Possible Causes of the Boeing 777 Crash Landing at Heathrow?

[chroot]

Andre, I'm sorry, but I don't think you have any idea what you're talking about. The ideal landing is very near stall speed in ALL aircraft, including big ones. You want to land as slow as possible to shorten your roll -- if the aircraft is operating properly. If you've lost your engines, you want to set the aircraft up in whatever configuration maximizes your glide ratio so you can actually obtain the runway. This configuration will result in a faster-than-normal landing in every airplane I'm familiar with. Futhermore, this glide configuration already produces the lowest possible rate of powerless descent, and that's how you want to land. Forget the flare, forget the nosegear, forget proper form and bonus points for imperceptible maingear touchdown. Just put it on the damn tarmac with everyone still alive.

The weight and thrust of the aircraft have nothing to do with these decisions. I really don't think I have any idea what point you're trying to make anymore.

- Warren

 

[Cyrus]

I don't know if its even possible to do a flaps up landing on an A/C that large. But as the saying goes, "Your altitude deterimines your attitude".

Andre, please don't fly any airplanes.

 

[Ivan Seeking]

I think Andre was a career fighter pilot or similar?

 

[Ivan Seeking]

Ah yes, here we go.

I'm a semi-retired fighter pilot and I published four articles about strategy and air power in a mean stream Dutch military journal.

https://www.physicsforums.com/showthread.php?p=1233209#post1233209

 

[russ_watters]

I don't know if its even possible to do a flaps up landing on an A/C that large. But as the saying goes, "Your altitude deterimines your attitude".

It is - it ain't pretty, but it is possible. You're just going too fast, it doesn't necessarily mean a hard landing.

United 232 (Soux City) came down without flaps and probably would have been fine if it hadn't caught a gust right before touchdown. Given the condition of the plane, though, it was miraculous they got that close to a safe landing.

edit:
Wow, never heard about this one: http://www.pages.drexel.edu/~st55/Research/DHL.pdf

An A300 landed safely at Baghdad airport after being hit by a SAM that severed and drained the hydraulic lines. The pilot was able to control the plane partly because a few months earlier at a seminar, he heard a talk about how they (almost) landed United 232!

It's a little unclear about the landing speed (either 225 or 180kts), but it sounds like the flaps were retracted, but slats were extended (in takeoff configuration, as it was hit on climbout).

 

[russ_watters]

I remember hearing once that to fighter pilots, energy is life - but they were talking about dogfighting, not landing.

Anyway, the fly-it-into-the-ground technique warren's describing seems to have been pretty much validated by the outcome of this crash. It destroyed everything hanging below the plane, but they made it to the airfield and no one died.

 

[chroot]

Well, I'm only a private pilot, still working on my IFR, with no air carrier or (gasp) dog figthing experience. All I can say is how I'd fly my own aircraft were it powerless. I'd set it up for maximum glide ratio all the way to the threshold. If I didn't make the threshold, there's no way I'd even consider a flare. I'd land it dead flat, do everything I could to keep the wingtips from hitting the ground, and let the destruction of the undercarriage slow me down. Warren 1, plane 0.

- Warren

 

[Andre]

Don't you think that a 15 tonnes fighter aircraft also has to obey simple basic aerodynamics on final approach? There might be a slim change that a semi retired fighter pilots knows what he is talking about after having done that ...oh... perhaps 3000 - 4000 times?

The 600 feet / 2 miles resemble a 3 degrees glidepath which is pretty standard nowadays, without engines a modern jet glides about a nautical mile losing 1000 feet. The difference is thrust. Thurst also helps in the flare, when the trust vector is pointed more upwards, the vertical component adds to the lift. A standard trick for a super soft landing is adding a bit of thrust a second or two before touchdown. However the aircraft owner is usually not amused by that because it wears down tyres pretty fast.

The point is though that without engines you need more speed to be able to break the rate of descent. And for a jumbo sized flying contraption that's not for the comfort of the passengers but has to do with the difference of dropping a mouse or an elephant from 20 feet. Mouse runs away Elephant is likely dead. Pure inertia.

So in this case, without engines and 'behind the power' curve, already losing a lot of speed there was no option than dropping the nose, pick up speed and concentrate on the flare, which they seemed to have done rather professionaly.

