Monday, September 5, 2011

An upgraded simple three phase generator design.

After my failed attempt at getting an output I deemed acceptable out of the simple three phase generator I built I went ahead with a redesign. Instead of four large disc magnets, I used 8 smaller block magnets. Instead of six 500 turn coils I used twelve 200 turn coils. Once again I used 30 gauge magnet wire.

This generator becomes a rats nest of wires and is therefore less simple than the previous version. Still, it can be constructed fairly cheaply without expensive tools.

The first step was to wrap the coils. This time I wanted coils just barely larger than the magnets which would be rotating. Therefore I decided to wrap wire around the magnets. It is very important to wrap the wire in the same direction for every coil, and remember which way they are. If you flip the rotation on the coils, it will dramatically lower the output of the generator, perhaps even to no output at all.







After making 200 turns around the magnet I did one more turn where I wrapped the wire around the coil to hold the wire in place.

Once all twelve coils were constructed, I mounted them on a CD with super glue. This is the step you need to be careful that all are placed in the same orientation. If you install some upside down wiring the generator will become a real mess. It is also important that all the coils are evenly distributed. One coil should be at 0, 30, 60, 90, 120, 150, 180, 210, 240, 270, 300, and 330 degrees. Here is them just before I glued them.



Positioning the magnets was done on a saw blade this time, rather than the CD used in the design I based this on. This is for two reasons. The saw blade being steel acts as a backing iron and helps direct the magnetic fields through the coils it also is a little easier to rotate than the CD. I assume this generator would work with a CD, but would produce a little less power.

Similar to the toys from trash design a large washer was glued to the top and bottom of the hole in the center of the saw. A toothpaste cap with a hole drilled in it was glued to the top of the rotor. The magnets were placed at 0, 45, 90, 135, 180, 225, 270 and 315 degrees which was made easier by the holes in the saw blade. Once positioned they were glued in place, although I am not sure that was necessary with how strong the force between the magnets and saw was.

As before the magnets must alternate. The magnets at 0, 90, 180, 270 degrees will all have north pole facing up when the magnets at 45, 135, 225 and 315 degrees are all have north pole facing down.



This time instead of iron filings, which were quite a bit of work to collect, I decided to use steel wool for the cores of the generator. I simply cut up the steel wool into small pieces then mixed it with wood glue before filling the cores with it. This didn't seem to be quite as good as the filings, but it clearly performs the basic task.

The next step was wiring, which requires some thinking. A good description of the theory of wiring such generators can be found here. For this generator it was necessary to connect 4 coils together to make each phase. The coils at 0. 90, 180, and 270 degrees make up one phase; the second phase consists of the coils at 30, 120, 210, and 300 degrees; while the third is made from the coils at 60, 150, 240 and 330 degrees.

The coils must be connected together so the current adds rather than subtracts. First for each magnet you connect the low side of one coil to the high side of the coil on the opposite side. which side you declare to be low is arbitrary. I decided I wanted the right side wire on each coil to be low. Once you have connected each with the opposing coil, you need to connect the low side of one group with the high side of the group perpendicular to it.

I made some simple sketches showing the wiring, first of one phase, then of all three phases combined. Notice that the low side of all three phases gets connected together. That is where neutral connections are made. This makes it a typical Y connected generator, leaving me with three phases neutral. I could have done a delta connection but I wanted to have the neutral.








This time I decided not to use the sewing machine bobbin for a rotor. It just allowed too much wobble, making it too difficult to get the distance between the magnets and the coils as small as possible. So, I decided to glue a nut and washer together. I then placed additional nuts and washers in place to adjust the height. The rotor then simply rotated on a washer which I lubricated with WD40.







This generator worked a better than the previous version. It produces about 1 Volt phase to neutral, and 1.7 Volts phase to phase. I can light LEDs by connecting an it to any 2 of the four wires.

Still, the rotor could be a lot better. Two washers rubbing against each other do not exactly make the most efficient bearing. I will be keeping an eye out for a way to mount the rotor in a lower friction manner.
Another step which could be done to improve the generator is a backing iron on the coils. So far I just haven't found an easy way to modify this design to allow that. Unlike the backing iron for the rotor, the backing iron for the coils cannot be solid iron. Doing so would create eddy currents which sap the efficiency of the motor. So I would need to make a laminated steel or iron filing based replacement for the bottom CD. Nothing simple comes immediately to mind.

