Armchair air crash investigator • Hacker News Board

Re: Armchair air crash investigator

#151

Post by CamperBob2 » Mon, Jul 08, 2013, 8:02 PM UTC

nknighthb wrote: post #115

CamperBob2 wrote: post #113

Earlier quoted context omitted.

So, just to make sure I don't misrepresent your position, you're saying that the people who built and programmed this hardware ( http://www.space.com/16878-mars-rover-landing-sky-crane-guid... ) couldn't safely land a 777 in SFO (or, for that matter, an A320 in the Hudson) any day of the week. Is this the case?

What part of the skycrane experienced serious damage or malfunction that the software corrected for? The whole point is we're not talking about when things go according to plan. For that matter, how many times has the skycrane been used? What is its safety record? If you go ask the guys at JPL how many times out of 1000 they think it'll work perfectly, what will they say? Incidentally, not cool moving goalposts.

Sigh. Yes, you're right. Nothing went wrong during the Curiosity landing, so I can therefore not use it as an example of successful flight automation.

You win this one, it looks like. GG.

Re: Armchair air crash investigator

#152

Post by VladRussian2 » Mon, Jul 08, 2013, 8:10 PM UTC

marvin wrote: post #149

VladRussian2 wrote: post #118

Earlier quoted context omitted.

well, until report is out, we're here to speculate (or to read a good speculation if we're lucky :). From the Flightaware data the plane had 109kts at 100feet altitude before the seawall. The stall speed of 777 is about 110kts (its Vat is ~140, and it is 1.3 of the stall). And the speed was falling as the next recording is 85kts. While we can only guess about the original cause - pilot error or hardware, etc... - the…

Are you sure that the Flightaware data is accurate enough? When flying a glider, I have attempted to make similar judegments from GPS log data when doing an outlanding, and I'd be similarly careful when using radar data from a website. I don't want to be very negative here, but my point is I've made wrong calls about this when speculating in the past. So currently I'm opposed to doing it for reasons other than entert…

> for reasons other than entertainment.

well, i'd be surprised if there are people here for any other reasons :)

the flightaware data correlates very well with what is on video. Personally i spent a bunch of time on Coyote Pt. so i can mentally correlate the video with how it looks for other planes/landings at SFO. As we aren't ordained NTSB officers, we're free to make wrong calls :)

Re: Armchair air crash investigator

#153

Post by RockyMcNuts » Tue, Jul 09, 2013, 3:39 AM UTC

fnordfnordfnord wrote: post #122

RockyMcNuts wrote: post #56

Earlier quoted context omitted.

some video of the crash - kind of looks like it was skimming the waves in ground effect trying to go around http://www.cnn.com/video/?/video/us/2013/07/07/vo-plane-sf-p... I stand by the parent having some useful data and analysis, and the OP being totally speculation, and in many respects uninformed. - number 1, if you say 'loss of power', makes sense as a possible cause. if you say 'icing in the FOHE', it's like sa…

>no icing conditions Icing in the fuel lines may be caused by fluid flowing through an orfice (pressure differential). All that is needed is water in the fuel.

water goes to the lowest point since it weighs more than fuel. then if it's not drained properly it gets sucked into the engine at the start of the flight. would be odd to discover water contamination after a long-haul flight.

I kind of assumed the FOHE was something that used external air, it wasn't water in the oil or the fuel. you descend through humid and coolish air ie not summer heat, water condenses and freezes as it's sucked through something from the pressure differential, or from actual icing conditions. but I could be wrong.

Re: Armchair air crash investigator

#154

Post by fnordfnordfnord » Tue, Jul 09, 2013, 12:04 PM UTC

RockyMcNuts wrote: post #153

fnordfnordfnord wrote: post #122

Earlier quoted context omitted.

>no icing conditions Icing in the fuel lines may be caused by fluid flowing through an orfice (pressure differential). All that is needed is water in the fuel.

water goes to the lowest point since it weighs more than fuel. then if it's not drained properly it gets sucked into the engine at the start of the flight. would be odd to discover water contamination after a long-haul flight. I kind of assumed the FOHE was something that used external air, it wasn't water in the oil or the fuel. you descend through humid and coolish air ie not summer heat, water condenses and freeze…

>water goes to the lowest point since it weighs more than fuel

While that's true, it takes time, and it won't get enough of the water out of the fuel[0]. Also, the aircraft must not be disturbed for the water to settle out. Operation demands for aircraft don't allow that. Even if they did, some water will remain in the fuel.

While the engines are operating, fuel is constantly circulating which disperses any water within the fuel. Airport fuel supplies at major airports are usually tightly controlled, but it's still be possible for water to get in.

>I kind of assumed the FOHE was something that used external air,

The FOHE is a Heat Exchanger, sort of like the radiator in your car, but instead of engine coolant/air the working fluids are Engine Oil/Jet Feul. Both the fuel and oil systems are closed systems, with one exception. As fuel is consumed in flight, the volume of that fuel must be displaced by outside air. Fuel tank to atmospheric pressure must never exceed some low differential pressure. This is one source of moisture ingress that must be dealt with, and can cause water to accumulate in tanks if left unchecked.

>you descend through humid and coolish air ie not summer heat,

Are you aware that the air temperature at altitude 35,000 ft is around -55C?

[0] http://www.petrolplaza.com/technology/articles/MiZlbiYxMDI3N...

edit: Here is a bunch of related photos if you're interested. http://englishrussia.com/2013/06/20/how-aircrafts-are-fueled...