Just when I thought I was out-they pull me back in
Boeing 777 accident at London's Heathrow
« on: 2008-01-19 14:42:04 »
Kudos to the pilot and crew of the Boeing 777 that crash landed at London's Heathrow this past week.
The caption on the above picture is what is wrong. The pilot ABSOLUTELY did everything RIGHT, given the circumstances. --Walter
That aircraft was no doubt a handful of trouble on that particular approach.
Here's a pretty good estimate of the situation the flight crew found themselves in: ------------------------------------------------------------------------------------------------------ Lets assume the engines quit. The airplane will descend to the ground quite rapidly. The instinct of a pilot at that point would be to keep the nose up as long as possible and manage what energy he has left (energy being airspeed and altitude) and hope that when he runs out of both, he will be at the runway. In this case, he was short. Anyone put into this situation would have come up short. The pilot did a great job at least getting the jet to the airport property. If this would have happened outside of airport property, this would have been a far uglier accident.
The pilot would be adjusting pitch (nose up or nose down, but most likely nose up) to manage what energy he had left while keeping a margin above stall until he was in a safe spot to put the aircraft down. Judging by the very short landing roll, I am guessing that the aircraft touched down very near to an aerodynamic stall, which is fine ... just what you would want to do in this situation. The landing gear likely sheared off when it encountered the lip of the runway.
From the outside, witnesses probably saw an airplane coming in very low over the fence with the nose poised way in the air ... much more so than what the typical approach body angle would be.
Keep in mind, all of this took place in a very short period of time.
Here's a short Youtube I found that may give you an idea of the time element involved. The event seemed to occur around the time the crew would have heard the "FIVE HUNDRED" call. The "PLUS HUNDRED" call is 100 above decision altitude, or 300 feet above the ground in the case of this video. The rest should be common sense. This type of landing as viewed in the video clip is about as normal and unevenful as parking your car in the garage after coming home from the grocery store.
As you can see, it would have been all of 20 seconds or so between the onset of the event (i.e. no response to throttle up) and when everything came to a stop.
Re:Boeing 777 accident at London's Heathrow
« Reply #1 on: 2008-01-20 03:57:13 »
I don't know what happened here, when or what went wrong and if there was aircrew involvement in the event sequence, but as Walter observes,there are a lot of things pointing to this as a nice piece of work. Note that in the picture of the crash scene, the flaps are deployed. This suggests to me that up until the engines failed everything looked normal to the crew.
Modern airframes are extremely slippery, so in order to descend at the sharpish rate liked by ATC, the airframe involved would likely have been flying in under thrust to keep the airspeed up, but in a fully dirty configuration while approaching the go-around decision point, with wheels, slots, slats, flaps, air brakes and spoilers deployed in order to shed altitude in a hurry. In the event of a go around all of this (but the flaps which help prevent a stall by reducing wing loading and increasing the acceptable angle of attack ie permit slower flight and steeper climbs) would be retracted and the engines spooled to full emergency thrust. In a normal landing, the slots and spoilers would be retracted and the engines would be managed to control the airspeed.
Contrary to the commonly held idea of aircraft "gliding in" to a landing, in the case of the modern airliner they are under power all the way; not least because it takes significant time to spool up an idling turbine, with some risk (small in modern engines under FADEC, but still not insignificant) of a flameout, and because irrespective of wind changes, the pilot's principle task during a landing is to manage the heading, the landing is inevitable after the decision point, as the airspeed is predefined for the airframe type and mass to attain a predetermined descent rate (and normally the airspeed would be controlled automatically by the flight computer, which also maintains level flight, acting on both the throttle and the angle of attack, but in the 777 the throttle position can be overridden by the pilot moving and holding the throttle quadrants at whatever level he demands.
The thrust selected provides the power to fly the airframe during the final approach when manoevering is taboo. Airframes are landed with a significant margin of safety above their stall speed, but this is dependent on thrust being available all the way to the threshold. Significant thrust is needed after landing to provide reverse thrust, and an attentive passenger will notice the engines being spooled up to provide this. Normally the wheel brakes are deployed automatically once the main struts are fully compressed and the main undercarriage is spinning at full ground speed, and the thrust deflectors are deployed, also automatically, once the nose strut is compressed. As hard braking overheats the pads and damages the tires, so far as possible, they are avoided. For the thrust deflectors to function as effective brakes, the engines need to be producing thrust. This setting to do this is called the reverse thrust setting..
When the engines are ineffective, all one can do is get rid of as much drag as possible (but keeping the flaps to try to prevent a stall and because they take a lot of time to operate and not messing with the undercarriage because there just isn't time to retract it ), keep the nose as far up as possible while maintaining some margin above stall speed and hope for the best. As loss of control or a stall caused by too little airspeed is far more likely to cause breakup of the airframe than a controlled flight into off field terrain, when there is danger of loss of control or a stall, the nose is lowered to gain airspeed at the expense of altitude. Only after the decision point, there isn't much in the way of altitude to lose. Here is a vertical diagram of the flightpath.
The Boeing 777 has an unusual but helpful instrument mode (and optionally an independent display) inherited from military aircraft where control of energy levels can become critical, which reflects the computed effective angle of attack on the PLI and speed tape. This is a much more direct (and helpful) input than simple-minded minimum for control and stall speed, as it reflects the safe flight envelope for the aircraft configuration and mass (both Vmcl and Vs/Vso speeds are pretty much meaningless numbers reflecting worst case situations, as actual handling is dependent on mass, density altitude, degree of bank, wing configuration and most critically, on the angle of attack). While I haven't seen it mentioned, possibly because the use of it becomes subconscious, this may well have been one of the reasons why, in the circumstances, this was such a soft landing.
With or without religion, you would have good people doing good things and evil people doing evil things. But for good people to do evil things, that takes religion. - Steven Weinberg, 1999
Just when I thought I was out-they pull me back in
Re:Boeing 777 accident at London's Heathrow
« Reply #2 on: 2008-01-20 14:17:10 »
Quick follow up:
BTW Hermit. Outstanding analysis on that last post. Best to you and yours!
LONDON (Reuters) - Initial investigations showed that the engines of a British Airways Boeing 777 failed to respond shortly before it crash-landed at London's Heathrow Airport on Thursday, investigators said on Friday.
Thirteen people were injured when BA flight 83 from Beijing came down well short of the southern runway and the 136 passengers were evacuated down the aircraft's emergency chutes.
"At approximately 600 ft and two miles from touch down, the Autothrottle demanded an increase in thrust from the two engines but the engines did not respond," the Air Accidents Investigation Branch (AAIB) said in a statement.
"Following further demands for increased thrust from the Autothrottle, and subsequently the flight crew moving the throttle levers, the engines similarly failed to respond."
The AAIB said the flight had been normal until the plane made its final approach for landing. It said 13 people were injured, one passenger suffering a serious injury.
It said the investigation would now focus on more detailed analysis of information from the flight recorder and other recorded information, and examination of "the range of aircraft systems" that could affect the engines.
Captain Peter Burkill, given a rousing reception by staff at the airline's headquarters, refused in a brief statement to reporters to make any comment on the cause of the incident.
But he praised his crew for showing "the highest standards of skill and professionalism," thanked the passengers for "their calmness and good sense" and praised the emergency services for reacting so speedily.
A team of specialists from Boeing arrived to join air accident investigators and experts from Rolls Royce, the aircraft engine manufacturers.
(Reporting by Paul Majendie, editing by Tim Pearce)
« Last Edit: 2008-01-20 14:21:24 by Walter Watts »
VH-Cheer Up 26th February 2007, 04:25 That MD-80, fus. 909, had less than 1,000 hours, and was still in Douglas Fright Test program when an FAA pilot made that great landing.
Douglas Fright Test? Aptly named program. Upon acquisition of MD by Boeing, I believe it was superseded by the Boeing Brown Trousers Program. So-called, because excessive rotation 10kt below V1 and landing at vertical speeds of 500fpm+ seems to leave enormous skidmarks. (On the runway).
