Windmill Generator - why won't a small brush-less motor work as a windmill generator?

I bought a small brush-less motor that was designed as a motor for a small RC plane. I tested it by spinning it with a drill. I was surprised to find that the motor did not generate much in the way of voltage. Why not?

It's too bad - many small brush-less motors are so nice. No brushes, ball bearings, and low cost. I would have thought that forcing a brush-less motor to spin would have generated power.

Thanks for any insight.

Picture of Windmill Generator - why won't a small brush-less motor work as a windmill generator?
Arx2 years ago
I know this is an old thread, but in case anyone is interested, the problem is simply RPM
KV is the number of RPM the motor develops per volt, and conversely indicates the number of RPM you need to spin the motor at to produce a volt.

If you're wanting to use a small BLDC motor as a generator, you'll need to find a way to spin it really fast, or alternatively find a much lower KV motor. Some very large RC motors can be found in the 100-200 KV range, and would be much more appropriate for use as a wind generator. (they're quite a bit larger and more expensive though)

They're going to be much larger, stronger motors though, so you'll probably need to run a bigger prop to get over the cogging. The upside is that a bigger prop will be collecting a lot more energy.

jj.inc3 years ago
Well, brushless doesn't mean a motor without brushes to transfer electricity to a inner drive. To me it seams that it uses triple phase AC. This means that it has 3 sets of coils that run with electricity in high speed switching directions. You may be able use a bridge rectifier to make it into usable DC.

A brushed motor has a rotor inside with a magnet ring on the outside, this allows DC current to be made because of the way the magnets rotate relative to the coil. In a brushed motor, coils are wound and mounted parallel to the rotor, this means that the coil is stimulated whenever passed, in a brushed motor the coil is wrapped like lights on a tree, around the center. This allows a constant flow by pulling energy around and around. It is like pulling a magnet around a table with one underneath. If you move the bottom one left to right and the top one is only allowed to go up and down it will be stimulated a different direction with every pass, if the top magnet is allowed to go left to right then it will fallow the bottom in one long stream.
valmic jj.inc3 years ago
I am doing a science fair project that involves a solar panel, connected to a charge controller, connected to a 12V battery, connected to a switch system that can either go to a flywheel with a 12V DC motor or straight to a DC to AC inverter which leads to light standard light bulbs. The system worked (light bulbs lit) when I connected the panel to the controller to the battery to the inverter. However, when I had the battery power the flywheel and then have the flywheel output its energy to the inverter, the inverter shut down and would not work (red alarm went off). The flywheel is producing more than 10V (starts at about 11V) for at least 30 seconds, and the inverter's low voltage shut down in 10.5 plus or minus 5 volts. So why doesn't it work? Is the inverter messed up or is the flywheel outputing AC power? Also, when I connected the solar panel to the charge controller to the flywheel, the charge controller completely shut off. Are charge controllers not able to be directly linked to motors? And if I tried to directly connect the solar panel (which has 20V) to the flywheel (with a 12V motor), without a charge controller in between, what would happen? valmic3 years ago
I think you may want to try your original experiment with the lights again to see if your inverter is messed up, also your inverter may have a higher wattage than your motor can produce, this may send it into the same thing low voltage would. You may need a bigger DC motor. If your solar panels are 12V DC and your motor is to I don't see any need for a charge controller, I don't think you need one in the first place, most people use them only for dealing with a battery bank.
jimk3038 (author)  jj.inc3 years ago
I've never had any good luck with power inverters. I recently tried to run a 1/4 hoarse motor off a 1000 watt inverter. Should work in theory - in the end it just let the smoke out of the inverter.

My guess is that the inverter draws a surge from the batteries at start up. A solar panel would not be able to supply that needed surge.

In terms of the motor / solar panel - you gotta match the solar panel to the DC motor. A 15 watt panel is not gonna start a 1/4 hoarse DC motor. The solar panel will have a sticker on the back that specs. the amount of current. Try running your motor on a variable power supply at that same current. If the motor does not run then it sure won't run on the solar panel either.

