DIY Electric Car Forums banner

1 - 20 of 23 Posts

·
Registered
Joined
·
4 Posts
Discussion Starter #1
I am very new, and I am not sure where to start.
Do the inboard SRM motors ever use the 5 phase 10/8 arrangement?

I guess I am asking what is the current industry thinking on SRM pole count and arrangement?


What systems, if any, use on outboard rotor? thinking of a hub style or in wheel motor.

I am trying to correlate what I learned in college, with where the EV practice is headed. in school, we basically learned reluctance motors where simple and then spent all the time on induction motors or sync generators or condensers.

any direction on the EV practice or roadmaps where it might be headed is great.
thank you and have a good day
 

·
Registered
Joined
·
5,028 Posts
By SRM, I assume that you mean switched reluctance motors. While there is some heavy equipment using switched reluctance motors, I am not aware of any in road-going EVs, so there is no established common practice for SRM configuration in EVs. There are good reasons that they spent that time on induction and synchronous motors... ;) Some synchronous motors are described as a hybrid of permanent magnet and reluctance (because their rotors have distinct paths of reluctance), but they're not switched reluctance at all; TM4 calls this "reluctance-assisted permanent magnet".

While EVs of a decade ago commonly used 3-phase induction motors, normal practice for the current generation of EVs is 3-phase permanent magnet synchronous motors... usually with internal (rather than surface) magnets, and always in an inboard mounted "inrunner" (outer stator) configuration. There is a bit of return of induction motors for selected applications such as one axle of some AWD vehicles which power only one axle under low-load high-speed conditions, since the induction motors are more desirable when spinning without power.

While many companies (none mainstream) have promoted various types of in-wheel motors, none make it to mass production. They're too heavy, and have too many practical issues, for cars and light trucks. There are some axle-mounted motors for buses and trucks, but even they are not in-wheel. But yes, there would be some logic to using an outrunner (outer rotor / inner stator) configuration for high torque in this case. The large low-speed motors from TM4 are outrunners (they call this "external rotor topology") but this topology is rare.

The only trend that I have noticed (but I'm not an authority at all) to change motors from the current internal permanent magnet 3-phase synchronous designs - other than the induction motors noted earlier - is the trend to reduce the amount of rare earth material used for the magnets; this is due to the cost an environmental issues of rare earth production.


I had to look up the "5 phase 10/8 arrangement" to realize that it is a description of the pole count. Rotor pole count is twice the phase count, and stator pole count is rotor pole count plus two. A "10/8 arrangement" has 8 rotor poles and 10 stator poles, making it a 5-phase motor. So I learned something today. :)
 

·
Registered
Joined
·
5,028 Posts
https://teslamotorsclub.com/tmc/threads/switched-reluctance-motors.49916/page-2

I thought everyone was switching to reluctance motors to get away from rare earth dependence since china is the leading exporter
What "everyone" is moving to is not to any switched type (they're still synchronous), and not to straight reluctance. As I mentioned, there is a trend to reluctance-assisted rotor designs, which are still primarily permanent magnet.

That linked discussion quotes a tweet from Musk. It is possible that this is a switched reluctance motor assisted by magnets (which is interesting if the intent is to get away from rare earths), but even if it is, it is a exception to any trend. Believe it or not, Tesla is not "everyone". In two substantial ways, Tesla has been the exception to general trends:
  • cylindrical cells (everyone else uses pouch cells or prismatics)
  • induction motors (almost everyone else has used synchronous PM for a decade)
In both of these cases, they were just using the technology which could be readily purchased a decade ago when they started.

Whatever switched reluctance tech Tesla might be using, the details will be hard to find. Musk made a big deal about opening sharing technology... so maybe it's out there somewhere.

I also mentioned the desire to reduce rare earth materials. This can be accomplished by:
  • smaller magnets due to better rotor design (including the reluctance assistance)
  • magnets with less rare earth content, and even some ferrite type (no rare earths)
  • use of induction rather than synchronous PM rotors in some cases (such as the front of the dual-motor Model 3, according to the Musk tweet quoted)
 

·
Registered
Joined
·
1,478 Posts
good investigation, i learned something too.

"You must spread some Reputation around before giving it to brian_ again."
 

