HyMag tech boosts magnetic flux density of permanent magnets by 10-30%

neptronix

Administrator
Staff member
Joined
Jun 15, 2010
Messages
17,534
Location
Utah, USA
This just in today.. Argonne lab has invented a new kind of magnet that uses far less rare earth materials and can increase power per lb. on our motors. This might lead to some EV makers backing away from induction motors.. as permanent magnet motors will get a significant edge in power density and efficiency..

https://www.greencarcongress.com/2019/01/20190120-hymag.html
 
Hey cool and thanks for sharing that. This is a refreshing break from all "new battery" breakthroughs we're used to hearing about. There are several other aspects of the EV drivetrain that are all deserving of improvements and it's good to see permanent magnets getting some R&D love too!
 
There's also PM wire announced a few months ago as well, seems to be some sort of wire that can be pressed into shapes and selectively magnetized or something. https://www.designworldonline.com/pm-wire-magnet-manufacturing-technology-could-change-industries/

They didn't really explain a great deal it's unclear is they are manufacturing it now or just trying to get some large OEM to use it.
 
https://www.anl.gov/article/stronger-lighter-greener

Possible link to the original article? OP link roboplagiarized to the point of nonsense. I mean, "Neodymium" somehow changed to "Niobium"? Same things only completely different, nobody cares if we change a few facts...
 
https://www.electronicsweekly.com/news/research-news/rare-earth-free-magnet-made-from-cheap-materials-2016-05/

Its nothing new that various thin films can be made twice as powerful as bulk Neodymium Iron Boron. But the trouble is they are just thin films.

Many rely upon strain, which cannot happen except with a different substrate many times thicker. Doesn't create a useful bulk magnet.
 
neptronix said:
This just in today.. Argonne lab has invented a new kind of magnet that uses far less rare earth materials and can increase power per lb. on our motors. This might lead to some EV makers backing away from induction motors.. as permanent magnet motors will get a significant edge in power density and efficiency.
Biggest change I see here for Ebikes is that you will be able to get efficient hub motors at lower RPM's, since flux density limits maximum torque at low speeds. Good for motors on 29" rims.
 
justin_le said:
Hey cool and thanks for sharing that. This is a refreshing break from all "new battery" breakthroughs we're used to hearing about. There are several other aspects of the EV drivetrain that are all deserving of improvements and it's good to see permanent magnets getting some R&D love too!

NP! what i find amazing is that there's still room to improve on electric motors after around a century. Apparently 97% peak efficiency isn't enough. :mrgreen:
 
neptronix said:
justin_le said:
Hey cool and thanks for sharing that. This is a refreshing break from all "new battery" breakthroughs we're used to hearing about. There are several other aspects of the EV drivetrain that are all deserving of improvements and it's good to see permanent magnets getting some R&D love too!

NP! what i find amazing is that there's still room to improve on electric motors after around a century. Apparently 97% peak efficiency isn't enough. :mrgreen:

Well I'm still waiting for room temp superconductors to get us to a nice round 100!

Really for a while it looked like there was a moore's law of rising super conductor temperatures being discovered, and then it sadly just plateaued...
 
justin_le said:
neptronix said:
justin_le said:
Hey cool and thanks for sharing that. This is a refreshing break from all "new battery" breakthroughs we're used to hearing about. There are several other aspects of the EV drivetrain that are all deserving of improvements and it's good to see permanent magnets getting some R&D love too!

NP! what i find amazing is that there's still room to improve on electric motors after around a century. Apparently 97% peak efficiency isn't enough. :mrgreen:

Well I'm still waiting for room temp superconductors to get us to a nice round 100!

Really for a while it looked like there was a moore's law of rising super conductor temperatures being discovered, and then it sadly just plateaued...


RTS dev is still making breakthroughs! A couple of exciting options are in development!

https://phys.org/news/2019-02-navy-patent-room-temperature-superconductor.html
 
That's insane.

While we're at it, here's a picture from CERN showing the difference between a copper wire and a superconductor designed to carry 12500A.

800px-CERN-cables-p1030764.jpg
 
I'd like to see some more halbach array motors. Yes, they are more expensive because they are more complex to put together, but they are much lighter because you don't need a metal armature core, thus no laminations. So you could have very, very, light rotors. The weight would be mostly from the windings.

A halbach array is a configuration of the magnets so that it creates a unidirectional magnetic field. Field strength is ideally 2x greater, but in the real world we aren't that lucky, lol.