Same goes for the fly-wheel. Gotta match the fly wheel output to the load. But again, the fly-wheel is not able to supply the surge the inverter needs.

Hope this helps,
overkillbbb3 years ago

get some ideas here, easy stuff

making my own out of neodenium magnets, no power needed to spin magnetic motor, but i have to buy 216 1"X1/2"X1" dowel magnets to embed them into acrylic rotors,
if you use the brushless motors you will have to add a lead and this involves separating the armature and adding another post then having a control or trigger, i could use a stepper motor using a small trigger device to keep it going thus allowing power generation with magnets over a coil, kinda like the Bendini Motor circuit, wish me luck cause tomorrow im going to turn the plates down
jimk3038 (author)  overkillbbb3 years ago
Good Luck!

Post some pictures when you get something working.

How many R.P.Ms does it give under a certain voltage/current? It should generate quite a bit less than that when spun at the same speed because these are not designed to be used as generators, and are much more efficient being used as a motor. Not that it won't work. I recommend going to some auto repair shop and asking for an alternator, I've gotten a couple for free and some for 10-15 bucks. Just wire it in with a diode and windmill away, make sure to use gears on the windmill to help it spin faster.
Bugilt3 years ago
Maybe this thing will do the trick.
Use a stepper motor from a printer or scanner they have six wires but can be pinned out with a LED. After you have found the two wires you can attatch them to the ~~ side of a diode bridge and hook what ever you want to power to the + and - sides and then use that as a windmill I know this because I have tried and this has worked. For blades use a desk fan blades attatched to the shaft of the stepper. Hope this helps :-)
Try a network of diodes like this.
brushless to rectifier.JPG
jimk3038 (author)  steveastrouk4 years ago
I don't think the diodes are going to work. And here is why - I just spun the motor again while connecting my voltmeter between one phase. On the AC scale, I only measured about 0.25 volts at 1000+ RPM. Not enough voltage to be able to use diodes.

I also tried using the current scale on my voltmeter. Again, on the AC scale, I measured 200mA of current. So, with a dead short across one phase, I can get 200mA.

So the results are, open circuit, one phase produces 250mV @ 0mA. With short circuit, we get 200mA @ 0V.

I think maybe the problem is, the motor is designed to produce torque. The magnets are not that powerful. The coils produce the vast majority of the magnetism. The magnets just get pulled around by the coils. By back driving the motor, the weak magnets don't induce that much power into the coils.

I'm guessing, to make the motor good as a generator, the weak magnets would need to be replaced with very strong magnets. Also, I worry the coils don't have enough windings. Again, the coils are designed to produce magnetism. Instead, the coils need to be designed to accept a rotating magnet field and induce a current.

I'm just guessing on most of these points. What I can tell you is, back driving a brush-less toy motor does not produce much usable power.
No. That's a 350 W motor, it has a motor constant of 3500 KV, and its designed operating voltage is from 4 to 14.8 V.

That means that the thing can run between 14000 and 49,000 RPM, unloaded. Its not TERRIBLY surprising you can't get more than 0.25 volts off it at 1000RPM.

You really need to read up on motor operation ! A motor is a generator. Its even a generator whilst its being a motor. You DON'T need to change anything, you just need to run it the way it needs to be run.

jimk3038 (author)  steveastrouk4 years ago
OK, OK, I get it.

My question, after all, "Can one of these little brush-less motors be used as a generator in a windmill application?" The answer appears to be no.

Also, have a look at this:

Now, this guy has done some homework. It's gonna take me some time to digest this.

JIM: After reading some of the remarks and the replies, I see the basic problem that is not addressed properly. The Motor designed to operate at extreme RPM for small airplanes in the areas of 15,000 to 20,000 RPM with 4 to 12 volts or so -- with a constant of 3500 RPM/Volt is indicating that the motor to operate as a generator has to be spinned above 8000 RPM to produce enough voltage to light an LED, this is 1.8 to 3.8 votls depending on the LED type from RED to Ultraviolet ( white).
A quick glance and that instructable looks great.  Lots of good info there.  For any motor to be an efficient generator, it should spin at or near it's operating speed.
See the difference ? Its all to do with operating RPM. The hub motor is NOT designed to spin at many K rpm !

bertzie4 years ago
Spin it faster!
nepheron4 years ago
 You can make a generator out of any motor. See my instructible "mini dynamo". It shows you how to turn a motor from a walkman and an LED into a quick hand crank light.
jimk3038 (author)  nepheron4 years ago
Yep, I've done that before too (connect a LED to a toy brushed motor). I recently showed a bunch of Cub Scouts and they were all amazed. Being boys, they all had to compete and try and see who could get the LED the brightest.

However, what I was shooting for was a brushless motor. I loved the idea that it would last a really long time since no brushes were used. Plus, the brushless motor seem to be better made (mine uses ball bearings).

But sadly, the toy brushless motors made for remote control (RC) stuff are just all wrong. They are small in diameter, have delta windings, and have cogging problem. Just no good for a small windmill application. Note, there are larger brushless motors available for RC stuff. But the price goes way up.

My next idea is going to use a power steering motor. Delphi make an electric power steering system that uses a large brushless motor. The Delphi motor is 3 1/2 inches in diameter and four inches tall. There are millions of these motors out there in junk yards! A bit large for my tastes, but since I already have one I guess I'll make do.

Thanks for the comment,
scratchr4 years ago
Brush-less DC motors are like a stepper motor but have internal electronics.
You could tap and rectify each coil.
Not sure if i have already posted on this question if not here goes, I have a 24 volt DC brushless motor that will produce voltage but It has to be spun at a very high RPM inorder to do it when spun at approximately 30,000 rpms as with a dremel tool on high it will produce approximately 96 volts which is enough to make a  110 volt AC light bulb light up fairly good!
I'm pretty sure brushless motors don't work period. It has to have a brush for it to have the same configuration as a generator. Brushless motors work differently than othe motors.
 LOL :D   sry no offence but the "I'm pretty sure brushless motors don't work period." sounded funny to a guy skimming through the ible site. :)
jimk3038 (author)  imshanedulong4 years ago
Have a look a this guy's instructable:

I learned a bunch from reading about his project.

One of the neat ideas from reading his work, a motor is really a transducer. It converts between torque and electricity. The transducer works in both directions too. This is what makes regenerative braking work on hybrids. In other words, the hybrid uses the motor to push the car forward. But, when braking, the motor operates as a generator and converts torque back into electricity that is pumped back into the batteries.

In terms of using a brushless motor as a good generator, the jury is still out. My toy BLDC (brushless DC) motor was not wound well for use as a generator. The windings were wound in a delta pattern using five parallel strands for a total of six turns on each tooth.

I've recently ripped all the windings off my motor. My plan was to reuse the copper to rewind the motor. However, I kinda messed up the strands by pulling them off the stator. So, I've ordered new (multi colored) magnet wire. I'm expecting the new wire any day now.

My plan is to rewind the motor in a star (Wye) pattern using a bunch of turns. I expect to get at least 30+ turns on each tooth. This should make a big difference in the output of the motor / generator. Once rewound I'll back-drive the motor again and see what we get.


jimk3038 (author) 4 years ago
Well, I finally mustered the never and cut into my motor windings. I removed all the old copper windings and replaced with brand new colored magnet wire from The new magnet wire cost me $10 bucks, by the way. However, I now have a ton of magnet wire (120' of red, green, and gold). 

The old windings were wound in a delta pattern using multi-strand bundle of 5 wires with 6 turns on each tooth. So really, there were 30 turns of wire on each tooth.

The new windings were wound as a star, or wye, pattern using 25 turns on each tooth. 

The old winds produced 0.25 volts rms across one phase, open circuit. The new windings produced 2.0 volts rms, open circuit. Both results were achieved using a drill motor to spin the motor at 3600 rpm.

So, going from 6 turns delta to 25 turns wye gave an eight fold increase in voltage, at the same rpm.

With more practice, I think I could get all 30 turns back onto each tooth. That might boost the voltage to maybe 2.5 volts rms.

So finally, to answer my own question, the reason small brushless motors are poor windmill generators is because they are too small. Even rewinding the motor into a wye pattern and using the maximum number of turns the motor still needs to be spun at over 3000 rpm to light an LED. And gearing is not so good because the motor already has a lot of cogging. Startup torque would be awful if there was a 5:1 ratio in front of the motor. 

Anyway, it's been fun. Nothing like rewinding your own motor to make you fell like a true EE!

Thanks for all the comments,

knex_mepalm4 years ago
 What doy uo mean by  brushless motor?
jimk3038 (author)  knex_mepalm4 years ago
Do a google search on "BLDC motor". The first link to Wikipedia is good. 

The best part of BLDC is, of coarse, there are no brushes. The down side is, there are no brushes. So, kinda good and kinda bad. 

With no brushes the motor will last a very long time. There is only one moving part. So, as long as the bearings are good the motor will work. With no brushes, there is no friction either.

However, without brushes, the motor must be driven with a microprocessor. The micro switches the windings on, and off, to pull the magnets around. By controlling the switching rate the speed of the motor can be controlled. Many folks call the micro control box an ESC (electronic speed control).

It's easier to deal with the BLDC motor as a generator. The motor, when forced to turn, outputs a nice three phase waveform. You would need a scope with at least three channels to see this waveform. Without a scope, just use good-old google on "three phase waveform". Again, Wikipedia is the first link and they have some good info.


Your brushless DC motor works in a way more similar to a synchronous three-phase AC motor than like a single-phase brushed DC motor.  It is still fed by a DC power source, but notice that in order to operate it requires a different type of ESC. The waveform of the brushless ESC is very similar to a three-phase AC, but with a lot of harmonics (more square wave, sorry if I am getting too technical).  When used in generator mode, expect the output to be similar to a three-phase synchronous generator. There are many ways to play with the output voltage of the generator. The easiest way is to connect the windings in Wye rather than in Delta.
jimk3038 (author)  MidlifeCrisis4 years ago
My hope was to simply use a three phase rectifier (six diodes) to generate a rough DC voltage. The three phase rectifier suffers because of the diode drop inherit in the device. I think there is two diode drops involved. So the generator output looses a volt (assuming each diode drop is about 1/2 volt) just getting over the diode drops. Still, it's dead simple.

I had not thought about Delta / Wye. I went to wikipedia and tried to understand Delta / Wye. Here is the link:

I also found some stuff on this site:;col1

An answer did not jump right out. However, sounds like a Wye connected motors will generate more voltage for the same amount of turns. Correct me if I'm wrong.

Thanks again - good comments,
VadimS jimk30384 years ago

A much better page.
Yes star (or wye) will give you higher voltage then Delta for the same number of windings.
jimk3038 (author)  VadimS4 years ago
OK, you convinced me - I'm going to rewind my motor. I only bought the motor with the intent of using it as a generator anyway. So, if I mess it up there is no loss.

The existing magnet wire should be OK to reuse. I'll just rewind it using a single strand instead of the 5 / 6 stands in the original.

I'll play with both Wye & Delta just to see the difference. With all the wires hanging out of the motor it should be easy to make some measurements both ways.

Thanks for the encouragement,
jimk3038 (author) 4 years ago
Just a couple of updates. My goal was to find a little toy motor I could potential build into a toy windmill. I thought it would be cool to have some glowing LEDs when the wind blows. Using a brush-less motor is a big plus if the windmill is going to operate for years and years. This motor is also made with ball bearings which is another plus. Last plus, in theory, the windings are water proof too.

The actual motor I'm playing with is a Alpha 400 from Exceed RC. The motor has a Kv of 3500 - which (roughly speaking) means 3500 rpm/V. As you can see in the attached pictures, I connected one phase of the motor to my Fluke 189 voltmeter. The Fluke can display both AC voltage and frequency at the same time.

Spinning the motor at full speed the Fluke displayed 60Hz at 0.25 volts. So, 0.25 volts rms * 1.414 * 2 = 0.707 volts peak-to-peak. But the motor constant would suggest voltage should equal (60 rps * 60 sec/min) / 3500 which equals 1.0 volt.

So, measured (no load) voltage equals 0.707 volts and theoretical equals 1.0 volts. I wonder if I'm missing something?

The Alpha 400 can be purchased in three different configurations. In other words, with three different motor constants. The three different motor constants available are 3500, 920, and 740. If I had purchased the 740 I would have gotten 3600rpm / 740 which is 4.9 volts (theoretical). This is much better for my application - higher voltage / lower current is a good thing when your trying to bring a LED to life.

I believe the only difference between the different motors is the way the windings are wound on the stator. As you can see in the attached picture, my 3500 uses parallel strands (5 or 6 strands grouped together) around each pole.

If I rewound the motor with a single strand around each pole I believe that would greatly lower the motor constant - which is my goal. I just haven't mustered the nerve to cut into my motor yet! I wonder how many parallel strands the 740 uses?

For the math inspired, the master BLDC equation I'm basing my assumptions on is:  "T = 4 * N * B * L * R * i"   So, for a given torque input from the windmill, the current will be i = T / (4 * N * B * L * R)". Bottom line, more turns gives me less current (more voltage) with a given torque. Note, the turns on the stator is the only knob I can control in this equation given an existing motor.

FYI, the other motor (Delphi) in the picture below was used in a automotive power steering application. I worked for Motorola who built the amplifier for the motor. I got involved in building some end-of-line testers for the amplifier.

Anyway, spinning the Delphi motor I got 4.2 volts rms at 3600rpm. It's just such a large beast - I smoked my voltmeter's 3 amp fuse by fooling with this dumb thing! However, I was able to get a LED to blink just by hand spinning the motor. Plus, the motor has angled magnets - there is almost no cogging which is important in a power steering application. Low cogging is also important when a windmill is just starting up. This motor would probably be good if making a large windmill. I remember the amplifier could dump 100 amps into that motor for short durations! It's just not the motor I'm looking for.

Thanks for all the comments,
VadimS jimk30384 years ago
The reason you're getting .707v instead of 1 is because of inefficiency. In this case your motor has a kv of 3500 when used as a motor but when run as a generator it is approximately 4950.

Re-winding is an option but I would recommend just getting a cheep motor out of china.
380kv 40$ and about 10$ for shipping.
jimk3038 (author)  VadimS4 years ago
How did you get to 4950? Looks 3500 * square root of 2. Is this a rule of thumb? So, the motor in your link has a motor constant of 380 (as a motor). In generator mode I can expect 380 * sqrt(2), or 537. Is this correct?

The motor spec says "6 turn" as one of the line-items. Does this mean the motor also uses multi-strands around each pole? If so, I'm curious how dramatic rewinding the motor would have on the motor constant.

I know there is trade off if the windings were rewound. No way the motor could handle high current. But, as a generator, it would be better to have higher voltages to work with. As an example, if the motor output was applied to a full-wave bridge diode block, the diodes would eat half a volt with their diode drop.

Thanks for the good comments,
VadimS jimk30384 years ago
Please ignore the numbers I gave you, my reasoning at the time was wrong.
What you want is a motor with a high kv, hence high number of turns.
Most low voltage rc motors use malty strand because it fits better on the core and reduces skin losses. AC has a tendency to use only the outer portion of a wire, this effect is increased at higher frequencies (higher rpm) and larger wires.
I recommend trying to rewire it with a single strand of wire something like 18 gauge and giving it a try.
But these motors just aren't suited for use as a slow speed generator.
onefreewalk4 years ago
Look, if you want a decent generator, you need to up the RPM, obviously there are many different ways to to that, but simple gearing would be the fastest solution. ie, find a prop, attach it with a larger sprocket, or even set up a system of sprockets, and get it spinning that motor at a much higher speed.
There are lots of places online to find those gears, or gear-boxes for that matter.

Then on the motor side, put a real small pinion gear and light up a bulb, or charge a battery!

Just make sure you don't give up till you get what you want!

jimk3038 (author)  onefreewalk4 years ago
Yep, gearing would work. However, I worry that gearing such a large jump would create a lot of reflected torque. Note, I'm not a gear-head so I don't know what to call it. But I do know that any resistant in the motor would get amplified by the gearing. My little brush-less motor has significant cogging. With gearing, that start-up cogging would become a real problem.

I think I'm going to bit-the-bullet and rewind my little motor. It would be really neat to see what a single strand wound on each pole would produce. I'm only at risk of trashing a $25 buck motor. Worth the risk.

Karletto5554 years ago
wow 25 answers
I have done this with a 24 volt brush motor that came out of a motorized skateboard and the output at the highest rpms i could achieve direct drive was around 98 volts which will power a house light fairly good! The more it takes to power the motor the more volts you can achieve!
sorry spelling was wrong that was suppose to read limited too.
pwer input is lower for your motor to run ,their for when your trying to create pwer your output is limit to the motors spec.  .
Acepilot424 years ago
brushless motors dont have brushes (obviously). therefore, there are no brushes to toutch metal, creating that electrical current.
Read up on brushless motors for your homework. ;-)
i tried
Concentrate on BRUSHED motors first to really understand what's going on, and learn about the basic motor equation V=IR + back emf

i am a nitro guy
Is that some kind of literacy/numeracy problem ? Before you give answers to people who might believe them, you really should understand what you are talking about.
hay, i was bored and thaught i might give some answers.
"Read up" seconded.
smitec084 years ago
it is most likely due to how you are taking your voltage reading. a brushless motor is usually run with three square wave signals (one per wire) and based on the timing difference the speed of the motor is controlled, to use the motor to produce energy you would have to rectify (per say) the output from the three wires into a stable DC voltage.
have a read of the following forum thread its got a lot of information on driving brushless motors then think about the reverse of their conversation. XD

Hope that helps
jimk3038 (author)  smitec084 years ago
I tried connecting a low ohm resistor (10 ohms) between one phase and then back driving the motor with my drill. I was hoping the resistor would get hot. Nope, nothing.

I just don't think the magnets are good for a generator application. The weak magnets are designed to get pulled around my the coils in the motor.

Oh well, the little motor only cost my $20 bucks mail order.

It is interesting, while spinning the motor by hand, if two of the wires are connected together the motor generates a big drag on the rotating motor shaft. Connecting all three wires together makes an even bigger drag. Interesting.
You wrote "It is interesting, while spinning the motor by hand, if two of the wires are connected together the motor generates a big drag on the rotating motor shaft. Connecting all three wires together makes an even bigger drag. Interesting."

That's because when a generator is generating a current there is a force created in the opposite direction, that's why it takes more power to generate more power.
JFYI, and continuing electromagnetic theory 202 that's "Lenz's law" ....its a bit easier to follow than Earnshaw's theorem, from yesterday's tutorial....

Now that I could understand.

I this going to be on the final? sells a lot of info on makin' generators out of motors ,  I aint sure if "brushless" is th' same as " induction" when you are talkin' electric motors but I do know that it can require a capacitor in th' wirin' in order to get an induction motor to "kick" after you got it up to speed !
A BLDC motor has a MAGNETISED rotor, an induction motor, in general, does not. There is an AC motor called a synchronous motor, where the rotor IS magnetised, but they don't work on the same principle as induction motors.


lemonie4 years ago
I go with smitec08 in the first instance - it should produce AC. Have you tried connecting a load to this like a filament-bulb?


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!