Hacker News new | past | comments | ask | show | jobs | submit login
New Electric Motor Could Boost Efficiency of EVs, Scooters, and Wind Turbines (ieee.org)
104 points by amynordrum 63 days ago | hide | past | web | favorite | 50 comments

These guys have definitely spent some money on PR since closing their investment round in April. They have placed 4+ articles with many of the same talking points, pictures and video over the last 2 weeks:





Funny to see the VC money being spent so tangibly and the timeline for creatives, relationships and placements.

And they don't really have anything useful. Their claims dont make any sense to someone in the field. Now I have to file ieee spectrum under "willing to pedal bullshit for money" like so many other publications.

Yea, not a good look for IEEE to be posting articles on buzz word vapor ware with pretty pictures.

It's probably not even money: it's just a reprinted press release. Once you know the style, it becomes obvious just how many articles everywhere about a new development from company X are just prints of company X's press release with some mild rewording.

It's mostly harmless; companies need to be able to announce products somewhere.

My own tell-tale about those is the use of the word "could" in the description. Not an outright disqualifier, but raises the likelihood of the content being a PR fluff piece.

> These guys have definitely spent some money on PR

And wasted it. Jalopnik? InsideDevs.com? CNET? Seriously?

I would have agreed... but if that all helped them book IEEE, then it was worth it. From a reputation standpoint, IEEE could give them quite a bit of cache.

I'm all in favor of more cache, but I think that was supposed to read 'cachet'...

Ha... thanks.

10% is nothing to sneeze at but it also does not mean it is a game changer. I like the way the coil ends are described, they are 'dead' weight in a normal motor because they don't contribute to the flux but they do contribute to the resistive losses in the motor. But that 10% of the length of the winding that sits outside of the coil chambers is not responsible for 10% of the electric consumption of the motor, which is not resistive in nature in an AC drive motor.

I do like the fact that they believe they can do without a transmission entirely, if that works it will save a good bit of energy through weight reduction, mostly during acceleration and deceleration. The bit I don't get is where they talk about DC-DC conversion, most electric cars that I know something about the guts of use tri-phase AC drive.

Making this work across a wide range of scales will be an interesting challenge, in my experience increased torque comes with its own challenges (such as shafts snapping or rotors slipping on or shearing off the mounts of the shaft they are mounted on).

From the end of the article: The company estimates that, in a large, 8-megawatt wind turbine, the HET could save 90 tons of weight, millions of dollars in costs, and lift efficiency by 3 percent. So maybe this is more about making electric motors suit their applications better and needing less support infrastructure than just looking at the % efficiency.

And then there is the 'could'. So it doesn't really mean much of anything without more substantiation. Btw, the 90 tons is based on gear box driven wind turbines, but SOTA has been direct drive for many years now.

> 10% is nothing to sneeze at but it also does not mean it is a game changer.

They give a quote on a practical application:

> The resulting Segway enjoyed a 50-percent increase in range, and four times the torque, which boosted the vehicle’s top speed and allowed it to climb a steep 20-percent grade.

If that scales up linearly to electric cars, that's quite a boost.

That's a lie. Or, at least, there's another existing normal motor that can achieve at least nearly the same results. Existing motors are already 90+% efficient; it's not possible to make a 50% increase in range.

I agree that they are at least misrepresenting their achievement. However, for the sake of argument, electric motor efficiency goes down quite a lot as overall power potential and RPM both drop. So it's possible they've come up with a sizeable efficiency improvement in a worst-case scenario, i.e., improving efficiency of a small motor in low RPM, part throttle situations.

If that's the case, however, it doesn't really bode well for them, as their efficiency gains will not translate to the high-RPM, high-power motors used in cars. I'm not EE, but I do know that most engineering involves making trade-offs. So I wouldn't be surprised to learn that techniques to improve low-power, low-RPM efficiency probably sacrifices high-power, high-RPM efficiency.

they did call out scooters and drones, so id guess you’re right

Possibly. You can get motors running 90+% efficiency, but only in a more or less narrow power band. If they can improve the efficiency in the ends of that bell curve, even if it's not an overall efficiency increase across the entire power/torque/rpm/"throttle" range, they might realistically be getting a 50% increase in "real world" use... Especially if they can do that in a way that removes the need for a gearbox (to keep the motor in it's highly efficient range).

That's a pretty big if. A Segway is very much not a car. The relative weight of motor + batteries is a much higher proportion of the total mass of vehicle+occupant than it is in a vehicle. So it may be a boon to lightweight electric vehicles where that is true, but the same will not hold in proportion if the vehicle were much more massive and the weight of the batteries were much more massive. And Segways are speed limited, as are most other vehicles, top speed is rarely even an issue for anything on wheels, we can go up to speeds where you have to use aerodynamics not to end up flying.

