Hacker News new | past | comments | ask | show | jobs | submit login

Over 100 successful ejections per year? That's wild! That sounds like an awful lot of hull losses. And how many people weren't saved?

What's the risk calculus look like for a plane with an ejection seat vs. a plane without one? War obviously has one risk profile, but military vehicles and their operation differ wildly from day to day commercial operation.




I thought they have a monopoly on ejection seats, but I've found at least 1 other manufacturer: https://www.collinsaerospace.com/what-we-do/industries/milit...

Martin Baker's website is a bit wild, you can join their "tie club", but to do so you need to have ejected using their seat: https://martin-baker.com/tie-club/

And they list the ejections they know of https://martin-baker.com/ejection-notices/ , but the records are a bit jumbled, the latest entries include ejections in the past that they've discovered, and they don't all have dates, even "Someone successfully ejected today". Today when, man?!


You can even buy a special watch with colored case that signifies you ejected from an aircraft (using a Martin-Baker Ejection Seat of course)

https://martin-baker.com/watches/mb-i/


I collect watches, I'd love to have one of these but they are probably super rare to find second hand. Obviously I'd never be in a position to buy one from retail...

There are a couple of pretty special pilot only watches. I like IWCs top gun too, the one with the real top-gun logo on only avialable to pilots that have graduated Top Gun.

Available to top gun graduates: https://www.iwc.com/content/iwc/americas/us/en/articles/jour...

The retail version: https://www.google.com/url?sa=i&url=https%3A%2F%2Fwww.watche...

The people that actually get these watches don't seem to want to sell them, understandably.


And since these companies probably don't sell to Russia, Russia has their own manufacturer: https://en.wikipedia.org/wiki/NPP_Zvezda . Interestingly Collins' origin seem to be seats designed by Zvezda: https://archive.is/www.nytimes.com/1998/12/02/nyregion/metro...


The Collins equivalent seems to be the "Grasshopper Club": https://www.collinsaerospace.com/what-we-do/industries/milit...

They mention the B-2 as a platform, that one was the most expensive aircraft crash in history according to Wikipedia.


> found at least 1 other manufacturer

that's Zvezda K-36


When I was in USAF, the pilots said, "When your fear of staying in the plane exceeds your fear of the ejection system, it's time to eject." That wasn't really a joke.


I've seen comparisons of American vs. Russian ejector seats. It appears that Russian seats are better designed to protect the pilot from injury while ejecting. The issue of ejector seats has been used to support the idea that pilot comfort and safety is a low priority for American designers.


Soviet designs tend to be more artisanal. Sometimes it gives them oxygen-rich rocket engines and better engineered ejection seats, other times it malfunctions and flip over ISS or send a space probe back down to earth by accidentally cryogenically freezing out a computer.


Do the Russians have zero-zero systems?

I know the US seats have started taking into account pilot weight to modulate thrust to limit the Gs to what is necessary


They had those a long time ago. One of the few parts of their airplanes that was actually quite good.

https://www.youtube.com/watch?v=jklGQxAOoo8&t=9m45s


"Russian pilot's cam ejecting from shot down Su-25SM3": https://www.reddit.com/r/CombatFootage/comments/ybdldt/insan...


Since even a successful ejection will hurt you due to the immense forces at play (an explosion will push you out the plane) even in the best case, and crush you, if something goes wrong, no wonder that joke is for real.


Yeah ejection is no joke. It's pretty much guaranteed to get at least spinal damage due to compression.


>Over 100 successful ejections per year? That's wild! That sounds like an awful lot of hull losses.

I remember calculating a few years ago that more than 10% of the ejector seats sold by Martin-Baker have been used to eject.

Military aircraft operations, even in peacetime, are much more dangerous than the vast majority of civilian fixed-wing flight (except airshows/racing and maybe cropdusting). I don't know how much of that difference is the flight profile being inherently more dangerous (closer to terrain), and how much is it being harder on the aircraft.


They're more dangerous mostly because they're optimized for maneuverability which requires trading off aerodynamic stability and glide ratio. All aircraft have problems mid-flight and most pilot training is about how to deal with that but when something goes wrong, fighter jets just don't have much time to recover before they plummet. That maneuverability also makes it a lot easier to get disoriented and lose control.

Trainer jets like the L39 albatross have a less maneuverable design in exchange for being a lot more survivable in event of a problem. The engines themselves aren't that much less reliable - some engines like the Garrett TFE731 have a TBO of 4,000 hours which is significantly more than many GA engines like the Continental in an entry level Cessna.


>some engines like the Garrett TFE731 have a TBO of 4,000 hours which is significantly more than many GA engines like the Continental in an entry level Cessna.

That shouldn't be a big surprise: jet engines are turbines with basically only one moving part that just spins, whereas those crappy GA engines are reciprocating piston engines with designs that haven't changed since the 1950s.


This incredible reliability is the justification for TOPS and ETOPS policy in which suitably qualified aeroplanes with only two jet engines are allowed to fly over an ocean. Before TOPS this is why you see jet aeroplanes with three engines, they don't need three engines to fly, but the regulations required that if two engines fail the plane must get to safety, with TOPS that's no longer required (hence the joke expansion for ETOPS: Engines Turn Or Passengers Swim) and so new planes with three engines didn't make sense.

With piston engines two engines failing is a completely reasonable thing which will sometimes happen to say, a transatlantic flight. In the early days of such travel, the demands are such that all the practical aeroplanes are four engine models anyway, so limping to an abort airfield on two engines was a reasonable strategy if two failed, but once jet engines on passenger planes were a thing, their enormous reliability makes this unnecessary hence TOPS and then ETOPS.


The mentions of TOPS above are a brain misfire. ETOPS was altered repeatedly but it was always named ETOPS, at first allowing 90 minutes, then 120 minutes and eventually 180 minutes of assumed single engine range. A modern twin engine jet can go a long way in 180 minutes at design single engine performance.


Military trainer jets have two seats so most crashes count for two ejections. Martin Baker has been fitted into 200 different aircraft types in 90 countries so averaged out that's only 0.5-1 hulls lost per customer per year.


Probably some years with much higher numbers due to wars that would skew the average.


That includes the Vietnam and Korean wars - lots of pilots got shot down




Join us for AI Startup School this June 16-17 in San Francisco!

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

Search: