first test with a coil and magnets

[jacquesm]

Today I ran the first test with a coil and 12 magnets, this is only one of two rotors, so the results come out plenty low.

Before being able to test this at all I had to find a safe way to insert / remove that rotor. I've been playing with neo's long enough to know that 12 of them (1x2x.5" blocks) can clip your fingers off quite handily and I really don't feel like losing any of them.

The solution I came up with is pretty stupid, but works. I bolt the motor frame to the carriage on the lathe, then use the 4 jaw to center the rotor in the right spot and crank it in with the carriage wheel. You can really feel the point where the magnets are partially into the stator, because the cranking suddenly goes extremely light.

After that I put the end bells back on (that was quite tricky, those magnets really don't want to let go once they have a good grip on something...

I placed the frame back on the lathe and clocked the shaft until it was centered, then blocked the motor from taking off on my with some 2x4's and a tie down strap.

This test was not done at any powerlevel, so I figured I could get away with such a light mount.

The voltages produced are here (Hey DanF, can we please have tables and the 'pre' tag ?, I don't see how that is a risk to anyone... ) (open circuit only, 60 wdg testcoil):

I didn't dare to run it any faster because the magnets are still held on by their own grip, no glue or anything and I'd hate to wreck the stator if one came loose.

I may be able to squeeze another 5 windings in there, but that's about it (say another 10% gain).

So, now this is all set up I have the perfect test bed to see if I can somehow get that pulsed fet trick to work before I decide on the exact number of windings and wire size. Maybe if it works well enough I will be using thicker wire and less windings, this would give all kinds of options on the high end of things

Another problem is that I still need to figure out how to drill a hole through that shaft, the longest drill I have is about 4 inches, and the shaft is a lot longer than that !

Anybody with hints, tips or very long drills ?

Shaft with rotor and magnets mounted in 4 jaw chuck, motor frame mounted to the carriage:

Slowly easing it in:

End view:

The rotor in place:

Clocked in and ready to run, all it needs is a way to block the housing from spinning:

[baggo]

Hi Jacquesm,

One way to make your own long drills is to turn down the shank of a normal drill and braze it into the end of a length of suitably sized tube or a length of bar drilled at the end to take the reduced shank of the drill. You have to keep withdrawing the drill to clear the cutting swarf. This increases the time taken to drill the hole but better than nothing and a lot cheaper than buying very long drills!

John

[Drives]

Jacqesm:

I have heard of long drill bits called "noodle" bits because they look like a long piece of twisted pasta.  A machine that uses these bits is called a "pecking drill" because just like John said earlier, you have to drill a little, and back out of the hole a little to remove the chips.  (Just like a chicken pecking.)  I have seen these bits used extensively in gun barrel manufacturing.  I hope some of this helps, Good Luck!

[monte350c]

Hi Jacques,

Princess Auto has them:

and you can shop online in case there isn't one nearby (and I don't think there is)

www.princessauto.com

Ted.

[monte350c]

oops! hit the post button a bit too fast

What size is the wire you made the test coil out of?

Thanks,

Ted.

[ghurd]

I made them too.  Drilled 31" into a 32" hex bar. It stayed straight.

I would recommend going a hair small on the bit. Mine was 1/2, but the hole ended up 0.540, top to bottom.  It was a muzzel loader barrel. Just lucky that 54 cal. is also a common size.

G-

[Victor]

Hi Jacquesm,

 How is the cogging?

Victor

[jacquesm]

the cogging is pretty good victor !

A tiny little bit, but I assume that's because the magnets got jarred a bit during insertion, once they're fixed in place that should be gone.

Still trying to figure out how to pulse those fets properly...