Magnetic gearing?

amberwolf

Administrator
Staff member
Joined
Aug 17, 2009
Messages
40,463
Location
Phoenix, AZ, USA, Earth, Sol, Local Bubble, Orion
http://www.engineerlive.com/Design-Engineer/Power_Transmission/Magnets_offer_advantages_as_an_alternative_to_mechanical_gears/22078/
 
Great find, Amberwolf. The company developing this is here:
http://www.magnomatics.com/Technology/Pseudo-Direct-Drive.aspx
I'm wondering if this could be developed for ebike use. They mention hybrid vehicles.
"Advantages of PDD®

* Significant size reduction over conventional direct drive machine
* Ultra high efficiency removes the need for ancillary cooling
* Reduced maintenance and improved reliability over mechanically geared drives
* Inherent torque overload protection
* High power factor (typically >0.9)
* Employs standard power electronic controllers
* Possibility for two output shafts with different rotational speeds"

Can this technology be used to combine the power of the pedalling with the power from a battery; would it be better than the "mechanical drive mixing" that Miles and others are trying to achieve here:
http://endless-sphere.com/forums/viewtopic.php?f=28&t=13532&hilit=splined#p200663
?
I'm wondering if "Possibility for two output shafts with different rotational speeds" could metamorphose into "Possibility for two INPUT shafts with different rotational speeds" ie and RC motor input and a human pedaling input?
 
So is this just basicly a multi layered motor like Luke is proposing in the axial flux thread with different windings and the motor just switches between two different winding counts to get the desired motor speed? It does sound awesome, but I am wary about the efficiency. That motor looks badass, why can't Astro make thier motor look like that? 8)
 
etard said:
So is this just basicly a multi layered motor like Luke is proposing in the axial flux thread with different windings and the motor just switches between two different winding counts to get the desired motor speed?
No, it's an additional system.

Ref: http://v3.espacenet.com/searchResults?locale=en_EP&PA=magnomatics&ST=advanced&compact=false&DB=EPODOC&submitted=true
 
I wish I knew more about magnetics...I am wondering if neo magnets out of hub motors and the like would be sufficient to transfer the forces needed in such a gearing system on my bike.

Perhaps some experiments are in order, after I figure out how it is supposed to do what it does (assuming I can grok it).
 
Amazing that they would publish a video with such detailed conceptual and structural info. I bet someone with a fair amount of mechanical and machining know how could replicate this in a shop. Shoot, I may want to have a crack at it myself when I finish my upper division physics classes.
 
fireraisr said:
Amazing that they would publish a video with such detailed conceptual and structural info. I bet someone with a fair amount of mechanical and machining know how could replicate this in a shop. Shoot, I may want to have a crack at it myself when I finish my upper division physics classes.
It's already covered by patents, that's why they're free to.........
 
Miles said:
fireraisr said:
Amazing that they would publish a video with such detailed conceptual and structural info. I bet someone with a fair amount of mechanical and machining know how could replicate this in a shop. Shoot, I may want to have a crack at it myself when I finish my upper division physics classes.
It's already covered by patents, that's why they're free to.........

I figured as much, I meant for personal use not corporate. Most innovations like this aren't disclosed so openly. It really is fascinating tech though.
 
Hi,
fireraisr said:
Amazing that they would publish a video with such detailed conceptual and structural info. I bet someone with a fair amount of mechanical and machining know how could replicate this in a shop. Shoot, I may want to have a crack at it myself when I finish my upper division physics classes.

If its half as good as they claim thats a great idea (up to 50:1 reduction, quiet, efficient, CVT or multiple speeds, a good match for thin axial motors…). Sounds like a perfect reduction unit or combo reduction unit multi-speed gearing for an RC motor (built-in or external) or Axial:
http://www.magnomatics.com/Technology/magnetic-gears.aspx
Contactless, high-efficiency, high-torque transmission with inherent overload protection

The high-torque magnetic gear was invented and demonstrated by Magnomatics co-founder Dr Kais Atallah in 2001. This pioneering research has been extended by Magnomatics to provide a mature range of gear technologies that are suitable for a very wide range of applications.
Advantages over mechanical gears
* Reduced maintenance and improved reliability
* Lubrication free
* Higher efficiency than conventional gears
* Precise peak torque transmission and inherent overload protection
* Physical isolation between input and output shafts
* Inherent anti-jamming transmission
* Significantly reduces harmful drivetrain pulsations
* Allows for misalignment/vibration of shafts
* Very low acoustic noise and vibration

A magnetic gear uses permanent magnets to transmit torque between an input and output shaft without mechanical contact. Torque densities comparable with mechanical gears can be achieved with an efficiency >99% at full load and with much higher part load efficiencies than a mechanical gear. For higher power ratings a magnetic gear will be smaller, lighter and lower cost than a mechanical gear. Since there is no mechanical contact between the moving parts there is no wear and lubrication is not required. Magnetic gears inherently protect against overloads by harmlessly slipping if an overload torque is applied, and automatically and safely re-engaging when the fault torque is removed.

Magnomatics has developed a range of magnetic gear technologies for achieving low and high ratios and a linear gear variant. The magnetic gear concept has been extended to provide both an ultra high torque density pseudo direct drive electrical machine and a variable ratio gear topology for continuously variable transmission systems.
Low ratio is up to 50:1!:
http://www.magnomatics.com/Technology/low-ratio-magnetic-gears.aspx
Contactless high-torque magnetic gears with ratios of less than 50:1 with inherent overload capability

A low ratio magnetic gear uses a series of steel pole-pieces to modulate the fields produced by two permanent magnet rotors with different numbers of magnetic poles. The magnet arrays rotate at different speeds with the gear ratio determined by the ratio of magnets in each array. The transfer of motion between input and output shafts is passively achieved using just high power permanent magnets.

A high degree of magnetic coupling between the rotors allows for torque densities comparable with mechanical gears to be realised. Gear ratios of 50:1 down to 1.01:1 with almost zero torque ripple are readily achievable. The low ratio magnetic gear forms an integral component within Magnomatics’ new type of permanent magnet machine, the ultra high torque density pseudo direct drive, PDD®.

Low ratio magnetic gears can be realised in a conventional radial field or in an axial arrangement when a thin “pancake” design is required. Unlike conventional mechanical gears the magnetic gear can be arranged to provide geared linear movement.
http://www.magnomatics.com/Technology/Magnetic-CVT.aspx
Contactless, high efficiency continuously variable transmission system with inherent overload protection
Advantages
* High efficiency
* Contactless, lubrication-free variable transmission
* Inherent overload, torque fuse capability
* Speed of prime mover matched with variable speed load

The high torque density low ratio magnetic gear concept has been extended by Magnomatics to create a variable ratio magnetic gear which can be used as an integral component in a continuously variable transmission (CVT) or Infinitely Variable Transmission (IVT). The magnetic CVT allows a variable speed drive to be connected to a fixed speed load, is highly efficient and compact, requires no cooling or lubrication and is suitable for applications as diverse as hybrid vehicles and wind turbines.
 
:shock:

Freaking cool! Now all we need is magnetic chain drives!! :lol:
 
Well...actually... how about a magnetic "friction drive". Put magnets on the rim similar to what they seem to be using for their magnetic gearing, with a drive "gear" that engages and pulls them along. Do this on both sides of the rim. Then you can keep the whole thing narrow for a bike, assuming the motor is forward (or aft) of the wheel, rather than to the side of it.
 
amberwolf said:
Well...actually... how about a magnetic "friction drive". Put magnets on the rim similar to what they seem to be using for their magnetic gearing, with a drive "gear" that engages and pulls them along. Do this on both sides of the rim. Then you can keep the whole thing narrow for a bike, assuming the motor is forward (or aft) of the wheel, rather than to the side of it.
Yeah I had an idea for something like that sometime last year. It's basically a U-shaped solenoid and the bike rim has permanent magnets fixed to it. The problem is low torque but high speeds, I thought it would be good for a roady style bike. I got the idea thinking of ways to replicate the japanese electric motorcycle prototype that was shown a few years back.
 
dequinox said:
:shock:

Freaking cool! Now all we need is magnetic chain drives!! :lol:

If you know me, you will know that I like to throw in daft ideas in case some genius is listening and can use it as lateral thinking fodder.
So, dequinox's "daft idea" suggests this to me:
If you had a rubber belt with a continuous series of magnets stuck on the outside, with the said belt arranged just like an ordinary bike chain over chainring (toothed to engage with belt) at the front and "derailleur" at the back (maybe 3 toothed cogs to engage belt), you could have a tube around the top of the belt with electromagnets imbedded in it and a controller that would switch them on and off in such a fashion to drive the belt forward (bit like a linear motor).
No need for dual freewheels at the rear and no need for a secondary system separate from the pedal-power transmission. Regeneration is easy and "cogging" is no problem, because when you aren't using the motor the electromagnetic coils are not energised.
What do you think?
 
Nice find Amberwolf! Very simple idea. I wonder if external power is actually needed if you just want a fixed gear setup. Seems like a simple fixed gear setup can be done this way like the first half of the video without adding power. I'm tempted to play with this idea as well. Need to find a good souce for such long magnets.
 
