YokelessAndSegmentedArmature(YASA) topology- GT 30N·m/kg

MitchJi

10 MW
Joined
Jun 2, 2008
Messages
3,246
Location
Marin County California
Hi,

I'm hoping Todd or Miles or Luke will make a small run of Ebike sized YASA Motors :):
http://www.greencarcongress.com/201...ic-torque-above-30nmkgyasa-20100928.html#more
Oxford YASA Motors Shows Specific Torque Above 30N·m/kg
28 September 2010

Recent results from certified dynamometer testing has demonstrated that the 23 kg Oxford YASA motor produces a peak torque of 700 N·m (516 lb-ft), pushing its specific torque to more than 30N·m/kg. The company has also received the second tranche of £1.45-million (US$2.3 million) investment funding.

The Yokeless And Segmented Armature (YASA) topology is a new type of axial flux motor that has no stator yoke, a high fill factor and short end windings which all increase torque density and efficiency of the machine. The topology is based around a series of magnetically separated segments that form the stator of the machine. The novel design is enabled by using powdered iron materials that enable complex magnetic parts to be manufactured easily.

The YASA motor shows a step change improvement in torque density—initially with 20N·m/kg, or typically at least 2 times better than the best alternatives, according to the company. The improvement in specific torque comes from the combination of patented improvements in the magnetics, the cooling and the packaging of the motor.

We already have our sights set on the next generation YASA motor, which will push specific torque towards 40N·m/kg. This will enable the company to deliver a range of exciting direct-drive products with unparalleled performance.

—Nick Farrant, CEO of Oxford YASA Motors

Direct-drive electric motors are increasingly viewed in motor racing as the most efficient method of delivering quick acceleration as well as capturing a high percentage of energy from braking, th company said. Over the next 12 months, Oxford YASA Motors will be installing systems into new racing vehicles and TTXGP motorcycles.

The company is continuing to develop new markets for hybrid and electric vehicles, through ongoing collaborations with Delta Motorsport and Morgan Motor Cars in the UK, plus Electroengine in Sweden.

Oxford YASA Motors was founded in September 2009 to commercialize the YASA electric motor, developed by Dr. Malcolm McCulloch and Dr. Tim Woolmer at OXford University. McCulloch, head of the Department’s Electronic Power Group and Woolmer, then a PhD student in the group, originally devised the electric motor for the 2008 Morgan Lifecar.
6a00d8341c4fbe53ef0133f4ac4faf970b-800wi
 
MitchJi said:
Direct-drive electric motors are increasingly viewed in motor racing as the most efficient method of delivering quick acceleration as well as capturing a high percentage of energy from braking,


u reely otta delete this part for the good of the community.
if the non-hub crowd gets wind of this they're liable to slit their wrists.

oh yeah,
told ya so. :p
(along with many others in Costa Rica)
 
MitchJi,
My little axial motor design falls straight into the yokeless clasification & I think the project Kingfish is working on in the "doing the math" thread is the same. I should get back to my motor build after the hunting season.

Regarding marckcycles hub powerd motocrosser...I have a challange:
Bring that bike to Michigan with any rider you want to put on it....(I don't care if you hire Carmichal or Stewart) I will race it for 1 full lap on any full sized motocross track in Michigan, take your pick.....
On my 1974 125 Elsinore.

Better yet bring a hub motorcycle to toucson & run it against the non hub Zero's the track owner has.
It will be a great advertising oportunity for your hubs.
If you want to really test that re-gen we can go for a full 45 minute moto :lol:
Beat this fat old geezer & I will buy one & build my own.
Get beaten & you will have to only put these shamfull advertising plugs in the "for sale" section where they belong....realy...re-posting that VIDEO in this thread is truly self serving.

Other than Aj's constant slap fight with Hyena, I don't know why there is this rift between the hub guys & non hub guys....sure I bait Aj & am a member of m.e.n.s. But thats only because I KNOW a hub motor will FRY with the corner to corner style I ride the majority of my builds with. Mostly Off road & stop & start.
pffff.. methy had one smoking just lugging around in his drivway on his trike....
If I was going to comute I would really consider a hub as it is the "cleanest" way to power a bicycle.
Any where I ride a bicycle to in my neighbor hood I normaly grab My specialized & just pedal my azz there & back.
 
Thud, thanks for the encouragement :)

Recent results from certified dynamometer testing has demonstrated that the 23 kg Oxford YASA motor produces a peak torque of 700 N•m (516 lb-ft), pushing its specific torque to more than 30N•m/kg.
…
The YASA motor shows a step change improvement in torque density—initially with 20N•m/kg, or typically at least 2 times better than the best alternatives, according to the company.

Q: How is the mass measured? If we go from the image provided then perhaps it is constrained to:
  • Magnets
  • Windings
  • Iron
  • Electric Steel and/or Rotor Side
Yes? :)

I believe We can do better than that. I just ran a crude calculation using one of the earlier studies “Plan-D”:
(Read from here going forward)

  • Torque (Ï„) = 33.9 N•m per stator.
    Arc Magnet length = 20 mm with an OD of 200 mm, and 3 mm high.
    32 Poles; Halbach = n * 2 therefore 32 * 2 = 64 / side. Arc Width = 5.6* & mass = 3.93 g each =>
    64 * 3.93 = 251.52 g / rotor or 0.503 kg / motor
Copper Windings:
  • 30 Teeth (Windings) / stator; 3 sets of turns / Winding, each length being 0.945 m for 11 turns.
    Therefore 30 * 3 * 0.945 = 85.05 m of 24 AWG Flat Wire; Mass of 24 AWG = 0.00219 kg/m =>
    85.05 * 0.00219 = 0.186 kg / stator
There is no Iron in this AF design.
Rotor and Stator Plates:
  • Material of choice is Aluminum 7075 has a density of 2810 kg/m³.
    I used CAD to create solids modeling and calculated the volumes.
    Rotor Side Volume = 114019 mm³; Mass = 0.000114019 * 2810 = 0.32 kg
    Stator Volume = 54240 mm³; Mass = 0.000054240 * 2810 = 0.152 kg
Total Mass Calculation:
  • Magnets = 0.503 kg
    Copper = 0.186 kg
    Rotor Faces = 0.32 kg * 2
    Stator = 0.152 kg
    Therefore 0.503 + 0.186 + (0.32 * 2) + 0.152 = 1.481 kg
Thus a single stator AF Motor having Torque of 33.9 N•m, the Torque Density = 33.9 / 1.481
= 22.9 N•m/kg

That’s not bad for a single stator. :) The thing is – I am actually considering a double-stator, so the math changes just a little bit:

Internal Rotors:
Have much less mass because the amount of magnets is ½ and the size of the internal rotor has an OD the same as the Rotor face, but has a huge hole in the middle; we only need the facing to be large enough to capture the magnet.
  • Therefore the Internal Rotor Volume = 35966 mm³; Mass = 0.000035966 * 2810 = 0.101 kg
    Mass of extra magnets = 32 * 3.93 = 125.76 g = 0.126 kg
    Mass of extra Copper = 0.186 kg
    Mass of extra Stator = 0.152 kg
    Thus, for each additional stator, the mass = 0.101 + 0.126 + 0.186 + 0.152 = 0.565 kg
    AND we must include an additional Torque (τ) = 33.9 N•m
THEREFORE, a dual-stator AF motor as I have been discussing of late would have a Torque Density of:
(33.9 N•m * 2) / (1.481 + 0.565) = 67.8 / 2.046 = 33 N•m/kg

… and I have room left in the hub design to add at least one more stator for a bicycle.

But we’re not done with the math and we haven’t cut metal. So without much ado, I crawl back to my cave... :mrgreen:
Cheers, KF
 
KF,
I love the DTM thread (wish I had more to contribute there) but from LAR (looks about right) your #s seem optomistic by a factor of maybe 3. (SWAG)
please don't take that as a wet blanket but as a:project grounding-devils advocat,or-keeping it real, or maybe a optomistic- sketic to provide motivation? :lol:

I am already impressed at the level you have taken this too & can only imagine a certain troll's frustraitions as he tries to digest all of your good work. (hi safe!) (you sick little tourist) I don't know what access you have to machine tools, but count me in if you need any parts machined or cast. That much I know I can do & would love to have my name added to the credits list on a project of this magnatude.
 
:cry: mmm <sniff> How could I be off by a factor of three??? :oops:

I does the math. Pretty simple math too. :)
CAD don't lie - only people lie :wink:

If ye sees a mistake, I shall gladly correct. 8)

...and yes, I yam an optimist. I have to be. <nods>
In this day of Age, someone has to step forward and light the candle of imagination and hope!
It's not me though: I'm just the candle-bearer :)

...and I'm not too bad at scraping wax off the floor either...

~KF
PS - Thanks for the contribution-machining offer! One never knows where this will lead...
 
KF, don't take my litle slip of negativity to heart.....I am the guy who blunders way more than I bubble. (maybe your biggest fan!)
I have no worries with Eric checking you on the math portions...but experiance has shown nothing works as well in hand as it does on paper. Cad is only as good as the operator (3d parametric modeling is prety dang good though). That said, I am rooting for an earth shattering, uncomprehenceable leap of kind of break through. (at the very least another viable option for the masses) It would be so cool to see it come from an mass think-tank forum such as this, over some privetly or gorvernment funded project.

...and yes, I am an optimist. I have to be. <nods>
In this day of Age, someone has to step forward and light the candle of imagination and hope!
It's not me though: I'm just the candle-bearer
This is why I am certain we will be good friends. I struggle to maintain true optimism every day. I really try though & that has to count for something. I have several freinds who come to mind as I read your posts, they are allways looking past the hear & now & into the future & asking the improtant questions on some point of interest. I admit I can be slow on the uptake but be certain I am rooting for the win.
cary on, I will clean up the wax & let you get back to more serious issues. Like building a controller I can't vaprise in a nano second with low resitance super motor. your faithfull minion, Thud
 
That looks very interesting. Powdered iron pole pieces should make for low eddy current and hysteresis losses while providing the extra flux you get with a core. They might be brittle though. I wonder what they use to hold the core pieces to the frame?

Now how do you make those in the garage?
 
another article with video.

http://green.autoblog.com/2010/09/29/oxford-yasa-motors-axial-flux-motor-yields-money-monster-torque/
 
Hi,

Heres the video:
[youtube]7_0X1kpYhHE[/youtube]
Early bench testing of two 500Nm traction motors, running one (drive) against the other (regen). Motors designed by Oxford YASA Motors in partnership with Delta Motorsport.
 
Sounds less like a 'cordless drill' electric motor, and more like a subaru boxer motor!!
But with the ft-lbs of a large 7.0L+ v8 in 50lbs.. that's just crazy.

I like it! Looks like water cooling may be required though.. so no hub motor version i bet :(
 
Not really adding much to this discussion in terms of my own work (I'm still trying to learn this electronics stuff - I'm a mech eng student) but i did do a bit of research into the YASA stuff a while ago and found this white paper written by someone who helped on the design, it's pretty interesting and may be of help...
http://kron1.eng.ox.ac.uk/media/papers/Woolmer_paper_2007a.pdf

also, this guy this was inspired and tried to build his own YASA...he too couldn't work out how to do the powdered iron cores and went for the laminate (?) route...
http://web.mit.edu/scolton/www/SCThG.pdf

be warned, that second pdf is huge...he explains a lot of the basics (which helped me massively) and it also describes how he built his own controller
 
Searching around, I found some powdered iron cores that are already made. These are round, not pie shaped, so wouldn't be optimal, but round might not be too bad. They're the inductors used in crossovers for stereo speakers. They might need to be rewound, but you might find some that are already wound with the right wire.
Here's an example:
42410686.jpg

From Farnell. http://cpc.farnell.com/1/1/1072-3-3mh-high-power-crossover-inductor-lsif-330.html

Using these would be a lot easier than trying to make your own. There's also a few places like Micrometals that sell tubes rods and discs that can be combined to make a spool shape. If the the end discs were large enough, they could be machined into pie shapes. The pieces can be glued together with epoxy or bolted somehow.
 
I think those inductors are ferrite cored, not powdered iron. The flux density and so also the torque density of the motor would be low.
The permeability of powdered iron is low so open shaped cores may be hard to find, usually they are toroids or other closed shapes.
 
Hi,

MitchJi said:
I'm hoping Todd or Miles or Luke will make a small run of Ebike sized YASA Motors :)
Thud said:
MitchJi,
My little axial motor design falls straight into the yokeless classification & I think the project Kingfish is working on in the "doing the math" thread is the same. I should get back to my motor build after the hunting season.
What I was hoping for was a small run in which Mitch gets to buy one for a nominal price :D.

marckcycles said:
Done correct direct drive in wheel motors can impress

Thud said:
Regarding marckcycles hub powered motocrosser...I have a challenge...

Get beaten & you will have to only put these shamfull advertising plugs in the "for sale" section where they belong....realy...re-posting that VIDEO in this thread is truly self serving.
Mark seemed to miss the whole point of this motor. Gary said when he switched from a Crystalyte 5000 series to a Cyclone driving through a 3 speed hub he got better performance with 2,800 watts than the Crystalyte with 5,000 watts. Of course having a Crystalyte is like having a small boat anchor in your rear hub. The main advantages that YASA claims make this a viable hub motor are the light weight and high power. This would overcome the lousy Crystalyte performance and the boat anchor weight. If I were trying to sell a Crystalyte based hub motor this is one of the last places I would post a video claiming its "Done correct".
 
Ok.

I like this design. :)

And it looks too easy to make to pass up making it.
 
hmnn.
I posted a link to this design & a technical write up on it in the Axial motor discusion thread. I guess some guys missed it. I think the point of signifigance is the lack of the full steel laminated core all sharing the copper winds thus reducing any cross hysteies & other hi-tech terms that don't come to mind of this second.

It looks like a decent design & given the lack of a central stator I kind of grouped my mule, & lukes design into the same "yokless" catagory.

I think the design Miles has follows a bit more traditional axial stator philosophy. with the coiled stator laminations.

Mitch, if my litle motor works & actually functions as a viable bicycle motor, you wll get the 1st unit for testing. look for more updates in that thread after we get an inch of snow here in MI. Especialy since I have my cnc running now to produce stator blanks with the accuracy I need. I am getting inspired again by KFs threads & am learnig way more than I am contributing again (if ever LOL)

I guess no one saw my challange but mitch......it was total spam re-posting that in this thread. :mrgreen:
 
Hi,

Now, Fully Charged, the excellent video series from British plug-in car enthusiast Robert Llewellyn takes us inside the company's shop for a closer look at these axial flux motors...

A quick visit to the YASA motor works to see the incredible developments in electric motor design and construction...
[youtube]6m_iIbX0gmA[/youtube]
 
Back
Top