4 ms·
When a helicopter loses power it glides, safely, to the ground - assuming it's being flown by a qualified pilot. https://en.m.wikipedia.org/wiki/Autorotation h
by underdeserver 4y ago
When a helicopter loses power it glides, safely, to the ground - assuming it's being flown by a qualified pilot.
https://en.m.wikipedia.org/wiki/Autorotation https://en.m.wikipedia.org/wiki/Autorotation
- jackpeterfletch 4y agoI guess the question is - can we maintain the standard that is ‘qualified pilot’ as it is today while making it accessible enough for this to scale. Learning to drive and learning to fly isn’t the same bar today. And if Lilium think Palo Alto to San Francisco could be $50, they’re gonna need _alot_ of flights to balance their costs.
- jstummbillig 4y agoI think it's safe to assume this class of vehicles (and likely any new class of vehicles from here on out) is not designed to be piloted by a human past its prototype stage.
- kevin_thibedeau 4y agoThat is a fantasy that almost never works out in real life. Autorotation is only feasible with enough forward air speed and control authority. Most helicopter crashes happen outside that narrow zone of survivability. Even if an autorotation is semi-pulled off it's likely to be a hard landing that causes significant injuries.
- sokoloff 4y agoAutorotations are regularly trained (and often taken all the way to touchdown during training). The checkride for private and commercial rotorcraft has a power failure at hover taken all the way to touchdown as part of the standard. Autorotations are by no means a "gimme", but if every autorotation "likely" caused significant injuries, there wouldn't be enough helicopters or pilots to go around (no pun intended).
- vanattab 4y agoWell to be fair those training exercises are probably done in near ideal conditions of altitude, landing zone, forward airspeed, and you know it's coming have studied mentally ahead of time. It's one thing to pull it off during training and another when the engine fails at a random point during a flight in LA/SF/NYC.
- HPsquared 4y agoDon't you think pilots would tend to try and stay within that "safe autorotation" envelope most of the time while flying? It seems mostly a case of having sufficient altitude and/or speed, similar to what fixed wing pilots try and maintain - a safety margin.
- Deritio 4y agoWhy would they? They will probably want to fly across non easy terrain because otherwise you could just use some other means than an expensive helicopter. Hight perhaps but you will fly across water, woods, power lines, cities.
- sokoloff 4y agoWatch a helicopter take off sometime. It (almost) never lifts into a hover and climbs straight up. Instead, it lifts slightly off the ground and accelerates forward following the recommended takeoff profile. That’s so the craft can stay in the “autorotation possible” section of the HV diagram as much as possible.
- robonerd 4y ago> That is a fantasy that almost never works out in real life. Can you quantify this? > Most helicopter crashes happen outside that narrow zone of survivability. AFAIK the majority of helicopter crashes are not caused by a loss of engine power, but autorotation is relevant specifically in that context. It's not relevant to helicopters crashing into hills during storms, or hitting wires with the rotor, or anything like that. We're discussing what happens if the engine stops, not all accident scenarios. > Even if an autorotation is semi-pulled off it's likely to be a hard landing that causes significant injuries. Yes, but it's better than being dead.
- acid__ 4y agoI’m not at all qualified here, but it sounds like autorotation might not be involved in crashes because when autorotation works, there is no crash :P
- _moof 4y agoAnd you need a place with enough room to land--extremely difficult to find in a city. I really don't like flying single-engine aircraft over large cities.
- marcosdumay 4y ago> Autorotation is only feasible with enough forward air speed and control authority. Control authority yes, but altitude is always a substitute for forward air speed (and is an even better substitute for the eVTOL planes than helicopters).
- kevin_thibedeau 4y agoMost helicopter mishaps start at low altitude.
- p_l 4y agoAutorotation generally works just fine. Most helicopter crashes involve things unrelated to possibility of autorotation or not, and sometimes autorotation won't help when you hit something too hard at the end, with all of it happening because client pushed too hard and overworked the crew (I might be quoting a specific crash here...)
- bombcar 4y agoAlso I suspect the majority of autorotation failures are from using the helicopter for things only a helicopter can do like hovering. This is why you rarely see them do that, instead they circle - which any small plane could do but the helicopter can hover if needed. It’s just the most dangerous part.
- fsckboy 4y agoif a helicopter hovers high in the air it's not dangerous because they have time to turn that height/energy into forward motion. see "dead man's curve" https://en.wikipedia.org/wiki/Helicopter_height%E2%80%93velocity_diagram https://en.wikipedia.org/wiki/Helicopter_height%E2%80%93velo...
- p_l 4y agoHovering is the signature helicopter move for laymen, but it's also considerably hard, and "high hover takeoff/landing" is, IIRC, one of the hardest maneuvers (it's been years since I've read the pilot's textbook, though) There's also a lot of other things that can impact the event - for example, the crash I mentioned involved bad weather, tired crew, and losing both engines at once due to icing. They autorotated and nobody died... but they fell into a forest, landed hard enough to destroy undercarriage, and the first officer survived by accident due to helping get passengers into position when a tree smashed his station. EDIT: corrected captain to first officer, as visible in this photo: https://static.polityka.pl/_resource/res/path/bc/08/bc086fae-d035-4dd3-b5ec-354eded29f08_f1400x900 https://static.polityka.pl/_resource/res/path/bc/08/bc086fae... (it's Mi-8, so captain sits on the left, not on the right as is normal in western helicopters)
- cjg 4y ago
- Aeolun 4y ago> Most helicopter crashes happen outside that narrow zone of survivability. That’s kind of a given, since the successful autorotations do not qualify as crashes.
- carabiner 4y agoWrong. Most helicopter crashes are non-fatal: https://generalaviationnews.com/2021/02/10/fatal-helicopter-accidents-down-20-in-2020/ https://generalaviationnews.com/2021/02/10/fatal-helicopter-... 2020: 94 accidents, 19 fatal accidents, 35 fatalities, 22% decrease in accidents compared to 2019, 36% decrease in accidents compared to 2013 2019: 121 accidents, 24 fatal accidents, 51 fatalities 2018: 122 accidents, 24 fatal accidents, 55 fatalities 2013: 146 accidents, 30 fatal accidents, 62 fatalities
- Deritio 4y agoAnd it then glides onto what? Humans? A uneven roof? A busy road? And for whom? The rich? People who can afford a few hundred dollars for a quick flight from the airport to the city? How about no?
- cjg 4y agoI don't know about other parts of the world, but in the UK, helicopters are quite carefully controlled so that they can land safely in such an event. For example, to fly a helicopter over London requires pilots to fly only over well defined routes - although those with multiple engines are allowed more scope.