The crash of ValuJet Flight 592 in 1996 was recently brought up in the Dreamliner thread. I remember when that crash was in the news, and the Wikipedia article sums things up pretty well. The plane crashed pretty much straight down into a swamp at 500+ MPH. The swamp had a bedrock base at its bottom, resulting in the plane and all aboard being rendered into small pieces. Recovery of debris and body parts was made very difficult by the swamp.
Investigators ultimately concluded that the crash was caused by an in-flight fire, which in turn was caused by some oxygen generators that had been improperly packaged for transport.
Question: how in the hell did they figure that out? Clearly there was a fire on board (as evidenced by reports from the crew captured on CVR before the crash), but given that there were no survivors, and that the plane and everything on board were reportedly shattered to bits by the impact, how did they determine that the O2 generators were the primary cause of that fire?
They rebuilt the aircraft like a puzzle taking the shrapnel and components and placing them in relation to one another. Experienced mechanics, engineers and air safety investigators can pretty much see an aircraft component and figure out what it is, particularly in their area of expertise (eg fuel, doors, hydraulics etc). From that you look for evidence of fire and fire propagation and see what is more severley damaged and where the fire source was. You pretty much know it can only be in certain origins anyways. The flight data recorder records the various fire detection loops and failure of sensors and components with a time stamp so you can “see” the damage propagate and use existing design failure analysis modes to confirm it.
Google the accident investigation report from the NTSB- it’s technical but covers all the evidence. Accident investigation reports for civil aviation are public information in most jurisdictions and are easy to find.
In the course of the investigation, they examined all parts of the wreckage that were able to be retrieved, did interviews with ground crew, listened to the in-flight recorder, and got witness accounts of the crash.
They found signs of heat damage (soot, melting, etc) in the forward hold, but not the aft hold. Further, the heat damage was not consistent with signs of an electrical fire. They knew oxygen canisters were on the manifest and actually found several pieces of damaged canisters in the wreckage. From there it only required testing with similar materials to see if they could cause a fire when packaged in the way they’d been on the plane. (Yes.)
NTSB investigations are pretty impressive, in my opinion. The report goes into a lot more detail. Their recommendations as a result of this investigation include everything from better fire detection systems in cargo holds to changed procedures for listing “lap children” on passenger manifests. Very thorough.
I was perusing that Wiki entry a while back too. Isn’t the CG image of the plane about to crash completely wrong? Didn’t the jet quite literally ‘auger-in’ nose first? Not go down straight & level as depicted?
No, it essentially dropped straight down out of the sky, and made a plane-shaped hole in the swamp vegetation. I remember some pretty dramatic pictures of the crash site taken from above.
The NTSB determined that the wreckage indicated the plane struck the ground in a “right wing down, nose down attitude.”
Further, witness reports indicate “that they saw a low-flying airplane in a steep right bank. According to these witnesses, as the right bank angle increased, the nose of the airplane dropped and continued downward. The airplane struck the ground in a nearly vertical attitude.”
You are correct and the Wiki graphic is wrong. I don’t know enough about the inner machinations of Wikipedia to be able to tell you how to approach having it changed, although I did notice that someone in the Talk page for the graphic already pointed out that it is incorrect.
Found this animation, apparently used in subsequent legal actions; it appears to have been made based on FDR data. Things start to go to hell around 8:45, with impact around 9:30. in the end, the plane appears to hit with about 90 degrees right roll and the nose maybe 10-20 degrees below horizon. That’s just the attitude, though; it’s not clear on what the actual path of motion is, though I’d guess it’s closer to vertical than the attitude would suggest. Either way, it’s nothing like the CGI pic on the wikipedia article.
OTOH"
The video is not consistent with this eyewitness report re: attitude.
