PDA

View Full Version : Recent invention: Magnetic gears that work



Evan
05-25-2008, 10:09 AM
I feel a new project coming on. I stumbled on this while looking for drive coupling options for my electric bicycle. Time to order a handfull of high power magnets and do some experimenting. The power handling capacity is impressive and the construction isn't complicated. According to the info the power density possible is 70Knm per m^3. To translate that to more familiar units it can handle up to around 50,000 foot pounds of torque per cubic yard of machine. Break this down to a reasonable size and you have a magnetic planetary gearbox with two non contact moving parts and no wear at all that can transmit 50 ft lbs of torque in a device the size of a 1 quart (or 1 liter) can. It's totally scalable and a unit the size of a car transmission can easily handle 500 ft lbs torque. [edit] I forgot to mention that best of all is that if you overload it absolutely nothing bad happens.

http://vts.bc.ca/pics3/maggears.gif

Lots of interesting ideas in this pdf including the above.

http://www.prizz.fi/sivu.aspx?taso=3&id=290

Analysis here:

http://www.infolytica.com/en/coolstuff/ex0106/

A.K. Boomer
05-25-2008, 10:22 AM
That is ingeniuos, its about the coolest thing iv seen in along time --- if I catch it correctly they are simply using magnetic effects in a planetary structure to come up with different "gear" ratios?
The possibilities are endless --- and what about in an AC application with 3 phase?

So many times you see something and have that "why didnt I think of that" remark hit you on the forehead.

dp
05-25-2008, 10:30 AM
One nice thing about machining magnets - no swarf or dust flying around the shop :)

The animation is interesting. Lots of field rotations so it looks like it could get warm even when doing no real work. It also looks like it should have a lot of vibration from magnetic attraction to the stationary ring. I wonder too about the effects of the load driving the driver in deacceleration.

One of the links is from 2001 - what's happened since?

Evan
05-25-2008, 10:32 AM
Unfortunately many of the papers on this are behind a cash register. It's really a sorry state of affairs now. I wonder if any of the authors actually make any money?

Lew Hartswick
05-25-2008, 10:43 AM
We have had magnetic coupling through the wall of vacuum chambers
for a lot of years. It's a wonder someone hasent though of this before.
A very slick gear reduction. At least one of the shafts must be
cantilevered to alow for the non rotating member so bearings will
have be hefty on it. It'll be intresting to see how long it takes to
show up in a device. I remember it took a long time for winglets to
show up on aircraft after they were "discovered". :-)
...lew...

dp
05-25-2008, 10:56 AM
The illustrations show a concentric ring configuration but there's no reason I can see why they can't be axial. And with a bit of cleverness you could create a multi-speed transmission by sliding magnetic collars between stator rings.

Evan
05-25-2008, 12:33 PM
It's already in a device. It's being used in mud turbine powered reduction units for down hole drill bit and pump drives. It's an ideal application because it's simple to make a coaxial opening in the device for the slurry to flow and it is automatically torque limiting so you don't snap something when you hit a small asteroid inclusion.

Nothing needs to be cantilevered. Both ends of everything can be supported either on the inside or the outside at either end. There is no reason you can't have rolling element bearings in concentric rings. Also, input and output can be at the same end.

Dennis, axial is a working configuration as is linear.

Forrest Addy
05-25-2008, 12:44 PM
Hm. I can see ratio changing as simple as electro magnet "sun gears"

OTH I foresee resonance problems and compliance problems in drives required to be tortionally stiff. Then again a soft-acting utterly shock resistant drive has many advantages.

Weston Bye
05-25-2008, 01:00 PM
I wonder what the efficiency is. Heat will be generated in the form of eddy currents in the pole pieces as dp pointed out. Heat could be the limiting factor, cooking the magnets.

Neodymium magnets give the highest energy, but Samarium Cobalt, somewhat more costly, have a higher Curie temperature and will operate at higher temperatures without suffering reversable and non-reversable losses in magnetism. Maybe a reasonably sealed gearbox with circulating coolant?

Evan
05-25-2008, 01:10 PM
I don't recall seeing efficiency numbers but it is hard to beat a well contructed regular gearbox. They have efficiencies in the mid 90s.

Fasttrack
05-25-2008, 09:42 PM
I actually stumbled across these during my eigth grade science fair project. I believe they were more or less in their infancy at that point.