I have a toy quad copter that runs for about five minutes on a small LiPo battery pack. The larger multirotor units that I have seen also run on batteries and have flight times on the order of 15-30 minutes max. I have the vague sense that the energy density of chemical fuel is still greater than that of batteries, but that doesn’t take the mass/volume of an engine and generator into account.
If I wanted to build a quadcopter where the prop motors were powered by electricity from a generator hooked to an IC engine, how small could I scale such a thing? In other words, how small of an engine/generator/fuel unit can I make that will provide the equivalent amount of power, for longer, than a high-end battery pack of the same mass? Would one big engine be more efficient than a couple of smaller ones?
I’d love to see some little glo-plug engine and a tablespoon of fuel powering a quadcopter for half an hour, but as a total WAG I would guess that it would require something more like a lawnmower engine and a gallon of gas.
Chemical fuels have way higher energy density than batteries, but tiny motors are also very inefficient. It’s not obvious which wins, so let’s do some math.
By the numbers I can find, a Cox 0.049 motor produces about 42 watts and goes through a 5 cc fuel tank in 2-3 minutes (let’s say 2.5 minutes). I can’t get density figures on the fuel, but it’s a mix of castor oil, methanol, and nitromethane, so let’s say the density is 0.9 g/cc. Multiplying out, we get a final energy density of 0.71 g/kJ.
A modern LiPo battery might be 2.2 A-h @ 11.1 V and 201 g (taken from a semi-random selection from a hobby store). Multiplying out, we get 2.3 g/kJ.
So the fuel is about 3 times as energy dense here, even taking IC inefficiency into account. However, it ignores the mass of the motor and generator, as well as inefficiency of the generator.
On balance, I’d be surprised if you could make it a net win. The motor needs to be a minimum size just to lift your fuel+copter. I think this would irretrievably eat into the mass ratio, putting it back in favor for the battery.
One possibility is to use a combo multicopter/winged aircraft. Wings are very power efficient. You could get VTOL in pure electric mode, using a small battery just for takeoff, and then using the engine+generator as a sustainer for long distance flight. Still pretty dicey but more plausible than a pure VTOL hybrid machine.
Of course you can, once you get to a sufficient size. There’s a reason that all heavier-than-air vehicles which carry people use burnable fuel instead of batteries.
So, if there’s a large size where fuel is better, and there’s a small size where batteries are better, there must be some intermediate size where it changes over. The OP wants to know what that size is.
Or in other words, make a list of payload sizes, and then for each size, determine whether a battery or fuel would be better. At the small end of the scale, it’s going to be battery. At the large end of the scale, it’s going to be fuel. How far down your list do you get before it changes?
Without getting into the math there are bigger issues with quad copters. They have to be precisely controlled and that is not easy to do with a gas engine. I’m not saying it’s impossible but I would think it extremely difficult. It would have to be done with propeller pitch on individual blades that are fed by a single connected power source. The other way would be as the op suggests to have a gas generator feeding a conventional electrical motor system and that would need to be done so it doesn’t create a rotational effect.
The OP is asking about a hybrid design, though–a chemical engine, a generator, and electric motors. As far as I know, no full-size aircraft does this. Furthermore he was asking specifically about “little glo-plug engines.”
Multicopters in general don’t really scale very well. They maneuver by differential throttling, which stops working when the rotor gets too big (too much angular momentum). Chemical engines also don’t throttle nearly as quickly. Hence, helicopters use totally the different mechanism of tilting the blades for steering and lift. Even multirotor machines like the V-22 Osprey use this mechanism instead of differential throttling.
There is ongoing NASA research on the idea of using a chemical fueled engine to power a generator and batteries and then have the batteries power electric motors which turn propellers or rotors.
The idea of the hoped-for gain is similar to that of a Prius: by sizing the chemical engine(s) for the cruise case and not the takeoff case they can be smaller & lighter & more efficient. As well, by using multiple smaller motors & propellers instead of 2 or 4 huge ones they improve handling and reduce the severity of a single failure.
Right now the specific efficiencies don’t work out. The entire power train is much heavier than a turbine engine directly driving fans or rotors. They need about a 3x increase in power/weight of the electric motors and about a 10x increase in the energy/weight of the batteries before it works out.
And everything they’re doing is aimed at medium or large scale aircraft. The problems get worse as they get smaller. They can’t even use subscale models to test the concept since they can’t get those to even fly; they’re too heavy.
So bottom line for the OP: At the current state of the art, using a Li-ion battery pack is far better than a small IC engine driving a generator & batteries.
But lots of very smart folks are working on exactly the idea the OP proposes. And not just for toys.
The old two-stroke helicoptors were about the same size as a quadcopter, and had an engine and a fuel tank each a bit bigger than a large fist. So something between a tablespoon and a lawnmower.
The were also much more powerful, dangerous, difficult to fly, and much, much much noiser. The technology has been replaced, not because the engines became less efficient, but because electric motors have become more efficient, batteries have become more weight-efficient, and quad-copter control has become simple.
My guess is that the minimum size of a comercial drone is set by the size of the required payload and range. So if you were willing to accept a smaller payload and range, you could have a liquid-fuel quadcopter smaller than a commercial drone, but bigger than an all-electric quadcopter.
The exact size would be somewhere in there inbetween.
I race R/C boats, and even with the inefficient nitro two-strokes we use, they have one big advantage over the electric boats- I can re-fuel my boat in 10 seconds. The electric folks have to wait at least half an hour.
I looked up some numbers and it seems that modern turbofans are in the ballpark of 6 hp/lb, while electric motors are more like 4 hp/lb. So already that’s a loss, and that’s not even counting batteries or the generator. Presumably the generator can be undersized compared to the motors, but it’s still going to be significant.
Arguably, you can drive electric motors at arbitrarily high power levels if you can cool them sufficiently. So maybe they can get a significant increase if they use a fancy liquid cooling system or the like. Hard to say.
Batteries are getting better… slowly… but they’ll need a revolutionary change to achieve a 10x increase. I think there’s probably another 2x in Li-ion with relatively incremental improvements but definitely not 10x.
We’re optimizing for power/weight ratio here, not power/size or whatever. A modern brushless motor is not too different from a solid lump of copper, and the overall weight is going to be a strong function of the weight of the conductor.
Wikipedia has a nice table of conductors sorted by their “resistivity density product.” As one might expect, aluminum does better than copper in this regard; it has lower conductivity but is much lighter. Lots of aircraft already use aluminum wiring for this reason.
At the top of the list is sodium, with under 1/3 the resistivity density of copper. You can’t build anything out of sodium… can you? It’s too soft, and melts at too low a temperature.
I suggest using this to your advantage. Build the stator+rotor out of some non-conductive material with internal channels where the wiring would have gone. Fill the channels with sodium.
Once it warms enough (~100 C), you can now pump the sodium through the channels, and ultimately out of the motor, where you can send it through a heat exchanger and ultimately dump the heat overboard via radiator or other means.
Crazy, yes. I’m aware of what sodium does when combined with water :). But maybe crazy enough…
OP checking in here - I appreciate folks’ input so far.
Yeah, I’ve seen the YouTube videos of people working on quadcopters with variable-pitch rotors driven directly by IC engines, either via belts or shafts. Those are clearly larger designs with big engines (multiple engines in some cases), and they clearly must have separate electrical systems run off of batteries to operate the receivers/servos to control the variable-pitch systems. So now you have to refuel and recharge batteries. I’m looking for the smallest size for a workable system that doesn’t require the batteries at all; although if you can spec out a system where small batteries are included which get charged by the engine/generator system, I’d be OK with that too as long as the whole shebang produces the same power for longer than the same weight in modern high-density batteries.
I brought this up in the context of quadcopters because that seemed like an application where the electric motors used to power the props could be justified because of the control requirements, but the same question could be asked about any battery-dependent system, like a laptop or artificial limb. Of course, no one wants a laptop that farts two-stroke exhaust, so a quad seems like a good application for such a IC/electrical hybrid tech. There are of course many others as well.
I’m not married to the idea of glo-plug engines; that was just a neat notion. Some more google research has pulled up stuff like the MICE generator or Toyota’s free piston linear generator. Like I said, I suspect that the smallest effective motor/generator system would be comparable to a lawnmower engine. As Chronos said, I’m wondering what the crossover point is between batteries and fuel-burning engines, on a weight basis.