 

[Andre]

Just to explain in general:

without engines and 'behind the power curve'

This is a generic aircraft "power curve"

showing how the drag is speed related, at the high end drag is evident, at the low end, the induced drag is the mainly result of increasing pressure differential under and above the wing apart from some other effects like turbulent boundary layers. The solid drag curve also resembles the energy required to maintain that speed. So the approach speed is an trade off between these effects, somewhere around the minimum drag point, and you would try to avoid getting into the speed region where the induced drag becomes dominant, because slower speeds require more thrust. that's called "behind the power curve". In a slow decent things change hardly, the engines deliver not enough thrust to maintain level flight and you get the additional energy required to maintain speed by losing height, or actually potential energy. It's all a matter of energy.

So without engines, reducing speed gets you behind the power curve where the induced drags kills your energy pretty fast. Inevitable result is a stall and followed in this case by disaster.

You need to maintain the speed to remain in the lowest part of the drag curve. That's why speed is life. That's also true for a B777.

 

[pitot-tube]

Just to explain in general:



This is a generic aircraft "power curve"

showing how the drag is speed related, at the high end drag is evident, at the low end, the induced drag is the mainly result of increasing pressure differential under and above the wing apart from some other effects like turbulent boundary layers. The solid drag curve also resembles the energy required to maintain that speed. So the approach speed is an trade off between these effects, somewhere around the minimum drag point, and you would try to avoid getting into the speed region where the induced drag becomes dominant, because slower speeds require more thrust. that's called "behind the power curve". In a slow decent things change hardly, the engines deliver not enough thrust to maintain level flight and you get the additional energy required to maintain speed by losing height, or actually potential energy. It's all a matter of energy.

So without engines, reducing speed gets you behind the power curve where the induced drags kills your energy pretty fast. Inevitable result is a stall and followed in this case by disaster.

You need to maintain the speed to remain in the lowest part of the drag curve. That's why speed is life. That's also true for a B777.

Induced drag results from vortices that move forward in the direction of the aircraft.The aircraft slows down because of Newton's Law - every action has an equal and opposite reaction.Minimise the size and momentum of the forward moving vortices and you minimise the induced drag.If you hold a ruler vertically, with one end in some still water, and move it quickly you will see the vortex it sheds moves in the direction of motion of the ruler.

 

[pitot-tube]

So in this case, without engines and 'behind the power' curve, already losing a lot of speed there was no option than dropping the nose, pick up speed and concentrate on the flare, which they seemed to have done rather professionaly.

The best way to land a large aircraft such as a boeing 737 is to increase thrust a few seconds before touchdown - a compromise between getting the
aircraft to move slowly without allowing it to drop out of the sky.
Without engine power the pilot could dive to get the aircraft's speed up.But given the low altitude of the aircraft he couldn't have done a very significant dive.The aircraft landed at the perimeter of the airfield and this may mean that the pilot kept the aircraft landing as he normally would have done knowing it would be a close call -would the pilot have risked any tricky diving manoeuvre or adjustment of control surfaces over people's houses without engine power? Unlikely.

 

[Andre]

We not talking about more than a few degrees of course. Furtermore, if you don't change the pitch attitude of the aircraft, the slowing down will automatically increase angle of attack, meaning that the rate of descent increases, but in this situation a flare at slower speeds without engines will be rather futile.

 

[Art]

Would it have helped gain more airspeed and thus greater glide capability if the pilot had retracted the undercarriage or was it too late for that?

 

[Andre]

Aerodynamically yes but on the other end the landing gear absorbs a lot of impact energy while collapsing, reducing the damage to the rest as Warren observed.

Induced drag results from vortices that move forward in the direction of the aircraft.

I don't understand that. Yes there are wingtip vortices generated by the pressure differentials also known as 'wake turbulence' or even 'jetwash'.

Source NY Times

But that's more the result of the induced drag.

 

[rewebster]

great photo, Andre, reminds me:

has anyone here tried to win the million?


http://www.claymath.org/millennium/Navier-Stokes_Equations/

 

[Art]

Aerodynamically yes but on the other end the landing gear absorbs a lot of impact energy while collapsing, reducing the damage to the rest as Warren observed.

With the priority on actually clearing the airport perimeter fence wouldn't increasing glide ability be the number 1 priority? Plus for a landing on grass I'd have thought the SOP would be with undercarriage up to avoid the landing gear punching holes in the wings with the attendant risk of fire?? (at least that's what they always did in the old WW2 films :) )

 

[Andre]

The SOP's I know call standard for gear down emergency landings. Holes trough the wings would only happen at zero speed. At typical touch down speeds over 100 knots in soft soil undershoots would act as strong brakes until the gear collapses torn off to the back, sure there will be damage to the wing but that was inevitable anyway.

