In a retro-future version of Earth, one of the main modes of air travel is by gasoline-powered biplane (design circa 1914).

enter image description here

I want these to land in places where in real-life Earth only helicopters would be able to land. For example in a small area on top of a skyscraper.

The proposed method is that they fly into a large wind tunnel on top of the building. This way their ground-speed can be zero whilst the airspeed remains high enough for stable flight. By using normal landing procedures, co-ordinated visually by an expert operating the wind tunnel, planes are able to land and take off vertically within the wind-tunnel and deliver and load passengers or cargo.


Could this work in theory? What snags could prevent it from working?


I am aware that a wind tunnel has a large fan blocking the entrance/exit at one end. This is catered for by turning the plane through 180 degrees and reversing the direction of the fan for departure. To be clear, I don't suggest a tail-wind at any point. The plane is always flying into the wind from the fan.

Additional information

I envisage there being a turntable supporting the wind-tunnel. It could be powered, or may even act as a huge weathervane. This will eliminate cross-winds.

  • $\begingroup$ Comments are not for extended discussion; this conversation has been moved to chat. $\endgroup$
    – L.Dutch
    Commented Sep 18, 2020 at 2:51
  • $\begingroup$ You can get rid of the fan issue with a blade less fan. They’re a bit gimmicky, but maybe for this society this is a good use for them. Something like a giant version of this? amazon.com/gp/product/B00KX1OD1S/… $\endgroup$
    – oeste
    Commented Sep 18, 2020 at 4:09
  • $\begingroup$ Does the world extend to 1940's planes like the Vought "Flying Flapjack" and "Flying Pancake" ? en.wikipedia.org/wiki/Vought_V-173 etc, They could do stupendous/stupid impossible manoevers. $\endgroup$
    – Criggie
    Commented Sep 18, 2020 at 7:59
  • 1
    $\begingroup$ @Criggie - That's a great-looking aircraft! Unfortunately, this future is one that is imagined by someone living in 1914 and I'm only allowing biplanes, at least as far as this question is concerned. I could maybe introduce this as a startling new prototype that is seen flying above the city to everyone's amazement. $\endgroup$ Commented Sep 18, 2020 at 8:36
  • $\begingroup$ With technology currently approaching usability on Drones... why wouldn't something like that be an option? cleantechnica.com/files/2018/07/evtol-3.jpg $\endgroup$
    – WernerCD
    Commented Sep 19, 2020 at 7:23

9 Answers 9


The bush plane STOL records are your friend here.

No additional external technology or infrastructure, just carefully designed STOL light aircraft. The current record for landing is 9ft5in (2.87m) and this sort of thing is a bush pilot speciality.

Your world is going to have something similar, but they're going to be the tower-top pilots specialising in landing on the flat tops of tower blocks.

  • 22
    $\begingroup$ I think this is a case where the real life solution would look less credible to a layperson audience than some steam-punkish contraption. $\endgroup$
    – user412
    Commented Sep 17, 2020 at 9:59
  • 4
    $\begingroup$ @EikePierstorff, I mean, I don't disagree, the real world is often weirder than we imagine. $\endgroup$
    – Separatrix
    Commented Sep 17, 2020 at 10:51
  • 10
    $\begingroup$ When bush flying in arctic Scandinavia (Piper PA-18-150 Supercub, without special STOL modifications, but retro-fitted with the 180 hp engine) we routinely landed/took off from soccer pitches. About 100 meters, 125 if the wind allowed using the long diagonal. That is half the official required distance, but perfectly doable with some practice. $\endgroup$
    – Tonny
    Commented Sep 17, 2020 at 13:29
  • 4
    $\begingroup$ @chasly-reinstateMonica the Fokker Dr.I and Sopwith Camel both stalled at around 50 mph, modern STOL bush planes seem to be in the realm of 30 mph at the low end. For landing, all you really need is good brakes - a plane of the same mass at 50 mph has about 3x the kinetic energy as one at 30, so roughly triple the braking distance given in the answer, which is still under 10 metres. Doubling that for a modest safety margin, plus another factor for weight gives you the distance needed $\endgroup$
    – llama
    Commented Sep 17, 2020 at 23:30
  • 5
    $\begingroup$ @llama, it's not just good brakes. These aircraft tend to be tail draggers to keep the nose up and assist with short take off, if you brake too hard you risk flipping it, if you watch videos of this sort of thing, most of them are braking as hard as is safe. It's all very carefully balanced and there's a lot to go wrong. $\endgroup$
    – Separatrix
    Commented Sep 18, 2020 at 7:23

