4 ms·

But the piston engine design really have to be slower? For example, some WWII fighter aircraft could be quite fast, easily comparable to the claimed top speed

But the piston engine design really have to be slower? For example, some WWII fighter aircraft could be quite fast, easily comparable to the claimed top speed of Tomahawk missile that can be found on Wikipedia and since you'd have less drag I don't think you'd need a huge piston engine. I'm guessing you'd want something like 500 kW out of 200 kg of engine, but that's similar to the claimed performance of the apparently not-quite production ready Zoche Aero-Diesel.

xyzzyz 2y ago
Looking this up, Zoche Aero-Diesel were something like 1-3 horsepower per kilogram. You can buy a jet engines for model planes today that will generate 150 HP of exhaust gas power while weighing under 4 kg, giving you something like 30-40 horsepower per kilogram. The difference is just tremendous. Piston engines are not even that much cheaper, and the only reason they are is that we produce literally billions of them every year, so it pays to invest in tooling. High performance piston engines are more complex, have more parts, and more complex assembly than simple jet engines designs that handily outperform them.
- impossiblefork 2y ago
  Yes, but those engines have tremendous fuel consumption. 90 ml per minute for an engine weighing 2.4 kg. Over an hour that's 5.4 litres, 2x the weight of the engine. But with a heat exchanger to heat the compressor air with the exhaust the situation will be changed, so I still you're right to some degree but I am not sure about cost. Competing turbine engines are much more expensive than piston engines and you can actually build pistons engines at reasonable cost. Turbotech in France produce a small fuel efficient jet engine, but it's like 90 kW weighs 90 kg and probably costs US$50,000+. I think there's a reason a lot drones have gone with Wankel engines.
  - xyzzyz 2y ago
    > Over an hour that's 5.4 litres, 2x the weight of the engine. This perspective is completely wrong, because it actually penalizes smaller, lighter engines, all else being completely equal. Imagine you take a 100 kg diesel engine that produces 200 HP, and uses 10 kg of fuel per hour, and through insane feat of engineering, you minimize it to only be 5 kg, while producing same power and using as much fuel. This makes the engine obviously better, but by your logic, it now uses 2x its weight in fuel, versus just 0.10x as before. This logic is obviously wrong. You need to instead ask: how much fuel does each engine need to produce equivalent amount of energy? On this metric, jet engines are not significantly worse than piston engines. > Competing turbine engines are much more expensive than piston engines and you can actually build pistons engines at reasonable cost Yes, if you compare a slightly modified off-the-shelf automotive piston engine produced in millions, to a small batch jet engine, then yes, the former is cheaper. However, if you need your drone to be both very fast and fly very far (and these two are very important in military context), then off the shelf automotive piston engines are really not an option, and instead you need to invest in something very custom, which would destroy the per-unit cost advantage of piston over jet, and that’s ignoring benefits like smaller form factor allowing for much bigger payload (extra 100 kg available for warhead does make a difference).
    - impossiblefork 2y ago
      Yes, and that is in fact not my perspective, I only mentioned the ratio in order to illustrate that the fuel weight is substantial. The right condition is of course J/kg for the fuel + engine + propeller/shroud/etc. package, and also taking into account cooling drag and frontal area. >Yes, if you compare a slightly modified off-the-shelf automotive piston engine produced in millions, to a small batch jet engine, then yes, the former is cheaper. However, if you need your drone to be both very fast and fly very far (and these two are very important in military context), then off the shelf automotive piston engines are really not an option, and instead you need to invest in something very custom, which would destroy the per-unit cost advantage of piston over jet, and that’s ignoring benefits like smaller form factor allowing for much bigger payload (extra 100 kg available for warhead does make a difference). The problem is that jet turbines of the required efficiency and weight are actually expensive. They have high temperatures internally, and continuously, and have all these blades and bearings. I think it might be feasible to make them much more cheaply than they can be manufactured today, but a Wankel engine can be mass produced with present technology and is not incredibly bad.

xyzzyz 2y ago

Looking this up, Zoche Aero-Diesel were something like 1-3 horsepower per kilogram. You can buy a jet engines for model planes today that will generate 150 HP of exhaust gas power while weighing under 4 kg, giving you something like 30-40 horsepower per kilogram. The difference is just tremendous. Piston engines are not even that much cheaper, and the only reason they are is that we produce literally billions of them every year, so it pays to invest in tooling. High performance piston engines are more complex, have more parts, and more complex assembly than simple jet engines designs that handily outperform them.

impossiblefork 2y ago
Yes, but those engines have tremendous fuel consumption. 90 ml per minute for an engine weighing 2.4 kg. Over an hour that's 5.4 litres, 2x the weight of the engine. But with a heat exchanger to heat the compressor air with the exhaust the situation will be changed, so I still you're right to some degree but I am not sure about cost. Competing turbine engines are much more expensive than piston engines and you can actually build pistons engines at reasonable cost. Turbotech in France produce a small fuel efficient jet engine, but it's like 90 kW weighs 90 kg and probably costs US$50,000+. I think there's a reason a lot drones have gone with Wankel engines.
- xyzzyz 2y ago
  > Over an hour that's 5.4 litres, 2x the weight of the engine. This perspective is completely wrong, because it actually penalizes smaller, lighter engines, all else being completely equal. Imagine you take a 100 kg diesel engine that produces 200 HP, and uses 10 kg of fuel per hour, and through insane feat of engineering, you minimize it to only be 5 kg, while producing same power and using as much fuel. This makes the engine obviously better, but by your logic, it now uses 2x its weight in fuel, versus just 0.10x as before. This logic is obviously wrong. You need to instead ask: how much fuel does each engine need to produce equivalent amount of energy? On this metric, jet engines are not significantly worse than piston engines. > Competing turbine engines are much more expensive than piston engines and you can actually build pistons engines at reasonable cost Yes, if you compare a slightly modified off-the-shelf automotive piston engine produced in millions, to a small batch jet engine, then yes, the former is cheaper. However, if you need your drone to be both very fast and fly very far (and these two are very important in military context), then off the shelf automotive piston engines are really not an option, and instead you need to invest in something very custom, which would destroy the per-unit cost advantage of piston over jet, and that’s ignoring benefits like smaller form factor allowing for much bigger payload (extra 100 kg available for warhead does make a difference).
  - impossiblefork 2y ago
    Yes, and that is in fact not my perspective, I only mentioned the ratio in order to illustrate that the fuel weight is substantial. The right condition is of course J/kg for the fuel + engine + propeller/shroud/etc. package, and also taking into account cooling drag and frontal area. >Yes, if you compare a slightly modified off-the-shelf automotive piston engine produced in millions, to a small batch jet engine, then yes, the former is cheaper. However, if you need your drone to be both very fast and fly very far (and these two are very important in military context), then off the shelf automotive piston engines are really not an option, and instead you need to invest in something very custom, which would destroy the per-unit cost advantage of piston over jet, and that’s ignoring benefits like smaller form factor allowing for much bigger payload (extra 100 kg available for warhead does make a difference). The problem is that jet turbines of the required efficiency and weight are actually expensive. They have high temperatures internally, and continuously, and have all these blades and bearings. I think it might be feasible to make them much more cheaply than they can be manufactured today, but a Wankel engine can be mass produced with present technology and is not incredibly bad.