No Magnets, Big Power: BMW’s Fifth-Generation Electric Motor (2022) | Hacker News Reader

No Magnets, Big Power: BMW’s Fifth-Generation Electric Motor (2022)

motortrend.com

76 pointsclouddrover64 comments

These brushed motors generate fine dust like a brake pad in very slight but constant use:

> We asked BMW about the life expectancy of those brushes and commutators, and what happens to the dust as they wear. While they couldn't give us a lifetime estimate for the brushes, they did reassure us that the brush modules are "in an enclosed and sealed compartment, eliminating dust contamination inside the stator/rotor wiring." Anyone who has worked with brakes or even brushed DC motors knows just how fine those dust particles are. It will be interesting to see both how long they last and just how much dust those modules will actually contain (or release).

Interesting. I cynically view this (brushed motors with reduced lifetime as compared to brushless, sold as a benefit b/c free of rare earth) as just another self-serving decision automakers can make in the push to reduce the usable lifetime of EVs.

The longevity of ICE/hybrid vehicles has long been a bugbear for auto manufacturers. One of the convenient things for them in the push for EVs is that EVs will inherently have less utility at 10 - 15 years than ICE vehicles.

With manufacturers' standard EV warranty including 70% bat capacity at 10 years or XYZ miles, it's almost like an ICE manufacturer saying your gas tank will get 30% smaller after a decade. Except that an aging battery also has reduced current draw and thermal allowances, so there is also less instantaneous power available as the battery pack ages.

This is an engineering reality of battery EVs, to keep them safe as their packs age. The power limit could be viewed as consistent with older ICE's not performing at nameplate HP, which none of them will without significant maintenance.

But this is why I'm cynical about automakers and EVs: EVs lower their cost burden in the medium to long term with simplified production, and "enhanced" planned obsolence via wear components that are much more expensive and can significantly limit the utility of the vehicle in the 10 - 20 year time horizon, as compared to ICEs and maybe hybrids.

This doesn't pass the smell test.

They can already do that - they arguably are already doing that (*) - by making combustion engines which aren't reliable in the long term. And there are so many other things that you can design to be unreliable in the long term, especially in modern cars which have lots of features.

They're warrantied at 70% for 10 years because that's what the law requires. It doesn't mean you're going to lose 30% over 10 years. We could also see manufacturers put a better warranty on their vehicles as a selling point.

I haven't done much research into this, but I believe the existing EVs from various manufacturers have had very good battery degradation characteristics over time and over mileage. Not anything remotely like 70% over 10 years. If you look at technical documents and presentations that manufacturers create for industry trade shows, training their technicians, and in press releases for new models, they seem to genuinely care about battery degradation and mention all the things they're doing to ensure a long battery life. Obviously the situation might change in the future and companies may decide to become greedy, but for now this doesn't appear to be an issue. And as battery technology (& capacity) advances over time, this is going to be less of a problem in the future.

(*) Edit: That was poorly worded. I'm not saying it's common practice to intentionally design things to break, but that manufacturers who get greedy and cut costs are naturally going to end up making drivetrains that don't last 10+ years. Nissan's CVT debacle is a perfect example. The companies who would skimp on spending $$$ for battery management are already doing the same things right now with their ICE products. But what we've seen as an industry trend is that manufacturers for the most part don't intentionally sabotage their products; in fact cars have been slowly but steadily getting more reliable over time. Going from ICEs to EVs doesn't change this situation at all.

Another example is anti-corrosion coatings on the frame and underbody. They could easily save $$$ here with no noticeable effect to the customer until 10 years down the line. But across the industry, we see that luxury brands choose to spend extra $ on advanced coatings (when they could use the cheaper value coatings which the economy brands use), and the economy brands have also been upgrading to the newer coatings as cost comes down (rather than sticking with the older cheaper coatings). Just like how engines have become more reliable over time, the frames of modern cars last much longer than older cars. The industry in general is not sabotaging their products, they are genuinely trying to make good quality products.

>> making combustion engines which aren't reliable in the long term.

Who? Which engines are not reliable over 5/7/10+ years? Fiat? Some custom thing built in a German garage? All the big manufacturers are churning out amazingly reliable IC engines. Almost anything from Honda/Toyota/BMW/Merc will last decades. Maybe some sensor needs replacing or an electrical circuit has an issue, but with normal maintenance modern IC engines go forever, far longer than most new car buyer ever keeps the vehicle. The air conditioning, cruise control, radio, suspension, transmission and seat cushions will all be dead long before the engine gives up the ghost.

What breaks are things like sensors and control systems, little inexpensive parts that are generally worthless compared to the effort needed to diagnose and replace them. And that isn't going away with EVs. They have basically the same number of drivetrain sensors and systems as any IC car. Taking the car into the mechanic because of an engine light will still be a thing.

Maybe you will call the following anecdotal but my car has 280000 km and is completely worn out. Lots of expensive parts already have been replaced, including engine wiring. It's not a cheap Fiat but a more expensive Mazda. No, cars in general (the medium price segment, family cars) are not made to last much more than 10 years. One good reason for that fact is that the used technology becomes obsolete after that time. Another reason is fashion, people don't want to be related to an old car. Another one: they need to be as light as possible, hence sturdy enough but not military grade. Another, labour is expensive, so preferrably maintainance every 20k km or once a year. Cars have lots of pieces that don't need maintenance, e.g. ball bearings, but instead need replacement every x years. Same with the engine, nobody wants to bother with an expensive revise of that part. Electronics, the PCB boards/components are not produced anymore beyond 10 years. The list goes on. ICE or electric, the same arguments apply. Thinking of it, the same life expectancy holds for many appliances.

That begs the question why cars up until roughly the early 90's had an expected life time of 10-15y and have increased life expectancy now. Surely, it can't be hard to engineer parts of the engine or the chassis in a way that it breaks/corrodes after 10y. In fact, before zinc was used to protect the metal, that was one of the limiting factors.

I think the answer is competition, and I don't see why there should be less competition for BEVs which are simpler to construct than ICE cars.

Additionally, some managers claim that real-world battery life is much longer than expected, see https://www.forbes.com/sites/carltonreid/2022/08/01/electric...

This is nonsense. BMW has said the brushes in the motors should last roughly 15 years and the motors have been engineered to make replacing the brushes a fairly simple service.

Unlike bake pads, these brushes are conductive. So their dust should react to magnetic fields, which are everywhere in these motors. I would expect the dust to settle or cake onto any number of other metal surfaces rather than drift through the air like brake pad dust.

Even the title says "No magnets"

No magnets doesn't mean no magnetic fields. I believe they were referring to permanent magnets. An electric motor without any magnetic fields would be a very odd device.

And a motor with a static magnetic field is just as weird.

Ion drive thrusters. Not sure how one would translate them into a car, although there have been some on model aircraft, but they are a type of motor with a static magnetic field.

Thanks for the clarification!

Conductive metals does not necessarily also mean magnetic. Aluminium is just one example, conductive but not magnetic.

The brushes are almost certainly graphite, which actually isn't a metal.

Also graphite is diamagnetic, so strong magnets actually push it away.

Since we're speculating, the brushes could also be some harder material, with Carbon Nanotubes impregnated, or graphene infused and coated, etc.

Also, although I don't have the numbers for your issue about strong magnets pushing it away, I've found you need REALLY strong B Fields to do this. In other words, are we sure this is an effect that is at play here? (I'm asking)

Or add magnetic plugs to capture the dust. https://www.lislecorp.com/oem-products-division/magnetic-plu...

Makes me wonder how "enclosed and sealed" this compartment is. On various types of motors there is actually an access hole that allows you to change out the brushes. If I was evil I would just seal up that huge motor, weld it shut and never allow the brushes to be swapped out, but I'd put a check engine light when the brushes start to wear down so that the customer has to buy a whole new motor.

While true, there's other stuff that generates much more tiny dust particles - tires. Heavy vehicle tire rub is cause to roughly 1/3 PM2.5+ air polution, considering all sources. (Used to work for a exhaust emissions systems manufacturer, I remember reading studies about this, can't give you source now). So, the small particle dust from such motors, I imagine, is a minuscule source to air polution, compared to tire abrasion.

There’s a brushed electric motor in your refrigerator. How often does that need to get serviced?

