# How Can 335 Horses Weigh 63 Pounds?

Koenigsegg, the Swedish car company, has a history of unusual engineering. The latest innovation is an electric motor developed for its Gemera hybrid vehicle. The relatively tiny motor weighs 63 pounds and develops 335 horsepower and 443 lb-ft of torque. Dubbed the Quark, the motor uses both radial and axial flux designs to achieve these impressive numbers.

There is a catch, of course. Like most EV motors, those numbers are not sustainable. The company claims the motor can output peak power for 20 seconds and then drops to 134 horsepower/184 lb-ft of torque. The Gemera can supplement, of course, with its internal combustion engine — a 3 cylinder design.

The motor uses advanced materials including hollow carbon fiber to reduce weight. The company believes the motor will also find use in marine and aerospace applications. They also announced the availability of a drive unit consisting of two Quarks and a planetary gear shaft that can produce 670 horsepower in 187 pounds.

Weight-to-power ratio for electric motors is tricky because you might not include the weight of the batteries and other supporting equipment. Still, this seems like a lightweight and small powerhouse.

The last time we looked in on Koenigsegg, they were 3D printing production vehicles. We imagine the Quark won’t be cheap, but it would be a great crate engine.

## 110 thoughts on “How Can 335 Horses Weigh 63 Pounds?”

1. BobH says:

I think you need to include the weight of the inverter/driver at least because it is such an integral part of a drive motor. Batteries you could argue parallel the weight of the fuel tank of an IC engine.

1. The+Gambler says:

“Batteries you could argue parallel the weight of the fuel tank of an IC engine.”

no so much, a gallon of gas weighs approximately 6 lbs. so in a 20 gallon tank you have 120 lbs roughly. the tesla s on the other hand the weight of its battery pack is 1200 lbs. that is quite a large difference and really in no way are parallel.

1. Pat says:

And you’ll never really be able to reach weight parity, either, because 75% of the mass of the stuff you use to generate the energy in an ICE isn’t stored in the car.

1. Thrashy says:

Lithium-air batteries are forever stuck in the lab, but the theoretical energy density combined with the better overall system efficiency of electric drive would put them solidly ahead of liquid fuels in terms of range vs. mass. I’d say they’ll *probably* show up before fusion power, but at this rate it could be a close race…

1. Dude says:

But then solid oxide fuel cells would burn liquid fuels at comparable efficiencies and win the race again.

Lithium air batteries get heavier as they discharge because the “waste product” is a solid oxide and oxygen atoms are relatively heavy. It also means the electrode changes volume drastically between empty and fully charged state, which is part of the reason why they haven’t come out of the lab. The mechanical stresses are enormous.

2. Pat says:

Yeah, lithium air battery stuff doesn’t exactly fill me with joy because I have *no* idea how in the world you’d ever manage to run them in an uncontrolled environment. Generally it’s a bad idea to shove random ions/chemicals through the battery. And of course, the high energy density’s gotten in the same way you get it for the ICE – huge portion of the chemical weight’s not in the car.

If lithium was super-cheap and common and you could make the manufacture/refurbishment of batteries easy, I could see it making sense in terms of battery swap. But, well… lithium isn’t super-cheap and common.

2. Clancy says:

Parallel as in equivalent component, not similar weight. The inverter is a mandatory component of the motor/drive system, the battery is a separate component that can be modified semi independently of the peak power output of the motor (again, not perfect, but analogous to how you can get the same power out of the same engine with either a large or small fuel tank).

2. Drone says:

@BobH said: “Batteries you could argue parallel the weight of the fuel tank of an IC engine.”

For the most part batteries weigh the same as you fill or drain them. That is NOT the case with a tank of gasoline.

3. Six677 says:

That drive unit quoted at 187lb’s actually includes their new inverter dubbed “David” as part of the package

4. Yes and no. One opportunity for electric vehicles is to mount individual motors at the wheels. Suddenly you don’t need a “drivetrain” – no differential, no axle linkages, no need to design a suspension that co-operates with an axle. Two reasons why this isn’t usually done: many carmakers already have drivetrain-accommodating designs as they have legacy ICE based vehicles, and it’s cheaper to adapt a known design than start from scratch; and the weight of the motor is problematic on rough roads. If we can make the motors lightweight and put them in the wheels, we can save weight elsewhere, gain efficiency, and implement stability control without having brakes fighting the motor(s).

1. Menga says:

Is not that automakers want to keep making drivetrains for electric cars for legacy reasons.
Motors in the wheels, no matter how light you can make them, are unsprung mass.
And that’s terrible for any suspension system.
Tesla solve it quiet well by putting the motors where the differential would be

2. Dude says:

Well, it can’t.

