Tesla switches to permanent magnet reluctance motors and sees range gains

[neptronix]

https://electrek.co/2019/04/05/tesla-model-s-new-electric-motors/
https://insideevs.com/news/348504/tesla-improves-motor-efficiency-increase-range/

Apparently Tesla got a ~10% range increase by switching from induction motors to what basically looks like the BLDCs we have in everything..

Seems that the ebike world was right all along..

 

[spinningmagnets]

Here's some info on that type of motor...

https://electricmotorcycle.com/switched-reluctance-motors/

 

[neptronix]

That's quite the post!

How about the cliff notes version? These motors seem awfully similar to BLDC.. what's the difference between the two?

 

[liveforphysics]

That's quite the post!

How about the cliff notes version? These motors seem awfully similar to BLDC.. what's the difference between the two?

Our ebike motors use the PMs for making the motors whole torque.

The Tesla motor just has a tiny amount of weak PMs, so it can cruise down the highway at light loads without having to build the rotor field through slip (like induction). When you want full torque, its using the rotor iron reluctance to make that big torque value. The beauty is, unlike our PM ebike motors that are always eating the hysteresis core losses of passing the strong magnets by the stator teeth iron, a SR/SRPM gets to impose minimal hysteresis and core losses when spinning at light loads (highway cruising).

Its also a synchronous style motor, so no slip% losses like an induction rotor requires to create the rotor field that the stator interacts with to make torque.

 

[neptronix]

That's a very good TL;DR, thanks.
Makes total sense for a car application.

Aside from reducing magnetic drag massively, are the motors more efficient when power is being applied, or about the same as our trusty permanent magnet motors?

 

[liveforphysics]

That's a very good TL;DR, thanks.
Makes total sense for a car application.

Aside from reducing magnetic drag massively, are the motors more efficient when power is being applied, or about the same as our trusty permanent magnet motors?

PM motors at full load can achieve ~97% when everything is done perfectly. When cruising at a light load, they might be <~90%

SRPM motors can achieve ~97% when everything is done right at full load, but then also cruise at light loads with minimal efficiency hit.

Why you don't want one for an ebike is rotor mass gets big when you want high shaft torque compared to PM motors.

If you were making a dragster, PM is still the champ, but for something you want to accelerate hard and then sip watts at cruise, SRPM is pretty amazing.

 

[Lebowski]

That's quite the post!

How about the cliff notes version? These motors seem awfully similar to BLDC.. what's the difference between the two?

Just like BLDC, switched reluctance motors make a backemf. Problem is that it disappears when there is no phase current flowing. The nice thing of small magnets is that they maintain some amount of backemf for the sensorless controller to sync to...

https://endless-sphere.com/forums/viewtopic.php?f=30&t=97880&hilit=ipm&start=50#p1435404

In the picture of the first post you can see that the inductance variation for the motor is very large (lots of big gaps between rotor and stator iron, no 'T-heads' like with BLDCs)

 

[bigmoose]

Pure genius the SRPM architecture. Great efficiency gain when a large portion of use is at part load. Saves on rare earth materials. The person who thought of this marriage of technologies has paved the way forward with a giant step. Thanks for the insight on losses Luke.

 

[Punx0r]

Saves on rare earth materials

This has got to be a big advantage besides the efficiency. Though it does make it harder for critics to argue EVs won't scale or will destroy the planet.

 

[nicobie]

Great thread and posts!

Threads like this are the reason I check ES every day. I might be getting old but I can still learn new stuff.

 

[neptronix]

Great thread and posts!

Threads like this are the reason I check ES every day. I might be getting old but I can still learn new stuff.

Only here would a discussion of motor topology bring so many great minds in the EV DIY space together.

 

[Grantmac]

So what are the disadvantages or issues implementing this tech into smaller vehicles?

 

[Hillhater]

Great thread and posts!

Threads like this are the reason I check ES every day. I might be getting old but I can still learn new stuff.

Hmm,? you must be getting old,...you missed this last year.. :lol:
https://endless-sphere.com/forums/viewtopic.php?f=30&t=95956&p=1405420&hilit=Tesla+motor#p1405420
And..
https://cleantechnica.com/2018/03/11/tesla-model-3-motor-in-depth/

 

[toolman2]

PM motors at full load can achieve ~97% when everything is done perfectly. When cruising at a light load, they might be <~90%

SRPM motors can achieve ~97% when everything is done right at full load, but then also cruise at light loads with minimal efficiency hit.

Why you don't want one for an ebike is rotor mass gets big when you want high shaft torque compared to PM motors.

If you were making a dragster, PM is still the champ, but for something you want to accelerate hard and then sip watts at cruise, SRPM is pretty amazing.

That's very nicely put Luke, 100% agree.
So what if rather than these 18 000rpm motors that are designed for around a multi stage reduction unit in a car, you designed an SRPM motor that was optimized for maximum torque per Kg (concentrated winding's not overlapping, higher pole count, grow the diameter less stator width etc, etc).
-Would this not be the way forward for general use and even racing (although non-dragster) bikes?

 

[liveforphysics]

Bigmoose! Always a pleasure to hear from you kind Sir!



Toolman- It might be...

The trick is in creating as short of flux linking path through the rotor, while maintaining the aggressive change in reluctance.

If you make a rotor that when viewed from the side has a bunch of "U" shaped iron segments radially laid out with the tips of the "U" facing toward the stator, you can get the rotor mass lighter for a given specific torque requirement. This also reduces the torque ripple which is another SRM drawback. Each time another set of poles is added on the rotor, the hysteresis loss increases for a given RPM though, so you don't want to get too overkill on distributing more rotor poles. Definitely has potential as a motor to hotrod, but trickier to drive than most.

 

[Hillhater]

...... designed an SRPM motor that was optimized for maximum torque per Kg (concentrated winding's not overlapping, higher pole count, grow the diameter less stator width etc, etc)......

There was a long discussion a few years ago about the Tesla motor regarding a similar thought.
As i remember, the conclusion was simple physics dictates that ..(all other things being equal , materials, input power, etc)..
.you can double the torque by doubling the axial length of the rotor and the weight increases directly in proportion (doubles in basic terms)
.. but if you make the same increase of torque (2X), by increasing the rotor diameter (2X) , but the weight increases 4X ( square law for volume)
This is likely why it is preferable to use a smaller diameter, higher rpm, motor , with reduction gearing to achieve the high torque at the wheels....and why few have successfully produced a DD or “wheel” motor with sufficient torque for an EV and at a realistic weight.
Obviously there could be design improvements/changes to offset some of that, but basic physics is against you.

 

[Lebowski]

.. but if you make the same increase of torque (2X), by increasing the rotor diameter (2X) , but the weight increases 4X ( square law for volume)

would 2x diameter not give you 4x torque ? 2x because of twice the arm (diameter) and again 2x because there are twice the amount of teeth ?

If correct this means torque goes by volume...

 

[liveforphysics]

Torque grows at the square of radius, linearly with length. Mass doesn't have to increase at the cube of volume if the shape is hollow inside (thin-gap, Joby, YASA). You can get any combination of electrical frequency you were getting with your gearing stage through increasing pole count and increasing radius in a single stage. Sometimes for unique applications it gets impractical to package.

 

[MitchJi]

PM motors at full load can achieve ~97% when everything is done perfectly. When cruising at a light load, they might be <~90%

SRPM motors can achieve ~97% when everything is done right at full load, but then also cruise at light loads with minimal efficiency hit.

Why you don't want one for an ebike is rotor mass gets big when you want high shaft torque compared to PM motors.

If you were making a dragster, PM is still the champ, but for something you want to accelerate hard and then sip watts at cruise, SRPM is pretty amazing.

Hi Luke,

When sandy Munro took the magnets apart they exploded because they were laminated together opposing each other. He thought that the reason for the performance boost was because the magnets were part of a hallbach array.

What would be the impact of the magnets opposing each other?

 

[flippy]

as fas as this black magic stuff its is done to direct/focus the magnetic field into the right area.
but contructing such motors is a pita due to the magnets wanting to implode. so you hardly see such motors as nobody wants to reinvent the wheel when there is a off the shelf alternative that is cheaper to design around.

because tesla started with a -real- blank slate and got designers and enginerds that are not constricted by fossil fuel company internal rules means you can push design limits and go into radical changes like this. no way you will see a motor like this on a volt or leaf in the next 5 years.