·
Registered
Joined
·
4 Posts
Discussion Starter #6
are the magnets used the SH type? I would think induction motors would be a better choice from the outset.
How do they prevent the motors from loosing magnetism in high temp areas like Arizona?

another question, what RPM range do the power trains typically run the electric motor at?
thank you again have a great day
 

·
Registered
Joined
·
5,028 Posts
How do they prevent the motors from loosing magnetism in high temp areas like Arizona?
I don't think the ambient temperature matters much, because all modern EV motors run high enough power density that they need active (normally liquid, often oil) cooling; their internal operating temperature can get pretty high, even in a temperate climate.

I haven't noticed anyone monitoring the internal temperature of a PM motor in use, but the builders of the Tesla-powered Cobra race car in this forum have been fighting overheating issues, and have reported internal temperatures of their motor's stator, which should be similar (after all, an induction motor and PM motor can have identical stators). Both rotor designs experience heating. When the internal temperature of the motor is more than hot enough to boil water, it doesn't matter much if the ambient temperature is typical of Alaska or Arizona.

I assume that one of the significant challenges of permanent magnet material design for motors is high-temperature performance, but that seems to have been managed in production EVs. Sorry, I don't know what the magic formula for this is.
 

·
Registered
Joined
·
5,028 Posts
another question, what RPM range do the power trains typically run the electric motor at?
Individual designs vary, but PM motors in production EVs and hybrids tend to run up to something in the range of 8,000 to 10,000 RPM. Induction motors can have an advantage at high speed (without unwanted excessive rotor field force causing high back EMF and requiring high supply voltage to the stator), so I guess it's not surprising that Tesla motors can run faster... up to 14,000 RPM if I recall correctly.

There are exceptions in production EVs; for instance,
  • very large motors (such as the Sumo series from TM4) are designed for lower-speed operation
  • motors directly coupled to engines in hybrids (such as the old Honda IMA and various add-ons to conventional transmissions that house a motor with the clutch or torque converter) will only run at engine speed, even though they might be capable of higher speeds; they tend to be large-diameter and axially short
  • the Chevrolet Spark had an unusual configuration with very high torque, and only less than 4:1 gearing to the axle, so it turns at about half the speed of a typical EV

Motors intended for aftermarket hobby use (that is, members of this forum ;)) are typically configured for lower speed, in part to reduce the need for high battery voltage.
  • brushed DC motors with series-wired field windings are typically salvaged from old forklift trucks, or built for conversion use but to the same design as "forklift" motors; they are presumably also limited by the brushes and commutator, and are typically unusable by 5,000 rpm (the forklifts run them much more slowly than that)
  • the induction motors offered for this market, such as from HPEVS, seem to be intended to replace those brushed DC motors, with similar operating speeds for peak power but better an extended usable range, up to roughly 8000 rpm; due to a lack of drive voltage, they can't produce rated power a high speed, so they would not normally be operated at the top of that speed range
  • the one synchronous PM motor aimed directly at this market (distributed by NetGain as the Hyper9) appears to be generally matched in speed and voltage to the HPEVS induction motors, since their goal is to provide an easy upgrade in low-voltage moderate-speed applications which have been using "forklift" motors

There are some motors which are somewhat available aftermarket and not used in production EVs, which can be various somewhat unusual configurations (such as the YASA motors, which are synch PM, but axial-flux). Larger-diameter "pancake" designs typically run below the typical 10,000 rpm of more common motor configurations.

High-voltage production EV systems enable production of rated motor power over the majority of the speed range, so they do not need multiple reduction ratios, and typically use a single-ratio two-stage reduction gear train to put the motor in this speed range. Lower-voltage DIY conversions, especially with brushed DC motors, often use the original transmission (the one which came with the gas engine) to do the first stage of reduction, and may be shifted to maintain performance over the road speed range; however, some builders with a large enough motor use only the single reduction stage of a final drive unit (as from the rear of a rear-wheel-drive conventional car).

In general, the maximum operating speed of any motor without brushes is limited by magnetic and electrical considerations, not mechanically. They can spin faster (although may need more expensive bearings to do it), but don't work faster (without design adjustments and typically more voltage).
 

