubed)

Example: air density = 1.23 kg per cubic meter at sea level. Swept area = pi x r squared. Our 2 foot blades = 0.609m, 4 ft = 1.219m. 10 mph = 4.4704 m/s, 20 mph = 8.9408 m/s.

How much power is in the wind: 2 ft blade, 10 mph winds = .5 x 1.23 x 3.14 x 0.609squared x 4.4704 cubed

= .5 x 1.23 x 1.159 x 89.338 = 63.7 watts

With 4 foot blades and 10 mph winds = .5 x 1.23 x 4.666 x 89.338 = 256 watts

With 4 foot blades and 20 mph winds = .5 x 1.23 x 4.666 x 714.708 = 2051 watts

That's the MAXIMUM power in the wind. However, it's impossible to harvest ALL the power. The Betz Limit tells us that the maximum percentage of power we can harvest from the wind is 59.26%.

Thus our maximum power from these turbines would be:

2 ft blades, 10 mph wind = 37.7 watts

4 ft blades, 10 mph wind = 152 watts

4 ft blades, 20 mph wind = 1,215 watts

In parting even the motor from a direct drive washing machine can and is used for a wind generator; the design is very similar to what you show everyone in your posting. All I am trying to comment about is allot of people may not have the time or ability to duplicated what you did.

Get More Out of Instructables

Already have an Account?


PDF Downloads
As a Pro member, you will gain access to download any Instructable in the PDF format. You also have the ability to customize your PDF download.

Upgrade to Pro today!