So that array configuration, with HyMag modified magnets, using PM-Wire for ease of construction and maybe new shapes (curved for more surface area?), could make a pretty awesome PMSM/BLDC motor.
 
atarijedi said:
I'd like to see some more halbach array motors. Yes, they are more expensive because they are more complex to put together, but they are much lighter because you don't need a metal armature core, thus no laminations. So you could have very, very, light rotors. The weight would be mostly from the windings.

I think you're confusing the stator iron with the back-iron iron. The halbach array eliminates the need for back-iron behind the magnets, but it doesn't get rid of the laminated iron in the stator core which is where most of the iron weight in a motor is located. For instance, in our Grin motor, the back iron ring is about 500g, while the stator iron+copper is about 2kg. You still need some support ring for the magnets in a halbach array for mechanical reasons if not magnetic reasons, so the overall weight savings in the end is pretty small.

Ironless motor as really cool, not just for being light weight but more importantly for not having any iron losses. Maxon in switzerland builds motors like this
https://www.maxonmotor.com/medias/sys_master/root/8815461597214/DC-Technology-short-and-to-the-point-14-EN-30-31.pdf?attachment=true
And I believe that their one ebike hub motor product is similarly based on an ironless technology
https://www.maxonbikedrive.com/en/home.html

The downside is usually that much higher motor RPMs are needed to achieve a given power density.

neptronix said:
While we're at it, here's a picture from CERN showing the difference between a copper wire and a superconductor designed to carry 12500A.
file.php

Man if only we evolved on a planet where ambient was like -180oC
 
Halbach is no magical replacement for backing iron. Nd2Fe14B is 72.3% Iron by weight. Twice as many magnets are then needed to form the array. So much for getting rid of iron...

"Iron losses" aren't a problem in permanent magnets or their backing. Because a constant flux doesn't generate eddies. If an eddy might begin, it fights to further hold the flux constant. Nd2Fe14B conducts more like pure iron than silicon steel, but doesn't want lamination or insulation. Shorts inside and between permanent magnets are not a problem. Rings could even be added. The nickle plating for example, hurts nothing.

If you choose to back permanent magnets, just go with plain iron. Saturates higher than silicon steel, and flux pinning eddies swirl with less resistance. Low flux change = low hysteresis and magnetostriction. Where eddies here are helping, no reason to laminate or resist them into non-existence.

Save laminations for the stator. Where huge changes of flux purposely must occur, thats's where shorts and eddies work against us.

-------------------------

Maxon doesn't appear to be ironless either. Look carefully at the housing #3 (magnetic return), there is your iron. But none in the rotor where flux alternates. Almost wrote stator to be consistent with paragraphs above, but Maxon windings spin, so its a rotor.

However, Maxon's gap is both wide and broad. Flux in such a gap is not concentrated. Going to need amazing RPMs to make up lost torque. Fine, except how to gear high rotor speeds down to reason without mechanical noise and further losses?

Precious metal commutator brushes, seriously? Can't help but think Maxon should have tried unbroken rings like an alternator. Unbroken rings wear brushes slowly, because they do not commutate. Better done by MOSFETs. Spin those, and you only need 2 rings. Or conductive roller bearings if such things exist. Motor shaft wants two bearings anyway.

------------------

If you wanted to replace half your backing iron with cobalt, that might make actual sense. Cobalt barely weighs any more than iron. Supermendur is such an alloy, 49%+49% with 2% vanadium for extra strength. Can handle 2T, so it may not need as much weight to complete a magnetic circuit. Compare to 1.4T for silicon steel. Or 1.5T for mild steel, reasonably close to pure iron.

Yes, I edited this a lot of times, its not a miscount...
 
justin_le said:
Man if only we evolved on a planet where ambient was like -180oC
Then our battery tech would probably be a mite problematic. ;)

(but then, we might have started out on a totally different basis for that, too)
 
So, my suggestion of roller bearings in place of Maxon's brushes might not have been completely daft. Not sure how much current Bosch intends to push through these? Perhaps this illustration was just about bleeding some unwanted voltage that accumulates.
bosch.jpg
 
KD5ZXG said:
"Iron losses" aren't a problem in permanent magnets or their backing. Because a constant flux doesn't generate eddies. If an eddy might begin, it fights to further hold the flux constant. Nd2Fe14B conducts more like pure iron than silicon steel, but doesn't want lamination or insulation.

