My working pulse motor

[What_The_Flux]

This is my first post although I've been visiting and learning from the group for several months.

Background:
After messing around with pure magnet driven system for a few weeks, I realized there is no future there, and started working on a Bedini SSG.
Like new experimenters, I was in a hurry to build anything so I cut a few corners. My main mistake was thinking I could build a sufficient coil with Radio Shack's magnet wire package. There definitely isn't enough wire in there, and I found  that when spinning some neodymium magnets by the coil, it could only generate a few hundred millivolts, not enough to drive the base of a power transistor.

Step two was to release the E.E. in me to design a circuit which could drive the 2N3055, and give me an adjustable pulse width.
I ran the drive coil into an op-amp, which gave me a square wave, and then ran that into a 74121 so I could shape the pulse. That worked wonderfully, except that output pulse would not drive the transistor either. Trying to use an intermediate transistors to generate enough drive current didn't work for me (yet) because the small signal transistors I have did not have enough drive current. I'll revisit this idea later.

The other major problem with this setup was the freakin FEEDBACK. I know some people want to build a Bedini oscillator, but TELL ME HOW TO NOT BUILD ONE! The spikes from the drive coil bleed into the trigger coil which bleed into the op-amp and cause mayhem.

So just to get back to basics and understand the workings of the pulse circuitry, I decided to simply use a magnetic reed switch to trigger the 2N3055. The results were pretty impressive. When I modified the hardware and used an inline skate wheel with Swiss bearings, the results were Wow! Yesterday I had that thing spinning at over 4000 RPM. Using 4 magnets, that means the reed switch is giving me about 275 clean pulses per second. This was an old reed switch from a home security system. I had no idea they could switch that fast.

The setup:
As I mentioned, I'm using an inline skate wheel, with 4 magnets, each one being two .5" neodymiums countersunk into the wheel.
The coil is a two conductor spool with about 40 feet each of 22 and 26 gauge wire. Since I abandoned the trigger coil, I put both coils in series for more drive and more impedence. I'll attach the circuit below. The circuit is being driven by a 16V DC power supply. The back EMF pulse goes into a 12V drill battery.

Observations:
I was pleased to see my battery (a good one, though uncharged) go from 0.6V up to over 12V over the course of me experimenting with it. But I was very disappointed in the performance of the charge. As soon as I hooked it up to drive the circuit or the drill, the voltage dropped within seconds to an unusable level. Perhaps this is attributable to the small size of the coil, and the corresponding size of the EMF pulse.

When running at full speed my input current ranged from 300-800mA depending on the duty cycle of the pulse, and the charge current (going to re-charge the battery) was around 30mA consistently.

So anyway, that's my first pass at a motor, and I look forward to hearing from anyone interested. I can post a video of this thing hauling ass if you would like. My next interest would be in fine tuning a drive circuit that would be more efficient.

[TinselKoala]

Looking Good!

I can make a couple of suggestions: a power MOSFET like the IRF540 will switch faster and cleaner than the 3055, and will switch at very small currents to the gate. And you can get Hall effect switches that will last longer than your reed switch and also switch much faster.
You might need the jitter from the reed switch to see "Bedini" type results, though.
To deal with the feedback, if you see it as a problem, you may need to insert some Schottky diodes into the circuit to keep the spikes out.
Using the Hall sensor, you can vary the timing of the pulse either by digital delay methods or by physically moving the sensor. If you are going for max rpm, the timing will need to be different than if you are going for max torque, and the optimum pulse timing is likely to vary as RPM increases.

In the bottom view of the Marinov slab motor below, the Hall sensor can be seen, mounted on an adjustable disk for timing changes. The MOSFET circuit is mostly on the other side of the circuit board.

[What_The_Flux]

Here is a video of it in motion, and some waveforms:

http://www.youtube.com/watch?v=nNloAJMkjO8

[spinner]

Hello, WTF! 

Hmm, what is this circuit of yours supposed to do? Charging a battery from BEMF "free energy"? Aha... Bedini stuff? Ok....

This circuit has just a few percent efficiency...  I'm sure you are aware of this?
Why don't you look up some (low cost) mass-produced "made in China" chargers? They're usually over 70% efficient....
Cheers!

[What_The_Flux]

Hello, WTF! 

Hmm, what is this circuit of yours supposed to do? Charging a battery from BEMF "free energy"? Aha... Bedini stuff? Ok....

This circuit has just a few percent efficiency...  I'm sure you are aware of this?
Why don't you look up some (low cost) mass-produced "made in China" chargers? They're usually over 70% efficient....
Cheers!

What is it supposed to do? It's supposed to make that little wheel go around in a circle. (See top of page where it says this is the section on 'Pulse Motors')

The charging battery is supposed to keep my transistor from burning up, and provide free energy for all the cities and underdeveloped nations around the world! Oh, and stop war too!

I already have a charger for the batteries for my drill.