5 comments:

Anonymous said...

This is really fantastic and a great way to learn about the generation of electricity!

A few questions:
1) If you increased the gauge of the wire, how would that effect your voltage?
2) What about stronger magnets?
3) Is the steel wool better or worse than magnetite?
4) is there a way to calculate the current produced?

I also found this explanation of Faraday's equation for # of turns of wire.
http://aesthetec.net/magnetgen/

Thanks again!

glmory said...

1) If you increase the gauge of the wire(make the wire smaller) you will increase the voltage output. The resistance of each coil will be higher though so you will probably decrease the current output at the same time. If I was to do this again I probably would work with 30 gauge wire again since it works well to light LEDs and any smaller gets really hard to work with because it is so fragile. If you instead want to run brushless DC motors it might be better to use larger wire than 30 gauge, my most recent generator used 24 gauge which seemed to work alright. Even there, probably 30 gauge is sufficient for demonstration purposes if not an actually practical generator.

2)Stronger magnets should in theory help, but actually add a lot of complications. The problem is that the magnets will try to pull tight to the iron cores. If there is any wiggle in your rotor the magnets will stick to the cores making it impossible to turn the generator. So your options when working with stronger magnets are make a much more robust design (perhaps requiring access to a machine) or use cores which do not have iron.

3) the steel wool turned out to be worse than the magnetite, but not by much. If I did this again I would skip both and make up for the loss by using larger magnets. Having a core helps if you want a really useful generator but for a demonstration it ends up adding a surprising amount of unneeded complexity.

4) I agree with the basic idea in the link you gave. Faraday's law is certainly the fundamental law to use and the basic procedure they followed would be the same that would need to be followed to figure out the output of this motor. He stopped at finding voltage though, to get current you will need both the resistance of the coil and the resistance of the load along with that voltage and ohms law. Also, iron cores will complicate the calculation since they will increase the total magnetic field passing through the core. I have never actually performed this calculation to compare to the output I saw with this generator so I cannot say how closely the calculation comes to matching the output of such a simple design.

A few other pages may be of interest to you.

I later built a three phase inverter to run this generator as a motor:
http://glmory.blogspot.com/2012/07/three-phase-6-pulse-inverter.html

I also built three other generators.First, there were two single phase generators:

One which copied this design: http://www.amasci.com/amateur/coilgen.html

and an upgraded version of that design:
http://glmory.blogspot.com/2012/03/more-powerful-simple-generator.html

Then there was a much more robust three phase generator:
http://glmory.blogspot.com/2012/10/another-three-phase-generator.html

In addition you may be interested in these two websites which I drew heavily on for inspiration. The phasing I use is mostly taken from this design: http://www.stanford.edu/~hydrobay/lookat/pmg.html

The basic structure started with this generator:
http://www.arvindguptatoys.com/toys/multiplegenerator.html

glmory said...

If instead of a simple demonstration, you want to build a generator which can actually produce useful power, than you may also be interested in this link:

http://www.instructables.com/id/DIY-1000-watt-wind-turbine/?ALLSTEPS

Beware if you do this though. Before using magnets of more than 25lb pull I strongly advise you read up on how to safely use them. They become quite dangerous once they are strong enough to jump off the table and crush your hand.

Panchito said...

hI THERE, I UNDERSTAND THAT THIS COULD NOT BE THE CORRECT WAY TO ASK BUT IAM IN TROUBLE
I DESPERATELY NEED A 3PH SPWM 6 PULSE 60 CYCLE ARDUINO CODE (MEGA IF POSSIBLE) TO FINISH MY PERSONAL PROJECT THE PULSES WILL BE DIRECTED TO A PC817 OPTOCOUPLER AND THEN TO A IR2112 AS A DRIVER, ANYBODY KNOWS ABOUT IT? MAY BE FOR SALE?

best ed pills said...

This is very fascinating, You're a very skilled blogger. I've joined your rss feed and look ahead to looking for more of your fantastic post. Additionally, I have shared your site in my social networks