The motor is one thing, but the battery to supply the power is another. 335 HP = 245 kW and if we assume a modest 5C discharge rate for the battery cells, so as to use li-ion instead of li-po cells for the better energy density and longer life, that would require at least 49 kWh of batteries to supply the full power. For 100 Wh/kg total assembled battery mass not including mounting, support structure and armoring, the mass increases to 490 kg or 1078 pounds, plus 63 lbs for the motor itself, plus another 100-200 lbs for the inverter and copper buses as thick as your wrist.

1. sdfdfsdfsdf says:

“and if we assume a modest 5C discharge rate for the battery cells”

The battery needs a *peak* discharge rate to match the motor. Very few EVs need full power for more than a few seconds.

“plus another 100-200 lbs for the inverter and copper buses as thick as your wrist.”

EVs these days are 400-800v and do not need “busses as thick as your wrist.”

Stay within your lane, friend. EVs clearly aren’t one of them.

1. Dude says:

That’s besides the point. Look at the Nissan Leaf drive unit – it’s as big (and heavy) as a small petrol engine. I also forgot to count the charger and BMS into the equation, which is again more stuff you need besides the actual motor.

https://timedm.com/wp-content/media/2019/10/Nissan-LEAF-2019-emotor-battery-PDM.jpg

And the peak discharge rate for the li-ion cells that Tesla uses is something like 8C. You have to leave some margin, or else you end up like Tesla where their “ludicrous mode” only works when the battery is 95% full. After all, as the voltage goes down, the current goes up to maintain the same power.

1. X says:

The leaf is definitely not a cutting-edge design. And afaik this motor is not yet part of a fully-integrated vehicle design. So they’re not all that comparable at the moment.

But there are several really easy ways of using a motor like this to design a better car. Not all of those ways are “clean green” pure EVs, but that’s neither here nor there.

An example: if you simply take a moderate-sized internal-combustion generator system (which can be RPM-optimized, unlike an automotive prime mover engine), you can benefit from a better-optimized engine, from the fact that electrons are easier to transport than “spinning torque” is, from the fact, and from the fact that you can redesign the differential to take advantage of a much wider torque-vs-rpm-tradeoff band.

The resulting car would potentially be superior (for whatever metrics you’re focused on) to both traditional and full-ev designs.

A good motor isn’t the only thing that matters in your design, but it’s another degree of freedom while you’re designing whatever it is you’re designing. And that has significant value to any engineer.

I am not a car connoisseur, but I could imagine a conventional car also has lots of heavy crap you remove when going electric, so their weight is not doubled (I read around 30% more).

If we are able to halve the weight of the batteries as a combination of better infrastructure (quick battery swap stations) and better battery chemistry (a matter of time) soon enough an electric car could weight less.

1. Pat says:

You can’t really make the comparison. I mean, I can make an EV super light if it only goes 5 feet, and I can make an ICE vehicle super light with absolutely no torque. If you want torque/power on an ICE, you need weight, and if you want range on an EV, you need weight. It’s just a different parameter space. Can’t really make fair comparisons between the two.

That’s partly the issue with talking about hybrid vehicle weight penalties, for instance, because you can’t really compare them.

3. To be fair, Tesla pushes their cells upwards of 20C. So that “modest 5C” is a bit low in comparison.

And last I checked, Koenigsegg is even more silly since they seemingly don’t use 18650 cells but rather the flat type that is easier to achieve much higher discharge rates with.

Since flat cells aimed at high discharge rates have many parallel plates with their own typically wide tabs, so 50+ C isn’t “unrealistic”, though discharge efficiency is abysmal… After all, one major thing holding cells back is conductive losses. One reason Tesla is looking at their new cells with a more or less continuous tab.

1. Dude says:

The ludicurous mode hits top acceleration with 100% battery level and goes down from there, which means the power is limited by the current the battery can (is allowed to) put out while the voltage keeps dropping with the charge level. With a 100 kWh battery they’re producing 615 kW which is roughly speaking about 6C.

They’re not pushing it anywhere near 20C.

1. Going back in time to one of my by now really old comments… https://hackaday.com/2020/10/05/teslas-new-tabless-batteries-unlock-new-levels-of-performance/#comment-6283940
I actually calculated it and seem to recall the right number, but the wrong unit. (units are though very important things to remember correctly, especially prefixes.)

“And that car (Tesla Model S) uses typical 18650 cells.
This is configured as 74 cells in parallel, 6 of those sets in series to form a group.
And 16 of those groups in series to form the complete battery pack.
This means each cell needs about 4.17 volts each for our total of 400 volts. (Though, that 400 volt will drop as the batteries discharge. Requiring even more current to get our 615 kW power output.)
But it also means that our 1.5 kA is only shared between 74 cells. Or 20 amps per cell when fully charged.

If we are generous and say that these 18650 cells have 3 Ah of capacity, then they are discharged at 7C, quite a bit above 2-2.5C where discharge efficiency is already crawling up the wall….”

So the “6C” you quote is reasonably close to that calculation of 7C.
Though, I don’t account for any losses in the calculation, so the 615 kW motor power is actually going to lead to a bit higher currents on the battery. So maybe it gets closer to 9C worst case.

But I guess it is called “ludicurous mode” for a reason.

2. Nick says:

My ICE car goes better on a cold day. And, ironically, accelerates faster if the fuel tank is almost empty. I’m never going to own a Tesla so I don’t really have skin in this game, but the “diminishing performance” aspect of Ludicrous mode on the Tesla has analogies in “traditional” cars too.

2. abjq says:

So if I’m thinking of a hybrid upgrade, and ran the cells at 50C, using Dude’s calculations above, that around 49Kg for the cells.

Hmmmmm….

1. Dude says:

At these current levels, you’ll see significant capacity drops after 100 cycles.

At 50C you have about 72 seconds of acceleration for each full load cycle.

1. abjq says:

When you add 300bhp to the existing 240bhp I already have, I reckon I’ll only need 2 seconds of acceleration t.b.h!

In fact the challenge will be putting the power down, so I’d have to scale it back, and only put out full electric torque during dead/lower power periods e.g. on a gearchange and for the next 300ms of acceleration after that.

Of course my idea of ‘hybrid’ may differ from yours. :-D

4. hap says:

Samsung 40T3 21700: 12.5 C continuous rating, 210 Wh/kg cell level. You gotta update your numbers my man, technology keeps moving

1. Pat says:

In general, charging and discharging cells at their rating limits *severely* degrades their life due to the heat generated (and other mechanisms). Hence the reason he said “modest” and “longer life.”

However since this is just a short-term energy draw it’d be fairly insane to build the entire battery pack around it.

1. Dude says:

You also can’t load a partially discharged battery at the full rated current because the cell voltages would crash below the safety limits and the BMS would shut you down. That’s why the top acceleration of Tesla cars is a gimmick.

5. Clancy says:

NMC batteries are capable of 10C peak discharge easily and the best available cells are hitting 260Wh/kg, more than double your assumed value.

3. three_d_dave says:

A T58 gas turbine engine weighs about 280 pounds and generates a continuous 1200 HP, so about the same HP/lb, but as for as long as one cares to operate it or roughly a few thousand hours of operation.

Sometime I saw a 800-900 HP turbine engine at 150 lbs. Add another 100 lbs for the reduction gearbox.

Plenty of warm air to defrost the windows too.

1. AKA the A says:

don’t forget the several gallons of fuel per minute needed to feed the beast, the tank size severely limits the operating time ;-)

2. looopz says:

Yeah continuous that’s a huge difference. As for aerospace you need 100% available at least 5minutes and then go down to 75 or 80%. So this engine would have be rated very differently to go inside an AC. If we take that 134 is 75% percent then that would be a 175hp engine. Airplane will need gas only thing efficient…

4. Chris Nobel says:

And in civilised units:

600 Nm of torque, 250 kW of power, and 28.5 kg in weight.”

1. charles says:

+1
I was just about to convert it from hamburgers-per-freedom to something normal.

1. Pat says:

There’s actually a greasemonkey script that does unit conversions for you in a tooltip. Heck, you could easily have a button which swaps article units, so we can all read articles listing speeds in furlongs per fortnight.

2. Comedicles says:

Must use. But “It can also handle nonlinear conversions such as Fahrenheit to Celsius ” bugs me when tech and software people use words that don’t mean what they think they mean. Now I have to check all the source code!

1. Criatura says:

Funny thing is that the official weight and measure units in the US are metric, but there are no laws stating we have to use it, so we’re still stuck with using pots of chili per superbowl party measurements.

1. Pat says:

All US units are metric units. They’re just not standard SI units. But they’re still metric units. Hell, several SI units are actually defined… by measurements and methods from the US.

There’s this huge insane idea that the “good thing” about SI and the metric system was the units they used. Which units you use are pointless. The “good thing” about SI was that everyone agreed on a way to *define* the measurements.

1. > All US units are metric units.

1 foot = 12 inches
1 yard = 3 feet
1 mile = 1,760 yards = 5,280 feet = 63,360 inches
(source: https://www.mathsisfun.com/measure/us-standard-length.html)

None of the above are “metric units”. Specifically they are not part of a “metric system” because they measure the same thing but are not “metrically compatible”(? ;-)) among each other.

2. Pat says:

Yes. Absolutely. 100%. All units in the United States are officially based upon SI standards. US units of length measurement have been defined with respect to metric units since the 1800s.

“because they measure the same thing but are not “metrically compatible””

That’s not a thing. Do you mean “related to each other by powers of 10”? There are plenty of units accepted for use by SI which are not related to other units by powers of 10.

Is a kilowatt-hour a metric unit? It’s not related by a power of 10. What about a third of a meter? If I take a meter stick and cut it in 3 equal parts, suddenly it’s now no longer a metric measuring stick?

3. What? One core part of the metric system is:
“For any given quantity whose unit has a special name and symbol, an extended set of smaller and larger units is defined that are related by factors of powers of ten. …; the unit of length should be either the metre or a decimal multiple of it; and the unit of mass should be the gram or a decimal multiple of it. “[1]

How in FSMs name are foot/yard/inch and what not units from the “US Customary Units” system supposed to be metric (floz anyone?)?

Just because they’re now (since when?) defined through metric Si units? Nope!

KWh? because of h=3600s? yes, this derived unit may not be pure metric. That’s why we have Joule (or KWs if you want).

Your “1/3m” examples make no sense because it doesn’t define a new “special name and/or symbol”. Your even using the base unit “metre”.

If on the other hand you were to define a new unit of length – let’s call it a “mard” – with 1mard=3metres, the mard would not be a unit in the metric system.
Same with the “foom” if e.g. 41fooms=1mard. <- Not in the metric system.

[1] https://en.wikipedia.org/wiki/Metric_system (2nd paragraph of the intro(!))

4. Pat says:

I forgot that we have kilohours, mega degrees and picolightyears. (And, incidentally, people use kilofeet all the time).

Anything which is defined as an exact conversion from a metric unit is metric. That’s what it means to be part of a measurement system. Names are meaningless. 52.3 inches is just a weird way of saying 1328.42 mm. Measurements care about their reference, not the words. If I say 0x342 mm, it’s still metric.

Customary days (86400 s) are metric. Sidereal days (time between 2 transits of the first point of Aries) are not.

5. Sandro says:

Standardization is good yes, but you’re underrating the benefits of uniformity in standard SI units.

6. Comedicles says:

Any system of units and measure is a metric system. This is why I tend to say “French Metric” or “SI units”. ANSI/ American Customary is a metric and British Imperial is a metric.

Personally I find it offensively imperial that SI uses call it The Metric System. And incredibly imposing and disrupting to this day to not standardize the yard, and define the meter out of thin air! What were they thinking? I know, a new World built on the ashes of revolution and freedom and -oops, a new Emperor! Again, what were they thinking?

Maybe SI should be cancelled for producing dictatorship.

7. Common usage is not definition says:

Response is actually to limroh, but…

All US units are metric units, because a “metric unit” is simply a “unit of measure”. US units are NOT SI units, but the linkage between “SI” and “metric” seems to have been driven primarily by efforts to promote it as a “common system of metrics for Europe”, which resulted in referring to a country’s proces of switching to SI as “metrication”.

But the word “metric” applies to *any* consistent unit of measure, no matter how it’s defined.

8. Pat says:

No, I’m really not. Users standardize units based on whatever’s convenient, and it *does not matter* what those actual units are called, just that the reference is defined.

Acting like the whole “imperial vs metric” thing causes problems is just nuts. NASA didn’t lose a mission because of mixed imperial/metric units, they lost a mission because of improperly defined standards, period. You’d get the same problem if someone mistook cm instead of mm, or if someone took a diagonal dimension instead of a linear dimension. Or, hell, if someone misread tolerances!

What other complaints do people give: screw/bolt sizes? That’s a *totally different* standard. That’s not “metric versus imperial,” that’s ISO vs SAE, and powers of 2 versus integer steps. Wire gauge? AWG is logarithmic spacing, ISO is “let’s try to make nice round numbers.”

If SAE had defined things in quarter-millimeters and AWG was defined with a minimum size of 0.1 mm vs 0.127 mm, you’d still have the same issue.

Even SI doesn’t result in people using common units. Meters/second isn’t related to km/h via a power of 10. 1 kWh isn’t related to joules via a power of 10. People are happy to accept those factors of 3.6 sitting around. Other conversion factors are no different.

It’s hilarious that people talk about SI units as if “it’s what scientists use.” No, it’s not. Astronomers talk in dex units, parsecs, and solar masses. Physicists talk in millibarns and electronvolts and use 300 K as a reference standard. And then they use unit systems that get rid of constants to make their math easier (which is the *same thing* as working off a standard gauge spacing!).

Saying “I want all drawings in metric units” does not mean you get ISO wire gauges and screw sizes. 3.96875 mm is a metric unit and a 5/32″ head size. 4.2 mm is a metric size, and you’ll swear up and down at an engineer who makes a 4.2 mm hole size.

9. dave says:

It’s more a problem that I dare say most Americans refuse to use metric units.
So when something is clearly designed in metric and is say 10mm, many people will still refer to it as 3/8th which it is not, or if they want to be accurate, 0.393701 inches.

A good example of this is Abom79’s video here: https://www.youtube.com/watch?v=FLYdhfgF6Pg
He doesn’t care what the customer is measuring in, but it would certainly be better if they’d measured it as 100mm not in inches.

Some countries (people) can embrace both metric and imperial and use the ones which best suits for the job.