WHBM 26th February 2007, 09:27 That MD-80, fus. 909, had less than 1,000 hours, and was still in Douglas Flight Test program when an FAA pilot made that great landing. One broken ankle was the only human casualty. A few years later, that plane was repaired and put back in flight test program, and later sold. Just a minor correction here but #909, although repaired, was never delivered to any customer. Swissair, who had ordered it, refused to accept it and nobody else would take it either. McDD used it for test purposes for about 10 years, then by about 1990 it was no longer required and scrapped. #917 was also lost in the other testing accident described at around the same time. Douglas built another two replacement aircraft for Swissair.
> >Looking for the reg carried by the MD-80 un-ducted fan demo aircraft > >(MD-UHB DEMO on the fuse) flown in the late 80s. > > Registration N980DC, serial number 48000, line number 909, and the > first MD-80. > > >What happened to this aircraft after the project was discontinued? > > I noticed it parked at Mojave in April 1990. The unusually long #1 > pylon with no engine attached to it first caught my attention, then > from an angle the distinctive livery eliminated any doubts one could > still have had. Apparently is was sold to International Airline > Support Group on October 4, 1991, who re-registered it N560MD. They > stored it at Grayson County Airport in Sherman, Texas. I thought I > had heard that it had been broken up, though I'm not sure.
Within Douglas this aircraft was known as '909' and for many years was the only flight test aircraft that DAC owned. It was originally intended for delivery to SAS (? not sure), but during flight test an FAA flight test pilot landed a bit hard. Main gear touchdown seemed OK, however when the nose wheel hit one can see a ripple flow backwards then the empennage breaks off. <opps> For some odd reason the customer didn't want that particular aircraft anymore, so DAC got a flight test vehicle. (The usual procedure to get flight test aircraft was either to lease one back or, as in the case of the MD-11 program, sell them as used to FedEx.)
"Brian A. Reynolds" <bareynol@dhinternet.com> wrote: >flight test pilot landed a bit hard. Main gear touchdown seemed OK, >however when the nose wheel hit one can see a ripple flow backwards then >the empennage breaks off.
If I recall correctly, hard landings were being demonstrated for the flight test program when this happened. The hard landing was intentional. BUT, for a reason that escapes me at the moment, the signals got crossed between the test engineers and the pilots with the result that the plane landed about 3 times as hard as intended. The video I've seen clearly shows the forward fuselage buckle in downward bending forward of the wing as the main gear hit, the empennage breaks off at the aft pressure bulkhead, then the forward fuselage buckles again in upward bending as the nose gear hits. During rollout, the fuselage again looks normal, with no trace of buckling visible. The pilots, being up front and away from the action, and knowing it was going to be a hard landing, didn't know that anything had gone wrong.
I've heard the plane was flying again in just a few months, after repairing the damage concentrated on the few locations that took the stresses, empennage interface and forward fuselage skins that had buckled. There was no damage to the landing gear or its interface with the wings -- a testimony to the ruggedness of Douglas' structural design.
My reading is that the pilots did exactly what they were asked to do. NTSB still blamed the pilots despite the "contributory factors". Probably based on historical principals.
Title: Loss of tail empennage on landing, McDonnell-Douglas Corporation, DC-9-80; N980DC, Edwards Air Force Base, California, May 2, 1980 Micro summary: This McDonnell-Douglas DC-9-80 experienced a loss of its tail empennage in this landing during the test-flight program.
Event Time: 1980-05-02 at 0634 PDT File Name: 1980-05-02-US.pdf Publishing Agency: National Transportation Safety Board (NTSB) Publishing Country: USA Report number: NTSB-AAR-82-2 Pages: 25 Site of event: Edwards AFB runway 22
Departure: Edwards Air Force Base, California, USA Destination: Edwards Air Force Base, California, USA Airplane Type(s): McDonnell Douglas DC-9-80 (MD-80) Flight Phase: Landing Registration(s): N980DC Operator(s): McDonnell Douglas Corporation Type of flight: Certification Test Flight Occupants: 7 Fatalities: 0 Serious Injuries: 1 Minor/Non-Injured: 6 Other Injuries: 0
Executive Summary: About 0634 P.d.t, May 2, 1980, a McDonnell-Douglas, Inc., DC-9-80, N980DC, crashed while trying to land on runway 22 at Edwards Air Force Base, California.
The aircraft was on a certification test flight to determine the horizontal distance required to land and bring the aircraft to a full stop as required by 14 CFR 25.125 when the accident occurred.
The aircraft touched down about 2,298 feet beyond the runway threshold. The descent rate at touchdown exceeded the aircraft's structural limitations; the empennage separated from the aircraft and fell to the runway. The aircraft came to rest about 5,634 feet beyond the landing threshold of runway 22 and was damaged substantially. Seven crewmembers were on board; one crewmember, a flight test engineer, broke his left ankle when the aircraft touched down.
The National Transportation Safety Board determines that the probable cause of this accident was the pilot's failure to stabilize the approach as prescribed by the manufacturer's flight test procedures. Contributing to the cause of the accident was the lack of a requirement in the flight test procedures for other flight crewmembers to monitor and call out the critical flight parameters. Also contributing to this accident were the flight test procedures prescribed by the manufacturer for demonstrating the aircraft's landing performance which involved vertical descent rates approaching the design load limits of the aircraft.
Returns fond regards with interest.
PS I suspect that this just may be the most instrumented crash in aviation history. The 777 has a beautiful data recording system. Everything is logged in full Technicolor detail and compared by intelligent algorithms against a database of expected behavior and performance. Parameters which exceed a certain threshold of tolerance are automatically flagged for attention. Additionally the log can be used to feed a simulator, to control an engine test bed or even to perform a test flight.
With or without religion, you would have good people doing good things and evil people doing evil things. But for good people to do evil things, that takes religion. - Steven Weinberg, 1999
Re:Boeing 777 accident at London's Heathrow
« Reply #5 on: 2008-01-20 17:46:25 »
The best article on the event written for a non-aviation audience I have found is this.
Hunt for fatal flaw of Flight 38
An electronic systems failure is emerging as the prime suspect in BA’s brush with disaster at Heathrow
Source: The Times Authors: Richard Woods, Steven Swinford and Paul Eddy Dated: 2008-01-20
After 10 hours of flying, Speedbird 038 was almost home. First Officer John Coward was preparing to land the British Airways flight from Beijing to Heathrow. In front of him on the right side of the cockpit were three screens displaying flight, navigation and engine data; another three were arrayed in front of Captain Peter Burkill, who sat on the left.
Beyond, through the cockpit windows, Coward could see central London, one of the most densely populated areas of Europe, stretch into the distance. It was Thursday lunchtime and they were approaching Heathrow from the east. There is a preference for flights into the airport to cross the capital in this direction because it is quieter than having them take off over the city. Below them, millions of people were going about their business, never imagining that a plane might fall out of the sky.
At eight nautical miles from the airport, BA038 was down to about 2,400ft in a shallow glide. At 7½ nautical miles, the plane lined up with the instrument landing system that would guide it into Heathrow.
“At 2,000ft you lower the gear,” said a former 777 pilot, referring to the undercarriage. “That’s the procedure.”
The plane was lining up for a “category one” landing in good weather, being guided in by two radio beams, one horizontal, one vertical. They were drawing the plane down a three-degree glidepath onto the southern runway, known as 27L, at Heathrow. Everything appeared normal.
The Boeing 777 is one of the most advanced passenger jets in the world, crammed with highly sophisticated electronics. Over 12 years it had established a remarkable safety record - more than 600 of the planes had gone into service and not one had crashed.
To Coward, a 41-year-old career pilot with BA, it was a routine flight and the plane was taking the strain. The autopilot and autothrottle were engaged and making the necessary adjustments.
You can see the throttles moving themselves. It’s as if they have a ghostly hand on them,” another retired 777 pilot said. “The gear is down and the flaps are down. In most cases you’d see the runway at this point. The aircraft would be holding a speed or even slowing slightly.”
As the plane approached, Coward, according to former pilots, would have announced: “1,000 radio. Man land 200.” This meant the plane was at 1,000ft, in its final approach, and that Coward was going to switch from autopilot to a manual landing at 200ft.
Once such a procedure was set, the plane would continue under automatic control until it reached an altitude of 250ft. Then a female computer voice would say, “Decide.”
That’s decision time,” said a former 777 pilot. “The co-pilot would take the autopilot out. He’d say, ‘Man land 200, I have control’.”
At two miles out and 600ft up, the plane was “established on late finals” - it was less than a minute from landing. The crew and passengers must have thought that they would soon be inside the terminal.
As Coward stared at the controls, the autothrottle demanded more thrust. It was a normal procedure, a small adjustment intended to keep the plane at the correct speed and height. Nothing happened. The computer system again ordered more thrust. Again, no response.
In the central control displays of the 777 is a screen that shows the engine pressure rate (EPR). One indicator shows the “command EPR”, the level ordered by the pilot; another indicates the actual EPR. “The pilots would have noticed that the achieved EPR was drifting back,” said one former 777 flier. “That would have been one of their first indicators.”
Coward or Burkill - it is not clear which - tried to increase thrust manually. Still nothing happened. [ Hermit : Almost certainly the Captain, Peter Burkill, as normal cockpit protocol is that the pilot in control's task is to maintain heading, attitude and spatial awareness, and it is the person who is not flying that watches the MDIs and manages airspeed, throttles, flaps, gear and other checklist tasks (which the 777 helpfully steps one through). The non piloting officer will also call distance and height checks, and monitor fate of descent. ]
At high altitudes, planes that lose power can glide for distances of up to 100 miles, according to Boeing, helped by starting at cruising speeds of more than 600mph. At less than 1,000ft and at much slower speeds, they can drop like a stone. [ Hermit : This is somewhat of an exaggeration I have a friend, a senior ATP (for a different airline) who was fired for gliding a 747 home to a perfect dead-stick landing from 11 miles out, "just because" (After watching a shuttle landing he felt that "if that brick can glide", he had to prove that he could do it). He was only caught because a random check of the flight data recorder happened to intersect with his experiment. What was really funny was that the real consequence for him was a significant and immediate increase in pay.]
Below 600ft the “ground proximity” warning is inhibited so that it does not distract pilots from an ordinary landing. But other signs and sounds would have rapidly filled the cockpit.
Through his clenched hand, Coward would have felt the control stick judder. “If a stall is imminent, the aircraft analyses its airspeed and gives a preliminary shake to the yoke,” said one former pilot. “It won’t let you miss it. We call it the ‘stick shaker’. An ‘airspeed low’ warning will also flash up on screen.”
Coward had only seconds to respond, only seconds before the plane would hit the ground.
Just 100ft or so below, Pym Reehal, whose house lies in the Heathrow flightpath, had gone outside to his car. He had an engine problem of his own and was trying to jumpstart the vehicle.
As he tinkered, BA038 hurtled overhead. “I had a quick glimpse and saw it was coming in at a very weird angle,” recalled Reehal. “It was sat at such a strange angle.”
A short distance away, John Rowland was driving his taxi just outside the airport perimeter fence. “It looked as though it was just missing the roof of my cab,” he said. “So low you would think you could lean out of the window and touch it.”
The plane skimmed the road and missed the perimeter fence by a few feet. Further down the Heathrow airfield another BA flight was preparing to take off. One of the air crew watched as BA038 hurtled towards the ground.
“I thought: oh my God, something’s terribly wrong,” said the witness, who asked not to be named. “The angle was all wrong. It just looked like it would be a disaster.”
Inside the cockpit of BA038 Coward and Burkill had no time to issue a Mayday, no time to warn passengers to brace. In the few remaining seconds they just fought to keep the plane flying.
The failure of the engines had cut the main power. The 777 does not have cables connecting wing flaps and rudder to the pilots’ controls. It is all done by sending electronic signals. However, the plane has several back-up batteries that enable the instruments to work until the emergency power units kick in.
“If they had done nothing, the autopilot would have tried to fly the glide path,” said a former pilot. The plane would probably then have stalled and crashed. “So they have to lower the nose to maintain speed, then lift it just before hitting the ground.”
Coward, who lives in Valbonne in the south of France with his wife, said yesterday that he had thought it was “the end” for him - that the plane would land with an “almighty crash”. He added that “some thanks has to go to the man upstairs for giving us that little lift at the end”.
But he and Burkill, said one former pilot, did “a brilliant job”. If the angle is right, a plane gets the benefit of “ground effect” - the wings in effect trap a cushion of air underneath them that softens the landing.
The Boeing 777 has the biggest landing gear of any craft, with six wheels instead of four on each assembly. As the plane slammed into the grass before the start of the runway, the landing gear on the left side smashed upwards through the wing; the right-side landing gear was torn off.
“It wasn’t just one thud,” said Coward yesterday. “It was a series of thuds.”
The plane ploughed great gouges in the earth as it skidded hundreds of metres, skewed right and came to rest on the edge of the runway.
“I felt like I was in a washing machine,” said Jason Johnson, one of the passengers. “The wings were making cracking sounds. You think of your family and your loved ones.”
Others barely noticed that the plane had crash landed until oxygen masks fell from the overhead storage lockers.
The passengers fled down the emergency chutes that had been triggered when the crew opened the doors. Amazingly, nobody was killed and injuries were relatively minor.
How did one of the safest aircraft ever made come to crash-land, narrowly avoiding disaster, and why did both engines apparently shut down at the same time?
THE 777 is the first plane to be designed entirely on computers, with input from pilots from the outset. It is a long-range fuel-efficient workhorse. Since it started commercial operations in 1995, 777s have made more than 2m flights. Boeing claims that operators enjoy a “99% dispatch reliability rate”. In other words, 777s rarely go wrong.
According to one database, there have been only a handful of recent 777 incidents – and two of those were hijackings. The only fatality had been that of an airport worker who was burnt to death when a refuelling operation went wrong in 2001 in Denver, Colorado.
The pilots got much of what they asked for in designing the plane, including a highly automated “glass cockpit” that does away with all analogue gauges and presents flight information on bright, software-driven LCD displays. Integrated into the software is EICAS (engine indication and crew alerting system) which monitors the two engines and alerts the crew to any abnormalities.
Nancy Novaes, an American pilot who flew 777s until she retired last year, said: “This is a great plane to fly. It’s highly computerised, highly logical. It knows what it needs . . . and EICAS tells you what it wants.”
However, the 777 and its computers are not infallible - as investigators knew before last week’s crash.
Early in the evening of August 1, 2005, a Malaysia Airlines 777, en route from Perth, Australia, to Kuala Lumpur was climbing past 38,000ft towards its cruising level when the flight crew were confronted with what the official report on the incident described as “a situation that had previously been considered not possible”.
On the EICAS screen a “low speed” warning appeared, suggesting the plane was approaching the speed at which it would stall. But almost simultaneously, the primary flight display screen warned exactly the opposite: that the aircraft was approaching the overspeed limit, the maximum speed at which it is designed to operate.
Too slow or too fast? Before the crew could decide, the autopilot pitched up the nose and the 777 climbed for 3,000ft, while the air speed dropped from 270 knots to 158 knots - at which point the stall-warning horn correctly sounded and the stick-shakers activated.
The pilot prevented disaster by disconnecting the autopilot and pushing the nose down. But then the autothrottle kicked in, commanding more thrust from the engines. The nose pitched up again and, of its own volition, the aircraft climbed another 2,000ft until it was brought under control.
The plane landed safely but, as the Australian Transport Safety Bureau report put it, the combination of a failed sensor and “a software anomaly” had created an “unexpected situation that had not been foreseen” and for which the crew had not been trained.
The sensor was a tiny “accelerometer” - a device used to measure the plane’s acceleration that is similar to those used in cars to deploy airbags.
