A Question about Generators

Home Model Engine Machinist Forum

Help Support Home Model Engine Machinist Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Andrew Pullin

Jun 26, 2019
Reaction score
Wodonga, Australia
Hi All,

My Mind works in a very convoluted way sometimes. It occurred to me that Steel Flywheels
would be perfect to simply stick Magnets to, and if I have a coil(s) of copper wire that these
magnets rotate past then it would produce a voltage and current. This is basic Faraday's Law
and the Internet is full of Science Fair type experiments on how to do this. The problem is that
most Science Fair experiments are used to light a light bulb or do some other simple task, which
is fine, but what if you want to do some actual USEFUL work? Unfortunately all I can find out
about this stuff tends to be Science Fair or Power Turbines for Hydro Stations and such like, so
the info is either too simple or too complex. Ok, so what do I want?

I can easily buy Magnets with a known Magnetic Flux. I can easily buy Copper Wire with known
electrical characteristics. What I need to know is how do I calculate how much Voltage and Current
I can produce with a known coil and magnet so that I can then work out how big and how fast to
make a flywheel and how far from the flywheel I should place the coil?

If I want to make say 12V at 1A what do I need basically to achieve this. The actual voltage and
current is not relevant as this is a theoretical exercise. What I want to avoid is building a test rig
BEFORE I know what I am testing.

Thoughts Anyone?


I don't know of any simple method to get the info you're seeking. You're dealing with a magnetic field changing over time and to properly calculate a resultant voltage requires surface integrals over the area of the conducting coils (basically some fairly intricate calculus). Electromagnetism was one of my least favourite parts of my physics degree and I've managed to expunge a huge amount of that knowledge from my brain. If I had to do what you're contemplating I'd probably just whip up a test rig and work it out from there. As you can easily change the speed of the flywheel and distance between the magnets and coils, you can tune it quite a lot to get it to where you want. The physics data you've looked up for Hydro Power stations, etc. is the exact same stuff you need for this exercise. Just because you only want to deal with 12 volts doesn't make the calculations any easier. Sorry.

Of course someone may have a starting point for you, or some real-world hack/shortcut that I'm not familiar with (they only taught me the theoretical stuff).
I think that the PDF was written by someone who's native language was not English !

For what its worth, I would design a generator empirically. Useful things to remember are that the output will be an alternating voltage, the frequency of which will be a function of the rotor speed and the number of magnets. The actual output voltage will be a function of the number of turns of wire on a coil and the strength of the magnet used.

So if we have a single magnet and two similar coils and we rotate the magnet past the coils at 3000 rpm, we will generate an output voltage with a frequency of 50 Hz.

Exactly the same as an induction motor, two poles at 50 Hz = 3000 rpm, four poles would be 1500 rpm. To get the voltage you want you would simply wind a coil with a known number of turns, measure the voltage at the rotational speed you want to run at and adjust the turns on the coil to suit.

The amount of current that you can draw is determined by the wire size and the strength of the magnetic field.

This of course is a very simplistic description but it works !
On top of what Cogsy say one have to define and resolve the magnetic circuit.
Just magnets spinning is not the entire story, the flux must have a return path of low reluctance to have a sufficient flux to efficiently generate the voltage. The magnetic circuit better be laminated otherwise the result is a magnetic brake more than a generator.
What you started to describe sounds a lot like the radial flux permanent magnet alternators often built by the DIY wind energy folks. These are usually powered by low RPM wind turbines generating low voltage AC which is rectified to DC and then modulated at a usable DC voltage. I think you will find what you are looking for on the DIY wind energy websites.
This is where Maxwell's equation are your friend or more likely your albatross.

You're talking about real electrical engineering stuff here: answering the questions answered some 130 years ago.

I'm a mechanical engineer, not electrical, so I'm not much help here. Perhaps start with some small DC motors and run them in reverse to get some empirical knowledge?

(Partial differential equations suck beyond belief.)

Personally I would not worry too much about voltage and frequency. Rather once I had my generator working I would just rectify the current (convert to dc) and run it thru a voltage regulator to achieve what I wanted. If AC were needed you could then use a small camping type inverter in line.

While not the most efficient setup it would make use of readily available and inexpensive off the shelf components.

Current will be controlled by your available horsepower assuming you have a somewhat efficient generator design.

I once worked with a cruising generator company. Our rule of thumb was with average efficiency an ac generator would require about eight horsepower per kilowatt.

Remember Volts X Amps = Watts. Figure about 50% efficiency from heat loss in the various components and you can roughly calculate what you can expect from a given engine of known output power.
1HP=745Watt 1KW=1.34HP
8HP for 1 KW give 16.6% efficiency A company producing such generator would not last 24 hrs.
If 50% of power went into heat losses the unit would burn up in 15 minutes.

V x A = VA Apparent power
V x A x cos(F) = Watt Things get complicated with current distortion.
If you want to make around 12W look at this link.
I am not suggesting you use it but it may give you an idea of the dimensions.
Bear in mind that this model is actually very well designed mechanically, magnetically and electrically. Any deviation from optimal design will take a penalty in size to achieve the desired power.

Many ways to make a generator BUT it won't be a simple job to build your own.
Use of dc motor or stepper motor or brushless motor all have their drawbacks and mostly no where near the 12w you describe.
Probably the best approach with at least SOME degree of output would be to mod a bicycle dynamo, bearing in mind the speed requirements to produce useful power.
Mod to a low kv brushless motor may get some results but the details of a rewind are too complex for a beginner.
It certainly is not going to happen with a couple of magnets popped onto a flywheel and a quick knockup coil.
It will depend on the gauge wire the length of the wire and rotation rpm plus what is the magnet gauss. T cal all I do not know but I do know this all effects the voltage and amps or watts. Time to do more research or build up something and see what you get. Basically you are talking about building a stator maybe this link will help you out.
1HP=745Watt 1KW=1.34HP
8HP for 1 KW give 16.6% efficiency A company producing such generator would not last 24 hrs.
If 50% of power went into heat losses the unit would burn up in 15 minutes.

Excuse the confusion, what I was speaking to is the use of a solid state voltage regulator and voltage inverter,

Those devices coupled with a small scale model generator outputting maybe a few amps would indeed waste half the energy in heat if not more.

You are correct if a large scale, say 20kw , gen set were to create that much wasted energy in the form of heat I would think it’s life expectancy would be more like fifteen seconds.
If I want to make say 12V at 1A what do I need basically to achieve this. The actual voltage and current is not relevant as this is a theoretical exercise. What I want to avoid is building a test rig BEFORE I know what I am testing.

Thoughts Anyone?



Hi Andrew,
Just pull an old battery operated drill apart and use the motor, that will get you close to what you want.

If you just want to play about, pull an old 3.5" floppy disk drive apart. The motor in those will get you 8 to 10 volts at 50 ma or so.

For those that say my empirical method won't work have never done it for themselves !
I would look at generators that people have built for small wind mills. Those are easy designs since most are built by hobbyist. An easy to use program to design your own in FEMM software http://www.femm.info/wiki/HomePage . It has been an open source FREE program that has existed for decades so a search of the web will find many supporting information and likely someone has designed a generator. It is a 2D finite element modeling program. And ~ 90% of magnetic machines types have been designed by professional with such programs just a few decades ago, ~ 1960's to 1980's.
Hi All,

Some good feedback and interesting suggestions. A couple of points to clarify (but thanks to those that made suggestions otherwise).

The point of this exercise is not so much to build the generator but to work out how I would build it. As mentioned, there are many
Science Fair projects out there in Internetland and the theory is simple and straight forward - wind some coils and move them through
a magnetic field. My theory of sticking Magnets to a Fly Wheel and spinning them past a coils should work no problems.

For those that suggested pulling apart or acquiring motors and spinning them some way - I agree that it is simpler and easier and almost
certainly cheaper but only gives me the end result. What I want is the satisfaction of working out how to do it, designing a machine that
will do it, and then seeing how close I get to the theory with actual test measurements.

For those who suggested the wind power generator path, that is sort of what I had in mind when I dreamed this up. I am investigating this
now, but many of these projects also use the "get a motor" method. Some use a method similar to my method but they are more of the giant
Science Fair project type - they work fine but you get what you get, there is no theory behind the designs to calculate what you get.

I have made some progress. I am not interested in PHD level explanations of how this works, just a working knowledge, and I am happy
to plug numbers into a formula. My interest is doing the exercise and having an understanding of what any formula actually calculates
and why it needs certain inputs, e.g. there will be some trig function involved because the rig spins - I get that. I will leave any heavy lifting
to a computer, I don't need to know Quantum Field Theory so that the arcane quirks make sense.

For those interested, I was put on to these sites:

Hi Andrew,

The generated voltage can be calculated using the first link below. The amount of power that can be generated (and hence current flow) is related to the amount of power provided by the force causing the rotation and generator efficiency. The wire gauge and resistance are inter-related and determine the amount of current that can be generated either due to voltage drop and/or fusing of the wire. The wire gauge does not affect the generated voltage until there is a load applied to cause current flow.


More information in these links.





John Clarke

Thanks to John Clarke fro Silicon Chip Magazine in Australia.

I am greatly enjoying all you input. Thanks for all of your help.


Andrew, if you are still interested, I have a couple of books on this subject from my University days. I can get you some formulas if you want. Unfortunately, they are in engineering terms and require an understanding of calculus and electromagnetism. This is not a simple subject. The magnetic circuit has the primary impact on the design. Using unknown characteristic ferrous material will cause problems that can only be solved emperically.
I checked my book and there are sample designs for both DC and AC motors, which can be used as generators if rotational power is input. These are much larger than you need - 15HP (11kW) for DC and 200HP for AC. The basic electromagnetics will still apply. Both designs use field windings instead of permanent magnets (I predate rare earth magnet motors). To give you an idea of the complexity, the design parameters cover 24 pages for the DC motor and 19 pages for the AC motor with references to figures and calculations in other chapters.

Latest posts