And it's is esential to have a few feet of breaking the sink speed, if that can be accomplished by the gear, that's a whole lot better than having the tail or the engines touch first, because those are going to deform also the same couple of feet with high risks of more structural damage than necesary.

 

[chroot]

This is all nonsense, Andre. You're arguing second- and third-order effects. This is a discussion about making a runway without any engines. Your mention of turbulent layers, wingtip vortices, induced drag, etc. is just unnecessary complication to make it sound like you've won the argument. I know this is how you usually conduct arguments, though: you throw jargon and complex language at people until they just give up and decide you must know more than they.

The bottom line: If you don't think you're going to make a runway, you don't give up altitude to gain speed so you can do a nice flare and protect the aircraft -- you keep it in best-glide configuration all the way into the ground, hoping all the while you might miss the super-highway, make the threshold, and keep your passengers alive.

- Warren

 

[Andre]

I guess we are repeating moves. You consider the flare unnecesary to keep the passengers alive, I think it may be correct for smaller aircraft but you underestimate the inertial effects of large aircraft

http://findarticles.com/p/articles/mi_qa3744/is_200505/ai_n13642230

One of the Class A mishaps involved a C-5 mission well into a long duty day, on a night VFR, idle power descent, tactical arrival into an airfield in the AOR. During the final turn, with throttles still in idle, the airspeed decayed to 10 knots below approach speed with a high sink rate. The crew was late in recognizing the aircraft's negative energy state. The pilots advanced the throttles 3-5 seconds prior to touchdown, but failed to arrest the sink rate. The aft fuselage contacted the runway causing extensive damage. This is the worst case scenario: slow, high sink rate, with throttles set to idle. If thrust is set to idle and maintained at idle, no energy is available immediately to recover from a low-speed condition or to initiate a go-around.

Point is failed to arrest the sink rate - extensive damage. And the sink rate in a glide is considerably more than a normal approach. This 777 was hardly damaged that way.

This guy agrees with me:

http://yarchive.net/space/launchers/horizontal_vs_vertical_landing.html

When landing an aircraft with the power off, the pilot must keep the airspeed high until he or she is ready to bring the flight path level with the ground and land.

For an F-16 normal approach speed is around 160 kots weight depending but in a flame out pattern we use 220-230 knots.

 

[chroot]

You don't understand -- you have two choices:

1) Slow the rate of descent by giving up altitude, and crash into the freeway outside the airport. Enormous property damage and many deaths result.

2) Fly the plane into the ground in best-glide configuration, missing the freeway and clearing the airport perimeter fence. The airplane will be a hull loss either way, but you might actually make the threshold and keep everyone alive.

-Warren

 

[Andre]

Almost agree, 1) but slowing the rate of decent by giving up altitude is an implicite contradiction. Perhaps you intend to indicate 'stretching the glide' by increasing pitch attitude and work yourself into serious trouble

2) That's what I'm trying to say too. Problem with this kind of aircraft is that approach speed is likely lower than the best glide speed, following the logic of the power curve. In the case of the F-16 the approach speed of a Cessna 172 type A/C fits in between the two. In the case of a B777 I would assume the difference would be in the order of magnitude of 10-20 knots. At those higher best glide speeds much more control authority is avaible to flare and break the rate of descent, which is about 3000 ft/min for the F-16.

In this mishap the loss of energy that demanded more power only aggravated the situation. But if you begin adjusting speed at that point obviously there is very little altitude to trade. In hindsight my gut feeling says that in real double flame out situation, there would have been no chance to get anywhere near best glide speed in that -behind-the-power-curve position. It occurs to me that the engines may still have been running at lower power setting but just did not respond to inputs as the story says. That would explain why they did not loose the energy so rapidly and were able to break the descent enough to prevent wrecking the aircraft on impact.

But I agree, it's only speculation.

 

[Art]

Update on likely cause of crash

Autothrottle problems suspected in Heathrow 777 crash
Possible clues from previous engine failure incident?
By Lester Haines → More by this author
Published Thursday 24th January 2008 11:35 GMT


Investigators probing last Thursday's Heathrow Boeing 777 crash may be able to glean useful information from six previous engine failures on the type, one of which could prove highly significant in pinpointing the cause of the incident.

The Air Accidents Investigation Branch (AAIB) has apparently ruled out bird strike and fuel starvation as factors in the accident, and its update says it's examining why "neither engine responded to throttle lever inputs during the final approach".

Although it was initially believed both engines on BA flight 038 had failed, it's now been revealed that they "did not shut down and both engines continued to produce thrust at an engine speed above flight idle, but less than the commanded thrust".

http://www.theregister.co.uk/2008/01/24/777_autothrottle/

It would make you nervous flying on 777s at the moment thinking there is an unresolved computer glitch which could cause catastrophe at any time

 

[Andre]

I'd though "aha!" reading that, but yet it seems to be a bit different now:

The great escape of Flight BA038

The keys:

Martin Green, another airport worker, told Sky News: “It came in at a very high angle and just dropped like a stone — I would estimate 200ft.

“It seemed to be flying fairly slow and it had a very high angle of attack. The nose was high up in the air, which is very unusual.”

Mr Green, who has worked at the airport for 23 years, added: “It’s the first time I’ve seen anything like that, and I hope it’s the last.”

and

Antonio De Crescenzo, 52, from Naples, said there was little warning that the plane was in difficulty. “We were coming into land but the plane felt like it should have been taking off. The engines were roaring and then we landed and it was just banging.

and

Mike Zihni, who lives less than 50 metres from Heathrow’s southern perimeter fence, had just woken when he heard the roar of the engines. Having heard thousands of planes descend through a channel slightly to the right of his house, the taxi driver knew instantly that something was amiss. “It was very loud, as if the plane was a lot closer to the ground. I don’t normally notice the sound of the planes, but this time it was weird. It was like a reverse thrust,

Obviously the stories don't add up. I have not a good feeling about this. Sounds an awful lot as behind-the-power-curve.

 

[grant9076]

First, I must say that I am no expert on the B-777 because I have never flown one (yet).

However, I have type ratings on the B-737, B-757, and the B-767. In addition, I have flown into Heathrow more than a few times in the past (more on this later). So hopefully I can provide some useful input.

I have to agree with Andre on the sink rate issue and here is why:

An airplane that is stabilized on a 3 degree glide path at about 140 to 150 Knots will have a vertical velocity indication (VVI) of about 700 ft/min (give or take depending on groundspeed). If it is in the same configuration and decelerating through this speed at idle thrust, the pilot will either have to drastically lower the nose to maintain airspeed, or decelerate into the region of reverse command. Either case will cause a drastic increase in the sink rate.

A couple of years ago, I had the rare opportunity to practice some idle descent approaches in a full motion simulator for a B-707 type aircraft. The approach speed for full flaps was about 148 Knots, the Reference Speed was about 143 Knots, and the touchdown speed in this configuration was about 133 Knots. I had to maintain a final approach speed of about 180 Knots, a glide path of about 10 degrees (with a VVI of about 3000 ft/min), and aim about a mile short of the runway in order to break my descent and land in the touchdown on speed. If I did not break my descent, the 3000 ft/min VVI impact would deliver about 50 times the energy of a 400 ft/min VVI no flare impact. Furthermore, I flew the final approach segment at the optimum partial flap setting and only selected full flaps about halfway through the roundout which started about 700-800 ft above field level. Had I flown the entire approach with full flaps, the extra drag would have forced me to maintain at least an additional 15 Knots and a much steeper glidepath (and thus a much higher sink rate) in order to maintain it so that I can have energy to flare. If someone fails to flare with this sink rate, I am sure that it will do severe damage to the occupants of the aircraft. I believe that it is for reasons like these why Boeing has warnings about the possible lethal consequences of idle descent approaches in some of its aircraft flight manuals.

The approach controllers at Heathrow and many other airports like to vector aircraft in high and fast and force them to lose a great deal of their energy in the last 5 to 6 miles. We call this being "slam dunked" and it is a practice that I and many other pilots disagree with. Because many large aircraft have restrictions against using speedbrakes at low altitudes on approaches, this forces the pilot to use a low power setting. With this high rate of energy loss, it is very easy for "high and fast" to quickly become "low and slow", especially when combined with the normal delays in automation response and engine spool up. Incidentally, I have been "slam dunked" twice at Miami International and once at Dallas Fort Worth in the past couple of weeks. I deal with it by disengaging the automation and smoothly bringing up the throttles just prior to reaching approach speed.