Airlines were subsequently warned that they had to modify the plane’s software. Yesterday BA said it had received the directive and implemented it immediately.
Could a similar electronic or computer fault have occurred on flight BA038? Were the pilots not warned of engine failure or did a system erroneously shut down the engines?
One former 777 pilot believes that the accidental mechanical failure of both engines at the same time is unthinkable.
“There are separate autothrottles, a left computer and a right computer . . . everything is split. That is the philosophy of the plane,” he said. “For [both engines to fail] at the same time it has got to have been commanded. We are all aghast.”
In other words, it may have been an error in the computerised engine control systems. The Air Accidents Investigation Branch (AAIB) is examining all possibilities and downloading full data from the flight recorders.
Experts noted that its initial report said it would be “examining the range of aircraft systems that could influence engine operation”.
As investigators picked over the crash site, other hypotheses were being aired. One was that a “bird strike” had shut down both engines. The impact of large birds such as Canada geese hitting the fans inside the jet engine can cause engine damage; but no witnesses have recalled seeing any flocks of geese or other birds in the vicinity.
Authorities also monitor flocks by radar and said there were none on Thursday.
“The possibility of geese is remote at that height,” said one pilot. “And though they bend the blades, the engine keeps producing power.”
Pictures of the crashed plane also suggest that this was not the cause. The blades in the plane’s starboard engine were undamaged.
Turbulence is another possibility, particularly at busy airports. Sudden wind shear or even the wake left by powerful jets can damage the surfaces of planes, sending them out of control. But again there seem to be no signs that such forces were at work.
A fuel problem was another hypothesis. In the immediate aftermath of the crash, internet forums populated by pilots were rife with speculation that BA038 had run out of fuel.
Some airlines, although not BA, are known to run their planes with as little fuel as possible to reduce costs: heavier planes use more fuel. However, this policy can be problematic when planes face unexpected headwinds or delays in landing. BA had got its sums wrong, went the theory.
It holds little water, however. “It is well known that if all the engines snuff it, fuel is the most common cause,” said David Learmount, operations and safety editor at Flight International magazine.
“But fuel exhaustion is not the issue here because the AAIB report says fuel was spilt all over the [crash site]. Luckily it didn’t catch fire.”
Learmount suggested, however, that water could have got into the fuel, frozen at high altitude, thawed as the plane came into land and caused a slush in the tanks. This may have blocked fuel to both engines. Reports yesterday suggested that BA ground staff were warned to check the fuel mix in all its 777s.
Fuel starvation in certain tanks is another possible explanation, because large jets like the 777 have multiple tanks in both wings. Some aircraft have additional tanks in the belly and even the tail.
Transferring fuel between the tanks during the flight to maintain the aircraft’s equilibrium is a routine process, controlled by sensors, pumps and valves. Like all technologies, it is not infallible.
On February 7, 2005, a Virgin Atlantic Airbus 340, flying from Hong Kong to Heathrow, was passing through Dutch airspace when, without warning, one of its four engines - the outer engine on the port wing - went dead.
The crew quickly established from the Airbus’s sophisticated displays that the amount of fuel contained in the inner tank, from which the engine was feeding, registered as “0”. What they did not realise was that the automatic transfer system between the tanks had failed.
The outer engine on the right wing also began to lose power and the warning display showed that its tank contained zero fuel. The captain immediately realised there was a “fuel management problem” and opened the valves between the tanks to begin a manual transfer. It worked, but the crew still declared a Mayday and diverted to Amsterdam.
Some experts are sceptical that such a problem affected BA038. One former pilot on the 777 pointed out that during landing, fuel is going directly from tank to engine - there may be no transferring from one tank to another - and it is unlikely that both engines would suffer such a problem at the same time.
THE wreckage of the plane will be removed from the runway tomorrow and the airport will start getting back to normal. But crash investigators will not produce a definitive report into the cause of the accident for months. Until then, suspicion is likely to remain with the avionics.
So far other 777s have been allowed to continue flying, but the implications are serious. A senior airline industry source said: “I have heard that BA are going to have to check every single one of their 777s. They are not grounding them, but they will be checking every one because the AAIB has identified that it seems to be something connected with electrics and avionics coming from the flight deck to the engines, because the engines seem to be okay.
“They will go through the records of every single 777 flight looking for similar issues. They will go through all the engineering logs to find out if have had any similar problems between the flight deck and the engines.
“Each 777 will be brought in one by one for a maintenance check to look at whether there are any untoward signs. They will have to do that out of good practice as much as anything else.”
Whatever the cause of the accident, it was a remarkable piece of flying by Coward and Burkill and an extraordinary escape for their passengers.
The pilots, hailed as heroes, have been modest about their achievement. But Novaes, before she retired, had experienced the reality of crisis in the cockpit.
“These pilots have practised flying with limited controls,” she said. “They wouldn’t panic. Bravery is saying, ‘I’m afraid, okay, so deal with it’. Most pilots really only have to think about themselves. They are in the front. They are first on the scene. If they take care of themselves, everybody behind them is safe.
“But afterwards you feel a tremendous sense of relief. You are on the ground and you're almost giddy. The entire glut of emotions is probable.
“The pilot is a human being who may be better practised and controlled than others, but they’re still human.”
Why did both engines fail? The competing theories
Bird strike: A flock would have to have struck and stopped both engines. But there is no evidence this happened. Verdict: implausible
Pilot error: A mistake by those flying the plane must always be considered by air traffic investigators. But it is unlikely that BA would have allowed the pilots to be hailed as heroes if there were any suspicion that they had made a mistake. Verdict: highly unlikely
Running out of fuel: Last week pilots’ internet forums buzzed with speculation that the plane had run out of fuel. But the interim accident report said that “a significant amount of fuel leaked from the aircraft”, suggesting the plane was not short of fuel. Verdict: very unlikely
Fuel contamination: Reports yesterday suggested that investigators were focusing on the possibility that the plane’s kerosene fuel had been contaminated with water. A poor fuel mix has, however, never been held responsible for an air crash and one would expect other planes that were refuelled in Beijing to have been affected as well. No such problems have been reported. Verdict: unlikely
Fuel management problem: Fuel is transferred as required between various tanks in the wings and centre of large planes. In one incident, two engines on an Airbus 340 failed, even though the plane had plenty of fuel, because the transfer system malfunctioned. But it is unlikely to happen to both engines at the same time. Verdict: possible, but unlikely
Computer glitch:This happened with a Malaysian Airlines 777 and a former 777 captain told The Sunday Times that for both engines to fail at the same time “it has got to be commanded” - ie, it was computer error in controlling the engines. Verdict: possible and many experts’ prime concern
Boeing 777: how it measures up
There are about 670 Boeing 777s in service around the world. Last Thursday’s crash was the first involving the plane, although in Denver, Colorado, in 2001 a member of ground staff was killed while refuelling a 777 when the fuel hose malfunctioned.
It was the first plane to be digitally designed using 3D computer graphics. Its flight-control system also differs from earlier Boeings in using electronics to control the wing flaps and rudder, rather than mechanical systems such as cables.
The 209ft-long plane can carry between 305 and 440 passengers at a cruising speed of 615mph with a range of up to 8,300 miles.
When it came into service in June 1995 it became the first twin-engined plane permitted to fly three hours from the nearest runway. Until then, no twin-engined plane could fly more than 60 minutes from a landing spot. Airlines liked its reduced fuel consumption compared with four-engine planes. Double-engine failures are also rare.
The Boeing 777s have, however, been involved in at least 12 incidents where electrical systems have overheated during or before flights, four of which resulted in “major damage” to the power panels involved in controlling the plane, according to Britain’s Air Accidents Investigations Branch. Single engine failure or forced shutdown resulting in an early landing has occurred on 11 known occasions.
With or without religion, you would have good people doing good things and evil people doing evil things. But for good people to do evil things, that takes religion. - Steven Weinberg, 1999
Author
Logged
