The title says it all really,
Could the way drones fly (with four sets of rotors) influence the design of passenger carrying aircraft?
Are there any examples already in use? (I’ve only seen helicopters with 2 sets of rotors).
Yep. Self-piloted passenger-carrying drones are in development.
As shown, the idea works for smaller copters. They’ll be less efficient than a normal helicopter, but for short-range flight it doesn’t matter much, and hybrid designs with wings may solve that problem as well.
The approach doesn’t scale, though–quadcopters control their movement through differential thrust; that is, some rotors are sped up relative to the others, which tilts the craft in the opposite direction. This requires a fast response time, which is incompatible with the big rotors you’d need on a large passenger craft. You couldn’t change the rotor speed fast enough to maintain stability.
You could solve the problem with cyclic control as on a normal helicopter, but then you’ve just made things worse; each rotor is as complex as a normal one but now you have four or more of them.
Some kind of guard around those blades would be nice :eek:.
In the event of engine problems, planes can glide down, and helicopters can auto-rotate down. What are giant drones going to do? Assuming loss of only one engine, would the remaining three be able to compensate? And if so, be able to compensate quickly enough to avoid splat? I’m sure ballistic parachutes will be suggested, but as always, there is the death zone between the ground and sufficient altitude to deploy.
Yes. https://www.ted.com/talks/raffaello_d_andrea_the_astounding_athletic_power_of_quadcopters
3 bladed flight is possible and with near perfect control algorithms like demonstrated in the video, it would be possible for the drone to adapt to missing rotors almost instantly.
Also, a giant drone isn’t going to have 4 propellers. With giant drones, you scale up by increasing propeller size to some maximum size (probably around a foot long) and then you add more propellers thereafter. The reason is so you can change blade speed rapidly to achieve differential thrust for stable flight.
In essence, each rotor/motor/digital motor controller need to be independent modules from all the others. Probably packaged as such. The batteries need to be built into a mesh of wiring busses so that most faults and fires will only reduce current availability but not reduce it to zero. There would need to be several redundant flight computers, and a mesh of control wiring so that the mentioned fires and other failures cannot cut all paths to reach all the remaining motors.
These requirements are tough, but I think roughly achievable. You would increase your flight range via a hybrid design, where you would transition to conventional winged flight after a VTOL ascent. You also might have an onboard jet turbine engine to recharge your batteries during flight, though there are weight issues with this.
Are we going to see widespread numbers of passenger carrying fly drones? No idea. It’s a feasible idea in principle, and obviously the prototypes are already flying, but there are serious cost issues, battery lifespan issues, and the FAA requirements to get to a salable passenger aircraft are very onerous.
There are similar moves afoot with winged airplanes.
The idea is to decouple “engine” from “thrust”. IOW power production and thrust production are separated so each can be numbered and sized for maximum efficiency and maximum redundancy.
So one might have 3 very large engines producing electricity and 25 much smaller propellers producing thrust. The loss of any one engine has no impact on controllability and moderate impact on total power output whereas the loss of a small number of props has no impact on total thrust output and only modest impact on controllability.
The challenge, as always in aerospace, is keeping the intermediate systems light enough, small enough, efficient enough, and robust enough.
This is not greatly different in concept from a modern locomotive. A diesel engine produces electricity and electric motors produce torque on the wheels. One thing most (all?) locomotives lack today is battery storage and a cruise-sized power source. No doubt there’s some research on that too.
Many flavors of hybrid cars are another example of the same concept.
Two issues - reliability and human control.
Until lately, reliability was a concern. As others have mentioned, if one engine/rotor fails, then what? The more pieces you add, the more things to go wrong. Plus, allowing for the obvious, like running out of fuel. The Osprey, for example, (Two big tilting rotors, one at the end of each wing) had an extensive history of assorted failures. To compensate for engine failure, they added a connecting driveshaft so one engine could rive both rotors well enough to land. More things to go wrong… drive couplings that had to be light but handle hundreds/thousands of horsepower. And so on…
It’s only recently that computer control has been able to handle the 3 or more rotor control issues. Standard helicopters used mechanically linked controls, so a human could control one or two rotors and nothing was doing the job for them; but 4 at once? Now it’s getting a bit complex. But of course, a computer is just one more thing to possibly fail…
Yes this is already being done. The Ehang 184 can fly for about 20 min carrying up to 220 lbs. It has had over 100 successful manned test flights and supposedly will be used as an air taxi in Dubai this year.
http://www.ehang.com/ehang184/specs/
Test flight of full-size vehicle without passenger: - YouTube
While it looks like a quad copter, it actually uses eight independent motors and blades, and could probably fly with one or more failed, depending on which ones.
A multi-rotor drone cannot autorotate like a regular helicopter, and at first this seems less safe. However traditional single-engine helicopters can be less safe than first appears. If a piston-engine Robinson R22 is hovering at 400 ft, it cannot autorotate – it is within the “dead man’s curve” of the height/velocity chart: Helicopter Aviation
That is why medivac helicopters are often twin turbine: each engine is more reliable, and in a contingency they can fly briefly on one engine. Yet single-engine piston helicopters are legal and often are used by smaller municipalities. I occasionally see one hovering at fairly low altitude.
A small vehicle like the Ehang 184 could also possibly have a ballistic parachute, which is not possible for a regular helicopter. So the safety factors of a human-carrying “drone” can be better than first appears.
Since drone propellers are fixed-pitch there is some efficiency loss relative to variable pitch props, but this reduces complexity. The Bell V-22 used four variable-pitch ducted fans and was planned to carry 24 troops, plus pilot and co-pilot. So obviously “drone-like” quad-prop vehicles can work:
Here is another example of a manned multirotor “drone” helicopter:
http://www.volocopter.com/index.php/en/vc200-prototyp-en
Video here: https://youtu.be/OazFiIhwAEs
That site says it already has a ballistic parachute.
Rather than have a full cyclic control on each rotor, I wonder if anyone has tried variable-pitch multi-rotors. joema linked to the X-22, but that was 50 years ago.
Rather than a dozen small, fixed-pitch propellers, how about four larger props, with some degree of variable pitch to compensate for the slower response? When you move the stick forward, the computer speeds up the two aft motors; and because that takes a few seconds it briefly increases the pitch of those props, and then decreases it as their speeds come up.
Seems like it could be done; don’t know whether it should be done.
What about using something along the lines of vectoring to manipulate the thrusts with out needing instantaneous rotor speed changes?
Would that not be simpler?
You’ll still need servos and linkages or gimbals to direct that thrust and that adds complexity and mass, although I’d love to see something like the Avatar gunship fly IRL. I would agree variable pitch might be the way to go as far as control-ability.
Maybe, maybe not. Four hundred feet is a limit of the deadman’s curve for zero airspeed. Per the link, ‘Note that near sea level, an R22 can experience an engine failure in a 400 foot hover, and have sufficient altitude for the pilot to react and make a safe landing. If the pilot wants to operate lower than 400 feet AGL and still be able to make an autorotational landing if the engine quits, he needs some forward airspeed.’ So if your density altitude is near sea level, you can pull it off. There’s also usually a wind, so if you’re hovering, you still have forward airspeed. And so on.
I was going to mention this. I think that to be safe, all rotors in a quad-helicopter would need to be interconnected, and that there be sufficient power to make a safe landing on a single engine. If you have no power, then the rotors will need to be of a sufficient diameter to have a usable ‘driven section’ during autorotation, and enough inertia to initiate autorotation and the landing flair.
True. Speed, payload, economy. Pick two.
Hmm, where is it written that all the props on a multi-rotor have to be the same size? I wonder if you could build one with two large props (opposite rotation, constant speed, constant pitch, responsiveness doesn’t matter) to provide most of your lift, and four smaller ones (variable speed, responsive) to provide the rest of the lift and maneuvering.
Sikorsky had a similar idea when developing the single-screw helicopter. He considered 2 and 3 smaller horizontal tail rotors to control pitch and yaw. When he realized that pitch control wasn’t necessary he came up with the single vertical tail rotor for yaw control.
-
Note that after leaving ground effect in a vertical ascent on a helicopter is probably peak engine stress. If you’re going to have an engine failure, it’s quite possible it happens then.
-
We aren’t really talking about quad-helicopters. We are talking about copters with a *lot *of propellers driven by individual electric motors. What you are saying would be handled with a mesh of electric connections, where each motor has access to an electrical bus that is subdivided into sections across the battery pack. Protective switches of some type would activate automatically to isolate sections of the battery pack with too low or too high a bus voltage, so that the other motors still have power. (so a fire or other sudden battery failure would in theory only cause power loss to a couple of the motors. An isolation switch failure would still limit power loss to a small number of the motors)
Electrical redundancy like this can be better (and much cheaper) than a method using mechanical driveshafts. Of course, these vehicles do have a serious problem that batteries suck. Range is going to be limited, even if the VTOL has a forward flight mode that is much more efficient than a helicopter, and after landing, it’s going to take a while (and a BIG battery charger supplying hundreds of amps!) to recharge.
Here’s another project along the same lines:
Looks like 36 small electric fans, 24 of which are mounted on the wings and can be pointed down for VTOL and backwards for forward flight.
Do only military jets (or fighters only?) have thrust vectoring right off the assembly line?