Designing Compact Gasoline Generator Prototype For Drone Use

Lithium batteries and brushless motors helped make multirotor drones possible, but batteries only last so long. Liquid fuels have far greater energy densities, but have not  been widely applied in these roles. [Tech Ingredients] has been experimenting with a compact gasoline-fueled generator, with the aim to extend drone flight times well beyond what is currently possible with batteries (Youtube link, embedded below).

The build began with a single-cylinder, four stroke engine. However, torque spikes and vibration made things difficult. After some iteration, the design settled on employing two single-cylinder two stroke engines, fitted with a timing belt to keep them 180 degrees out of phase. In combination with a pair of balanced flywheels, this keeps vibration to a minimum. Brushless motors are used as generators, combined with rectifier diodes and capacitors to smooth the voltage output. The generator is intended to be used in parallel with a lithium battery pack in order to ensure the drone always has power available, even in the event of a temporary malfunction.

This is a build with plenty of promise, and we can’t wait to see what kind of flight time can be achieved once the system is finished and flight ready. We’ve seen others experimenting with hybrid drones, too.

35 thoughts on “Designing Compact Gasoline Generator Prototype For Drone Use

  1. Since they’re using two strokes anyway for reducing vibrations, might look into two-rotor wankel engines.
    Although patents be damned have limited their developments for years and left their glaring flaws compared to traditional two-strokes festering.

    1. I came here to say the same thing. Wankel engines have a long history of use in target drones. They might be relieved to be in a drone that isn’t being shot at.

  2. I would think that for the purposes of compact and lightweight power generation, a gas turbine (possibly based off an RC/hobby jet engine) would be a better choice. No heavy flywheels, much less vibration.

    1. The problem with gas turbines is the speed (30,000+ rpm). In order to develop torque in a turbine you have to run it at high speeds. You then need a gearbox to reduce speed to the generator. Back comes the weight.

    2. Last I heard (which was admittedly several years ago) turbine RC aircraft have very large fuel tanks and still pretty short flight times. Turbine generators are certainly a good option at large sizes, but I don’t think they scale down very well.

      1. Simple turbines don’t fare very well when scaled down, but there’s a way to remedy the low efficiency with a “recuperator” that recycles exhaust heat back to the intake. This is a critical feature of any microturbine if you want to have high efficiency at the shaft. If you just want thrust, then it’s not important at all.

    1. turbines are great and stable power source, but something extra is needed to counteract the thrust they produce. or you might put more of them in a special arrangement so they cancel each other’s thrust.

      1. Just use the thrust! A multirotor drone has a steady minimum lift requirement to not fall like the brick it is. The rotational forces should be within the flight controllers capability to best with the normal electric motors rotations. If dealing with fixed wing having a steady thrust is also not the end of the world, though going there it might as well be kept simpler and just be liquid fueled straight up.

  3. maybe I’m missing something obvious here but if you want to make a drone liquid fueled why not do it the direct way with good old combustion motors instead of going the generator rout? or why no try to go with fuels cells? no vibrations and no moving parts….

      1. Thrust control the traditional way is handled by blade pitch control – not by adjusting the motor. It’s actually faster than electric control and can achieve thrust control (vectoring) within a single rotation of the blades.

        It’s just mechanically more complex than just screwing a simple propeller onto an outrunner BLDC motor.

      2. Variable pitch propellers running at constant or near constant speed. Helicopters do that. A few years ago there was a demo of a quadcopter with variable pitch props showing off precision maneuvers with no overshooting, doing rolls without having to first jump upwards, even reversing pitch to fly upside down.

        Quite odd that such hasn’t become widely used, especially in racing. Combine variable pitch with swashplates and a quadrotor would have the best features of a helicopter, a quad with fixed pitch props, and one with variable pitch props. Should be able to hover while tipped over at a pretty significant angle by using swashplate and variable pitch.

        1. i beleive it is a durability issue. mechanical systems to adjust pitch have many parts to break even during a light prop strike. there are a couple kits available but they are more expensive than motor direct to prop copters.

  4. No doubt this guy is smart. But he is one of those theoretical type engineers. I.E. book smart, but when it comes to application no so much. Anyone who has spent any real time in the RC industry could have told him a few things, and saved him allot of time and money.

    #1) Ya, a brush-less motor will generate power. They were never designed to run allot longer than a typical battery would last. Not to mention he has no cooling to force air into them to help cool them. Moreover, to have it push that many amps for any lengthy period of time WILL cook the windings.
    #2) He made his own exhaust pipes. Again, we would have told him on a 2-stroke, the pipe needs to be a tuned pipe. Just whacking something on there to make it quiet will seriously degrade its performance.
    #3) Has he even stepped back and looked at how huge and heavy this whole thing is? Your going to need a mammoth size quad, or really an octo copter just to pick it up. If that’s the case whats the point. It would be better to take that octo copter throw an extra battery on it and have it carry that 12 pack of beer to me.

    1. 2-strokes do not need tuned pipes, they are happy to run on stock bin type mufflers or regular “can” silencer on model 2-strokes. Tuned pipes are just a must if you want to squeeze power.
      What 2-strokes don’t like is running with an open exhaust.

      1. The exhaust has to expand, otherwise the flushing action doesn’t work quite right and the engine may gulp back its own exhaust. In other words, 2 stroke engines don’t like a straight pipe. Either no exhaust, or an expanding “bell” of some sort.

        The point is that the expansion of the exhaust gasses out of the cylinder makes a temporary vacuum that gets filled by fresh fuel mix from the intake port. Usually the engine is constructed such that the piston also pressurizes the crank case on the down stroke, which then causes the cylinder to flush with fresh fuel mix rather than pulling the exhaust gasses back in – but if there’s too much backpressure in the exhaust pipe then the exhaust gasses just bounce back and the engine runs poorly if at all.

        For the tuning of the exhaust, that’s because the exhaust gas flow will draw some of the fuel mixture out the exhaust port (hence why 2 strokes are really dirty running), the tuned exhaust will have a backwards reflected pressure wave that stuffs the fuel back into the cylinder at a certain RPM, which gives you more compression and power at that particular speed.

    2. You talk about being the right kind of smart yet you don’t know the difference between your and you’re. Also it’s not “allot.” It’s “a lot.”

      And 2-strokes don’t “require” tuned pipes. Plenty run just fine with a random tube coming off them. They only need a tuned pipe if you want to get your HP/torque at a certain RPM.

      1. Which is what a genset you want to fly needs, maximum power output at a narrow RPM range. The engine needs to run at its optimum RPM while the electric motors handle the variable speed needs – just like a diesel electric locomotive.

  5. I was working on the same thing, but one of the first design criteria I addressed was not having it weigh 60 pounds. Seriously, with that sort of design choices, it’s going to be 20 years before he comes up with something useful.

  6. Walkera already have these in market. They’re pretty spendy but they advertise pretty long flight times.
    120mins with a 2kg payload (Only had a the briefest of reads – was supposed to be paying attention on a video call).

    Search and rescue, firefighting etc.

  7. Not that I don’t no much about, it seems to me for flight that doesn’t require hovering, fixed wing UAVs, would be the logical choice. Single rotor UAV where hovering is require? For looking over crops, and pasture, I’d think the fix wing could do what’s required at lower cost.

  8. Years ago, I knew a gentleman who briefly held the radio control endurance record for airplanes. His craft stayed aloft for well over 24 hours. I recall him saying that the engine was diesel since it was much lighter weight and the fuel had a higher energy density also saving weight. These are recollections of a kid, so I don’t have proof of any of it, but it is worth looking into.

Leave a Reply

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