Yes, and that motor operates in a practically constant external temperature and likely fairly constant humidity, with very few shakings of the environment in which it operates - it's about as optimal for electric motor use case that could be imagined.

A car's motor is in the elements, and gets bumped around and otherwise has to keep working in a non-optimal environment.

I'd say if motor survives 15 years, that's plenty, as the kinds of folks who buy BMWs tend to buy new at least every 10 years, I suspect.

The OTHER thing people I think are missing in this discussion is: how much OTHER car tech is going to improve over, say, that 15 year time? Advanced driver safety tech alone may make buying a new car a wise choice, vs servicing an older, less-tech-capable vehicle, for certain market segments.

I’m pretty sure they’re describing a wound-rotor synchronous motor.

In that case it wouldn’t have a commutator per se but constant-contact slip rings. These last a lot longer since they’re not subject to arcing.

The advantage is that unlike with permanent magnets it’s possible to reduce the strength of the rotor’s magnetic field at will (although there are some hacks that can be done with field-oriented control that provide the same field-weakening effect).

The automotive industry is already familiar with slip rings, used for decades in alternators.

I wonder if the supplier of this motor also makes alternators...

Is there even a need for commutators, could you not always use slip rings and electronically commutate within the rotor?

Once you have electronic commutation it usually makes sense to just make the magnetic field rotate in the stator and keep it fixed in the rotor. That way you avoid putting hot electronics in an already difficult-to-cool area that may also need to withstand some serious g forces.

But you do need some way to magnetize the rotor (setting aside reluctance motors) so that probably means permanent magnets, an induction-powered electromagnet, or a conduction-powered electromagnet (via slip rings).

Old-school DC motors use commutators in times and places where electronic commutation is too expensive or hadn’t been invented yet. I’m not aware of any configurations where the commutator creates a field that rotates relative to the stator, probably because you’d need to send the power into and back out of the rotor so there are twice as many parts to wear out.

Magnetless motors lag behind magnet-ed motors for efficiency.

However you design it, a magnet creates a magnetic field with zero power use, whereas any other source of magnetic field generation ends up using electrical power, generating heat (in a part hard to cool), and usually heavier for the same torque.

I'm disappointed that BMW doesn't give efficiency figures in their press release, because that is what has been pushing everyone else to the various kinds of BLDC/SynRM machines.

The article claims the opposite. I haven't looked at BMW's press release, so I don't know where the claim comes from.

"It's a big advance in terms of the efficiency of the electric motor, while getting away from the use of rare earth minerals."

Their reasoning:

"Permanent magnet machines typically rank as the most power-dense electric motors, but there's no turning permanent magnets "off." Their magnetic fields never change, so when they're not being powered by energy in the stator winding, they're generating energy in those windings, which creates drag."

Yeah, but only by adding a clutch or something to prevent the drag when the motor is coasting. I'm a little confused by this, I thought the motor would turn into a generator in that situation.

It would be theoretically possible to have a magnetic 'switch' which turned the permanent magnets off when not needed. Sounds impossible but it's possible[1].

[1]: https://www.goudsmitmagnets.com/en/solutions/magnetic-handli...

Yes, and.... the time constant of an LR circuit = L/R -- it takes finite time to 'turn off' the electromagnets. And, the collapsing magnetic field's energy has to go 'somewhere'; and it's of opposite polarity to the current that created it.

In other words, yes, possible, but more complicated than one might think. For example, we could build a circuit that "sucks out" the field by jamming very high currents in the opposite direction through the coils. To do this needs even higher voltages that the motor normally uses.

So, yes, it's possible. Not especially easy for optimal efficiency.

Motor regen is good to recover energy when you want to slow down. But apparently the most efficient way to drive is to coast as much as possible, rather than always trying to regen when off the gas. These motors let you do that without needing a clutch.

I think BMW's implementation of one pedal driving uses front radar and map info to adapt between coasting and regen as needed to get the most efficiency.

Their existing ICE cars also have a coast mode where they disconnect the engine from the transmission when you're not giving it gas (to eliminate engine braking). That function is only active when you put the car in eco mode. It's surprisingly smooth and responsive, you can't really feel the engine disconnect and reconnect as you're driving.

So, in the magnet-less design, are they saying that you are more efficient because your motor won't pull away energy/speed when not powered? As opposed to a permanent-magnet design, where you have to constantly add energy to maintain speed?

That's one quick explanation for how a magnet-less design would be more efficient...

After driving an EV for a while now, coasting is the number one thing that drives me crazy when driving an ICE car. It just isn't something that my mind thinks about anymore, so when it happens, it's a bit jarring at first. My wife's profile in my car is set to "coast" mode because she is the opposite. Her primary car is an ICE, so when she is driving my EV, it is jarring for her too!

I guess, for me, this would really have an impact on highway driving. And with non-highway driving (changing speeds frequently), I'd rather have the better one-pedal control. But what I didn't see what how much of an impact this would really have for energy efficiency.

Maybe the end game here is a magnet-less design that has a longer highway range. So you'll have two different range numbers: city vs highway ranges. Which would be very similar to the city vs highway MPG ratings we currently have for ICE cars. But those are the numbers I'd really like to see.

> So, in the magnet-less design, are they saying that you are more efficient because your motor won't pull away energy/speed when not powered? As opposed to a permanent-magnet design, where you have to constantly add energy to maintain speed?

Yes. And the magnetless motors are still able to regen when you want to slow down.

Specifically, they're talking about mechanical drag when the desired power is zero - ie. Freewheeling.

Turns out in typical car use, the efficiency gains when driving outweigh the losses from this.

Well, the iX is just about the most efficient EV on the market (in it's style), so they've clearly figured something out.

And yet, magnetless generators are the mainstay of power (as in your electrical service) generators everywhere.

Only synchronous power generation without modern control electronics.

Pretty much anything where the generator speed isn't 50/60Hz fixed uses something with permanent magnets.

Is one of the big advantages of these no magnet motors that they have a reduced need on rare earth metals? I thought this was a big advantage for Tesla's announcement recently - which has implications for relying on foreign miners / allows for cost reductions over time as supplies decrease.

Can someone confirm my understanding - thanks!

You can achieve the same type of independence with AC induction motors - that's even what early Teslas used. Overall efficiency is worse though.

There's a lot going on lately in the magnet/motor space - all to wean us off rare earths.

Got it - so my original supposition is correct that were really trying to reduce (A) dependence on rare earths which will (B) give us more direct control on pricing in the long run. The trade off with that is that we have a lower overall efficiency of use.

BMW are very good at innovations using old/unfashionable tech and are happy to sometimes be a bit eccentric/different from everyone else. I have a BMW motorcycle with a "Boxer"[1] engine which is the product of a process of continuous refinement of an original 1923 design. It's an amazing engine design that is almost completely unknown outside of BMW. On the other hand, my old BMW motorcycles had a design for the indicator controls that everyone in the industry found really weird and ended up being abandoned[2]

[1] Horizontally opposed twin cylinder https://www.carolenash.com/news/bike-news/detail/a-brief-his...

[2] I actually liked the old design but w/e

I think this might validate Neil Saiki's (no longer with the company) view on brushed motors for their electric motorcycles (Zero Motorcycles).

https://www.youtube.com/watch?v=0hQ7S3QPSpI

They later moved to brushed motors, but I believe that was more for customer demand and not a direct engineering reason.

Who knows of the exact solution for brush lifetime in the BMW motor, but one could absolutely have a sweeper, a reservoir and brushes that could last 1M miles or more. Maybe the solution is to have brushes in a cartridge that could be replaced. Only an engineering problem. If the major ingredient into your system is iron, your asymptotic cost is lower than $1/lb.

I really appreciate Neil's engineering ascetic, it seems brutally honest.

> ..there's no turning permanent magnets "off. " Their magnetic fields never change, so when they're not being powered by energy in the stator winding, they're generating energy in those windings, which creates drag. This is why some EVs powered by multiple motors either use a clutch to decouple and idle a permanent-magnet motor when it's not needed, or they fit one asynchronous AC induction motor (which suffers no such losses) and one permanent-magnet motor.

This part is really interesting from a physics energy (thermodynamics?) standpoint.

That's really interesting that they picked brushes; Renault's EESM motors are similar but use wireless power transfer to excite the rotor windings. Maybe brushes can pass more current cheaply?

Yes, brushless necessitates more rotating mass plus the loss in the wireless exciter. Longer life (no brush changes or slip ring wear), but lower efficiency.

I wonder if a hybrid rotor would pay off? It would still need brushes, but the current could be less to make them last longer. Probably a problem with demagnetizing the fixed magnets or something.

Replacing the brushes as they wear out: fine.

But what about the dust accumulating inside and sticking everywhere, adding additional wear.

I guess we'll see how many iXs are still on the road in 5 years.

It uses slip rings like your alternator, not a commutator like a typical brushed DC motor. I’ve had alternators that have lasted more than 20 years, and they don’t have dust problems. There’s no arcing and there is a continuous contact surface, so there isn’t really any dust.

I wonder how much of this is due to legal issues. By switching to brushless, do they avoid and of the recent patents re the brushless motors that have dominated EVs?

Has BMW stopped using plastic parts everywhere? Nothing environmentally friendly about a disposable engine in a luxury car.

Why couldn't plastic be recycled for this purpose? Engine components seems like a perfect application for recycled plastic, whereas it can be challenging when it comes to food grade packaging, etc.

There's nothing particularly environmentally friendly about vehicle production in general, so it kinda seems to me that this point is grasping at straws.

> Engine components seems like a perfect application for recycled plastic,

I would expect the opposite, at least for any component which might get hot when the engine itself is hot, because thermoplastics soften when heated. Thermoset plastics don't soften, but as a general rule thermoset plastics aren't recyclable.

The issue isn’t recycling the plastic, the issue is an entire vehicle going to the junkyard when the planned-obsolescence plastic engine parts inevitably fail.

Plastic is not environmentally unfriendly. Tossing plastic into the ocean is environmentally unfriendly.

I have no expertise in that field, so perhaps I'm gonna say something wrong, but isn't plastic comming from oil ?

Extraction, transport, refining...

That doesn't sound environmentally friendly to me.

What else would you use? Metal? That's not a whole lot better for the environment. "Natural" materials like wood, glass, rubber are not a suitable replacement for many uses of plastic, or would require changes that would overall make it worse.

We could make car bumpers out of wood instead of plastic. Then you'd have heavier cars which use more fuel and are less safe (and more expensive). The amount of waste generated by these uses of plastic is negligible and well worth it when you compare to the alternatives. There are many, many bigger sources of plastic usage & waste which we should tackle first before we focus on durable non-disposable plastics.

For a long time I looked at it this way. But let’s be honest with ourselves: plastic is not going anywhere. It’s too useful and too integrated into the global economy. Oil extraction is energy intensive and environmentally harmful, but so are many other industrial processes. As long as the carbon it contains does not end up on the atmosphere or ocean, I don’t see why we should view plastic differently than aluminum, steel, cement, or ammonia.

And tossing entire vehicles into the junkyard when the plastic engine inevitably decays and fails is VERY environmentally unfriendly.

What give you the impression these are disposable motors?

https://jalopnik.com/bmw-engines-are-gigantic-pieces-of-shit...

While I don't doubt this person's experiences, I've never have a BMW so can't comment. But this has nothing to do with the new motor being designed to be disposable and it seems to mention the brushes are serviceable...

>It's clear that the battle for magnet-less motor supremacy is only just getting started, and BMW is answering a key criticism frequently leveled against EVs: relying on rare, sometimes unethically mined materials makes EVs worse for the planet. We're truly stepping into an interesting time in electric motor development.

Right, what about the huge lithium batteries though? they're not exactly ethically sourced either...

Cobalt-free lithium batteries are expected to be the most common EV battery in the near future.

LiFePO4 batteries are already the most common EV battery in China. For high future performance applications, Tesla's high nickel chemistry doesn't use cobalt.

> LiFePO4 batteries are already the most common EV battery in China

Pretty common in the US as well, in terms of total EVs sold. I wonder if we will see almost all regular EVs switching over to LFP by the end of this decade.

Using brushed motors in 2023 is really stupid. Yes I know there is a semiconductor shortage, I know Silicon Carbide is expensive and silicon is not very efficient. But seriously. Batteries are also expensive. And brushed motors are very inefficient.

This is just another sign that BMW is way behind in electrification.

Reply on news.ycombinator.com