Muller dyno project

[Vorg]

Im planing to try building a Muller type dynamo. The first try will be using 8 3/8" magnets. I'm trying to work out a control curcuit. Attached is a schematic of what I have in mind.

What I'm trying to do with this one is all coil pairs will be full wave rectified And I want to be able to swtch any coil pair from generating to to drive. The curcuit will create a pulse each time a proper facing magnet passes under a coil. It will count the number of pulses from the last time a coil was fired so you can set how many to skip before firing again. It should be adjustable from firing every time to every 15th time.

Assuming its able to generate extra energy, 1 or more mosfets would be addeded, up to 1 per coil pair to disconnect the load from the drive cap when the drive cap voltage is below a set level and the coil voltage is equal to or higher then the drive cap. the extra mosfets would act as an overflow so the load can't pull the voltage down so far it stalls the system.

I was also thinking of swtching the coil pairs between series while generating and parallel while driving so it would work down to a lower voltage, but it would take too many mosfets. I thought about using a voltage doubler to do the rectifying, but good large caps cost more then power mosfets.

I'm not sure what I might be over looking with this setup for both extracting and driving with the same coils.

If this works out, I next want to try a 3 ring rotor using 4 magnets/3 coils for the first ring, 10 magnets/9 coils for the second and 16 magnets/15 coils for the outer. Only the outer ring would be used for driving as it would have the most leverage. With this layout each ring would be seperated by about the same space and the magnets are from each other. This should allow enough room for the coils.

[Tink]

How is the Muller motor/generator project going?

Most of the people here know the following sites I think, but just in case,......
http://www.mullerpower.com/
http://www.magnetricity.com/NeoG/NeoGen_Dynamo.php

[Vorg]

Very little free time to mess with it. I'm scraping the counter in favor of an ATmega AVR. They are only a few dollars and give a lot more control. To save money I'm doing a lot of reading first so I don't waste money on parts I can't use.

[Hydrostatic]

I am also in the midst of constructing a Muller based "Neogen Dynamo". I have construction of rotor, coils and housing completed, now it all comes down to timing control circuitry.? I have not yet decided on what microcontroller will be used, or how exactly it will be used. If you have any ideas, please share.

[Vorg]

I spent a long time deciding what to use for control to give the most option down the road. What I decided on is an array of ATtiny26'2. One per coil and 1 for master. Still not sure about master but likly will try it first. The reason is, the 26 provides both ADC and analog compair and has a USI. I don't think these things can really sample fast enough to use the ADC to monitor the wave form of even one coil with enough sample rate, let along a bunch of coils. But for what I want, I only need to know when the voltage is at certain points, probbly just 1 of 2. So I can use 1 port bit to select which votlage to check against and let the comparator run. I will still have several pins for controling mosfets to turn the coil on/of, etc. I plan to use the USI in TWI mode so  I only need 2 of the port pins, plus the first byte in a send is the address and data direction. A TWI interface can link up to 127 AVR's.

there will also be 2 IR photo sensors per coil. 1 will be for a slot that runs from the start of N pole to the start of S pole. The other will be for a string of holes to privide step signals. With 8 holes per magnet and 6 magnets and 1 step sensor per coil at 5 coils, you effectivly divide 1 rotation into 240 steps, because of the odd sensor/even hole setup.

This way, each AVR focuses on controling a single coil and teh master sets back and says when it needs coils fired and how long to run them.