·
Registered
Joined
·
5,028 Posts
Induction versus PM (and SRM?)

I would think induction motors would be a better choice from the outset.
Why? Just to avoid rare earth magnets?

Induction motors are cheaper, but the consensus appears to be that PM synchronous are more efficient, at least at moderate speed. There are advantages to every type, and since Tesla stumbled into using induction and yet became the leading EV manufacturer, they have been asked to explain their choice. There is lots of wild speculative babble in the Tesla fan forums, but more usefully there was an interview a few years ago with a Tesla motor expert, in which he explained some of the compromises... while carefully avoiding the question of whether Tesla chose induction, or just got that design feature as a consequence of the original business arrangement under which the company was formed.

In more recent public appearances, Konstantinos Laskaris as been identified as Tesla’s Chief Motor Design Engineer, and has provided comments on various motor designs.
Tesla’s chief motor engineer discusses the potential of next-generation motor technologies
Tesla’s top motor engineer talks about designing a permanent magnet machine for Model 3
If you dig online, you can find earlier statements from Tesla about induction versus PM (with no mention of switched reluctance), perhaps from the same person.

I note that in April 2017 (the first Charged EVs article) he essentially dismisses switched reluctance as just a future possibility, even though the Model 3 was supposed to be in production by mid-2017, so the Model 3 design must have been finalized, with motors rolling off a production line.

In a panel discussion reported in February 2018 (the second Charged EVs article) he describes the Model 3 motor (the rear motor, as this was before the AWD version was discussed in detail) as a "permanent magnet machine" without any mention of a switched reluctance design; in that discussion he has some comments about induction versus other designs.

The choice of design is not simple; as Laskaris said:
It’s well known that permanent magnet machines have the benefit of pre-excitation from the magnets, and therefore you have some efficiency benefit for that. Induction machines have perfect flux regulation and therefore you can optimize your efficiency.
It seems possible (likely?) that while all Model S and Model X motors are essentially identical induction machines (of two sizes),
  • the existing single-motor (RWD) Model 3 uses a synchronous PM motor, and
  • the coming dual-motor (AWD) Model 3 uses
    1. a PM-assisted switched reluctance motor (used full-time) at the rear, and
    2. an induction motor (used part-time) at the front
This may not be correct, but is the only way I can reconcile the widely-reported statements from Tesla (technical/management staff and Elon Musk).
 

·
Registered
Joined
·
5,028 Posts
are the magnets used the SH type?
You got me on that one - I had no idea what the "SH" type would be. Some searching...

... finds a page of
Neodymium Magnet Information from a supplier, which indicates that "SH" is a temperature rating class, suitable for operation up to 150 C. That seems reasonable for an EV motor, but I have no idea if this the class used in current production vehicles.

Not all rare-earth magnets use just neodymium, although that (NdFeB, or "NIB" for neodymium-iron-boron) is the type used in at least the Chevrolet Volt. A Motor Trend article claims that the Chevrolet Bolt motor uses dysprosium magnets; however, the same article incorrectly described the Volt and Bolt motors as switched-reluctance, so this is not a reliable source. It appears from a linked Motor Trend article that the author mis-interpreted a reference to "reluctance torque" in the synchronous PM Bolt and Volt motors as meaning that these were switched reluctance; similarly, the same author likely mis-interpreted a reference to dysprosium as meaning that these magnets used only dysprosium.

It appears from a quick search that dysprosium is used to substitute for only some of the neodymium in NdFeB magnets to improve characteristics, and that minimizing or eliminating the use of dysprosium has been an active area of development. According to a very informative Rare Earths 101 page from Ucore (a rare earth producer), of the many rare earth elements neodymium, dysprosium, holmium, and yttrium are all useful in permanent magnets... and dysprosium is especially useful but difficult to work with.

It is apparent that permanent magnet material technology is still evolving; the Wikipedia page for rare earth magnets doesn't even mention the use of dysprosium, even though it has become common in commercial production and is already on it's way out again (Ames Lab team replaces Dysprosium in permanent magnets with Cerium for lower-cost, high performance solution)!
 

·
Registered
Joined
·
5,028 Posts
"You must spread some Reputation around before giving it to brian_ again."
:D
Sorry, the tone of my second post may have come across as a bit... combative. badgers, you accidently hit a sore point of mine - the assumption that all things Tesla are automatically wonderful. :rolleyes:
 

·
Registered
Joined
·
4 Posts
Discussion Starter #12
thank you very much for the feedback. In response to your posts:

A "standard" rare earth magnet looses is field strength at 180F and in the presence of the stator field that will quickly breakdown the reminance.
if the ambient is 125F the cooling fluid will need to be some temperature above that. if you have a 15F Delta T that would result in 140F water. The rotor has to have some temperature above the 140 for heat to transfer.

The higher temperature rated rare earth magnets have lower strength ratings.
I would have presumed induction because I can get 1.6 Tesla in the airgap, where rare earth magnets can at best get 1 Tesla. Higher flux density acting on the rotor will allow me to shrink the size of the motor.
Vector drive technology has become cost effective and it allows us to extract maximum performance from the induction motors we use in industry. We also have easily 300F temperature ratings on insulation for induction motors.

I think the reason my perception is wrong is because I work in heavy industry and have no experience with what the practices are for EV.
I was surprised that the RPMs are so high.
I have to order special bearings and oil coolers for induction motors that I want to run up to the 10K range.
the US regulations puts a software cap on VFDs so we can't run the frequency above 300Hz so only the 2 pole motors get above 10K.
In practice I see more 4 pole designs so we only expect 8500RPM out of them. That seems to line up with your note about the brushed DC and their replacements.
Thank you again for all your feedback, have a great weekend.
 

·
Registered
Joined
·
5,028 Posts
I would have presumed induction because I can get 1.6 Tesla in the airgap, where rare earth magnets can at best get 1 Tesla. Higher flux density acting on the rotor will allow me to shrink the size of the motor.
I don't think that sufficient flux intensity is an issue for PM EV motors; to the contrary, excessive flux at high speeds seems like more of an issue, and reducing PM cost rather than improving PM effectiveness is a priority. While the first generation Voltec (GM transaxle for the Chevrolet Volt) used two motors with rare-earth magnets, the second generation uses rare-earth for one motor and ferrite magnets for the other.

Vector drive technology has become cost effective and it allows us to extract maximum performance from the induction motors we use in industry.
I think you'll find that the control techniques used for induction motors in EVs are comparable to those used in industrial applications. It's just software, so it scales down just fine. :)

Even if the intended controllers for low-speed vehicles and conversions were not sophisticated (but they are), the auto manufacturers would be right at the state of the art in motor control. A large auto manufacturer will spend amounts on development for a vehicle which would be the envy of major industrial suppliers, because their production volume is so high that they can justify it and the demands for performance are so high that they need to.

I think the reason my perception is wrong is because I work in heavy industry and have no experience with what the practices are for EV.
I was surprised that the RPMs are so high.
I have to order special bearings and oil coolers for induction motors that I want to run up to the 10K range.
the US regulations puts a software cap on VFDs so we can't run the frequency above 300Hz so only the 2 pole motors get above 10K.
In practice I see more 4 pole designs so we only expect 8500RPM out of them. That seems to line up with your note about the brushed DC and their replacements.
I can identify with the difference between industrial and automotive perspectives. Any automotive motor is small by industrial standards, but that makes higher speeds mechanically feasible. EV motors can be four-pole or even higher pole counts, combined with relatively high speeds - dedicated inverters are designed for the resulting frequencies.

The priorities and compromises of these different applications lead to different choices: EV motors are expensive per unit of output power, but also very compact for their power, because size and weight are very important and worth paying for. Industrial motors of several thousand horsepower are common, and work well, but for most applications wouldn't be economically feasible with the enormous amount of rare-earth permanent magnets that they would require.

Despite that, there are still industrial applications for rare-earth permanent magnets: wind turbines conventionally use gearing to step up the ~15 rpm rotor speed to a suitable speed for an induction machine in synch with the power grid (e.g. 1800 rpm + slip if a 4-pole), but some use a rare-earth PM synchronous design, especially the direct-drive wind turbines which run a machine of enormous diameter at that ~15 rpm. As of 2015, permanent-magnet synchronous generators accounted for a quarter of wind turbine production, and the fraction was rising. A typically sized (~2 MW) direct-drive wind turbine's generator - if using rare-earth permanent magnets - contains over a ton of magnets, and one-third of that is rare-earth material.
 

·
Registered
Joined
·
1,466 Posts
I was told by LTI that their srm motors were very closely held secret and they don't fit standard motor theory . And if he told he would have to kill me.
They have low inductance heating ( srm) on the rotor at start up as compared to induction.

They use a square wave ,so 10X less switching losses(95% eff. motor/controller combined) and can use a no shoot threw topology.

U.S.Navy subs are said to use srm.
They can run at extreme rpm
More phases less noise as they are loud.


LTI= Letourneau International
 

·
Registered
Joined
·
5,028 Posts
LTI= Letourneau International
As discussed when LeTourneau's switched reluctance motors came up a couple of years ago...

"LTI" is mining equipment company LeTourneau Technologies Inc, which was purchased by Joy Global, and now appears to now be part of Komatsu. They put those SR machines (generator and motors) into diesel/electric series hybrid wheel drives for machines such as the P&H L-950 Wheel Loader (their smallest model), which has a 1050 hp engine and an operating weight over 100 tons. I'm not sure how well motor characteristics scale down by a factor of a hundred, and concerns such as cogging are quite different in heavy equipment versus cars.

The specifications for the L-950 lists the B40A traction motor, and the SR motor shown in the brochure has 18 stator poles, so perhaps it is a 9 phase 18/16 pole design? Just guessing...

As a technical aside, I note that this machine is a diesel-electric hybrid, and Komatsu describes the energy storage as "Kinetic Energy Storage System", rather than a battery. Apparently their energy storage is short-term (it's not a plug-in), and they are using a flywheel.
 

·
Registered
Joined
·
8 Posts
I am very new, and I am not sure where to start.
Do the inboard SRM motors ever use the 5 phase 10/8 arrangement?

I guess I am asking what is the current industry thinking on SRM pole count and arrangement?


What systems, if any, use on outboard rotor? thinking of a hub style or in wheel motor.

I am trying to correlate what I learned in college, with where the EV practice is headed. in school, we basically learned reluctance motors where simple and then spent all the time on induction motors or sync generators or condensers.

any direction on the EV practice or roadmaps where it might be headed is great.
thank you and have a good day
My SRM (Switched Reluctance Machine) uses a U-shaped modular stator and a T-shaped modular rotor in a pancake configuration that will produce higher torque. The reason Tesla's SRM in the Model 3 uses magnets is bacause the location of the motors does not work well with a pancake configuration or, perhaps, they have not thought of that solution, placing the pancake motor on top of the differential. Besides the attach illustration, I have drawings and such [email protected] which I will be glad to download also if I get encouraged. I am an enthusiast but not in a position right now to start a project but will be glad to help where I can.
 

Attachments

·
Registered
Joined
·
5,028 Posts
The reason Tesla's SRM in the Model 3 uses magnets is bacause the location of the motors does not work well with a pancake configuration or, perhaps, they have not thought of that solution, placing the pancake motor on top of the differential.
If we can believe the description of this recent Tesla motor as a magnet-assisted switched reluctance motor, the use of magnets (in the stator) is a well-established SRM design feature, and would not be used only because the Tesla people "have not thought of" better configurations. While Tesla never does anything technically innovative in motors, they do have lots of staff who have lots of experience and spend lots of time and effort examining and evaluating all possibilities. The original question was about the direction of the industry - I suppose permanent magnets in the stator would be a current direction for SRM machines; pancake configuration is not as far as I have heard (which doesn't mean much).

The pancake motor on top of a differential (with a vertical pinion shaft) would package very poorly a car, and has no mechanical advantages. It is not an approach taken by any production vehicle, and I see no reason for anyone to consider it unless they are forced to use a large-diameter pancake motor for some reason.
 

·
Registered
Joined
·
8 Posts
For a "normal" SRM, the configuration, the quantity of stator/rotor poles is basically a matter of complexity and how smooth you want the motor to run. However, you may find that in the design that I have described, the SRM does not cog at all. I will attach some further details that you may find helpful.

Henry Keultjes
 
1 - 20 of 23 Posts
Top