50% range increase would be a huge boon to a vehicle like a Boosted Board. My V2 throws off a lot of heat and gets nowhere near the advertised 11km range.

> My V2 throws off a lot of heat and gets nowhere near the advertised 11km range.

Heat losses come in many forms:

- resistive losses in wiring

- resistive losses in the batteries themselves

(their Ri is nowhere near zero)

- switching losses in the powertrain

- slip losses between the field and the rotor

If a piece of gear does not get its 'advertised performance' that can mean a defective product, deceptive advertising or plainly bad design, or a combination of all of these.

Having a very expensive motor isn't suddenly going to solve any of that, you'll get an advertised range of 15 km while being able to do only 10 and you'll still feel cheated.

It probably would go 11km on a straight and level run with no stopping for traffic lights or anything else. It's just that that's not realistic in an urban commute setting.

I don't feel cheated, I'm just saying that even a modest bump in efficiency could make a big difference to this kind of vehicle, especially if affected both the propulsion and regen side of things.

The motors in a Model S make up about 5% of the car by weight.

Ten, maybe 15 years ago there was a spate of ACM topics about scheduling algorithms. I have some speculations that Apple incorporated similar logic into their laptops and phones, based on some brags they made about battery life improvements.

The observation is this: The faster you draw power from a battery, the less efficient the battery is. Therefore peak shaving goes a long way toward extending battery life. So a motor that is 10% more efficient may increase range by 12%. Or give you the same range with a battery 15% smaller (~100 lbs for some EVs).

My take was the transmission, DC-DC converter, and 10% efficiency gain were essentially the same conversation.

Transmissions and converters all result in transmission losses. They're also more components to purchase, install, and maintain. So if they can build a motor that embodies some of those qualities, that's an improvement, even if the resulting motor were slightly less efficient (which they say is not the case).

If an electric car is using three-phase AC, then there's an inverter in there (because batteries do not come with three-phase AC). I don't believe that's as much loss as DC-DC but it's not nothing.

The loses from AC/DC conversion must be pretty low. A typical solar installation with battery backup (e.g. Powerwall 2) would convert from DC to AC in the solar inverter, AC to DC when charging the Powerwall, DC to AC when discharging the Powerwall, then AC to DC when charging your electric car (and as you say, most likely DC to AC when driving it).

The original Powerwall was DC, so they must have run the numbers and found out the impact was minimal to switch to AC.

> The loses from AC/DC conversion must be pretty low.

For most applications on the order of 2-4%.

You can't run an electric car on straight battery voltage, unless you want the car to go a single speed. So you have two choices: DC-AC conversion to run a 3-phase AC motor, or DC-DC conversion to run a DC motor at variable voltage. Either one is going to have similar losses with modern power electronics.

Couldn't you though?

This motor has 48 coils, and the ability to change what order and groupings they get powered. Doesn't that change the current substantially without altering voltage? I don't think you would ever run one coil alone, but that leaves at least 9 other patterns where all coils fire evenly.

And what is 'battery voltage'? Battery voltage is something like 2.8 volts, and that's not consistent across all cells at all levels of charge. Any voltage higher than that is due to composition. From what I recall, recharging multiple cells in series leads to some pretty big problems when some of the cells are aging faster than others. Being hard-wired is not a given.

With 48 coils on the motor and a battery pack of 120 or more cells, there are a lot of voltages and currents you could run through the system by reconfiguring instead of inverting.

>This motor has 48 coils, and the ability to change what order and groupings they get powered.

Ok, great, so now you have a bunch of discrete speeds you can run the motor at. How smooth a ride do you think that's going to provide?

If you can't run a motor at any arbitrary speed within its range, it's useless for any kind of vehicle.

>And what is 'battery voltage'? Battery voltage is something like 2.8 volts

No, that's cell voltage. A battery is a group of cells. That battery in your car that produces 12V is a group of 6 cells each producing 2V, with the cells all connected in series. As I understand it, Tesla batteries are rather high-voltage.

>From what I recall, recharging multiple cells in series leads to some pretty big problems when some of the cells are aging faster than others.

Yet somehow it works just fine for car batteries over the last 100 years, and for most every other battery we use these days (laptops, etc.).

>With 48 coils on the motor and a battery pack of 120 or more cells, there are a lot of voltages and currents you could run through the system by reconfiguring instead of inverting.

Yeah, again, if you just want to run at various discrete speeds. How many people are going to want a herky-jerky car like that?

My point is it could be done, and these guys specifically claimed they don't need a DC-DC converter, so either they're lying, or there's something you and I don't understand. I'm just speculating on ways they could solve that problem.

And you're giving so much salt back that I have to ask, do you believe they are lying? If so then don't beat around the bush. Just say it. You think they are frauds.

> Yet somehow it works just fine for car batteries over the last 100 years, and for most every other battery we use these days (laptops, etc.).

Every time I've replaced my car battery, it's been because one cell failed after I let it deep discharge. Except once where 2 cells failed.

I thought Tesla did something more sophisticated with their packs, but I'm looking at a picture here and there are for sure ~40 cells wired in series. At best there are 4 batteries per array (or they may be connected on the back side). But you don't have to use that bus architecture. It's mechanically simpler and more robust. It's the default. But it's a compromise, favoring physical robustness for reduced chemical robustness. Some day I hope to see other solutions, perhaps like what I described.

They absolutely are frauds. Every time someone comes up with some new "breakthrough" like this, whether it's a new motor, or a new internal-combustion engine, it's a scam.

Also, I'm giving a lot of "salt back" because I'm an electrical engineer, so I can see how all this stuff is total garbage. You're obviously not. I've even worked on 3-phase "DC brushless" motors and their controllers, so I do know what I'm talking about. This stuff about not needing a DC-DC converter is BS. Of course you don't need a "DC-DC converter" if your motor is AC, which every motor without a mechanical commutator is. You still need some type of power control circuitry to drive the motor; you can't hook it up to a battery!

>Every time I've replaced my car battery, it's been because one cell failed after I let it deep discharge. Except once where 2 cells failed.

If you did a deep discharge, the other cells were degraded too, you just saw the cell that failed first.

>But you don't have to use that bus architecture. It's mechanically simpler and more robust.

Yes, actually you do, unless you want to have ridiculous currents at only 4.2V going to your inverter, which would be stupid because you'd need bus bars and would still have high resistive losses. This is the whole reason we have batteries with series-connected cells instead of parallel-connected cells.

Whenever i see a headline with a subjunctive mood verb (should, could, might, may,...) I think "marketing propaganda".

Previous conversations about this same announcement:


For those wondering what a seemingly marginal difference would mean in real world examples, take a look at this video from a conference on electric aircraft design a year ago:

High Torque, Low RPM Motors – Key Enablers of Quiet Flight https://www.youtube.com/watch?v=jXd4M_pZl78

I watched about half that. Interesting bit that newer high bypass turbofans need gearing to match the turbine optimal shaft rpm to the fan's lower rpm and higher torque.

Brings up

   High torque gears -> weight
   High speed gears -> maintence hassle
If I get this talk right the hope is the difference in weight penalty for an advanced direct drive electric motor vs a gear train is nil. This guy is pushing a very high efficiency design that minimizes magnetic and resistive losses while being capable of high torque.


Random OT trivia, but the first turbofan to fly was geared, and there have been commercial geared turbofans in use at least since the early 70s.

It's always been a trade off of fewer compressor stages vs. not having gears. Either of extra compression stages and gears add weight and reduce efficiency, so it's always a tradeoff.

Transferse flux motors (from the talk) have a "too good to be true" ring to them. I wonder what will be of them in 5-10 years.

Increasing the torque density compared to the current state-of-the-art would be a big deal for robotics applications.

In the latter part of the researcher mentions using different materials for the core of the motor to limit flux losses. Some of the efficient materials (cobalt in particular) are costly and would drive up the bill of materials cost of the motor compared to conventional motors, where copper is usually the biggest expense. One of his points was that a lot of the cost is currently around designing the motor to suit a given application, unlike existing electric motor designs where you can basically look in a book to see how it should be sized.

Aviation is probably the sweet spot application for this motor because decreased weight means increased payload and money. The increased cost is less of a concern than in, say, your cell phone. If you could run the motor without a gearbox and without active cooling, that would save costs in other areas and also decrease maintenance costs.

The "too good to be true" part of aviation is if it, together with battery technology, is able to keep airplanes in the air for a significant amount of time. The high energy density of aviation fuel means that we can build a "good enough" efficiency jet engine by building it like we do today, and just throw fuel at the problem to keep an airplane in the air. With electric propulsion the energy envelope is much tighter and we basically need these miracle solutions in battery tech, motor tech, or both, to make it viable. The presenter's premise was that an airplane that currently takes 180 horsepower to stay in the air, could do the same work with 90 horsepower if the motor and fan were designed differently. Many people are looking at the battery side of the equation, but not many people are looking at the motor side of the equation and there are significant gains to be had here. He then presents evidence indicating that designing a motor and fan differently is possible, and that current turbofan designs are optimized to work with the constraints of jet engines (ie, needing gearboxes to generate enough torque to run bigger and bigger fans).

>The company’s permanent-magnet tech requires no rare-earth metals.

Neither does any other motor. It will be larger without them.

A lot depends on information they have not yet shown us. We have seen some hypebole, but there are dozens of 'free energy' people on you-tube making all manner of expansive claims - again with no data. Their best bet is to partner with Tesla or another combustion based car maker. They speak of patents - that needs to be tested on a bench to confirm it. Tesla has many many patents that they share on a mutual basis - which means I share mine if you share yours. So if some car make wants to sign and exclusive with Linear Labs they may get cut off from the Tesla patents?? If someone is too greedy they will find no buyers as the old ways work fine. Look at how the Wright Brothers patent greed eventually killed them off as they were invented around. https://en.wikipedia.org/wiki/Wright_brothers_patent_war

So more data, more truth, more future

This whole thing sounds like a scam. Even worse, it's in an IEEE article. I'm really glad I ended my subscription to them so many years ago if they've fallen to these depths. Just look at this one line from their article:

>Torque is the amount of work that a motor or engine produces, typically measured on a per-revolution basis.

No, that's energy, not torque. Torque is rotational force. Torque times rotation is power, which is the rate of work. Work = energy. This is Engineering 101 here; it's inexcusable for the IEEE of all organizations to screw this up. I expect this kind of lousy journalism from a mainstream, layman news outlet like NBC or something, not something aimed at engineers.

>A typical motor’s copper content could be reduced by 30 percent, while generating equivalent torque, they say. So for a given torque level, the HET consumes significantly less energy than competing designs.

Very little of the electrical energy input to a motor is consumed by resistive losses in the windings; almost all of it is consumed by conversion to mechanical energy. Reducing winding resistance is nice, but it's not going to produce some enormous energy savings. Reducing the weight is helpful too, but again, those extra windings are only a fraction of the motor's weight, and unless the motor is being used in an aircraft, motor weight isn't really that big a deal in a vehicle, battery weight is.

>The company’s permanent-magnet tech requires no rare-earth metals.

This sounds like a total crock. Always be suspicious of new inventions when a single invention claims revolutionary advances in more than one field. Magnet technology is separate from motor technology, just like CPU technology is separate from mobile phone technology: it's a very necessary component, but still a component, one which is usually sourced from an outside vendor. Motor manufacturers do not make their own permanent magnets; there's companies that specialize in that. And making magnets with the highest densities of magnetic flux without rare-earth elements would be a pretty big game-changer, so why are they talking about motor windings instead of this?

>The system incorporates a purely electronic transmission

A what? Not possible: a transmission is a mechanical item that converts rotary motion into rotary motion at a different speed and torque level (e.g., high-speed, low-torque into low-speed, high-torque). This is just a meaningless sound bite. Most EVs use transmissions (usually single-speed "gearboxes", not multi-speed "transmissions") because they'd have to make the motor physically larger to generate the torque necessary to drive a car directly, and the extra weight and size isn't worth it; it's easier to drive motors fast and use a gearbox to gear it down, even with the efficiency loss.

>Many electrified vehicles must also integrate a DC-to-DC converter

Again, the "Institute of Electrical and Electronic Engineers" apparently don't understand basic electronics any more. Most EVs use 3-phase brushless motors, or induction motors (Tesla). The motor controllers are driven by DC because that's what Li-ion batteries produce, and they output waveforms to directly drive the windings on the motor in accordance with the current demand for power. I don't see how they expect to get around some sort of controller here.

Sorry, this whole piece sounds like BS. The whole thing is likely some kind of scam to get investor money and then disappear.

> >The system incorporates a purely electronic transmission

> A what? Not possible: a transmission is a mechanical item that converts rotary motion into rotary motion at a different speed and torque level

Other articles about this company explain what this electronic "transmission" is: variable windings. So they can switch from 6 phase to 4 to 3, which will change the RPM per volt. I think.


It's not unheard of for EVs to boost the battery voltage to a higher level before feeding into the inverter that powers the motor (e.g. https://www.eetimes.com/document.asp?doc_id=1281167). There are some advantages to running the inverter at higher voltage (cheaper switching devices and smaller wires), although I do find it surprising that this apparently sometimes offsets the cost of the boost converter.

But I believe this is much less common with lithium ion batteries (~3.6 V/cell) than it was with NiMH (~1.2 V/cell).

I'd like to see real comps with the Tesla M3 motor, which seems to be the clear market leader in cost/perf ratio.

You know it "will not" when the title says "could".

I wonder how easy it is to manufacture comparitivly. Also I wonder how well it works to generate electricity.

So what is the advance that they have made? I couldn't see anything in the article.

Amusing to me that "scooters" gets the number two slot here.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact