Hacker News new | past | comments | ask | show | jobs | submit login
Gyroplane Accidents 1985-2005 (nih.gov)
52 points by simonebrunozzi 5 days ago | hide | past | favorite | 62 comments

Well I have an oddly specific contribution to this thread: My first boss was absent all the time. He'd show up to the office and lock himself away, leave and then not come back for another 6 months. He hated software and computers and didn't want to be a part of the software product we were building-- which is self evidently strange.

He used to race experimental cars. He was in a horrific crash on one of those salt flats and his personality was permanently changed. He went from being pleasant to the person I knew him as-- cold, strange and reclusive.

Turns out he was coming in to the office just long enough to use Excel to run a calculation. He was designing a helicopter where the jet engines were on the tip of the rotors and you stood in a can above the whirling blades. We called it GarbageCopter.

Anyways, he eventually gave up and bought a gyrocopter. He flies that thing all day, every day. Just circling in the pattern at our airport at 1,000 feet and 20 knots.

He is very old and blind on one side. I have no idea how he got his health check. I am pretty sure he will be a statistic someday, but at least he's doing something he likes.

Tip jet [1] helicopters have been a thing, but it's never taken off commercially. The Fairey Rotodyne being a good example. [2]

[1] https://en.wikipedia.org/wiki/Tip_jet

[2] https://en.wikipedia.org/wiki/Fairey_Rotodyne

Tip jet helicopters have been a thing, but it's never taken off commercially.

Sort of like Wittgenstein himself.

Gee, [1] says they were invented by Wittgenstein! I never knew about that.

“ If people never did silly things nothing intelligent would ever get done”

I don't see the point of putting that quote here, sorry. Inventing the helicopter was a silly thing, or what?..

Traumatic brain injury is perhaps underappreciated. https://www.gwern.net/Questions#psychiatry (Maybe as death rate from other accidents drops, it rises in relative importance?) Among other amazing numbers from that page, it quotes one study that shows 45 percent of the residents of a homeless shelter had a positive screening result for TBI: https://www.gwern.net/docs/www/cmajopen.ca/03aff85438a259787...

Notoriously, Stanislaw Ulam and Steve Wozniak had major personality changes after head trauma.

Great comment. Was curious w/ regards to Ulam. Found this on Wikipedia:

In January 1946, he suffered an acute attack of encephalitis, which put his life in danger, but which was alleviated by emergency brain surgery.


Another friend, Gian-Carlo Rota, asserted in a 1987 article that the attack changed Ulam's personality: afterwards, he turned from rigorous pure mathematics to more speculative conjectures concerning the application of mathematics to physics and biology; Rota also cites Ulam's former collaborator Paul Stein as noting that Ulam was sloppier in his clothing afterwards, and John Oxtoby as noting that Ulam before the encephalitis could work for hours on end doing calculations, while when Rota worked with him, was reluctant to solve even a quadratic equation.


It's posited that Howard Hughes' increasingly strange behavior and eventual withdrawal into reclusion was precipitated by a series of TBIs he suffered in test pilot crashes.

There is also a theory that he began taking pain medication for back pain related to injuries suffered in crashes, and the dosage/addiction added up over time, all the while none of this staff or medical experts tried to wean him off it. Very much like the situation with Michael Jackson and painkillers.

and Michael Hutchens of INXS who got punched in the head while riding in Stockholm.

And even more famously, Henry VIII, who fell off a horse.

For a Gyrocopter you only need a Thrid Class Medical certificate or to comply with BasicMed neither of which list eyesight as a disqualifying condition.



The standards for a FAA third-class medical go beyond just what's on the official "disqualifying" list. It's not a walk in the park.

From personal experience, they care a great deal about your eyesight.

To start, you'd want to look through the AME Guide [1] and the rules in 14 CFR 67 Part D [2].

[1] https://www.faa.gov/about/office_org/headquarters_offices/av...

[2] https://www.law.cornell.edu/cfr/text/14/part-67/subpart-D

This reminds me, when I had my 3rd class medical exam at age 16, the examiner didn't have equipment to test hearing. So, he went into the adjacent room and whispered, "Can you hear this?" Then loudly, "OK, cover your left ear." Whispering, "Can you hear this?".... I passed.

It might be an ultralight. Don’t even need a pilot certificate, let alone a medical.

I thought you didn't need a license for a gyrocopter but when I looked just now it seemed like you did. Guess it's based on weight class so yeah he might be in the ultralight class, didn't mention it because I thought I was wrong about it being a no license class.

I don't know if ultralight gyrocopters are even a thing. I just figured if he's designing trash-can helicopters, he might be into an ultralight gyrocopter kit, if such a thing even exists. :)

(I also wasn't even sure about ultralights not requiring licenses. Had to look it up.)

It looks like you can build them small and light enough to qualify for the ultralight category it's just tricky and somewhat limiting since the combination of light weight and low fuel capacity is rough on a high drag gyrocopter.


Ultralight gyros are pretty popular. They are like motorcycles in the sky. Stupid dangerous and crazy fun. Also if you have an engine out you are already auto rotating so no big deal. From what I hear, the thing that kills pilots in those things is that people like to fly low and they are prone to pilot induced oscillation. Pilots panic, try to overcontrol, and then die.

" He was in a horrific crash on one of those salt flats and his personality was permanently changed. He went from being pleasant to the person I knew him as-- cold, strange and reclusive."

If it was not traumatic brain injurie, like others mentioned, I can maybe give an other point of view.

I guess it is the struggle of acceptance. Everyone knows accidents happen, but before you were in a bad one and have high self esteem, you likely believe, this won't happen to you. That would be unfair.

"Those things cannot happen to me" - well, yes, they absolutely can happen to you. Why not to you? Why not to me?

I had that with a actually not so serious sport accident, which still meant I could not do sport for years - which was a very serious crisis for me. WHY ME?

Well, why not me. In hindsight I would rather think I was very lucky, to not have that accident, while being somewhere far off.

Today I can even do (moderate) extreme sports again, ... but much more aware of what I am doing and how easy it is, to loose it all. So I can allmost say, today I am thankful for that accident, but I can also see, how such a accident can bring you to a very dark place of where you cannot get out.

Your former boss seems to need his flying to get out of it at least for some time. I guess there are worse ways to cope with it.

Fascinating. I have so many questions. How did this manager manage to not get his position terminated? Was he able to run Excel calculations for the GarbageCopter outside his office machine? I will stop here for now. Thank you.

There is a slow rotor variant of an autogyro that has the same efficiency as small single engine plane, which is pretty cool. They also claim it can take off and land vertically.


Interesting design. For vertical takeoff it first stores extra rotational energy in its rotor.

I'm not an aerospace engineer, but IMO this is the most innovative thing in aviation I've seen in the past 20 years.

I've read that Gyroplanes are dangerous to beginners because if you get in trouble the instinct to pull back on the stick like a traditional aircraft can be fatal to gyro pilots. They have to un-learn one of the most basic skills they learned as pilots.

Edit: I mean push on the stick. Sorry, had a brainfart. Good thing I'm not a pilot.

Here's a video of a Gyroplane "bunting over" or "pushing over" once the rotors become unloaded... the final impact is not shown but it's implied the pilot died:


An explanation:


> Fatal Sequence

> As the rotor is unloaded, both control thrust and the drag of the rotor disc fall off rapidly. With a high engine thrustline (relative to the vertical c.g. and center of drag) and/or with a momentary gust that can increase fuselage drag, the gyro can pitch forward and tumble. Such tumbling, also known as bunting over, can occur almost instantly. There is no possible recovery, and bunting over is invariably fatal.

From the description there, it seems like a simple "oh shit" lever to tilt the engine (or at least prop) upwards could fix this issue?

Should a lot of vehicles have fundamentally different controls now than when they were first invented? In particular, should we move away from controls which map onto how control surfaces or other physical parts move, and instead focus on controls that describe the desired motion? Quad-copters are pretty easy to fly, but they would be really really hard to fly if you were supposed to manage the thrust on each motor. I've never flown one, but I understand helicopters are challenging to fly, and controls do map onto actually actuating the swashplate and tail rotor.

I don't understand how a gyroplane flies, or why pulling back on the stick causes immediate problems. But ... should autogyros have controls where you describe the motion you want, and let a computer figure out the actuation?

I think there's an analogy to software here, where generally we're less error-prone and more productive if we can specify what we want by expressing higher-level, declarative relationships, rather than demanding that humans always think in terms of the low-level operations that will need to occur.

Airbus aircraft like the A320 are kind of like this. Roll and pitch is entirely fly-by-wire, and as I understand it, the basic scheme is that the sidestick position is taken as a desired rate of change and the avionics compute how to achieve that change using the control surfaces. This is somewhat necessary as a change in roll attitude is done using some or all of the slats, ailerons, and spoilers depending on the speed, pitch, etc. A cool consequence of this is that in the normal flight mode, pulling the sidestick full aft and keeping it there will result in maintaining best angle of climb. Under normal circumstances it is not possible to stall because the avionics will not allow control pressures that will leave the safe flight envelope. Changing the flaps, or more properly the configuration lever, not only adjusts flaps but also changes control parameters to modify the flight envelope for safer operation near the ground.

The thing I found funny when reading into that is that, for all that effort for roll and pitch, the rudder is just hydraulic. I suppose there's just not that many ways you can fancy up yaw control.

There are rare cases where you want manual control, I remember in the A320 that ditched into the river Sullenburger said the limits prevent the best possible landing[1]:

> However, Sullenberger said that these computer-imposed limits also prevented him from achieving the optimum landing flare for the ditching, which would have softened the impact.

That said the cases for this are probably very, very rare.

[1] https://en.m.wikipedia.org/wiki/US_Airways_Flight_1549

When Airbus introduced its fly by wire systems, there were quite a few airframe losses caused by incorrectly handling the moded interface, including their test pilot dropping the A320 at the Paris Airshow. The crases were universally attributed to pilot error.

If you have muscle memory trained in one system, you tend to revert to that behavior when under stress.

In aviation you want to avoid unnecessary complexity. You really want to avoid single points of failure. Most small aircraft are controlled via cables and pulleys because they're simple, can be redundant, easy to verify, and unlikely to fail.

What's your backup plan when your flight control computer crashes? Or if you have an electrical failure?

(There's an argument to have computers provide additional assistance on top of the manual systems, but we usually stop short of fundamentally remapping flight controls. And even then, we have to be careful otherwise we end up with situations like MCAS.)

I think what you really want to do in aviation (or anything safety-critical) is reduce risk of failure. Avoiding risk of mechanical failure at the expense of controllability is simply pushing the blame onto the pilot.

The general approach for critical hardware like flight controllers is triple redundant systems with two-out-of-three voting, driving redundant actuators.

MCAS is a prime example of this NOT being followed, because it relied on a single sensor input to generate a significant control output.

From my limited reading, my understanding is that modern fighter jets work in this way. They use their aerodynamic surfaces in ways that are not considered "usual flight", i.e. when we see the very cool looking J-turn. But all the pilot is doing in that situation is manipulating the joystick, the aircraft flight computer figures out how to manipulate the control surfaces.

IIRC, aircraft with forward swept wings are very unstable and can only be controlled with fly by wire controls. The control surfaces are not necessarily doing what the pilot controls say they should be doing.


It's called "relaxed stability".


Not sure about gyrocopters but fixed-wing pilots who later transition to, or add, a rotary wing license are intensely counseled to resist the baked-in urge to push the nose over to counteract a stall or imminent stall.

In almost all single-rotor helicopters, pushing the nose over — fixed wing style — runs a high risk of causing the tailboom to rise and collide with the main rotor. The rare rigid rotor ships (BO-105s, mostly) are less likely to have this happen.

That was alluded to in this old xkcd commic, though it doesn't go into the details: https://xkcd.com/1972/

I kind of assumed that "the one thing you instinctively do to escape a stall" is to push forward on the stick. Some answeres here seem to confirm that: https://aviation.stackexchange.com/questions/49865/why-does-...

The comic says it never stalls, but then says if it does you should not do the instinctive thing. ;-)

It says it crashes if you do the thing that you would do in a normal aircraft to prevent stalling.

From the other discussion I linked, it sounds like the issue is that the rotor normally creates forward drag, and earlier designs have the engine/thrust propeller mounted pretty high. Pushing the stick forward causes the rotor to have less forward drag than it normally does, so the engine thrust effectively causes the whole craft to pitch forward. Which is sort of what you'd expect to happen when you push forward on the stick, but I guess it happens in a way that's hard/impossible to correct from.

The issue is that the drag is what spins the rotor, and the rotor's rotation is what makes your controls work. So when you get close to a stall, the rotor's already going slowly, and then you push forward and it slows even further. Now you have almost no control authority and the only major force acting on the aircraft is the thrust from the propeller, which pitches you rapidly forward.

I'm not sure why (given sufficient altitude) you can't cut the engine and recover using autorotation the way a normal helicopter can, but I imagine it's pretty hard to do this in a normal helicopter too if you're falling upside-down.

I still don't understand. If it can't stall why would you ever try to prevent stalling?

It’s a reaction trained to fixed winged airplane pilots to the point where it becomes instinctual. If your airspeed is dropping low you push forward the stick. You don’t do it because you have run a fluid dynamics simulation which warned you about the approaching stall, but rather as a reflex to a set of sensory cues.

FWIW during my PPL pulling back during (un?)expected situations instinctually was very much discouraged.

edit: this parent comments makes more sense if instead the author meant push on the stick, as another pointed out.

>in trouble the instinct to pull back on the stick like a traditional aircraft

You mean push on stick? (that's what is done to avoid stall in normal aircrafts)

Anyway, https://xkcd.com/1972/

I'm not an expert, but yesterday I've heard from a pilot that most US gyros don't have a tail stabilizer (is that how it's called?), unlike the Magni models (Magni is a small Italian manufacturer of gyros), which don't have the problem you are describing.

If there's an expert, I'd love to hear if what I just wrote is correct.

Since there is no torque on the rotor (Except for maybe the part the spins up the rotor before takeoff) there is no need for a tail rotor. Typically autogyros will have a rear elevator and rudder like a traditional airplane. Since most have a pusher propeller that gives you even more control in a low velocity takeoff or landing which is what gyros are good at.

A good picture of this setup: https://en.wikipedia.org/wiki/Autogyro#/media/File:AutoGyro_...

Edit Addendum: Typically loss of control in a nose down situation for autogyros is because they don't have ailerons to control roll. They rely on the cyclic of the rotor disk.

Igor Bentsen[1] was my neighbor. My kids like to play on the dilapidated old prototype autogyro [2] in his old backyard.

[1] https://en.wikipedia.org/wiki/Bensen_Aircraft?wprov=sfti1

[2] https://imgur.com/gallery/MlJSqYK

Why aren’t autogyros more popular as drones? It seems like they would have a lot of advantages for photography and package delivery.

And they could be cheaper to build and easier to operate?

Drones are typically quad or hexacopters that can take off and land vertically even in a stiff breeze. If you're talking about some other kind of drone then maybe you could specify it. But compared to quad or hexacopter an auto gyro would have a lot more loiter time just as a fixed wing would, and it could come down vertically at least to a point, but would not be able to take off or land vertically.

There is one particular brand of autogyro that can take off vertically by pre-spinning the rotor and then applying collective at the same time as disconnecting power from it. They tend to leap off the tarmac. But that is not the norm. Standard autogyros require a takeoff roll just as a fixing aircraft plane does. People often look at the rotor and think that it is like a helicopter but in fact it is more like a fixed wing aircraft in the sense that air must be passing over the wings / rotor in order to produce lift.

That being said, an autonomous autogyro does sound very interesting. One could pre-spin the rotor with a typical brushless motor and then leave it unpowered during flight. It does require collective control to be able to do that but even hobby helicopters have collective and cyclic control, so adding just collective control seems very approachable.

We feel the same way about auto gyro based cargo delivery


Why would they be cheaper to build? A quad-copter is ridiculously simple in it's design - 4 motors is all the moving parts it needs. The rest is sensors and software.

Wouldn’t a gyro just have one pusher motor and maybe one motor to turn a rudder? So less motors at least.

I'm going to guess you'd have more software complexity to account for more dynamic conditions, as OP's link alludes to, along with more redundancy in safety. I'm pretty sure most 4 motor drones can function fine with just three blades/motors and possibly two even.

Source: I'm an expert flyer with dozens of seconds flying. I once bought my kid a tiny 2" drone and eventually took over as pilot. Now I didn't take it from him, he just stopped playing with it (What's wrong with kids these days? I would've loved it as a kid) but those little things are so fun to fly.

A gyrocopter requires a tilting rotor blade for control. This is massively more complex than the fixed blades of a quadcopter.

Quadcopters don't really scale past a fairly small point before the efficiency nosedives though.

It's mainly because they still behave more like airplanes. They can't hover (!), their turns have a certain radius, they have a take-off roll and usually a short landing roll, they need to actually fly a landing approach (albeit a very steep one).

Also they have a single large unducted rotor that's easy to destroy or hurt someone with, and you can't have automatic collision avoidance, because you can't easily stop an autogyro in mid-air.

When I was in high school, we initiated one as a senior class project in Mechanical Technology.

Then graduated early and left the building.

What does "perception-based performance breakdown" mean?

Refers to a problem where the pilot did the wrong thing because of disorientation or perceptual confusion.

Take controlled flight into terrain, a common enough problem that aviation safety writing just calls it CFIT. This is when an aircraft collides with terrain, but there was no loss of control. CFIT is a relatively common occurrence, but why does someone fly a completely controllable aircraft into the ground? Almost always due to perceptual confusion, disorientation either to attitude or location. Unintended entry into instrument conditions (e.g. into a cloud or fog bank) is one cause, but disorientation can occur for a number of other reasons as well, including in good weather conditions.

Most light sport aircraft are day VFR only, with an altimeter and airspeed indicator being the primary instruments. Some ultra lights don’t even have those, and you use visual cues and the “feel” of the craft to judge your energy/speed/height - and our perceptions are not the most reliable.

Fancy words for "pilot fucked up."

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact