Hacker News Reader

Comment by Implicated | Hacker News Reader

or the climb to 45k

Plenty of weird reasons to climb like that[1]. I think people dramatically underestimate how mechanical and/or electronic system failure combines with human misjudgement to make things operate in entirely unpredictable ways.

I've never been involved in an air crash investigation, but I have done many system outage investigations. The majority of them[2] involve a failure of a hardware/software system, followed by manual intervention done without complete information. The number of times I've seen small problems escalate dramatically by the combination of automatic systems and manual intervention is enough to make me assume accident when something like this happens.

[1] http://en.wikipedia.org/wiki/Air_France_Flight_447#Final_rep...

[2] Ok, the majority that don't involve the deployment of new software.

Completely agree. I'm surprised that, especially on HN, the large number of people who won't apply Occam's Razer to the situation.

I imagine that most air accident investigations begin like this; confused, competing information from numerous sources of varying reliability. Just with the internet and 24h news, everyone is following along with each revelation (see also the Pistorius trial).

Give it some time, let the investigators work and report their findings. I'd be very surprised if it's not a combination of system failure and human error in reacting to the failure.

I just blame aliens or a very localized rapture.

To be fair, it isn't at all clear if Occam's Razer doesn't point to a hijacking.

I think that the lack of clear & sufficient information makes Occam's Razer hard to apply.

> I imagine that most air accident investigations begin like this; confused, competing information from numerous sources of varying reliability.

Not to mention that but when things happen on the plane, pilots are given confused, competing from numerous sources. This is why I assumed it was an accident (like Air France 447) at first.

> Give it some time, let the investigators work and report their findings. I'd be very surprised if it's not a combination of system failure and human error in reacting to the failure.

That was my first impulse too. Google "IEEE Automation Paradox Air France 447." However, this is really hard to square with the engine information. So you have three possibilities: the plane flew an uncommanded course on autopilot for 5 hours following an accident, the flight data is wrong, or it is a hijacking.

In this case an investigation is hard because there is so little information. It took until the black boxes were recovered from AF447 to determine what happened there. Here? We don't even know where the plane is, much less the relevant recorders.

I don't know where you got your ATP from, but there isn't a good reason, in clear weather, to climb above the service ceiling for your air-plane.

Now, I don't miss-underestimate how failures happen, how they combine, but I know my procedures. If I'm pulling fuses for things, I'm going to first SQUARK 7700, have the co-pilot screaming on 121.5, whilst he is charged with flying the plane (following a heading, not dropping out of the sky, like AF447!) whilst I then follow the procedures for the kind of electrical fire. Even if it's only for 10 seconds, the 7700 will be picked up, the ATC will see it light up like an angry chirstmas tree. It will not be 'cancelled' by the transponder then being switched off, the operative would have to cancel it, which they wouldn't do until after contact had been reached. In the case of say 7500, they wouldn't take radio as reason to cancel it.

The point is, the transponder just happened to turn off at the perfect moment (between two ATC services) after the ACARS had been switched off long before.

If you had a fire, that had endangered physically separated, redundant systems in that manner, I doubt the plane would be able to fly for 7 hours. The 777 is a fly by wire plane.

However, if you wanted to drop off, un-noticed, it is the perfect time.

Occam's razor n all that. If I thought such failures were possible, as I've said before, I'd shred my pilots license tomorrow.

> I don't know where you got your ATP from, but there isn't a good reason, in clear weather, to climb above the service ceiling for your air-plane.

You are assuming the pilot was in command, but keep in mind a different B777 operated by Qantas had an uncommanded climb caused by twin air data unit failures. The pilots in that case recovered and returned to the airport for an emergency landing. However, I wonder how different it would be over the ocean and at night.

> The point is, the transponder just happened to turn off at the perfect moment (between two ATC services) after the ACARS had been switched off long before.

Yeah. Recovery over the water at night is one thing. Flying for hours after is a very, very different scenario, and that's what makes the difference for me.

Come on;. The Air France cause for climbing up was due to the copilot panicking and completely misunderstanding the situation at hand once the autopilot was disconnected. I would have a hard time to believe that a veteran pilot as the one in MH370 would make such an amateurish mistake.

I've heard arguments from private pilots that say exactly the opposite: The experienced pilots have so much "Nintendo time" flying just on autopilot that they are more likely to commit such gross flying mistakes. Newer pilots still have aviation basics in fresh memory because they have recently flown many hours in small planes with only manual controls.

The pilot in question had built a flight simulator at home. I doubt he was flying it mostly on autopilot. He also worked as instructor.

I am not a pilot, but I would expect both would provide opportunities to refresh memory.

Well in the AirFrance case it was clearly the inexperienced pilot who drove the plane to the water. The commander was sleeping at the time and when he came back in the cockpit he grasped what was happening but it was too late.

s/commander/captain/

s/drove/stalled/

s/sleeping at the time/had just retired to the crew cabin, taking a rest/

he was only out of the cockpit for 18 minutes.

http://www.bea.aero/docspa/2009/f-cp090601e3.en/pdf/f-cp0906...

If it was outside fire (tire), wouldn't it die out in higher altitude?

Possibly explaining the climb to 450? To remove more oxygen from the depressurized compartments to try to smother the fire? I don't know if this is a legitimate fire-fighting technique for an aircraft though (i.e. somehing that the flight crew would have been trained to do).

That's from primary radar. Those altitude readings are unreliable at that range.

Can you expound on that? Does it have to do with atmospheric curving of the radar beam?

> The use of primary radar requires a great deal of signal power, because objects further from the antenna will reflect or send back a weak signal. At longer distances from the antenna, radar becomes unreliable as a way to determine aircraft position with only reflected signals.

http://www.wisegeek.com/what-is-a-primary-radar.htm

I think it has to do with electromagnetic ducting or other clutter effects:

http://www.radartutorial.eu/07.waves/wa17.en.html

If the last effect on the page were in effect, you might get a return, know it's relative bearing well, it's range a little less well, and it's altitude not very well at all.

Radar derives altitude from measurements of distance and elevation angle. The greater the distance, the less sensitive elevation angle is to vertical displacement (simple trigonometry), and hence signal:error decreases.

Investigators flew the same exact flight path in another 777 and it should have been apparent if primary radar misreported altitude +/- such a large margin at the distances observed. If they observed primary radar lose no accuracy at that distance, why judge the 45k as inaccurate?

You are assuming that the altitude reading would always be off by (close to) the same amount.

one theory is that the aircraft was still climbing towards cruise altitude when all control was lost.

a stall, a recovery, and a randomly meandering flight followed.

It wasn't random, though. It was flying to waypoints.

maybe.

adjectives used of the radar data: "inaccurate", "undisclosed".

adjectives used of the flight path: "erratic", "uneven", "zigzagging".

waypoints are thick as thieves, the odds of a flock of birds not hitting three of them are small.

...plus hitting exactly those waypoints makes no sense: the middle zig should have been skipped as superfluous.

waypoints are thick as thieves, the odds of a flock of birds not hitting three of them are small.

Really? On a plane? On my boat they're typically within a few meters.

I don't know much about VNAV but couldn't have they configured a vertical ascent rate with no upper ceiling?

Reply on news.ycombinator.com