With all of that being said, I do not know enough about the B-777 engine systems to speculate.

P.S. Zero Flap landings in large aircraft often require very high angles of attack and pitch attitudes which make tail strikes (as opposed to nosewheel first landings) a major hazard.

 

[russ_watters]

Almost agree, 1) but slowing the rate of decent by giving up altitude is an implicite contradiction. Perhaps you intend to indicate 'stretching the glide' by increasing pitch attitude and work yourself into serious trouble.

I suspect he meant dropping the nose to keep the speed up (and losing altitude) before flaring to slow the decent rate before impact. The net result is you don't glide as far.

 

[russ_watters]

Update on likely cause of crash

http://www.theregister.co.uk/2008/01/24/777_autothrottle/

It would make you nervous flying on 777s at the moment thinking there is an unresolved computer glitch which could cause catastrophe at any time

It does sound like a computer bug. The 777 has a lot of flight time, so that makes it seem unlikely, but it is possible. It may have also been a sensor or other electronics problem that triggered the adverse reaction from the computer.

 

[russ_watters]

I'd though "aha!" reading that, but yet it seems to be a bit different now:

The great escape of Flight BA038

The keys:



and



and



Obviously the stories don't add up. I have not a good feeling about this. Sounds an awful lot as behind-the-power-curve.

No, they aren't consistent, but if that first one is correct, it would imply that the pilot did exactly the wrong thing - instinctively react to being too low by pulling back on the stick, causing a stall. But then, for an experienced pilot, instincts are supposed to be trained to tell you the right thing to do.

This will all wash out in the investigation, though.

 

[Cyrus]

As I don't fly Jets, I take back my criticism Andre. With 172's, altitude, as warren said, is important. I do know that on final approach, you can pitch up to slow down and increase the sink rate significantly if you are high on final. It certainly seems plausible that for a large A/C breaking this rapid descent rate can be more of an issue on flair, in which case coming in at best glide speed won't mean much if you are falling like a brick. The best glide speed just means you will optimize the sink rate without power. However, that does not mean its a sink rate you want to impact the ground with. The best glide speed is a velocity vector, and if most of it is pointing below the horizon, that's not an ideal situation.

Lets assume an excessive sink rate at 3000 FPM. Thats roughly 40mph impact down into the ground (forward airspeed will simply move the airplane forward). I am sure a big airplane like that would fare well and absorb the energy in the structure. So flying it into the ground at best glide speed should not be that big an issue. The airplane would be a loss, but I don't see it as killing you.

 

[Cyrus]

Update on likely cause of crash

http://www.theregister.co.uk/2008/01/24/777_autothrottle/

It would make you nervous flying on 777s at the moment thinking there is an unresolved computer glitch which could cause catastrophe at any time

This is the reason why there are airworthiness directives to aircraft owners. All aircraft have problems that only show up after many hours of flight time. Its the reason why a database is kept and notices are sent out when a problem is seen among many airplanes of the same type. If this were a problem on all 777s, an AD would have been issued and sent out. So to say that there is an 'unresolved computer glitch' which could cause a catstrophe at any time is nothing more than uninformed nonsense. As of now, there is no issue with all 777s, so don't expect them to start dropping out of the sky.

http://en.wikipedia.org/wiki/Airworthiness_Directive

A 777 is one of the best airplanes out there.

 

[Andre]

Lets assume an excessive sink rate at 3000 FPM. Thats roughly 40mph impact down into the ground (forward airspeed will simply move the airplane forward). I am sure a big airplane like that would fare well and absorb the energy in the structure. So flying it into the ground at best glide speed should not be that big an issue. The airplane would be a loss, but I don't see it as killing you.

Still underestimating the inertia. What for instance might have been the order of magnitude of the vertical velocity on impact of this 777?

Given: initial altitude 700, distant to cover 2 miles minus undershoot say 1,75 nautical miles, Say that the speed had been reduced to 120 knots then time to impact would be 52.5 seconds, covering that 700 feet to lose with an average sink rate of 800 feet a minute.

Grant says:

If I did not break my descent, the 3000 ft/min VVI impact would deliver about 50 times the energy of a 400 ft/min VVI no flare impact.

hence the 800 ft/min order of magnetude is four times the energy of the normal no flare impact. My estimation of a 3000 ft/min impact is a bit more pessimistic, I would not expect to walk away from that.