Hi,

kfong said:
I wonder if external power is actually needed if you just want a fixed gear setup. Seems like a simple fixed gear setup can be done this way like the first half of the video without adding power.

Do you mean external electrical power?

I only see references to permanent magnets:
http://www.engineerlive.com/Design-..._as_an_alternative_to_mechanical_gears/22078/
Improved field strengths and design techniques mean that smaller and lighter magnets can replace larger composite magnets or create new opportunities.…

Magnomatics' magnetic gears use permanent magnets to transmit torque between an input and output shaft…

Magnomatics_Fig4.JPG
 
[youtube]TtIkYbReMbk[/youtube]...all the parts of the puzzle are beginning to fall in place for an ultra efficient vehicle of the future... :twisted:
 
What they neglect to mention are the eddy current losses.

Those magnet holders an anything in motion through the magnetic fields, or anything with magnetic fields in motion against it all must be non-conductive or this thing is a giant eddy current brake.

Every one of the models shown would make excellent heaters, but very poor efficiency gear reductions for things moving in the 4 digit RPM range.
 
Good point, it seems composites with CF would be a good match for this kind of stuff.
 
When cars wear out, its most often the engine parts rather than a manual gear-box. Used diffs and manual transmissions are fairly cheap because of this as few people need them and there are so many available. Therfore I don't see the mag gearing having a wide application with gear-boxes, however...

The seal-less coupling through a wall was interesting, also I hadn't considered the 90-degree coupling. A few years ago I was forum-ing about Stirling engines, and choosing to use helium instead of air as the medium has certain benefits. It wasn't hard to sketch-up a DIY seal-less design with magnetic couplers.

Rodless magnetic linear actuators have been around for a while. Not much more expensive than a standard actuator, and useful in an air-driven piston where a rod-seal leak would require quick replacement, but would be hard to reach. No rod-seals, no possible wear leaks.

LFP, perhaps its possible to make a multi-plate mag-clutch that doesnt slip very much (for a car), also no-contact, so impossibe to fry or wear out? Maybe easier to prototype on a RWD (Nissan 240SX, Mazda RX-7?), but still...interesting idea, yes?
 
It is an interesting idea. It's going to need to be all some sort of plastic/fiberglass sort of stuff though. Carbon fiber as I've been learning through playing with it is unfortunately a lot better conductor than I had thought it would be. lol

If the two halves didn't eventually get to mate up, it seems like it would be hell on on the thrust axis bearings for both the tranny and the engine.

I think it's possible.

A number of aquarium pumps have used the magnetic coupling now for years to keep the motor on one side of the glass, and the pump on the other. It works, and as long as its passing through a non-metallic, non-conductive material it's pretty efficient. The examples shown in the video, and the images of the products they show on the other hand... looked like reductions and clutches with 98% efficiency at 10rpm maybe... very quickly turning into an eddy current heater as RPMs increase.

It's kinda neat, but when a gear stage can easily (and normally does) last the life of the vehicle with no hassles, and each gearing stage is also in the 98-98.5% range, and you can make it from, and locate it near metal/conductive parts, I just don't see much appeal.
 
paultrafalgar said:
dequinox said:
:shock:

Freaking cool! Now all we need is magnetic chain drives!! :lol:

If you know me, you will know that I like to throw in daft ideas in case some genius is listening and can use it as lateral thinking fodder.
So, dequinox's "daft idea" suggests this to me:
If you had a rubber belt with a continuous series of magnets stuck on the outside, with the said belt arranged just like an ordinary bike chain over chainring (toothed to engage with belt) at the front and "derailleur" at the back (maybe 3 toothed cogs to engage belt), you could have a tube around the top of the belt with electromagnets imbedded in it and a controller that would switch them on and off in such a fashion to drive the belt forward (bit like a linear motor).
No need for dual freewheels at the rear and no need for a secondary system separate from the pedal-power transmission. Regeneration is easy and "cogging" is no problem, because when you aren't using the motor the electromagnetic coils are not energised.
What do you think?

I think the problem with that would be the belt would tend to get sucked into the drive magnets unless there was a bunch of rollers or something to prevent it.

It might work better to use a disk as the motor (axial flux motor) and have that drive a second disk for reduction.

I've been using 'magnetic gears' for my expandable implant design. The eddy current losses are really not too bad in materials like stainless steel and titanium, at least at speeds around 1000 rpm. By directing the flux away from the metal, you can greatly reduce the eddy currents. In their disk-gear example, if the outer disks have iron backing between the magnets, there won't be a lot of flux passing through the metal between magnets. Still, some kind of non-conducting material would be better. I suppose you could use laminations too.
 
liveforphysics said:
What they neglect to mention are the eddy current losses.