2. X_sign_here___ says:

Well, if you’re actually feeding a football party crowd, using intuitive-for-the-purpose metrics can actually be a net win over using generally-applicable universal units..

My opinion of the utility of metric units is that if you’re constantly working in a wide spectrum of fields, having universal units lets you build and preserve an intuitive understanding that applies across your fields. But, if you’re mostly doing one specific thing, there is serious value in picking (or even creating) a set of units that directly simplifies the work you’re actually doing.

The value of directly-usable units was of course much bigger when most calculations had to be done manually. Simplifying or eliminating calculation stages had a direct impact on the time/effort/opportunity-for-error for the work. Nowadays, that’s less of an issue, of course.

So, for modern hobbyists, metric is quite useful. But it’s not always as useful in all contexts, especially when there are existing well-tested workflows structured in other units.

2. aaronfish says:

Hamburgers-per-freedom may be my new favorite unit. I’m going to need you to work up a conversion table so I can use it for everything. I’m fine having to specify dry or wet hamburgers-per-freedom if that helps, also prefixes are allowed but not metric ones because I feel it goes against the intent of the unit :D I want to be able to refer to both my fuel efficiency and how much sugar I need to add to this cake recipe in HpF (H/F maybe, but it might be confusing among audiophiles or radio operators).

1. Hum … don’t forget that a US Hamburger-per-freedom doesn’t weight the same as UK Hamburger-per-freedom, and of course Hamburger-per-freedom weight unit is not the same that Hamburger-per-freedom volume unit ;)
Hopefully, people in Hamburg use metric units ;)

1. You know what they call a Quarter Pounder with Cheese in Paris?
Royale with Cheese.

2. +1

Please provide metric units, I am an American in the US and have an engineering PhD, but that was all in metric (as it ought to be) and I have no idea what pounds, inches and fahrenheits are outside of human bodies and temperate American weather. My knowledge of forces and pressures outside daily domestic experience is quantified in metric.

3. Dude says:

When are you going to start measuring your televisions, computer monitors, cellphone screens, and subwoofer diameters in centimeters?

1. Nick says:

5. Fidel says:

Put it in a classic Mini Cooper.

1. JohnU says:

As soon as my lottery numbers come up it’ll be on my to-do list.

6. then says:

What is this sentence doing there: “The Gemera can supplement, of course, with its internal combustion engine — a 3 cylinder design.”

1. volt-k says:

What do you mean?

1. abjq says:

That it’s just like a Geo Metro?

2. Shannon says:

That sentence is explaining that the Gemera can supplement with its internal combustion engine which happens to be a 3 cylinder design, of course.

1. JohnU says:

3cyl 2 litre ~600hp, just as impressive as their electric motor

7. Julian Skidmore says:

It’s a Swedish company. They did everything in metric. Hackaday has an international audience, yet the consideration for them is ZERO. Instead, all their metric specs have been deleted and replaced with only imperial / US units. Why? Why can’t you even include the figures for proper units in brackets?

1. Pat says:

“Instead, all their metric specs have been deleted and replaced with only imperial / US units. Why?”

Because the entire point of the metric system is not to have One True Unit but to be able to perfectly convert between units. Imperial units are still metric.

The difference between 63 pounds and 28.5 kg used to be a real thing. Now you could make a JavaScript button on the website to swap between them. Now that would be a useful script…

1. Metric vs imperial is decimal ve a mix of binary, decimal, pental……. Whatever. Imperial is stupid.

1. Comedicles says:

Imperial/ANSI is based human scale and mental arithmetic with rational numbers – 1000 strides to a mile, 1 yard by 2 yards of cloth makes clothing for the average person, etc.

The (French) metric is based on the distance from the equator to the North pole divided by 10,000 (!) and knowing that Greek unit names are for small things and Latin names are for big things – and who wouldn’t know?

1. Pat says:

“The (French) metric is based on the distance from the equator to the North pole divided by 10,000”

Well, they *tried* to do that. But they screwed up the trigonometry, so it was only *close*.

Then after they did it, they realized they were complete idiots and decided never to do it again, just basing everything off of the really, really nice platinum-iridium standard they had made. Eventually the rest of the world (many of whom thought this whole “let’s change to a unit that’s 9% different for no reason whatsoever” was dumb) realized that the international prototype meter was *way* better than what they had, and so they just switched to using *that* reference, but converting it to keep all of their measurements the same.

The original French metric units came up with 1 good idea (a mass unit based on water and the length unit, democratizing weight measures) and 1 horrible idea (pointlessly redefining the length standard because of some insane idea of universal-ness).

2. Julian Skidmore says:

So, your argument is that it’s no sweat to convert. Then why didn’t the author just leave in the Swedish metric units and let you make the effort to convert? Why make all the effort to convert to metric and _then_ delete the metric measurements so that any metric user has to convert again? Why make it harder for metric users?

