Elon Musk has suggested that he'd look at implementing VTOL electric aircraft with a "gimballed thrust" system similar to the one used in rockets.

Aircraft have all these unnecessary things like tails and rudders and elevators - not needed. Just gimbal.. anyway.. gimbal the electric fan. For some weird reason gimbaling motors is normal in rockets and not in aircraft. Why not? [Do you have specific plans?] I've been sort of toying with the design for an electric supersonic vertical take-off and landing electric aircraft for a while, I'd love to do it, but I think my mind would explode

What are the challenges to implementing this? Is it technically feasible?

  • 19
    $\begingroup$ That's called thrust vectoring in aviation speak. And doing away with control surfaces makes the propulsion indispensable. You cannot even run out of fuel anymore. Technically easy, but it will be hard to find pilots who will fly this widowmaker. $\endgroup$ Nov 30, 2016 at 17:52
  • 3
    $\begingroup$ On a related note: feel free to check out the (many) questions about electric aircraft here, to get a bit of a feel how reasonable a supersonic VTOL electric aircraft is, gimbaled or not. Also related are dirigibles, which can use thrust vectoring, and can provide you with insight what happens when they lose thrust. $\endgroup$
    – Sanchises
    Nov 30, 2016 at 17:54
  • $\begingroup$ @PeterKämpf thanks so much, that term was exactly what I was looking for! $\endgroup$ Nov 30, 2016 at 20:29
  • 3
    $\begingroup$ i haven't figured out if Musk is losing his grip on reality or if he just enjoys winding people up and watching them act like idiots. Probably it's both. $\endgroup$ Mar 4, 2017 at 2:30
  • $\begingroup$ Wait - a tail is not needed? $\endgroup$
    – Koyovis
    Jul 19, 2017 at 0:00

4 Answers 4


Could you create an aircraft using an jet engine/propeller/ducted fan on a gimbal? Sure you 'could'; it is feasible by the laws of physics. But just because you can does not mean that you should.

Setting up multiple points of lift on a manned aircraft seems like a good idea with a lot of advantages but is a Pandora's Box of problems, the not the least of which is the grave risks of single point failure. Lift and control is totally reliant on balanced lift between multiple engines. Lose one engine and the craft departs from controlled flight and crashes. You better hope you have an ejection seat and/or parachute on board.

Some solutions to these problems come in the form tiltrotor designs like the Agusta Westland 609 and the V-22. But these aircraft have multiple redundancies in their design including a common transmission system between the rotors; in case of a single engine failure, the good engine has enough power to drive both rotors. Tiltrotors can also glide like regular airplanes and utilize control surfaces during this and cruise flight.

As a final note: Be careful in putting too much faith in Musk and other Silicon Valley whiz kids speculating about other design endeavors. The skills required to run a successful software company are often very different from that required to develop other artifacts like aircraft and automobiles, etc. Musk was a genius with PayPal but nearly went bankrupt starting an electric car company and his development work with SpaceX has been punctuated with a series of spectacular failures. He also quietly withdrew from his proposed 'Hyperloop' project, I suspect, because he really didn't understand what he was getting into there.

  • 1
    $\begingroup$ Thanks Carlo. Musk ad hominem aside, sounds like the fundamental issue with this approach is the fallback in case of a failure. Since electric fans are significantly smaller than regular engines, wouldn't you theoretically be able to include redundant units on board to avert this issue? $\endgroup$ Nov 30, 2016 at 19:12
  • 7
    $\begingroup$ No. Aviation gas turbines are roughly the same as the best electric motors in terms of power to weight and approx. the same size. We have also previously covered the (at least current) impracticality of manned electric aircraft in several other posts, primarily because of the energy density of batteries compared with existing aviation fuels. Finally my comments on Musk are based on his ideas, not on the man himself or his personal attributes. That's not an 'ad hominem' argument. $\endgroup$ Nov 30, 2016 at 19:22
  • 4
    $\begingroup$ I think your characterisation of Elon Musk is neutral to positive in my eyes. He did not start Tesla, he bought it from what he made with PayPal. By now his success has most likely bred such hubris that he now thinks the laws of physics don't apply to him. And a fawning press believes everything he says. $\endgroup$ Dec 1, 2016 at 11:42

Sure you could, but why would you want to? There's no inherent benefit to be had in vectoring thrust as opposed to using control surfaces. It doesn't simplify things to any extent, it just makes things complicated in other ways. In order to have effective roll control you'd need to space motors out on the wings which could vector in line with the center of the aircraft, which would be complex and add weight. Also, you couldn't use thrust vectoring unless you had at least 3 engines spaced around - one in the tail for pitch and yaw, 2 in the wings for roll control.

Of course there is safety as well. A loss of power or an actuator failure, even momentarily, would probably mean a very fast, catastrophic loss of the craft and anyone inside it. Imagine a software glitch which pointed the tail engine to the left for a couple of seconds while the airplane was at cruise speed - the airplane would go side on to wind at several hundred miles per hour and most likely break up. On a VTOL approach to land one of the engines loses power, causing the craft to roll and drop the last 100 feet like a rock, destroying the craft and its occupants.


Gimbaled thrust in 'rocket speak', is equivalent to vectored thrust in 'aviation speak'.

Many modern military aircraft use thrust vectoring in order to improve performance, particularly in tight turns, the F22 being an example. The harrier and the F35 also use thrust vectoring to enable VTOL (vertical take off and landing) and/or STOL (short take off and landing) capabilities.

Of course not only jet aircraft use thrust vectoring, nor is thrust vectoring confined only to military applications. The Boeing V22 Osprey uses thrust vectoring, while the XTI Aircraft Company are currently developing a ducted fan VTOL civilian aircraft.

Anybody with any aviation knowledge, be it in aerodynamics or that of piloting a craft, will know that failure is unlikely to result in an uncontrolled event of any kind.

Firstly, in forward flight wings will continue to generate significant amounts of lift. Aircraft do not require engines to remain in flight, only to maintain significant airspeeds of 500mph+. Think of multiple commercial airliners that have previously experienced complete engine failure, only to glide for many miles.

Secondly, for rotary wing VTOL aircraft, a failure in any stage of flight will likely result in autorotation just as for helicopters, allowing for a great deal of control. Assuming the craft also has surfaces providing lift, this acts to provide an additional benefit.

Lastly, the very first 'control surfaces' used by the Wright brothers were warping wings. Current control surfaces (ailerons, flaps, and the like) were originally brought into play because of the inflexibility of aircraft wings used since the early 20th century. We have used the same control surfaces ever since, but they are incredibly aerodynamically inefficient, hence why NASA has been developing alternative flexi-wings (https://www.nasa.gov/press-release/nasa-successfully-tests-shape-changing-wing-for-next-generation-aviation).

Offering a different perspective by controlling direction via engine management as oppose to manipulating airflow may improve efficiency yet further.

Beware of placing too much faith in armchair 'experts' and naysayers of online forums. Usually they have very little real experience or knowledge, let alone any vision of the future.

  • 2
    $\begingroup$ They do ask a couple of pertinent questions though, like how do you control the gliding aircraft without thrust. If wing morphing is the answer, why use thrust vectoring at all? $\endgroup$
    – Koyovis
    Jul 18, 2017 at 23:51
  • 1
    $\begingroup$ In aviation, naysaying is exactly what you need, for addressing Murphy's law. Aviation is very safe and accidents have a low probability, due to using proven methods with decades of experience. In aviation, a bug has deadly consequences. $\endgroup$
    – Koyovis
    Jul 18, 2017 at 23:59
  • 2
    $\begingroup$ I admit I'm not an aircraft engineer, but I have a really hard time seeing how a vectored thrust VTOL craft could autorotate. There just wouldn't be enough airflow through those (relatively) tiny fans to produce much lift. Helicopters can do it because they have those great big blades... $\endgroup$
    – jamesqf
    Jul 19, 2017 at 2:52
  • $\begingroup$ Koyovis: Search 'sailplane design' for info on how aircraft glide without thrust; aircraft do not require engines to remain in the air. Murphy's law can be stated thus: "anything that can go wrong will go wrong". This does not leave room for naysayers, rather engineers and scientists to solve the problems of today with better solutions for tomorrow. Jamesqf: Contrarotating fans (or multiple fans, see XTI project) will provide sufficient lift to autorotate. Contrarotating fans provide greater lifting force than equivalent non-contrarotating fans. Extra lifting surfaces will add to this. $\endgroup$
    – user23235
    Jul 19, 2017 at 11:11
  • $\begingroup$ @user23235: But doesn't it still come down to the amount of air that can pass through the area of the fan? So taking the VTOL fans to be about the size of a typical jet engine (because if they're much larger, we have a multi-rotor helicopter), we could try imagining what would happen if such an engine, powered off but with the fan free to rotate, was dropped from a height. $\endgroup$
    – jamesqf
    Jul 19, 2017 at 17:58

You won't like this answer but fixed wing VTOL aircraft with gimbals thrust actual fits many rocket/missile type of aircraft, which include:

1) High altitude anti-aircraft missiles, such as RIM-161 SM3 and S400 systems. Fins at high altitude doesn't work that well, so thrust vectoring is the main control authority at high altitude (while the body provides the lift).

2) SpaceX Falcon booster. If you argue missiles are VTO but not VL, consider the Falcon booster? VTOL, yes. Gimbals thrust, yes. Aircraft, absolutely. Fixed wing? Sort of, if you include lift body.

Within the scope of conventional manned airplanes, F35 and Harrier's reaction control is close (although differential thrust not gamble thrust).

The way you asked, "feasibility" misled many previous answers to "practicality".

For feasibility, it's a definitive yes, but for practicality, not many airplanes have done that, because of fuel economy and reliability.

Remember wings and control surfaces have a lift to drag ration greater than one, while thrust vectoring have a "lift to drag ratio" (control force to thrust lost ratio) of roughly one. For example, if your tail needs 10kn of down force, you have two options 1) use a tail plane with L/D=5 and pay for 2kn of drag, 2) divert your thrust upwards for 10kn and and lose a net thrust of 10kn.

This is the reason why thrust vectoring and differential thrust isn't used unless there are other choices, e.g. very slow(F35, Harrier, R73 AA missile) or very high (S300/400, SM3, etc), deep stall (F22, and various thrust vectoring demo and experimental airplanes, e.g. F18 HARV).


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .