News Boeing 767 cargo plane crashes outside Houston with 3 on board

[dbeachy1]

Quoting from https://www.usatoday.com/story/news/2019/02/23/cargo-plane-crashes-outside-houston-no-survivors-likely/2964491002/

HOUSTON — A Boeing 767 cargo jetliner heading to Houston with three people aboard disintegrated after crashing Saturday into a bay east of the city, according to a Texas sheriff.

Witnesses told emergency personnel that the twin-engine plane “went in nose first,” leaving a debris field three-quarters of a mile long in Trinity Bay, Chambers County Sheriff Brian Hawthorne said.

“It’s probably a crash that nobody would survive,” he said, referring to the scene as “total devastation.”

The cargo plane made a steep descent shortly before 12:45 p.m. from 6,525 feet to 3,025 feet in 30 seconds, according to tracking data from FlightAware.com.

The flight was being operated for Amazon by Atlas Air, according to a statement from the airline.

A CNN article is here.

 

[IronRain]

Quoting from https://www.usatoday.com/story/news/2019/02/23/cargo-plane-crashes-outside-houston-no-survivors-likely/2964491002/



A CNN article is here.

I've completely missed this one. Nothing in the news here.

 

[Urwumpe]

The video would be more interesting with the high altitude radar data... it looks like the plane flew straight into the bad end of a thunderstorm. 13000 ft altitude is pretty low there.

 

[garyw]

There has been some talk on reddit about the data showing a 7,000 fpm descent. I'm wondering if this is a windshear related accident.

 

[Marijn]

There has been some talk on reddit about the data showing a 7,000 fpm descent. I'm wondering if this is a windshear related accident.

Investigators are on site to recover the flight data recorders. So we'll learn what happened probably quite soon.

What troubles me is that there were no distress calls apparently. It's hard to imagine a plane over which the crew has little or no control descending 13000 feet in two minutes without any distress or mayday calls on the radio.

Besides hijacking, what could explain this? The radio could have been broken, but two special events happening at the same time is always highly unlikely. Maybe lightning disabled all systems. But I am not aware of something like that ever happening before.

A more likely scenario therefore is, imho, that the crew never knew they were in danger until it was too late. This happened twice in 2009 with Turkish Airlines Flight 1951 and Air France Flight 447. On both occasions, faulty information sent to the autopilot which eventually led to a stall which was not recognized in time by the crew.

 

[Linguofreak]

A more likely scenario therefore is, imho, that the crew never knew they were in danger until it was too late. This happened twice in 2009 with Turkish Airlines Flight 1951 and Air France Flight 447. On both occasions, faulty information sent to the autopilot which eventually led to a stall which was not recognized in time by the crew.

With Air France 447, they knew they were in danger, but it was the improper and uncommunicated reaction of one crew member to that danger that doomed them.

 

[Marijn]

With Air France 447, they knew they were in danger, but it was the improper and uncommunicated reaction of one crew member to that danger that doomed them.

True. But not recognizing the situation or reacting the wrong way to a situation boils down to the same thing: Pilot error. I wanted to avoid these words since it is pure speculation. But the absence of any radio calls just boggles the mind.

I've been a glider flight instructor for more than 15 years now. During these years, I've had 'students' in the front seat who were current and retired airline pilots of whom you might expect a complete understanding of flight and especially stalls. But to my astonishment, not all of them did seem to do so.

Recently, a retired pilot I know very well who is flying for an Asian airline now because he does not want to retire yet, complained that his new collegues did have the nasty habit of covering the windows with newspapers to block the light, eliminating the option to visually confirm the pitch of the aircraft. I'll never fly with that airline.

 

[Urwumpe]

What troubles me is that there were no distress calls apparently. It's hard to imagine a plane over which the crew has little or no control descending 13000 feet in two minutes without any distress or mayday calls on the radio.

Quite easy. Imagine being too busy to keep the aircraft flying than to call help. Also, if they had been inside a thunderstorm, they could have been unaware of their attitude and altitude until it was too late - you are flying instrument there then, with easy loss of situation awareness.

 

[Marijn]

Quite easy. Imagine being too busy to keep the aircraft flying than to call help.

Aviate, navigate communicate. Yes I know. But this plane was piloted by a crew of two, possibly three. Not all of them would have manual control at the same time. At least someone should bother about operating the radio.

 

[MaverickSawyer]

Crew would have been two pilots and a loadmaster in the jumpseat. The loadmaster would be little more than a passenger during flight, and would likely not even be able to reach the radios, let alone use them. As for what happened, I'm not going to speculate... there's very little publicly available evidence to work from. We've seen a very small portion of the situation, and are missing critical details.
Regarding the comparison to Air France 447, that's not a valid comparison. Airbus aircraft use sidestick controllers that are not physically interconnected, allowing two completely different sets of inputs to be generated. 767s have yokes that are physically interconnected by cables both between the yokes and to the servo valves that control the hydraulic boost units. There is no way, short of a complete severing of the control linkages, that the pilots could be inputting conflicting commands.
Regarding the quality of pilot training for airliners... agreed. They're being trained as systems managers, not pilots. Throw them an edge case emergency scenario that's not part of the required training and they're going to be at a complete loss.

---------- Post added at 08:27 ---------- Previous post was at 08:23 ----------

http://www.flightglobal.com/news/articles/video-shows-atlas-767f-in-steep-dive-prior-to-cras-456054/

"The aircraft is in the video… at a steep descent – [a] steep nose-down attitude,” NTSB chair Robert Sumwalt said during a press conference on 24 February. “I saw no evidence of the aircraft trying to turn or pull up at the last moments.”

The security video was taken from a county jail at a distance of slightly more than 1nm (1.9km) from the site where the Boeing 767-300ER Freighter crashed following a flight from Miami.

 

[Artlav]

two special events happening at the same time is always highly unlikely

Actually, many disasters are exactly that, since single special events are typically all accounted and planned for.

 

[Urwumpe]

Aviate, navigate communicate. Yes I know. But this plane was piloted by a crew of two, possibly three. Not all of them would have manual control at the same time. At least someone should bother about operating the radio.

Can you imagine how it would be like in a cabin of an aircraft suddenly accelerating to 7000 fpm descent rate? During a thunderstorm? Flying a controlled aircraft in a thunderstorm at 13000 ft is already a rough ride. When its leaving control of the pilots, its far worse and you can't trust your perception at all.

I really don't think this is a normal situation in which normal reaction times apply. Or in which CRM works perfectly. I could even imagine them to pull up banked until getting into a high-g stall - without them realizing the fact on the PFD until it is too late.


Also, being in a thunderstorm could of course also have effects on the cargo - what if the CG shifted when cargo was turned loose by massive turbulence? This is not a normal flight situation and small errors in securing the cargo, that usually mean no problem could be a serious problem then.

---------- Post added at 17:39 ---------- Previous post was at 17:36 ----------

BTW I was wrong... I somehow converted ft to meters and back again to often and came up with 13000 ft... actually it was just 6000 ft before contact was lost.

That means really in the bad end of a thunderstorm.

---------- Post added at 17:48 ---------- Previous post was at 17:39 ----------

And even more strange - it had been 10 minutes from loss of signal to impact. Not a sudden short dive.

 

[n122vu]

It's pointless to speculate until the CVR and FDR are recovered. For all we know the Captain or the FO either one could have been away using the restroom prior to landing and something went wrong. Or he could have been returning from the restroom and slipped and fell, knocking himself unconscious and laying on the controls, the sudden shift in pitch causing the not-yet-buckled-for-landing jumpseat occupant to be thrown forward against the back of the seat, unable to assist in getting the incapacitated pilot off the controls.

This scenario sounds ridiculous because it is, and so is speculating at this point.

 

[MaverickSawyer]

Agreed. I'm holding off on speculation until we have more information.

 

[Marijn]

Actually, many disasters are exactly that, since single special events are typically all accounted and planned for.

Agreed. That's why major disasters are rare in the first place. Multiple things have to go wrong at the same time before anything major is going to happen.

The same logic holds true when trying to answer questions such as why the radio wasn't used. Either the cause of the plane getting into trouble also compromised the radio, or the radio was working fine but not used for another reason. It's very unlikely that a separate event which has nothing to do with the plane getting into trouble broke the radio. That's what I meant.

---------- Post added at 08:08 PM ---------- Previous post was at 06:24 PM ----------

... actually it was just 6000 ft before contact was lost.

That's not very high when coming down at 7,000 ft/min. That's 50 seconds to crash. In that case, I can see how they were too busy to operate the radio.

But if there was indeed a longer period of several minutes between loss of contact and the crash, if the crew was aware of any trouble, I would expect that they would do a call about it at some moment during those minutes.

What do you mean with the 'bad end' of a thunderstorm? It's base? And the bad part is the very large rate of sinking airmass all the way to the ground?

 

[Gingin]

Really Sad, RIP and thinking to the family :/


Fire, mechanical linkage, Cargo shift,weather, overloaded situation,startle effect could have explain the silence...
Always a 1000 ways to speculate, just wait and see for FDR and CVR.

two special events happening at the same time is always highly unlikely

That a bit part of Reason process

https://en.wikipedia.org/wiki/Swiss_cheese_model

 

[Urwumpe]

That's not very high when coming down at 7,000 ft/min. That's 50 seconds to crash. In that case, I can see how they were too busy to operate the radio.

Well, 7000 ft/min was the peak vertical velocity. It accelerated downward, and that rapidly.

But if there was indeed a longer period of several minutes between loss of contact and the crash, if the crew was aware of any trouble, I would expect that they would do a call about it at some moment during those minutes.

Unless they had been unable to call - again, the last radio communication happened minutes before the sudden turn and the descent.

What do you mean with the 'bad end' of a thunderstorm? It's base? And the bad part is the very large rate of sinking airmass all the way to the ground?

No, the turbulent downdraft or updraft areas midway. There are no reports of a massive downdraft event and you also don't see it on on radar. it was far away from the thunderstorm front, but within the region of growing nimbus clouds.



But the radar data looks more like weather was, if at all, just a contributing factor, not the root cause. The final crash sequence started right after the turn to the north. As if something failed mechanically.

 

[Marijn]

midway

Midway is confusing me a bit. Midway between the top and the ground? Or midway between ground and cloudbase?

 

[Urwumpe]

Midway is confusing me a bit. Midway between the top and the ground? Or midway between ground and cloudbase?

Using this as reference, left picture, right above the cloud base.

 

[Marijn]

Ok. But yes, rain and turbulence don't bring a 767 down by themselves.

Finding the data recorders seems complicated. If they could track the transmitters, then they would have found them by now I would think. I assume the tides deposit a fresh layer of mud on top of them.