# Is it possible to take off backwards using reverse thrust?

Last night I had a wild dream where an airliner took off into the air backwards from the departure gate. I'd like to know just how impossible this is. Let's ignore the perils of starting a takeoff roll from the apron.

Is there any fixed-wing aircraft in which it's possible to use full-power reverse thrust to achieve backward flight? Either:

• To gain altitude using forward flight and then transition to backward flight, or
• To take off from a runway of any length?

I know that aircraft equipped with reverse-thrust have interlocks to prevent engaging the reversers in the air. Assume these interlocks are somehow defeated.

• I think you might want to ask whether backwards flight is actually possible at all.
– GdD
Oct 4 '21 at 15:25
• I doubt a Boeing 747 would takeoff that way, even if it accelerated to Mach 10 … Oct 4 '21 at 15:49
• Notice that nobody is even bothering to address the "transition from forward to reverse flight" part of the question? Oct 4 '21 at 17:39
• Did you mean "controllable" flight? If not, just about anything is arguably possible Oct 4 '21 at 21:16
• If you can do it with a model aircraft, then you can theoretically do it, which would then make taking off backwards "possible". I'm pretty sure some model aircraft somewhere can do that. Also, it appears to be possible to fly backwards to some extent with real-sized aircraft. I'm no expert so I won't try to answer this. But here's a link: bbc.com/future/article/… Oct 6 '21 at 0:22

Four reasons among many others it's not possible:

• Weathervaning effect from the vertical surfaces of the tail would turn the aircraft as it accelerates, the tail would not remain upstream of the nose.

• Taking off requires the engines to deliver their maximum power and the wing to convert this power into lift very efficiently to leave the ground. The poor efficiency of a reversed wing prevents generating required lift.

• Thrust available from reversers is significantly weaker than regular thrust.

• Takeoff rotation would be done with the nosewheel bearing all aircraft weight. Being not designed for this load, it would collapse.

Weathervaning effect

Perhaps the most simple to understand difficulty is the aircraft tendency to weathervane, to behave like a weathercock when airspeed is significant.

As the vertical surfaces of the empennage are large, having them upstream of the engines makes the aircraft instable in yaw. As soon as the aerodynamic force is significant, the tail is slowed down, and this tendency, unless taken into account in the design, cannot be stopped before the tail is the rearmost flying part.

The more a multi-engine aircraft is powerful, the larger is the vertical empennage, in order to be able to maintain the desired yaw in case an engine is inoperative.

A wing is not reversible

Engines are designed to deliver their maximum thrust when an aircraft takes off (TOGA -takeoff / go around- setting). This maximal thrust must be efficiently used by the wing, which is required to have a given lift capability at this time. This is the case when the aircraft is accelerated in the regular direction.

(I would tend to say a wing cannot work at all in the reverse direction. It is fitted with asymmetric moving surfaces, likely to disintegrate when not loaded as intended, hinges being now at the wrong place. This is also valid for the tail surfaces. But let's assume everything remains in one piece.)

When airfoil edges are swapped, a condition known as reverse flow. Lift cannot be generated except at small angles of attack. This condition can happen on helicopter rotor blades and is classical for tidal turbines though the applicable Reynolds number is different.

The problem is the airflow does not attach well to the upper surface when the leading edge is sharp, stall occurs at small angles of attack. By itself the stall problem is a no go for taking off, as takeoff require a high angle of attack (15°).

If we want want to ignore this, we face a second problem: Reverse flow also introduces a dramatic loss of efficiency (L/D ratio). Lift is at least divided by two, and drag multiplied by two. This too is a condition preventing taking off. (Note the factor would probably be more like 10, but I've no reference to support this figure. More on that aspect. By the way this could make a good question and we have aerodynamics experts able to answer.)

To compensate lift loss takeoff speed needs to be increased, this requires more thrust and a longer runway to accelerate. Compensating for higher drag requires more thrust too.

At this stage, even with we had the full regular thrust, this wouldn't be enough to make the aircraft airborne. Instead of two engines we would need four or more.

Reverse thrust is limited

Engines provide reverse thrust at the expense of significant losses. Reversers design is very ineffective in term of aerodynamics as airflow is strongly deflected.

In addition only a fraction of thrust is reversed and some direct thrust is still generated. We can estimate the effective reverse thrust to be only 70% of TOGA.

While more than normal thrust would be needed, the thrust actually available is reduced and totally insufficient to take off.

When the engine is used in forward direction, air enters the engine pushed by ram pressure. When used in reverse direction, ram pressure is actually negative, air is constantly moved away from the inlet, the engine is not fed correctly, the compressor likely stalls and stops functioning before reaching the takeoff speed.

In a turbofan, the reverser reverses the fan flow, but the regular core flow is still required to spin the engine. It leaves the engine at the exhaust, which is now subject to ram pressure. The turbine is not designed to work in these conditions.

Aircraft attitude to take off

Now the visual part. In a regular takeoff, at the end of the acceleration roll aircraft nose is raised to increase the angle of attack (rotation phase), so the wing can generate the amount of lift required to leave the ground. When the aircraft is used in reverse gear, this gives this:

Assuming elevators are effective to lift the tail, which is not granted, there is an impossibility to balance the aircraft on the nosewheel during the roll and to keep it on the centerline, which is anyway not visible by the pilot.