It's even more simple* than a wind-tunnel: simply set up the runway as a steep ramp on top of the building. The aircraft land into the wind, and the uphill nature of the runway will rapidly kill speed. Then when taking off, doing so downhill into the wind will allow a rapid gain in speed.

So, ideally a landing building would have a runway like a hat, a truncated cone, that could be approached from directly upwind from any direction, or alternatively a swivelling ramp that can be rotated to point in the required direction. The runway surface and structure should be a metal gridwork to allow wind to pass through it, otherwise the runway hat could cause disconcerting turbulence and loss of lift right before landing.

In the event of a missed landing, the pilot could go around or touch and go, where a wind tunnel would have a fan that would make that error fatal and costly.

*However, I can see that these rooftop landings would be dangerous and unpopular with experienced pilots and building owners. A landing that is missed low could result in plane-in-building syndrome, and while C1914 aircraft were light and flimsy compared with modern aircraft, some also had low-speed handling problems that would not have the advantage of ground effect to offset in such situations. Engine failure on takeoff would result in planes falling onto the streets, with likely fatalities amongst bystanders, plus disruption to road traffic.

So, while it is certainly possible, I would anticipate that it would not be something that a newly licensed pilot might attempt, but may require additional post-license training and certification, and be limited to certified models of aircraft. Of course, with sufficient demand, new pilot licensing may require rooftop landing skills.

  • 3
    $\begingroup$ +1 I was going to suggest the same. For supporting evidence, here's a video of a cessna 182 doing a uphill landing on a narrow, steep uphill strip. An artificial strip of those dimensions would fit on the roof of a skyscraper. Goes to show the idea is possible youtube.com/watch?v=hQkq2NKWrQ8&ab_channel=Backcountry182 $\endgroup$
    – Aubreal
    Commented Sep 17, 2020 at 13:51
  • 9
    $\begingroup$ Similar approach happened to be used by Polish WWII pilots in 1939. After one of the relocations, the landing strip was organised at the top of a quite steep hill. During landing the pilots had issues, if they went to the side, e.g. due to the cross-wind. Before all pilots relocated there were already several accidents. Yet, newly coming pilots noticed that rather than using the field landing strip all they had to do was just landing perpendicularly, up the hill. There were large meadows that were hardly suitable for farming due to the steepness. The results were amazing and accidents stopped. $\endgroup$
    – Ister
    Commented Sep 17, 2020 at 14:27

Too complicated

If something goes wrong, it would be difficult to abort. Just imagine a plane trying to land and the fan breaking.

The pilot also then needs to watch two walls and a roof as well as the ground so a crosswind while trying to enter the wind tunnel could be quite catastrophic.

Easiest solution is a landing cable like what is used on aircraft carriers. enter image description here

see How planes land on aircraft carriers

  • 3
    $\begingroup$ Note that in a regular or even cable-assisted landing, there's nothing in front of or above the plane, so it can apply takeoff thrust and go around again if anything goes wrong. In a "fan-assisted" landing, you can't do that because there's a giant fan in your way. $\endgroup$
    – Cadence
    Commented Sep 17, 2020 at 1:05
  • 3
    $\begingroup$ "Just imagine a plane trying to land and the fan breaking." That's not much different from a present-day helicopter landing on a rooftop and the rotor blade breaking. $\endgroup$ Commented Sep 17, 2020 at 8:19
  • 8
    $\begingroup$ @chasly-reinstateMonica, you are a little wrong - if helicopter engine or transmition fails in most cases helicopter can land with little troubles using autorotation. Once such a helicopter landed this way just infront of my job (we have a huge 100x100m lawn there). They were reparing it for three days just in place and it flied away by itself.If hypothecal fan fails in any way - it is a catastrophe in any case. $\endgroup$
    – ksbes
    Commented Sep 17, 2020 at 9:02
  • 4
    $\begingroup$ @chasly-reinstateMonica No. Suddenly you're flying too fast into wall and then crashing into the ground..... $\endgroup$
    – Thorne
    Commented Sep 17, 2020 at 9:54
  • 7
    $\begingroup$ I'd just say that in these kind of "retro-future" scenarios, the "If something goes wrong" bit is often part of the plot or the world's make-up. $\endgroup$
    – komodosp
    Commented Sep 17, 2020 at 14:18

Ан-2 airplane is known to be specially designed to be able to land in "helicopter way", i.e. in controlled stalling. In this mode it flyes at speed about 60-70km/h and need only 60-100m to stop. And since it is have a size of a bus (it is the largest comercial biplane) smaller version can have even better numbers.

This quite a harsh landing, BUT! You may point your fan up to reduce plain's stalling vertical speed - you would need wind of only about 5-10 m/s - enough to upskirt, but in other ways quite comfortable - people can walk during operation. And even this fan fails it would only lead to harder landing and couple of lost teeth, but not deaths.

Such a fan can also assist (or at least not interfere with) takeoff operations. An-2 needs 100-150m for takeoff (if you are not into "diving" takeoffs).

So combination of "contolled stalling" aicrafts (very bi-,tri-plane) and upward fans blowing through floor gratings is plausable (but totaly unpractical and risky)


enter image description here

Below is a video of an Antonov AN 2 taking off in about 30m and then flying ridiculously slowly!

Antonov AN 2 Tisted 2011 Short Take Off, Slow flight


  • 1
    $\begingroup$ This looks great but I don't understand the upward-blowing fan. Surely any stall from that height above the airstrip will result in landing nose-down (?). Are you saying that bipanes have a parachute-like stall that would avoid a nose-down scenario? $\endgroup$ Commented Sep 18, 2020 at 8:44
  • 1
    $\begingroup$ @chasly-reinstateMonica, some specially designed crafts (An-2 biplane in particular) can stall keeping normal position and land in that way. The only problem - it is an emergancy landing regime (say, when engine is off and only endless Siberian forest is below) and it has too high vertical speed to be completly safe (but the lowest total speed). But we can compensate this with your fan! Not all biplanes can do this, and not only biplanes can do this. It is a rare practice in RL avaition. $\endgroup$
    – ksbes
    Commented Sep 18, 2020 at 9:31
  • $\begingroup$ I've found a video of one taking off in about 30m! see the "answer" to my own question. If you want to copy this into your answer, I'm happy for you to do that and I will delete my answer. $\endgroup$ Commented Sep 18, 2020 at 14:00
  • $\begingroup$ @chasly-reinstateMonica - done. Edit my answer if you need to. $\endgroup$
    – ksbes
    Commented Sep 18, 2020 at 14:10
  • $\begingroup$ Done. You are also welcome to change anything you want about my edit! $\endgroup$ Commented Sep 18, 2020 at 14:21

Take a look at "parasite aircraft" on Wikipedia, there have been real world attempts to allow larger aircraft to capture and then launch smaller aircraft. Since the skyscrapers aren't aircraft (see, there's your problem), you might use a ring mechanism above the roof that would spin a tether ending in a loop that the aircraft would catch with a hook mounted on top of its wing. The ring mechanism would also be a crane that would pick up aircraft from storage and then spin the aircraft up to takeoff speed.

Either that or have multiple cantilever arms extending from the building that would catch aircraft, slow them down safely, turn them around when needed and then accelerate them off to takeoff.

  • 1
    $\begingroup$ I must say that this would be spectacular even though it's not what I asked for! The sight of a skyscraper whizzing a plane around itself and then letting go like David throwing a stone at Goliath would bring tourists to the city from far and wide. People would take to flying just for the thrill of it although a lot of people would be too scared ever to try. $\endgroup$ Commented Sep 18, 2020 at 12:49
  • $\begingroup$ Excellent out-of-the-box answer :-) $\endgroup$ Commented Sep 19, 2020 at 21:53

The AN2 biplane (produced until 2001) can fly extremely slowly without stalling (50km/h is quoted, but so is "no stall speed"). Unlike the WWI-era planes suggested in other answers, it has a useful payload of 12 passengers, so provides a good model. I'd ditch the fans have a rotating (to face into the wind) deck modelled on an aircraft carrier. Arrester hooks would be a better idea for dumping speed than an uphill landing, though the latter could be used.

Take-off may be exciting despite the short take-off roll of 560ft , though a carrier-style ski-jump would work. From an isolated skyscraper, taking off downhill into the wind is another option, using the height to gain speed.

At ground level the Empire State building is apparently 423ft long, so vertical sides on that scale gets you very close already to the take-off roll. Landing requires more length at nominal speeds, but we can use the minimal stall speed to our advantage even without arresters.

Show-off pilots could come in low, gaining height and losing a lot of speed on approach, and stall just over the deck, to land rather like a bird on a branch (though not one that relies on flapping on landing). Recovery from a miss would be unlikely even if there wasn't a building next to you.

You also need to consider what to do with the planes once they've landed - probably an elevator to an indoor hangar (again like a carrier) or the few planes you can store on top will be blown off in the first gale.

Something very similar has been done in 2023 using a modified Piper Super Cub landing on the helipad of the Burj al Arab. This was without payload or passengers, but in a plane with far higher stall speed than the AN-2. Takeoff, even modified to decrease weight and increase acceleration, involved losing a lot of height to gain speed.

  • $\begingroup$ Nice summary answer, but one thing is off. To "land like a bird", you'd need to be able to move your wings. Because that's precisely what birds are doing: They slow down gliding, increasing the angle of attack until they get near to stalling, then they start flapping their wings forward, increasing the speed of airflow over their wings above stall speed. I.e. while the bird is well below stall speed, its wings always move faster than stall speed. That's rather hard to do with a fixed wing plane. $\endgroup$ Commented Sep 19, 2020 at 21:49
  • $\begingroup$ @cmaster that's a good point, though some seem to land (on high perches) relying just on momentum and movable legs. It is of course possible to get above-stall airflow over (some of) the wing area without the overall airspeed being above stall - use the air flowing from the props. Or assuming you're landing into the wind, popping up from the wind shadow of the skyscraper would briefly increase your airspeed while starting to decrease your ground speed $\endgroup$
    – Chris H
    Commented Sep 20, 2020 at 7:15

Plausible - if your fans are top notch

A Sopwith Camel (circa 1916) can land at 42 knots. This is 77km/hr. Not that fast. A horizontal wind tunnel can be used to create a 45 knot wind. Lift is proportional to forward airspeed, it can land vertically (from the point of view of an observer on the ground).

Any imperfections in the airflow would be a problem. The airflow needs to go a considerable distance from the tower - as the plane needs to shed inertia without stalling. A minor crosswind would be a problem. Any loss in visibility and it'd be real dangerous. You'd want to combine this with WW2-era radio guided landing for night or bad weather operations.

You'd need to be careful where you suck the air from, if your fans are at one side of the building and pull air around the building you'll create dangerous vortices just in front of the building. You'll want a few meters of manifolds to ensure this doesn't happen. You need a long, horizontal, smooth, air stream.

Don't turn the fan around for departure. Taking off into a head wind is much safer than taking off into a tail wind. Put the wheels in a small pit, turn the plane engine on, and "take off" with the head wind (eg - "fly" towards the fan), increase air speed to stay above stall speed, but below the wind speed, you'll be able to climb up, then slightly reduce speed, to get backwards ground speed, get clear of the building, then go max power and climb over the airport.