1. Pat says:

“Then why didn’t the author just leave in the Swedish metric units and let you make the effort to convert?”

Same reason that an article author summarizes foreign-language articles in English rather than cutting and pasting the foreign language info here.

2. Instead of complaining, perhaps learning to quickly convert between units in your head would be a useful skill, just like every American child was taught to do back when I was in school (1970s-80s). I find it quite handy, and it keeps me from griping every time someone writes something in metric with zero consideration for the millions of us that are used to using imperial units.
You could, of course, also simply do a quick internet search.

1. Pat says:

If only we were having this conversation on some sort of a device that could do the math for us!

2. Tweepy says:

You mean for the 5% of the population using imperial units, 95% have to do the conversion?

Please HaD, stop using this stone age unit, move forward and adopt metric only. This is a scientist blog, there is no other measurement units.

1. Pat says:

“You mean for the 5% of the population using imperial units, 95% have to do the conversion?”

The majority of the people in the world don’t speak English, yet HAD’s still in English.

There are tools which will automatically convert measurements for you. All of these conversions are exact and defined. Imperial units *are* metric units. It’s literally just a translation issue.

1. Julian Skidmore says:

You know what’s meant. Imperial units aren’t SI units and we all know it. The standard is SI: it’s the only global standard. Also, English is the most popular language in the world including people who speak it as a second language – it is the Lingua Franca. Similarly, SI is the lingua Franca for engineering and science. If you don’t think conversion is a problem – do everything in SI units and let US citizens do the conversions – it’s easy, as you say.

But the issue here isn’t that one can do a conversion from Imperial to metric. It’s that the author already had it in metric and not only converted it to imperial, then deleted the original metric values – forcing non-Imperialists to do the conversion.

Except he left one thing out: the metric values are in the image itself: 250KW; 600Nm. Nice round numbers. Makes sense huh?

2. Pat says:

“You know what’s meant. Imperial units aren’t SI units and we all know it. The standard is SI: it’s the only global standard”

Friggin’ *hours* aren’t SI units. So what? The units don’t matter. The standards do.

“Similarly, SI is the lingua Franca for engineering and science.”

Nope. The SI *standard* is the lingua franca for engineering and science. Convenience units are used *everywhere*. The units don’t matter in science. In engineering, only the references/tooling matters.

What’re you going to do – pillory everyone who uses 2.54 mm pitch parts?

2. Julian Skidmore says:

Thank you! It’s about time we left the 19th century behind. Even Nasa in the 1960s had the sense to use metric units for the AGC, they only converted them for the AGC display.

1. Comedicles says:

The SI French Metric IS 19th century. Proposed at the end of the 1700’s and adopted at the begging of the 1800’s. And an idiotic disregard of the people and a massive disruption by choosing a unit of length out of thin air and unrelated to anything before. In the fervor of revolution and gathering of intellectuals they thought it was genius – aaaaand they had a new emperor. Doh!

3. abjq says:

Even though I use SI units mostly, what gets my goat when HaD convert to so-called Imperial, is that they use the colonial version of them, not the proper British Empire Imperial units.

The word “Imperial” is there for a reason, don’t you know.

8. Connor says:

Lots of comments about power/weight ratios and the downside of battery energy density, but to me the real advantage of these motors is the opportunity to relocate them closer to the wheel.

Having one on each corner, just near the wheel, would simplify a lot of the mechanics of an EV, and it’s even possible these would be suitable as part of the unsprung mass on a suspension system.

I wonder if the peak torque could be used in reverse as a substitute for friction-brakes. If so, future EVs could be dramatically simplified compared to their ICE buddies.

1. fiddlingjunky says:

I’m not sure why you bring up unsprung mass, the vast majority of drivetrains (ICE and EV) are unsprung (mostly, except for the outer CV and a portion of the half-shaft mass), as they drive the wheel through CVs and shafts. The exception would be hub motors, which are sprung, but those really are the exception rather than the rule. EVs don’t have any advantage here.

Having them on each corner, though, is an idea with merit, and has been done before (the first one that comes to mind is the SLS gullwing EV). The main disadvantage of this is that, for the most part, electric motors want to spin several times faster than a wheel does to make their peak power (though axial flux motors like this one usually run a little slower than equivalent radial flux), so they usually need to be geared down to make good use of their power. Gearboxes are heavy, and having one per motor instead of one per axle means a lot more weight, so most EV manufacturers have opted to go for one gearbox per axle, with torque distribution being handled by some kind of advanced differential.

Modern EVs already scavenge a ton of electricity under deceleration. I suppose one could drive the coils to actively brake, but that would use more energy for something that can instead be done by brakes, which are light and don’t require (much) electricity. I can’t imagine EVs ever ditching brakes, though I suppose it could happen.

1. I think you have “sprung” and “unsprung” backwards. Sprung mass is mass that is being carried by a vehicles suspension system, and unsprung mass is mass that is being carried directly by the wheels.
Unsprung mass is usually engineered to be minimized, as it contributes to poor handling. The more of the total vehicle mass that can be carried by the suspension system, the better as far as road handling is concerned.

1. fiddlingjunky says:

Regardless, just reverse it wherever I said it, and my comment makes sense. EVs still don’t present an unsprung mass advantage.

Thanks for the correction!

1. Jonas Eriksson says:

I don’t understand how the heavy unsprung hubmotor argument applies to normal cars driving on flat asphalt roads. For fast racing cars or off-road where the tires can see huge g-forces, I can see how low unsprung mass matters.
The low-friction tires used on EVs also points to that maximum grip is not the top priority but a compromise.

2. fiddlingjunky says:

@jonas it’s mostly hypothetical, as there haven’t been any significant road-going efforts with hub motors. It would likely be fine for a road-car, but the ratio of unsprung to sprung weight still matters for roads, as there are potholes and such.

3. thoriumbr says:

@Jonas, you would not want your precious motors hitting potholes all the time by sitting on the wheels. They better be stored on the vehicle body and be partly shielded from the vibrations by the suspension.

4. Jonas Eriksson says:

Electric motors are mostly solid metal, potted with tough epoxy and only one moving axle with ballbearings. It would have no problem handling the forces and vibrations from potholes. The disc/drum-brakes on the tires are more complex and they have no problems.

Also think of how the bigger and heavier truck tires with very stiff leafsprings have no problem with road handling.

5. Dude says:

It’s not so much a problem for the wheel or the motor in the wheel, but for the suspension system and the shocks which have to handle much more energy from the masses bouncing around, and consequently you get a rougher ride even on flat roads. More road noise, more rolling resistance etc. and also exposure to whatever dirt and debris, water, road salt… which is not a good place for high voltage electronics.

2. me says:

63Lbs (28.5 Kg) is quite a bit to add to the unsprung mass at each wheel. It would more than double the unsprung mass for many vehicles. That’s one of the reasons hub motors haven’t taken off for more than low speed equipment.

If you were to combine a hub motor with an integral suspension wheel however, I could see that combination having either less unsprung mass, or a significant part of the mass partially sprung by the wheel with the remainder taken care of by a conventional suspension.

That only leaves problems of bearings, sealing, single motor failure, etc.

To be clear… I want this. It’s not an easy path though.

9. Chris says:

Genuinely curious what the aerospace applications for this might be. 335hp is more than enough for small GA airplanes — 134hp is even enough for really small planes — but I can’t think of a use case where it’s okay for engine power to drop after 20 seconds. I can see why an electric car doesn’t need peak power for very long, but in a small plane, the only times you need full throttle are times you _really_ want that power to be sustained as long as you need it.

Little GA planes are where my experience is, though, and perhaps I’m missing some use cases in larger aircraft. (Jet engine or APU starters?)

1. fiddlingjunky says:

Rolls-royce’s recent electric show plane uses YASA electric motors, they were one of the first to build axial flux motors similar to this one. It looks like it has three bolted together in series, though.

You often want more power for a relatively short burst for takeoff and initial climb. Longer than 20s, but in the same order of magnitude. You definitely wouldn’t want to lose more than half your power like with Koenigsegg’s engine, but a more moderate continuous derating wouldn’t be the end of the world. It’s probably a thermal limitation, so if you didn’t push the motor quite so hard you could probably keep a more conservative peak power for longer.

2. X says:

The related aerospace test engine significantly derates the power and adds additional cooling. As far as I have seen documented, it’s nearly the same physical hardware (more like the Terrier than the Quark).

It appears that the aerospace engine works by downrating both of the Quark-like modules significantly (effectively 150 HP each), and by making the combined-package rating the same as either motor module by itself. This means that the whole package is *extremely* conservatively rated even if operated continuously for days at 100% rating. It also means that if either module fails, the combined motor still retains 100% capability, assuming the failure is not a friction event. Finally, it means that the combined module can provide a very high maximum takeoff power, and sustain that for long enought to get to maximum operating altitude before throttling back to rated power.

For aviation purposes, this is a 150 HP *cruise* configuration, and a ~300 HP *takeoff* configuration. It could also be provided with a ~600 HP “emergency: rebuild engine after flight” configuration, if that was considered useful.

For the aviation test engine, the duration limits appear to be purely controlled by heat factors; in the graphs, a larger heat exchanger directly (nearly linearly with dissipation) mapped to extended “maximum power” durations.

From what public info I can find, most of the patents involved in the YASA, Rolls Royce, Daimler-Benz, and Koenigsegg appear to be linked or at least related. I didn’t dig too deeply at the time, because it was incidental to my purpose, but these aren’t apparently “competing clones”, they’re “related developments”.

10. bluecat57 says:

So, everyone is Columbusing the Mazda Rotary Engine and declaring themselves geniuses.

There is nothing new under the sun. Ecc. 1:9

PS – I’ guessing this is not the same thing, but sheez I remember when we moved to Clemson 20-years ago Clemson was “printing” human skin. Then a couple of years ago some other university was declaring they “invented” it.

1. fiddlingjunky says:

This thing is in no way similar to a Wankel (which Mazda did not invent) other than that they both package as discs and turn a shaft.

11. el_ampo says:

Kinda missing the point in the discussion that we need to move away from combustion engines. Yeah, they are nice, energy dense and leaped mankind to the next level, but they are killing us to the point we can’t ignore it. Electric is the best choice right now until someone comes with a better solution (specially in the storage department) and adding a new small high powered motor to the mix is welcomed.

1. Plus we need to lower our energy needs. “Producing” electricity consumes resources: even “green” energy has downsides (building wind mills, or solar cells, produces waste water and air pollution). Not to say that nuclear power produces very long term hazardous wastes (even if the EU recently considered nuclear energy as “green”, which is a shame …).
So I think that we need to lower the size, and the weight, of our cars: is it really meaningful to go to work with a 1.5 Tons vehicle ? I think not. A 500 kg car, able to reach 50 km/h should be fine for such purpose. Moreover smaller cars would allow to have more vehicles on the same roads. So I think it’s a “win-win”.

1. Menga says:

True, a 1.5ton vehicle to move an 90kg human around…

1. Dude says:

In terms of momentum and energy transfer, I’d rather it be that than the other way around in a collision.

A Fiat 127 did not offer much for crumple zones and crash protection, and that was for a 688 kg car.

2. dave says:

So not an American then ?

2. Jonas Eriksson says:

There is nothing wrong with combustion engines, it depends on which fuel they are using.
Just as electric vehicles can be dirty if they are charged by coal powerplants.

1. Julian Skidmore says:

Isn’t that the same as saying: “fossil fuels make EVs dirtier, so fossil fuels are fine”? That’s always the crazy part of the argument; like saying “There’s nothing wrong with smoking, it depends on what you smoke. Besides, even non-smokers can die of lung cancer if they hang around heavy smokers all the time.”

Using the problem as an argument to perpetuate the problem, isn’t an argument IMHO.

1. Dude says:

It’s not that, but the fact that EVs don’t get rid of fossil fuels anyhow because they depend on an infrastructure that depends on liquid fuels to operate. It’s a non-solution, just like corn ethanol where you still use petroleum to make it – so it’s only a little bit less fossil.

If we fix that problem and source liquid fuels from non-fossil sources, then we also fix the problem for combustion engines, so why do we need the EVs?

12. Torrente says:

They probably learned from YASA that supplied electric motors to them, or bought the right to use the technology, putting only their brand

Nothing too much for those who already follow the market for Axial electric motors.

Data from two YASA models already considered old.

Parameter YASA 750 YASA P400
Torque (peak) 790 Nm 370 Nm
Torque (cont) 400 Nm 300 Nm
Power (peak) 200 kW 160 kW
Power (cont) up to 70 kW 20 to 100 kW
Axial length 98 mm 80.4 mm
Diameter 368 mm 305 mm
Weight 37 kg 24 kg

1. abjq says:

Interesting stuff so I started looking at the pricing of these things (not just Yasa)

It seems that they are double to treble the cost of a comparable power brand new ICE.

However, there is relatively little machining, moving parts etc. In short, they are simpler devices.

I’m scratching my head how we hacksters can (a) buy small/single quantities (impossible in many cases) and (b) at a reasonable price, except from the scrapyards….

1. X says:

They suffer from new-technology syndrome. They’re hard to come by in hobbyist quantities, and that will continue until they’re in much wider production.

They’re really simple (although a few bits require high precision). The real cost driver right now is simply that, to build them, you have to build a production line from scratch, since they’re not similar enough to anything already in production. This forces the wholesale price up. A product’s wholesale price is driven by cost-of-engineering (these are still in active development), cost of production line (these are enough different that they aren’t easily made by retooling existing production lines), cost of materials (similar to normal electric motors), and ongoing cost of production (shouldn’t be any higher than any other electric motor production line).

As they mature, the dev costs will taper away and it will become cheaper to set up production lines for them, and their prices will drop to be similar to existing electric motors, once supply catches up with demand.

But, until the dev and production line costs drop, they’re not going to be conveniently available at low quantities for reasonable costs.

Something to look forward to in the future, but not yet suitable for hobbyist use.

13. Chris Maple says:

The Terrier Drive Unit has some pipes on the outside. What are they for?

1. X says:

Some of those are cooling/lubricant pipes, others are large-gauge wire.

Electricity isn’t a magic solution to avoid friction, heat, and input-power… it just really simplifies the engineering required.