In fact there will be eddy currents induced within magnets and back iron, because the rotating magnetic field from the stator is spinning quicker as the rotor itself (ERPM vs. RPM).
Did you never had an outrunner motor where the rotor got warmer as the stator when it was spinning at no-load?

Shorts inside and between permanent magnets are not a problem. Rings could even be added. The nickle plating for example, hurts nothing.

Nickel plating usually contains also a layer of copper. The layers are quite thin (a few µm), but both nickel and copper cunduct electricity many times better as the material of the magnet and therefore the eddy losses within those layers will also be many times higher.
For this reason on really expensive motors sometimes magnets with "raw epoxy" coating are used.
 
KD5ZXG said:
So, my suggestion of roller bearings in place of Maxon's brushes might not have been completely daft. Not sure how much current Bosch intends to push through these? Perhaps this illustration was just about bleeding some unwanted voltage that accumulates.
bosch.jpg
The latter: What that is actually about is charge buildup (static, etc) across bearings that could potentially cause them to be "eaten" over time by effects that can look similar to galvanic corrosion (which is different metals in contact with each other), if arcing can occur between bearings and races (if there is any gap or insulation between them, like grease). If the lubricant itself is also sufficiently conductive, this shoudln't happen anymore.

You would nto want to pass significant current thru the bearings (although technically you could).


At one time I used my frame on Crazybike2 as system ground, particularly for the "12v" lighting. This eventually caused problems with the headset bearings because of the significant currents flowing for things like the car halogen headlight. At some point, for other reasons, I ran actual ground wires for everything, and replaced the bearings (with other used bearings), and stopped having this problem. At higher voltages, I expect the problems would have cropped up quicker and/or been worse.



If you did want to use axle bearings as brushes, you would have to electrically isolate the bearings from the axle (and the metal of the side or end cover, or frame), or you could not use more than one brush, and wouldn't have a current path. Once isolaed from the axle, you could then use one bearing on each end of an axle as a pair of brushes, but if you need more than two you'd need separate bearings for each "brush".

Since running significant current thru them would be problematic in them performing their job as bearings over time, youd' also want to put a pair of "real" bearings that provided the actual load support. Those would make a barrier to putting wiring to or from the brush bearings, so making the system more difficult to engineer.


It's much simpler and cheaper to use regular brushes, for any significant currents.
 
madin88 said:
In fact there will be eddy currents induced within magnets and back iron, because the rotating magnetic field from the stator is spinning quicker as the rotor itself (ERPM vs. RPM).

Stator field leads the rotor field by some angle of course, but is not spinning quicker. Maybe you are thinking AC induction motor.

Backiron eddies are helping. Lam those currents away, and permanent fields are more easily shoved out of the way. Giving less torque. Best you could do is plate those pieces in silver, keep those fields right where they belong. Since fields are permanent, no RPM slip is required to maintain them, but they can be pinned to make them more effective.

Why do the best loudspeaker motors (lowest Q, highest half-space efficiency) have unlaminated pure iron poles with shorting rings? Same reason.

Location, location, location. Don't want any shorted turn or eddy attached to a part intended to make an alternating field. There is hurting not helping. We recently saw in another thread. He'll figure it out eventually...

----

Too bad about the bearings. Anything has got to be better than mechanical commutation, no matter what exotic stuff you make the brushes from. Carbon brushes have always worked fine with a smooth unbroken ring. Four brushed rings or two, depending where you care to place the commutating MOSFETs. Spinning FETs might be more trouble than extra rings. I don't know offhand of any examples, and there may be reasons...
 
I'm working on it! Give me a minute. :evil:

Mercury was the 'bearing brush' of choice back in Michael Faraday's day. Toxic for sure, but might be a consideration,
for an idea.
 
KD5ZXG said:
Stator field leads the rotor field by some angle of course, but is not spinning quicker.
I have to disagree. The higher the pole count, the quicker the stator field will spin in relation to the rotor or motor shaft, which means the magnets and back iron does also see an alternating magnetic field.

Following paper describes the effect of slotted back iron and segmented magnets very well:

View attachment wills_reducing_2010.pdf

Backiron eddies are helping. Lam those currents away, and permanent fields are more easily shoved out of the way. Giving less torque. Best you could do is plate those pieces in silver, keep those fields right where they belong. Since fields are permanent, no RPM slip is required to maintain them, but they can be pinned to make them more effective.
Related to above mentioned circumstances, it would make absolutely no sense to use magnets with silver plating, and on 99% of inrunner or IPM motors you will find steel laminations used as flux return.. For a good reason!
Manufaturers put a considerable amount of costs to lower eddy currents in all parts within a motor, so i don't think they would do anything useful in terms of performance.
 
Monkey see monkey do. Lams are always safe. Properly placed dead shorts can work significantly better. Not impressed.

AC induction motors work entirely on eddies and nothing else. What would be the performance of this motor without them?


What I speak of is a solution that uses both PM and eddies. Because we have PM, no slip is needed. And any attempted slip is fought by eddies that make the magnets seem even stronger. With no slip, we do need a control circuit though, its not going to find its own lead angle.

Lams are needed where the field alternates, usually the stator. Not saying lams have no place. Only saying that place isn't anywhere, everywhere. As-if we had no power to reason where those places might be, and where they shouldn't be.
 
justin_le said:
I think you're confusing the stator iron with the back-iron iron. The halbach array eliminates the need for back-iron behind the magnets, but it doesn't get rid of the laminated iron in the stator core which is where most of the iron weight in a motor is located. For instance, in our Grin motor, the back iron ring is about 500g, while the stator iron+copper is about 2kg. You still need some support ring for the magnets in a halbach array for mechanical reasons if not magnetic reasons, so the overall weight savings in the end is pretty small.

Actually I'm not. Halbach array motors can be ironless, so you don't need rotor iron or stator iron. So no back iron or laminations.

Here is a company that makes such a motor > http://www.launchpnt.com/portfolio/transportation/halbach-electric-motor

Supposedly TFT FlightTech use an ironless halbach array BLDC in their UAV/Drone, but I couldn't find the press release they put out a while ago.

These guys talk about it a bit here > https://www.motioncontroltips.com/what-is-halbach-array-and-how-is-it-used-in-electric-motors/

Essentially you are making an air core. I read some papers on aluminum cores, and on plastic cores.

There is also a guy who has 3d printed his own halbach BLDC for a hobby plane, and he used magnetic PLA for the stator (outrunner), partially for the increased magnetic field, partially for better heat transfer.

Anyways, I think with a halbach array, and thicker magnets, it would totally be possible to make a hub motor mostly out of aluminum and carbon fiber, using no iron except in the magnets if they are NdFeB, or no iron at all if it's SmCo magnets.
 
Either type of magnet weighs more than iron. SmCo Gives a temperature advantage over NdFeB in return for brittleness and slightly weaker field.

Yes, we can get rid of backing iron. Replace with cobalt, or add more heavy magnets to make an array that needs no backing, but not always helpful to do so.

Again, alloying half iron with cobalt can give 0.5T improvement over either pure metal alone.
 
madin88 said:
KD5ZXG said:
Stator field leads the rotor field by some angle of course, but is not spinning quicker.
I have to disagree. The higher the pole count, the quicker the stator field will spin in relation to the rotor or motor shaft, which means the magnets and back iron does also see an alternating magnetic field.

I think one of you is talking about rotor-stator synchronisation (i.e. slip) and the other is talking about rate of stator field reversal (i.e. eRPM Vs RPM) :)

Interesting debate though. One the one hand you can argue for the status quo with "if it ain't broke, don't fix it", on the other hand a lot of established/standard practice is based on little more than "well, that's how we've always done it" rather than because it's been found to be the technically optimum solution you might expect.
 
Punx0r said:
Interesting debate though. One the one hand you can argue for the status quo with "if it ain't broke, don't fix it", on the other hand a lot of established/standard practice is based on little more than "well, that's how we've always done it" rather than because it's been found to be the technically optimum solution you might expect.
If laminated rotors would be just based on nothing more than
"well, that's how we've always done it" rather than because it's been found to be the technically optimum solution you might expect
then why should engine manufacturers make such huge efforts to lower rotor eddies by using segmented magnets and doing complex things like using ferrite as back iron (on outrunners)?

Tesla Model 3 motor does also have segmented magnets (4 single pcs per pole) to reduce the rotor losses.
It would not make any sense to do so if the rotor would not also see an alternating magnetic field.

KD5ZXG said:
What I speak of is a solution that uses both PM and eddies.
I have the meaning that eddies are nowhere good within a BLDC motor or motor that is using permanent magnets, but i might be wrong and there is something that i missed.
Do you have a link to a motor based on what you are saying, or to the source of information? It would help to understand :wink:
thanks
 
Back
Top