3D Printed Electric Motor Wants To Take Flight

Airplanes and spacecraft have a big problem. The more engine or fuel you have, the more engine and fuel you need. That’s why aircraft use techniques to have lightweight structural members and do everything they can to minimize weight. A lighter craft can go further and carry more payload or supercargo. Electric motors are very attractive for aircraft, but they suffer from having less efficiency per kilogram than competing technologies. H3X thinks they can change that with their HPDM-250 integrated motor and inverter.

Although the 15 kg motor is still in testing, the claimed specifications are impressive: a peak power of 250 kW for 30 seconds and continuous torque of 95 Nm and 200 kW sustained. The company claims 96.7% efficiency. The claims are for the motor running at 20,000 RPM, so you’d need to add the weight of a gearbox for practical applications, but the company says this adds a mere 3 kg to the overall weight.

The patent-pending innovation behind this motor is a 3D printed copper stator coil that improves current handling. The company claims these coils are 40% better than conventional coils. There’s also a 3D printed cooling jacket that contributes to the motor’s performance.

How real is it? We don’t know, but none of the claims seem farfetched or crazy. The company promises real performance data in the second quarter of 2021, so that will probably tell the whole story. Meanwhile, it is interesting the dream up what you could do with a motor that light and powerful.

While somewhat rare, there are larger electric planes out there. Of course, most hobbyist drones are electric, but there have also been some electric ultralights.

49 thoughts on “3D Printed Electric Motor Wants To Take Flight

  1. 1 – 0.967 = 0.033
    200kW * 0.033 = 6.6kW

    I always find it interesting when the engineering tradeoffs result in power electronics being bolted to a heater.

    Perhaps that’s why they have 30 second max power limit:
    250kW * 0.033 = 8.25kW

    1. It’s even worse when you do the arithmetic correctly. 200kW output requires 200/0.967 kW in = 206.82 kW, for 6.82 kW of heating.

      At 250kW output, assuming the 20 rpm is constant and the extra load is increased torque, requires a current increase of x1.25, or a I^2R increase of +56%. Assuming all the inefficiency is copper loss, that means the heating is now 10.6 kW. (in reality, much of the loss is windage and bearings, so the actual increase in heating might not be that much.)

      I’m actually much more surprised a 3 kg transmission can survive 200 kW.

      1. Here’s a guy who’s done some algorithms (and annoying GIFs) to optimize the efficiency of a BLDC motor:


        Forcing the efficiency up at the middle RPM range by limiting the I^2R losses reduces torque and power at the same time, so you end up with a power curve that’s more like a two-stroke gasoline engine – it picks up slowly and then peaks at the top RPM until it runs into the impedance limits of the coils and you can’t make it run any faster.

    1. It’s probably a really good metric though. If you consider that power applications are usually more efficient when they’re larger (eg a power station is more efficient than a car engine at turning fossil fuels into usable power). You could probably measure efficiency per surface area as well.

      1. Yeah, but it’s still meaningless. If you have an engine with 20% efficiency that’s five times lighter, then it’s the “same” as an engine with 100% efficiency – but knowing this has no use whatsoever. Considering your engine is going to be some few percentage points of the total mass you need to lift, you’re still going to be using almost five times more fuel.

        Electric motors have fantastic power density these days. That’s not the issue. The five tons of batteries is the issue.

        1. I started writing a detailed reply, but it’s pointless, you’ve pulled a load of numbers that don’t represent the situation at hand out of your arse. A car that’s five times lighter has always been an option, but it isn’t going to fill the same niche. It’s like comparing a Transit van and a bicycle. Equivalent vehicles don’t have a 5x mass disparity between EV and ICE.

          1. Nor did I say so, and it was exactly the point. I don’t understand what you’re complaining about.

            Here we’re talking about “specific efficiency” of the motor, which is essentially percentage points per kg, which is physically meaningless – and even if we allow that, it’s still useless as a metric because you are not changing the mass of the vehicle in significant amounts.

            In other words, a motor which is “equal in terms of specific efficiency” can still end up using five times, or any number of times, the amount of energy. The numbers don’t matter. I could have picked an engine that is 20% as efficient but 1/10th the mass, so apparently it has 200% the “specific efficiency” while actually using five times more fuel to move an equally heavy vehicle.

            The whole concept is complete gibberish. It doesn’t tell you anything.

  2. Integrating electronics into motors and actuators seems like a good idea, until you have to make it work. Heat, vibration, sealing and packaging all become more difficult and more expensive. It might make sense for a few specialized, low volume applications (the Mars Rover comes to mind) but not for most normal applications. I’ve worked on several projects in my career. None were cheaper or worked better. They were cool though!

    1. Seems like it could work well in consumer level sub-10W motors too, because you could mass produce the “Smart motors” with their own drivers and feedback, and just say “Give us 3v to 36v and a digital speed signal and we’ll do the rest”.

      Assuming you could get the drivers to be more reliable than the motors(Wasting a perfectly good motor seems a lot more pollutive than a driver board), it could have some repairability benefits, especially if the gearbox was an inline modular thing.

      Then you’d have all the stuff that normally fails, standardized and replacable, and mass produced enough for recycling.

      Everyone wants innovation and to invent the next big leap forward, but what we really need in most areas of tech is industry standards, even if they’re boring and arbitrary.

      1. What you’re talking about is a digital servomotor, and they already exist, and they’re “standardized” in a sense that you can get continuously rotating hobby servos in the same form factor.

        The thing is, they’re too expensive, and there are not that many cases where you actually need a sub-10 Watt motor with that much smarts. These tiny motors typically run toys or simple conveniences like fans and electric toothbrushes where the standard form factor would be an impediment for design.

    2. i have the same feeling. not only does it make some increment in complexity but it also ties the components together. if there are advances in the ESC or the motor, you won’t be able to simply swap one or the other out, you’ll have to wait for the next iteration of the whole unit. OTOH, i guess cooling might actually be simpler if you don’t need separate heat exchangers at the ESC and at the motor.

      1. interesting point. the only downside i know of to integrating the alternator’s voltage regulator is that sometimes you replace the whole unit when just one component dies. of course, it’s not exactly apples-to-apples because alternators are so commodity now-days. outside of maybe italian motorcycles i don’t imagine anyone is working to squeeze the last 10% of weight or performance out of an alternator. so the upfront cost of designing an integrated unit was paid decades ago and hasn’t been particularly revisited since.

    3. In regards to motors that needs to deliver a lot of power, then putting the driver onto the motor itself like this does make a bit of sense.

      The main benefits is that it reduces the cable length. Thereby both its parasitic inductance, but also its resistance.
      And considering that driving an AC motor up at 5-20 K RPM, then the skin effect will make a noticeable increase to the effective cable resistance.

      And the cables that we do have going over to the motor is mainly going to carry DC, with a fair bit of AC noise on top. But bulk current won’t be subject to the skin effect. On top of the fact that we don’t need 3 cables + power cables for the stator coils.

      There is though obvious downsides as well.
      Like it won’t be as easy to just upgrade individual parts in the system.
      Cooling can be a potential issue as well.
      Not to mention vibrations.

      Though, a lot of electric motors capable of delivering tens of kW tends to be water cooled regardless, so adding a bit of cooling for the driver shouldn’t be a major issue.

      I can see a solution like this make sense for applications where density is desired, for an example an electric car.

    1. Perhaps a little, but if this motor does pan out exactly as the specs say – which looks plausible to me – then current levels of energy storage will be more usable. If the motor is light and powerful you have more scope in your design to make good use of less energy dense or efficient storage mediums and can make better use of the energy you do carry – Not wasting as large a portion hauling the motor around means you can either afford a larger battery or possibly get the same/better performance out of a smaller lighter battery.

      Power to weight really is, for anything that has to move itself, a huge contributor to its efficiency – and why old economy aimed ICE cars can still do better miles per gallon than their modern equivalents (including fancy hybrids) – the old ones didn’t have as many creature comforts or mandatory safety features so weigh so much less that their smaller less efficient combustion engine has a really significantly lower amount of work to do.

    1. That will really depend on the scale you are talking about, and if you are including the energy sources mass. Or the energy required to refine the fuels etc – its not a really simple thing to give an answer to. In small scales its pretty hard to beat electric as making really small efficient combustion based systems requires vastly more precision machined interfaces with all the moving parts (not to mention the mass increase combustion engines get relative to their power output at small sizes).

  3. Just for the heck of it, I compared this with the good old Continental in a Cessna 172, that I’ve had the mixed fortune to spend several hundred hours behind. Let’s do a direct swap and see how it compares.

    The 140hp (100 kW) Conti drank 25 kg/hour of avgas to produce about 75 kW at cruise, and weighed about 130 kg. Normal tank fuel capacity was 115 kg (so, 4+ hours, though we had 7 hour tanks in ours, provided there were just two of us and no baggage on board).

    This new motor easily beats that old engine in weight and power output, so let’s replace the Conti and fuel with this motor and electric power: for the original engine + fuel’s 245 kg, you get the motor and a whopping 230 kg of battery. (though, too bad the battery doesn’t lighten up much as you burn it!)

    How long will 230 kg of battery last at 75 kW? Tesla’s Model S battery’s 6.3 kg/kWh is probably representative, so it looks like we can carry enough battery for almost a half hour of flight. Forgo baggage and passengers and you can get another 15 minutes.

    Oh, Yippee. Can’t even make the 45-minute reserve requirement.

    I invite someone else to do the same comparison with (say) a PT6a or a GE90, to pick two extremes.

    1. I mean, the obvious issue is you’re using complete motor and battery system for comparison. They aren’t claiming they’ve solved the “hydrocarbons are way more energy-dense than electrochemical cells” problem, they’ve just made a very power-dense electric motor.

      Everyone already knows batteries suck compared to gasoline/kerosene/whatever, this isn’t trying to address that problem.

      1. H3X says explicitly: “The HPDM-250 is an ultra-high power density integrated motor drive for electric aircraft. ”

        Unless there’s line of sight to provide the energy required, it’s bad business to expend resources into developing the actual hardware.

        Since “Everyone already knows batteries suck “, then one must conclude the product is not a business-viable piece of hardware, but actually hope, dreams, and smoke and mirrors.

        Just a swag, but I’m going to guess they’re actually going to pivot to military robotics after they’ve burnt this round.

        1. You don’t necessarily need to use batteries for electric propulsion. A hybrid system or hydrogen fuel cell are also viable energy sources. Hydrogen fuel cells show much higher promise than batteries for aerospace application due to their high specific energy. This is the direction the industry is moving. See ZeroAvia.You have to have both parts of the equation though for it to solve (the energy source + propulsion system).

        2. Batteries aren’t necessary for electric propulsion. Hydrogen fuel and a hybrid system are also both viable options. Industry is moving towards hydrogen fuel cell due to its high specific energy. See ZeroAvia. You need both parts of the equation for it to solve tho (high energy density and efficient energy source + high power density and efficient propulsion system)

          1. David Sanborn Scott covered hydrogen-powered airplanes pretty thoroughly in his book Smelling Land. TLDR: the inefficiencies and weight burden of fuel cells and electric motors makes hydrogen-electric air propulsion completely unworkable.

            In his rabidly pro-hydrogen view, he shows it’s possible you can make hydrogen-powered jets work by burning the stuff directly in a jet engine and dedicating a very large fraction of your fuselage to a liquid hydrogen tank.

            Despite his enthusiasm, he inadvertently shows that hydrogen-powered air flight is just a dumb idea, and dense liquid fuels are essentially required to make long-haul flight commercially viable.

          2. Hydrogen isn’t viable, but SOFCs can burn liquid hydrocarbons so the thing would still run on kerosene just the same.

            The trouble is, jet engines are surprisingly efficient at speed. Once you get up there, the fuel cell would be less efficient for the long haul. The further you go the less it makes sense to replace jet engines with electric propulsion of any kind.

        1. This isn’t “starting” from anywhere. Electric motors already have fantastic power density, and making a marginal improvement to reduce the weight of the motors is like blowing out candles while the house is on fire.

          It’s focusing on something you CAN do because it happens to be possible, instead of doing what you SHOULD do, which is not presently feasible, which means you’re not actually helping the problem at all. You’re just wasting time and effort.

        2. Btw. I’ve noticed a similar attitude with many of the “idea guys” who think about products in terms of what they should be instead of what they are.

          They’re basically designing the box to put it in before they have solved the fundamental engineering issues. They start thinking from what they can do, and the rest will solve itself, right? The same attitude then goes towards technologies made by others, in the sense that if anyone promises a nice thing that is not technically possible at the moment, they see it like the box that magically fills itself with the product.

          You just need to have the right box, the right framework, the right abstraction and paradigm, the right drive, and the rest is just a matter of time (and other people’s money) before some boffin comes up with the right stuff to put in the box.

    2. Bit of an unfair comparison to drop this electric motor in place of something with about half its performance – put it in place of a larger aircraft, one that actually needs something around this power level, and it will look much better. The engine it replaced would then weigh something like 200-250 kg, would drink more fuel so the tanks would be larger and heavier too, so you start looking at being able to have perhaps a ton of energy storage in whatever form as a drop in replacement.

      I expect it would have to be fuel cell of some sort not a battery, but with so much removed mass to replace I expect it would be internal volume for the energy storage that would be the trickier part – finding all that space, while keeping the COM correct and not annexing the human occupied spaces too much..

      Still don’t expect it to be better than ICE (yet), but if this pans out a motor with this much power to weight is a good step to making it viable.. With current battery/fuelcell tech I’d think in the right sized craft it would look viable already.

      1. unfair comparison to drop this electric motor in place of something with about half its performance – put it in place of a larger aircraft, one that actually needs something around this power level, and it will look much better.
        It scales just fine: the bigger airframe requires more power, more battery, so the runtime won’t change much. But just to try it, let’s drop this into a Cessna 185. Its 195 kW engine is 195 kg and drinks 41 kg of fuel per hour at 70% (137 kW) cruise. The tanks hold 186 kg, for 4+ hours of range.

        Dropping this electric motor in its place, you can carry 366 kg of battery = 58 kWh, which will last 25 minutes at cruise. Opt for the long range tanks and forgo cargo capacity, and you get an additional 8 minutes. Yipee.

        What’s really telling about battery-powered flight: you expend about a quarter of the battery energy just lifting the battery weight itself to altitude (and the glide back down doesn’t make up for the lift-induced drag from toting that extra mass cross-country).

        Just about the same as a Tesla: about a quarter of its stored energy is used to lug around the dead weight of the battery.

        1. By the specs you want to aim for an even larger plane still. I was thinking more in the range of the two or maybe four engine biplanes/ planes (the big boys of the lighter aircraft world). That probably need above the 130kW per motor cruising. (as I’d expect this electric to be most efficient above that, but as its not really fully characterized yet its hard to know and its got 250kW available for burst power so comparing it to an ICE that tops out nearer that number if fairer than one that tops out still below its continuous rating).

          You are not wrong lugging around an empty battery is a huge disadvantage to emptying the tank and being lighter as the flight goes on. But this motor being astonishingly light and performant (at least in theory) means you can see a cross over point where its possible a plane can use these and have useful range or flight time. Particularly if you design from the ground up for it.

          Being less massive means the frame can be less massive in many places as it needs to take lower loads, its also significantly smaller than a similar ICE so can have lower drag. With the size and relatively flat efficiency curve most electric motors have you could fit it to your glider style aircraft with minuscule battery (though a smaller motor still would make more sense there)).

          I’m not saying electric powered will ever match normal ICE planes with current tech, just that its getting to the point its a viable method – in the same way electric cars are becoming viable with hundreds of miles of range, good battery management allowing many cycles of fast charge without excessive wear etc. So if this motor delivers on the specs it promises it really helps make it possible.

  4. Would it really need the gearbox though? Of course it’s too high for a prop, but I think 20.000RPM is pretty decent for a small jet turbine. Couldn’t it be used to drive one of those? I’m not an aeronautical engineer so perhaps this is a stupid idea :)

    1. As far as I know jet turbines need to inject combustible fuel into at least one of the internal compression stages which effectively powers the earlier stages. I don’t know how they get the compression up to self-sustaining levels to begin with, but I wouldn’t be surprised if it’s some kind of electric motor already.

    2. The issue is that the power in a jet engine doesn’t come from the shaft spinning, it comes from all the expanding hot gases coming out the back end. Unless you’re running it like a generator, the amount of power being used by the shaft is a small percentage of the power it’s extracting from the fuel, so conversely when you’re running only the shaft you should expect to only produce a small fraction of the engine’s thrust

      1. That might have been true in 1950. But in a modern high-bypass turbofan most of the thrust and motive power comes from the fan, which is driven by the gas turbine powerplant. Replace that gas turbine and turn that fan with an electric motor of suitable power (100 megawatts, in the case of the GE9X), and you’ll get the same effect.

        Scaling up this H3X HPDM-250 motor to a HPDM-100000 motor will weigh 6 tons. Not bad, actually — very roughly speaking, about the same mass as the turbine and its associated support hardware. But it will burn about 630 tons of battery per hour. For each engine. A Dreamliner could carry almost enough battery to get to cruising altitude.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.