Those magnet holders an anything in motion through the magnetic fields, or anything with magnetic fields in motion against it all must be non-conductive or this thing is a giant eddy current brake.

Every one of the models shown would make excellent heaters, but very poor efficiency gear reductions for things moving in the 4 digit RPM range.

It appears they are working on a Kevlar encased magnetic gear box...page 8 http://www.magnetpowertransmission.com/web-storage/webstorage1/MGT%20applications.pdf hmmmm...10 years of R & D and a patent to cover the cost. Looking forward to see if this actually becomes a viable production unit.

I can picture where this frictionless gearbox would fit...imagine a supersport bike with a high speed compact clean burning turbine with air bearings like http://capstoneturbine.com/news/video/view/whisper.asp and a frictionless gearbox...imagine the torque and speed at 96,000 rpms and frictionless!

I used to commute to work 100 miles per day on a 1994 ZX-6e. In 6 years from 1994-2000, I logged a little over 58,000 miles on the odometer. I rode it hard at times when I used the power to speed away from sport cars that tries to kiss my back tire in the carpool lane...drop down to 5th gear at 70 mph and roll the throttle to 14,000 redline and it's see ya to that BMW M5 or Corvette. Running a motor at high speed takes a severe toll on the transmission and the reciprocating 4 stroke short bore motor...a little over 58,000 was all it could take before it started to burn oil and make clanky grindy noises when I shifted through the gears. I then purchased a 2000 Yamaha R1 that walked away from me ownership in 2 short weeks a la little sticky finger gremlins...but that was another long winded nightmare of a story that convinced me to give up the sport.
 
Hi Luke,

liveforphysics said:
Those magnet holders an anything in motion through the magnetic fields, or anything with magnetic fields in motion against it all must be non-conductive or this thing is a giant eddy current brake.

Every one of the models shown would make excellent heaters, but very poor efficiency gear reductions for things moving in the 4 digit RPM range.
magnomatics said:
Higher efficiency than conventional gears…

Torque densities comparable with mechanical gears can be achieved with an efficiency >99% at full load and with much higher part load efficiencies than a mechanical gear. For higher power ratings a magnetic gear will be smaller, lighter and lower cost than a mechanical gear.…

A high degree of magnetic coupling between the rotors allows for torque densities comparable with mechanical gears to be realized.…
Even accounting for some degree of hype you would think the efficiency should at least be somewhere in the ballpark of mechanical gears.

A possible fly in the ointment is what constitutes "higher power ratings" and if 5kw (for example) doesn't meet the criteria what are the size, weight and cost?

liveforphysics said:
It's kinda neat, but when a gear stage can easily (and normally does) last the life of the vehicle with no hassles, and each gearing stage is also in the 98-98.5% range, and you can make it from, and locate it near metal/conductive parts, I just don't see much appeal.

Their low ratio Magnetic gears go to 50:1 with:
magnomatics said:
Very low acoustic noise and vibration…

Gear ratios of 50:1 down to 1.01:1 with almost zero torque ripple are readily achievable.…

For higher power ratings a magnetic gear will be smaller, lighter and lower cost than a mechanical gear…

Low ratio magnetic gears can be realized in a conventional radial field or in an axial arrangement when a thin “pancake” design is required.
Even if the efficiency is worse than mechanical gears they could be very very useful.

Imagine Gary's bike with an 4 turn instead of 7 turn (with the higher reduction possible no reason not to go with less turns and higher rpm) Astro 3220 coupled to a thin, silent, single stage 21:1 magnetic reduction unit for an output of about 350 rpm (roughly 40 mph in third gear with 20" tires). More power, quieter, no 11t/~90t hassles…

Imagine a thin, silent, single stage 50:1 magnetic reduction unit coupled to your motor (you mentioned something like 30k rpm at high voltage). 600 rpm output! Even if its only 95% efficient with only a single stage it will be comparable to two or three stages at 98-98.5% per stage. Maybe you should build one for your motor :p . If you build a few in the range of 15:1 to 20:1 I'd like one :).

I haven't figured out how the CVT gearing works but, if its possible, a reduction unit integrated with a CVT or a 2 or 3 speed gearbox, could be a great Ebike reduction/transmission combo.
 
[youtube]EAELukfr2oY[/youtube] ...another magnetic gear with normal bearings...I wonder if high speed frictionless air bearings can be incorporated here :D
 
Back
Top