  • $\begingroup$ I wasn't suggesting a tail-wind takeoff. The plane would always fly into the airstream produced by the fan. This would necessitate reversing the fan as I suggested. $\endgroup$ Commented Sep 17, 2020 at 1:47
  • $\begingroup$ You wont need to reverse the fan, or even turn it off. The planes take off backwards - I may need to draw a diagram. $\endgroup$
    – Ash
    Commented Sep 17, 2020 at 1:49
  • $\begingroup$ I know what you mean but when you stall using your method, you will be facing towards the building. By reversing the direction and the fan, you are moving away from the building. $\endgroup$ Commented Sep 17, 2020 at 1:51
  • 1
    $\begingroup$ Ok, I get what you mean now. So your fans are sucking for take off, and the plane is turned around? I'm dubious, it's easier to blow a smooth air stream then to suck one. If the boundary at the top of the air is smooth rather than sharp it can power and climb out of it - the camel could do 100 knots, which is enough to power up and over the airport in a 45 knot headwind. $\endgroup$
    – Ash
    Commented Sep 17, 2020 at 2:04
  • $\begingroup$ 42 knots? Antonov 2: "Hold my beer." $\endgroup$ Commented Sep 19, 2020 at 7:15

An additional problem would happen at takeoff.

First, you will need the plane to accelerate to great speed in a very short runway.

Using the fan could be... complicated.

  • If it throws air in your direction your plane accelerates but losses lift.

  • If it throws air in the opposite direction your plane will be slowed but it will gain lift quickly (your plane moves fast relative to the surrounding air, even if the plane is not moving fast), but once you leave that gust of air (for example your plane leaves the tunnel) then you suddenly are in a slow moving airplane with little lift (the positive part is that it will accelerate as it spint out of control until it reaches the ground).

You could use a catalput as in air carrier, but that adds a lot of complexity and cost.

And the analogy of the air carrier is not perfect, either. Air carriers do move, and that movement reduces the relative speed between the air carrier and the plane, while keeping a higher airspeed of the plane. Buildings do not move.

The second problem is that, when exiting the tunnel, your plane will find itself very suddenly with different wind patterns. The pilot may find himself facing a strong lateral wind, with little time to adjust the controls to account for it and little speed and height to solve any situation. In fact some years ago something similar happened in Spain, a helicopter that was trying to take-off from a bull ring got a strong lateral wind when it cleared the height of the walls and crashed.

You would need very good reasons to put such a system of place, because it will never be either easy nor safe when compared to a conventional airport. And if you go with it, probably having your planes land on the roof(with arrestor cables and catapults) would be way better than in a tunnel.

Another issue could be structural. A building will probably have no problem supporting some hundreds or even a few thousands of kilograms of parked airplane, but resisting the stress when those airplanes bump into the ground (sometimes a bit too hard) may require to reinforce considerably all of the building structure.

  • 1
    $\begingroup$ An aircraft carrier's ability to move is mostly relevant in that it can rotate to give the planes taking off or landing the best angle of attack on the wind. A building could reproduce this by having just the landing platform on a rotating mount in order to get the most advantage from whatever the wind is doing at any given moment. $\endgroup$ Commented Sep 17, 2020 at 15:08
  • $\begingroup$ @Darrel Hoffman - Yes, that's exactly what I would like. It would be something like this railway turntable but with the fan where the shed is (and of course moving with the turntable). newmodellersshop.co.uk/images/Trains/Track/… $\endgroup$ Commented Sep 18, 2020 at 8:56

To add to the STOL - ish options, there's always the Pogo approach. VTOL is possible; you might want to 'redesign' this aircraft to accommodate more people, with rotating seats, maybe more engines mounted off the main tube, etc.

  • $\begingroup$ Thanks. The main problem is that I specifically want biplanes as mentioned in the question. If I allow other improved aircraft, I will just end up with what we have today and then it won't be retro! $\endgroup$ Commented Sep 18, 2020 at 12:58

You must log in to answer this question.

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