Newman Motor - tinkering and teething problems

[Spirality]

I'm experimenting with my own home-made Newman Motor at the moment, and just started reading other OU posts about it. (I attached a few images, I hope they appear when I publish the post)
My first version had a lot of windings, and the magnets were too close to the axle, expecting a lot of torque. (Newman2.jpg)
My second version has the magnets on a longer "core" but I suspect has too many windings for 9V (350+350 of 26swg, approx 2.5 inch diameter). I'm not getting any movement at all from that configuration. (Newman3.jpg)

I had wired up a Daftman circuit but again, there are probably too many windings.
I am using a reed switch but I suspect a big spark will just burn it out, so I am open to a rethink.
Going back to simplicity for the circuit until I can sort out the voltage-windings-torque-etc.

I am going to try larger (16mm) magnets and fewer windings (maybe around 100-200), or a combination.
Am also on the lookout for a Brass or Aluminium axle, so as not to divert the flow of magnetism. Certain plastics might do aswel, instead of the steel threaded bolt I am currently using, which is handy for assembly.
Do you think it makes a difference to have the coils wound in a circle instead of in a square?

I'd appreciate any comments, ideas, etc.
Thanks

[Spirality]

Using the simple reed/commutator switch, I just linked together 4 small 9V batteries in series and put the 36 Volts into the machine, it turns okay which means the basic arrangement is okay too.
When I spin it, it turns and goes for a bit, but slowly grinds to a halt. There is a lot of friction in my bearings, so I put some wd40 on them and it actually made the Motor go slower. Physics is mocking me

I will take a few readings, etc. before I change anything like magnets or that.
Even with just 36 Volts, the reed switch is lit up with a small spark. I know its a high voltage motor, but I'd still like to build something that runs off of 9v. I don't mind using a transformer with the 9v to get the high voltage either. I'd love to get some ideas about how to turn this 9v into around 100v or so.

[neptune]

I do not think square coils and round coils make much difference. When winding experimental coils, always add coil taps as you wind, so that you can change the number of turns.

[Spirality]

I know I said I would take some readings before modifying this motor, but I was at an Energy workshop recently and decided to go with the flow.
I made a few changes, 26swg wire, 150+150 turns. The device doesn't operate as well, so the next version will have many more turns of a thinner wire, and better magnets.
I also burned out my Reed switch so I made a new one at the workshop. I used a partially flattened Nail, tied loosely with stripped wire, and flat copper strip (plant-tag) wrapped around a matchstick.

Photos:
Newman4 = the latest arrangement, soon to be changed again.
Core = Magnetic Core made from 4 Neodymium Magnets, 3 Nuts and a cable-tie. You can also see a brass push-end to stop the coil wire from getting damaged by the axle.
Reed-Open-Closed = The new Reed switch in Open and Closed positions.

I do not think square coils and round coils make much difference. When winding experimental coils, always add coil taps as you wind, so that you can change the number of turns.

Thanks for the input Neptune, I will be using the square coils in the next version as its slightly more economical and the angle of wire-to-magnetic-stream is better.
I presume by "tapping" you mean to solder some short wires at various lengths along the coil wire? Great idea, I suppose every 50 or 100 turns should do it, or each time the winding layers reach one end of the coil.

[neptune]

Yes you got the basic idea on coil taps. Here is how I do it. After winding , say, 25 turns , double the wire back on itself for about 2 inches, and then twist the two resultant strands together. Then carry on winding. The result is a "spur" or branch sticking out from the coil. You only need to strip the insulation from the end of the tap if, and when you decide to user it.
     A useful idea is to add taps after 5 turns, 10, 15,20, and 25 turns. After that, tap every 25 turns. By using various taps you can vary the total turns over a wide range, and use the early, fine taps for fine tuning.