The regular rotation is eased by the main gears location and the main gears are able to bear all the weight until the aircraft is airborne. In your dream the weight is now fully on the nose wheel, and the nosewheel is unable to resist, it collapses.

There are other points, which would be boring to explain but which are as significant, e.g. how to aerodynamically control the aircraft during the takeoff and in flight, as control surfaces wouldn't be movable under the reversed ram pressure.

• Re "A wing is not reversible" -- in the case of a rudder-elevator-throttle (no ailerons) radio-controlled model airplane, I've seen a photo of the rubber-band-on wing installed backwards (trailing edge first). Word is it was flyable and controllable, with some loss in performance. (As I recall, very abrupt stall characteristics were reported.) Airfoil was probably flat-bottomed and probably somewhat along the lines of the classic "Clark Y". Will look for links. Oct 4 '21 at 21:30
• @quietflyer what is the wing loading of the model, though? A fair bit lower, probably low enough it didn't need an airfoil. Oct 5 '21 at 3:02
• You're obviously right but while An engine delivers its maximum power for takeoff, there is practically no way to get more from it. there is a way to need less power from it - take off empty with minimum fuel Oct 5 '21 at 9:09
• @ChrisH: This is right of course, and other optimizations in mass and balance could be done too if the reverse takeoff had to be tested. On the other hand assuming the lift is only half in reverse direction is way optimistic. The aircraft will never by light enough.
– mins
Oct 5 '21 at 9:46
• One more factor could be added, related to the first one: The center of gravity will be too far behind in backwards flight, so the airplane will be extremely unstable. Oct 5 '21 at 19:55

a wild dream (the Langoliers chewed up the runway ahead) so...

Theoretically, if you got the passengers to move rearward a bit and retracted the nosewheel to rotate, given the plane was somehow light enough and generating enough thrust, the resultant aircraft would be the latest in canard design with forward swept wings. With the right CG placement, possible, but ...

the ailerons, flaps, rudder, and stabilator are not designed to be reverse flown in 150-250 knot winds and would likely fail when deflected, as well as the wingtips.

Even if it got off the ground, a crash would likely rapidly ensue.

• Theoretically, yaw instability from forward vertical fin would make it nearly impossible. Stability on roll and pitch axes would be sufficient for a good movie plot. Essentially it would be a Pterodactyl in need of a computer for yaw control. Oct 5 '21 at 4:50
• good use of the word "nearly". Oct 7 '21 at 4:16

At current technology reverse thrust is not strong enough to push aircraft back even to take off speed. Use of thrust reversal at low speed can be harmful to the engine itself due to high risk of foreign object damage. Lastly wings are not design to efficiently generate lift in backward direction so you need more airspeed hence more thrust to have enough lift for flight.

If you go so far as to redesign the plane to efficiently fly backward the result would look very similar to a normal plane. You will not have anything new.

• Possible damage to the engines is not just from FOD. We once had a mystery problem where splines joining two parts of the main engine rotor were wearing out much faster for one airline than any others using the same engine/airframe combination. It turned out that their pilots were routinely using maximum reverse thrust to push back on the apron for departure, instead of waiting for a tug. Oct 5 '21 at 1:05
• I guess the next assignment is to design a "reversible" plane that can fly equally well either forwards or backwards. Step one is to put the engines on motorized swivel-mounts... Oct 5 '21 at 6:38
• @JeremyFriesner: I think a push-pull prop design would be simpler and work better. A bigger issue would be the center of gravity, but perhaps that could be dealt with by having cargo holds at both ends of the plane, with a requirement that whichever end would be in front have more cargo than the other end (using ballast if there wasn't enough useful cargo to be transported). Oct 5 '21 at 20:44
• Another concern with low forward speed use of reverse thrust is the engine ingesting the air blown out from the reversers. This is a concern at low forward speeds, but it's even more of a concern at all reverse speeds. There's a good chance that you'll have a compressor stall on your hands before you ever got to the point of rotation. Oct 6 '21 at 3:11

Interesting question but currently not a realistic possibility.

Because of the aerodynamics involved in airfoil design, propeller design, reverse thrust functionality, etc., I am very confident there are no current production airplanes that could achieve backward flight as described in your question.

There are Powerlift category aircraft (like a tilt-rotor for example) that have forward and rearward maneuvering capabilities, but they are not within the context of the question you have asked (being an Airplane category, using [conventional ] reverse thrust, etc.).

As a side note most jets (unless restricted by the manufacturer or by non-approval [e.g.,in the case of commercial operations] can "power-back" (move backwards) on the ground using reverse thrust (e.g. pushing back from the gate in the absence of a tug). Can be tricky/risky and not as common anymore.

Since it's only airspeed that matters, if you give up on the reverse thrust bit, you just need a strong enough wind blowing parallel to the runway.

So if you're taking off into a category 5 storm with 140 knot winds, you could be traveling down the runway backwards pretty fast in a small plane. And you'd climb out of the airport still going backwards in ground speed terms.

On a smaller scale, think of wind tunnels. Airspeed is what matters, so a stationary model can fly given sufficiently strong wind.

(Ignoring all the ways the storm would destroy your plane and kill you, of course).

• You can do the same with a kite in low wind. But this is still a forward flight. The question is about backward flight not about negative or null ground speed.
– mins
Oct 6 '21 at 11:42