Drone Flies For Five Hours With Hydrogen Fuel Cell

Multirotor drones have become a regular part of daily life, serving as everything from camera platforms to inspection tools and weapons of war. The vast majority run on lithium rechargeable batteries, with corresponding limits on flight time. A company called Hylium hopes to change all that with a hydrogen-powered drone that can fly for up to five hours.

The drone uses a hydrogen fuel cell to provide electricity to run the drone’s motors and other electronic systems. Thanks to the energy density advantage of hydrogen versus lithium batteries, the flight time can be greatly extended compared to conventional battery-only drones. Details are scant, but the company has gone to some lengths to build out the product beyond a simple tech demonstrator, too. Hylium touts useful features like the short five-minute refueling time. The drone also reportedly features a night vision camera and the capability to transmit video over distances up to 10 kilometers, though some of the video of these features appears to be stock footage.

Hylium claims the liquid hydrogen canister used for the drone is drop-safe in the event of a problem. Notably, the video suggests the company tested this by dropping the canister concerningly close to an active motorway, but from what we see, nothing went awry.

A drone that can fly for five hours would be particularly useful for autonomous surveillance and inspection roles. The additional loiter time would be advantageous in these roles. We’ve seen other aero experimenters exploring the use of hydrogen fuel cells, too.

82 thoughts on “Drone Flies For Five Hours With Hydrogen Fuel Cell

  1. I wonder how much this would be extended by fitting it with those toroidal quiet rotor blades and thin solar panels, perhaps some sort of vantablack/isotope teg, rf harvesting or just landing on a powerline occasionally. If it could pull in moisture from the air and break it catalytically, it might be able to fly iindefinitely on hydrogen alone.

    1. + Toroidal blades would be an improvement.
      – Solar panels would likely increase drag.
      – Benefits from a normal TEG would likely be marginal at best.
      ? The expense of RTGs tends to limit their use to niche applications in rare or special situations.
      – The weight of RF harvesting hardware will decrease overall flight time.
      – Landing on a powerline ends your flight… and you could just recharge a normal battery like that.
      – We don’t currently know a way to crack water without using more energy than it consumes. If we did then hydrogen generation would already be “green”.

  2. A 250 gram camping gas canister contains 3.4 kWh of energy. That’s about half a liter of liquid.

    A similarly sized liquid hydrogen container would hold 1.18 kWh of energy.

    If you swap out the PEM fuel cell for an SOFC, you can burn butane or LPG just the same and get three times the loiter time for the same fuel loading, and it’s safer to handle, and it’s easier to source the fuel, and you can carry the fuel and keep it without it boiling off constantly. Heck, even a small engine generator would beat the hydrogen drone in flight time using LPG.

    What’s the point of using liquid hydrogen? It’s just more trouble than its worth.

        1. If trash was valuable as raw material, they would pay you for it instead of making you pay a deposit or a recycling fee as part of the price of products.

          However, the main point about recycling is that the stuff wouldn’t end up in a landfill.

    1. Presumably because hydrogen generation is expected to become environmentally friendly (though I really doubt it). I agree but I would say use synfuel to keep it environmentally friendly.

      1. There’s portable 500-1000 Watt generators available for RVs and boats, but most of the companies are building them for grid power generators and cell tower backup systems.

        1. I saw a few, but if I have to call for a quote I assume the price is going to be high, and the ones that have a site full of old marketing I assume are mostly vaporware. And yeah, pretty low power especially power to weight. Great concept, but early days yet it seems for consumers.

          1. They usually come with a lead-acid buffer battery built in, because the SOFC doesn’t respond to transient power demands very well, so that’s what makes them heavy.

        1. It’s not really an issue. The hot fuel cell stack is housed in a double wall insulated box that isolates the heat from everything else.

          Regular piston engines too have to deal with combustion temperatures in the 500-1000 C range and we have no trouble making that work.

          1. It’s the same weight penalty as keeping liquid hydrogen in an insulated bottle. The fuel system becomes very much lighter when using butane or propane.

    2. Burning something requires an internal combustion engine coupled to a generator to turn the rotary motion in to electricity, or a combustion chambber and solid state peltier style device in thermoelectric generation mode (very poor efficiency). Hydrogen just needs a fuel cell to give energy straight away in the form of electricity. The fuel cell is much lighter weight than the engine could be. For something small like a drone this matters. So Hydrogen is giving 1.18kWh of useful electricity, a hydrocarbon fuel might give 3.4kWh thermal, but would give a lot less electric after conversion losses are accounted for, and would need heavy conversion equipment.

      1. > So Hydrogen is giving 1.18kWh of useful electricity

        It doesn’t. That’s just the heating value of the hydrogen. A PEM fuel cell or SOFC would both waste about 50% to heat.

        A small engine generator would have an efficiency around 25% and put out about 850 Watt-hours from the propane fuel, while the hydrogen PEM fuel cell would give about 590 Watt-hours out of the hydrogen.

        >The fuel cell is much lighter weight than the engine could be

        Fuel cells have very low power density compared to piston engines. It’s likely to be the opposite case.

        1. The DOE estimate for state of the art fuel cell systems in 2020 was 860 Watts/kg.

          Aviation engines usually produce between 1000-2000 Watts/kg and race car engines can reach 2500 Watts/kg, so the piston engine is much lighter for the same application.

          1. >You could use the information

            I’ve read it already at some point. One of the important points about PEM fuel cells like the Mirai is that the efficiency drops when the power output increases. It can achieve 62% efficiency at near idle, but at full load (i.e. high power density) it starts to drop as it overheats.

            SOFCs on the other hand like running hot. The ionic conductivity of the ceramic membrane gets better as it heats up. The harder you run it, the more power it gives you, which is a feature that lends itself better for miniaturization because you don’t need to over-provision the stack to maintain efficiency.

          2. >Still not piston engines my dude.

            If you want regular piston engines, go look at F1 racing engines. 7-8 kilowatts per kg power density. Of course it doesn’t compare because it’s built to last for just one race and the tolerances are so exact that the pistons are seized up when the engine is cold.

          1. The Mirai has 2,000 W/kg gross power density, but the net power density according to the DOE document I’m looking at is just 860 W/kg. From what I understand, the peak stack power can’t be maintained without overheating it.

            (DOE Hydrogen and Fuel Cells Program Record, August 31, 2020)

          2. > Give us a MORE powerful piston engine.

            Here’s one more powerful:
            Mazda 13B-MSP Renesis 1.3 L Wankel engine, 184 kW, 1500 W/kg.

            And here’s one with a better power density:
            O.S. Engines 49-PI Type II 4.97 cc Wankel engine, 0.934 kW, 2800 kW/kg

          3. To put things in perspective, typical helicopter turboshaft engines produce something between 3000 – 8000 W/kg.

            Scaling down a jet turbine to drone size takes a big hit on the efficiency though, but it would still be lighter due to the better power density, and achieve approximately equal run time on equal volumes of fuel, AND it’s a friggin’ turbine: just point it downwards.

      2. > would need heavy conversion equipment.

        Regular 50/60 Hz generators are massive, but when you’re making DC out of a high revving engine, the generator doesn’t need to be much bigger than the BLDC motors of a drone – about the size of two hockey pucks.

        1. Just want to say thanks, dude. Hackaday comments have grown rather dopey such that I’ve begun playing ‘moron or kid’. For example: propelling a yacht with a USB fan. Moron or kid? Anyway your comments are refreshingly neither, can’t believe a moron/kid blasted you here for overcommenting. Keep it coming

    3. if you have some source of room temperature or high SOFC, SPILL THE TEA!

      if you swapped the PEM for an SOFC you would need to add pretty significant weight in preheaters, insulation etc. 900-1050C operating temperatures just dont scream compact, light, nor drone friendly.

      The point of using hydrogen is that EVERY FUEL CELL uses hydrogen, some just have to make it before they can produce. Hydrogen is just GTG.

      1. SOFC can burn straight up carbon if you want to make it so. It pulls oxygen ions through the membrane rather than hydrogen ions, so it can directly burn hydrocarbon fuels without first reforming them into hydrogen.

        And these days SOFCs are reaching lower temperatures in the 400-500 C range. Makes them more efficient as well.

          1. Besides, the best insulation is nothing: a vacuum. If the fuel cell stack is inside a dewar bottle, it can be as hot as you please, and the insulation weighs nothing because it is literally nothing.

          2. The weight of containing a fuel cell which operates with compressed gasses at high temperatures inside it within an insulative vacuum shell is ZERO?

            So I guess your with the guy thats got the vacuum unfilled airship hes trying to fund around the bay. Good luck. Read more, comment less. Everyone will be better for it. Enjoy your bliss!

          3. A double wall stainless steel cylinder isn’t that heavy and the SOFC operates close to ambient pressure between 1-8 bars. The mechanics of it isn’t rocket science.

            Meanwhile, PEM cells need active cooling if you want to sustain high power, so you got to factor in a water pump and a radiator.

          4. More precisely, a SOFC stack is like a deck of cards with channels between them. The stack is compressed together around the perimeter with bolts, and this is placed in a thin walled steel cylinder. The hot parts and pressurized gases are inside the stack and the stack is insulated from the container by the empty space in between. It can be filled with inert gas, or indeed, a vacuum. The stack is glowing cherry red hot, but the container reflects most of the heat back and doesn’t get nearly as hot.

        1. >Besides, the best insulation is nothing: a vacuum.
          >A double wall stainless steel cylinder isn’t that heavy

          So you have a magical stainless alloy that despite having no mass itself can withstand ANY FORCE!
          Seriously MrWiztard, you should leave HAD and go save the world with all your brilliance. Stop wasting time patting yourself on the back with shill accounts! Enjoy your bliss!

          1. Don’t play stupid. Go to your local kitchen supply store and buy a stainless double wall thermos bottle. It ain’t heavy.

            And the insulation still weighs nothing, because it’s the vacuum that is insulating, not the steel bottle. You just missed the joke.

      2. >if you have some source of room temperature or high SOFC, SPILL THE TEA!

        Not room temperature, but Ceres Power appears to have their SteelCell series SOFCs which are meant for residential CHP and apparently as electric bus range extenders.

        1. Wow, they claim a lot, although it’s been a few years and I would like to see more in order to believe it. 3600 thermal cycles without significant problems and tiny degradation per thousand hours? Plus 60% from fuel to AC power is high too. That’s the kind of performance that I’m dreaming of when I think about replacing engines with fuel cells.

          1. The thermal cycles limit is a bit of a problem, since it still means you can only shut it down a limited number of times. Think in a car, two shutdowns a day means 5 years of useful life, which is too short. You may start and stop the car several times a day, which means the stack is a consumable item that needs to be replaced every few years.

          2. True, so you won’t get away from having a fairly good amount of battery capacity in frequently used things like cars, which you’d need anyway in order to buffer the power for acceleration. But those little RV propane SOFC units say something like 250 cycles, which is just awful, so this is still pretty good depending on specifics. Even for a car, if you can charge at home but you exceed the capacity of a small battery a bit more than six times a week (running errands on the sixth day, for example) you’re still looking at 11 years or so. That’s worse than an engine, although you’d have been saving money all that time and maybe it’d be easier to replace than an engine or a big battery. Ideally you’d have enough of a battery buffer to last the whole day, and your fuel cell might not even be used in a normal week, but when you did it would let you drive all day on a road trip with a very small amount of fuel. What I was thinking of in terms of replacing engines also includes replacing small engines in all sorts of things, particularly portable generators but also a lot of lawn/garden/farm equipment, things like atv’s, maybe occasionally welders and air compressors and things.

          3. Then again, if the stack is just a replaceable module that doesn’t cost a whole lot, then why not just keep replacing it? That’s what they do with the cell tower backup systems: the fuel cell is a bundle of rods that you replace every N months.

    4. An SOFC requires a running temperature of at least 600c and is made of heavy ceramic cells. The PEM fuel cell on that drone, however, could be made from acrylic with a few stainless shim pieces. The SOFC cells are heavy, bulky, need to be pre-heated, and can degrade from the catalysts being ‘poisoned’ by impurities in fuels

      They’re currently used for stationary power generation because i don’t think we’re ‘there yet’ on clever and compact production methods for the ceramic cells

      A good middle ground is to use ammonia for the fuel, but that requires more weight for the cracking device

      1. PMMA (Acrylic) has its glass transition temperature between 100 – 170 C and high power PEM cells get hotter than that. If you want to make one out of plastics, you won’t be getting much power out of it without melting the thing – or you need to add cooling channels and water circulation, and radiators to it, which defeats the weight advantage anyways.

        SOFC gets poisoned by sulfur in the fuel, not much else. They can deal with CO2 and CO. The reason why they are heavy and bulky is because they’ve been developed mostly for stationary CHP use so there has been no need to make one light enough. The fuel cell stack itself though is small. One playing card sized metal and ceramic wafer gives you 10-15 Watts of power and a kilowatt stack would probably fit inside a small beer can.

      2. You have to consider how much power the drone uses. If it’s a big drone, it may need as much as 5 kW of power at full takeoff load and climb. Obviously if the fuel cell is going to supply that, it’s going to produce waste heat in approximately equal measure.

        When you put 5 kW of waste heat into a stack of plastic membranes, it’s going to melt. That’s what limits the practical power density of polymer PEM cells. The attainable power density of PEMFC is 39.7 kW/kg while SOFC has been shown to work at about 2.5 kW/kg but that high number for PEM is only available at very low absolute power levels – in a tiny device that can be effectively cooled.

        Other types of proton exchange fuel cells exist. PAFC uses phosphoric acid suspended in a silicon carbide sponge, which works up to 200 C. The acid boils off at 212 C.

    1. I read an article a long time ago about a drone that flew around a city all day long and recorded everything – they actually managed to capture an assassination then were able to track the perpetrators back to a house by running the video backwards and watching where the car came from.

      So literally a time machine.

      Wonderful things could happen with small cheap all day fixed wing aircraft.

      Or significant malfeasance.

      1. I believe you may be referring to Gorgon Stare, or a similar panopticon system.

        It’s quite likely that the “actually managed” was fabricated to sell the technology. Were you watching a documentary or an advertisement? Or one dressed up as the other?

        How many unskeptical people wouldn’t be able to tell the difference?

    2. yep. drones are incredibly inefficient. you’ll see the better contenders in the field are hybrid vtol/fixed wing. I would love to see this cell on one of those.

  3. Strong proof of hydrogen’s superior energy density as compared to batteries, strong proof that we should be ditching designs for electric cars and building a hydrogen filling station infrastructure instead (the tech for hydrogen cars themselves is ALREADY mature).

    1. We have the infrastructure for ammonia which an energy dense fuel that is less tricky to handle than hydrogen. Another less studied fuel is Butanol that can be used in place of gasoline and can be renewably sourced.

  4. Imagine the bang and the shrapnel spray when somebody take a potshot at one of those. The sort of thing that could happen in many big US cities where there are enough people around to get hurt. There are safer and more energy dense liquid fuels that can be used in fuel cells, and they can be sourced via renewable processes so there is not an issue with CO2 emissions.

    1. Hydrogen still needs to diffuse some, and form a mix with air in the presence of an ignition source – for maximum bang.

      The bullet that ruptures the tank sparks too early, the boom won’t be as explosive as you’re envisioning.

      As for shrapnel, a properly designed tank works under tension and would puncture but remain in one piece.

      Now a piece of Chinese lithium might go up at any point in time, burn hot for a prolonged period, and be difficult to extinguish – the hydrogen fireball would be comparitively short-lived. Fire will cause more structural damage

  5. One of the biggest reasons why I don’t wish for vastly improved batteries is because the drones will fly longer, which would (unfortunately) be bad; for obvious reasons
    So I’m at least glad that this method makes them more vulnerable. but not vulnerable enough I fear.

  6. If you’re going to burn a chemical fuel, as opposed to an electric redox battery, it seems like it would be more efficient to just burn it in a turbine and use the kinetic energy directly in a turboshaft, as suggested up-thread in a nested comment I can’t reply to. Similar to a “conventional” helicopter. Sure, reynolds numbers say it will be less efficient than full-size turboshaft, but it couldn’t be worse than running a fuel cell to drive electric motors on a multicopter?

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.