Thursday, January 17, 2008

The BA038 incident at Heathrow

Like many people I watched the early reporting of the crash landing at Heathrow of a BA Boeing 777 in amazement.

It was also a great example of how technology has changed news. Most of the reports and pictures in the first phase of the incident came from members of the public to the BBC's News 24. The passengers had mobile phones and weren't going to be stopped by being corralled away from the traditional press.

It was slightly frustrating that the BBC presenters didn't seem to ask the right sort of questions of the passengers. ( If the plane was banking at 45' just before landing - why didn't the passengers notice. Due to force distribution they may have felt level, but looking out of window would have told another story ).

From their reports no orders to brace for a crash landing seem to have been given. The crew were perhaps too occupied the reported emergency of dual engine failure, but it must have been close to ground to not allow any time for a warning.

Now double engine failure is almost unheard of - except for fuel problems. Reports say the aircraft had sufficient fuel in its tanks after landing. ( Update an explanation for possible fuel contamination is provided here ).

Either way this is very significant as a lot of trans-Atlantic air travel relies on the low likelihood of dual engine failure.

We don't yet know how far out the engine failures happened. But one thing is sure official sources of information are being very quiet right now.

Terrorism has apparently been ruled out - but something doesn't fit right now. Perhaps fuel contamination which was only evident as the tanks emptied ? The electronic on the aircraft have to be a prime suspect with some sort of common mode failure being to blame. The proximity to the ground might give a clue - maybe the action of lowering the undercarriage caused a wider system failure or some other form of electromagnetic interference - the mostly metal body should protect the aircraft.

Still there is the near miraculous safety of the passengers and crew, and those on the ground, to be grateful for.

Still odd they allowed Brown to fly to China in what I assume is another BA 777.

PS It is worth speculating as to why all BA 777's and other airline 777s haven't been grounded given the reported total electrical failure of the aircraft. Perhaps something is known about the cause after all. This is the part of the story that is making the least sense to me right now.

Update the following is being reported on a number of sites including the Telegraph:

    One airport worker who spoke to the as yet unnamed pilot said: "He told me the aircraft shut down and lost power.

    "He glided it across and managed to get the nose up. It happened very close to coming in to land, probably over Hounslow.

    "Everything was normal and there was no warning or anything and suddenly 'boom’, the power’s gone.

    "This man deserves a medal as big as a frying pan. He’s done a fantastic job, he really has."

Also remember only a few days ago the incident with an Australian Boeing 747 losing power near Bangkok. Qantas blame a leak of water, but Boeing don't go along with that. Again all engines were unable to deliver generated power ( 4 of them ) though it must be assumed it happened at a much higher altitude as the aircraft flew on for 15 mins on batteries.

Again we have some form of common mode failure, though the distance from the ground seems to rule out ground based interference.

First indication from the Air Accident Investigation branch is expected towards the end of Sat. If it was a bird strike then the evidence will be obvious with the engines - but equally if they want to blame birds they won't be able to without the evidence as the engines look to have survived at least in part.

It is perhaps interesting to speculate if the aircraft actually did better landing of soft turf than on tarmac ? Obviously if the undercarriage had held tarmac would be better. But perhaps there's a lesson for some emergency landings here.

Update: The BBC have a page on speculation here. It suggests a range of options, which now include Attack and wind shear. It remains an oddity of this event that the aircraft was flying the reserve route that the British Prime minister was about to take on a similar model aircraft I assume. ( Tony Blair always had a BA 777 ).

I don't see how wind shear would shut both engines and lead to power failure - but I'm not an aircraft engineer so maybe I'm missing something here.

Thinking about what was reported perhaps the aircraft angle at 45' was down to the start of stalling on one of the wings ( meaning loss of lift from that wing and a subsequent rotation of the aircraft). The aircraft must have been at the very limits of its glide path. In which case we have been very lucky that the plane was not full as it probably would not have made the field of Heathrow airport.

Witnesses report high engine noise - I have no idea is windmilling engines could generate that or if some power was still available.

No one reports smoke from the engines or such a trail. ( Would that be an expected indication of a bird strike ? Or just a fire. ) Also you would expect some damage to the front of the engine which given the aircraft's location would be hard to hide from the media.)

Also does the lack of a fire indicate the engines were shut down on landing ?

What ever the truth it will be an episode on Discovery channel before you know it.

Mike Swain of the Mirror suggest early deployment of reverse thrust ( which explains the engine noise ). If that's the case I'm amazed the aircraft managed a controlled descent at all. Not sure how that relates to the loss of power.

Someone may start to look at any software involved in controlling the aircraft. Opening up the possibility of sabotage - but this must be very unlikely.

Update: BA have just had a press conference which presented the crew but gave no details of the technical side of the incident, citing the ongoing crash enquiry. I have no idea if that's normal procedure, but it seems odd. I guess BA have to be careful due to potential legal implications and the potential for legal action. However the staff and management at BA have given a very strong sign of support for the cabin crew.

Pictures of the engines are now available over at Flight. The interpretation placed on them is that one engine had no power on impact. The other is missing a blade, but no support for bird impact is given.

Update AAIB have reported (interim report), saying the aircraft suffered engine failure with a loss of power 2 miles out when the pilot attempted to power up. The power was being controlled by the auto throttle. The crew attempted to increase power manually didn't respond. Still no aircraft grounded.... ( This confirms the information provided in the comments by an anon source ).

Further there are some interesting points being made in the comments which are worth a look (a number of anon posting there ! )

20/01/08 Anon 10.09 has made a detailed description of might have happened. If its correct then I suspect heads may roll. Maybe he's right, maybe he's not ( that's the wonder of the internet you never quite know), but the description of fuel practices may raise some questions. It also might explains why there have been no precautionary groundings for checks.

24/01/08 The accident investigation now seems to be focusing on the fuel supply. Reports repeat that their was adequate fuel on the aircraft when it 'landed'. See this report on fuel incidents by David Millward in the DT.

15/02/08 Heathrow crash theories are revealed apparently based on a leak to a US pilots web site. In essence its common mode failure of either engines due to - fuel issues, hardware or software. The software issue is causing most concern due to its wide implications. However I still think due to proximity to the ground at the time of failure that electromagnetic interference should be considered, maybe from a hostile source - it will be interesting to see if it is specifically ruled out. ( Perhaps its something that would never be admitted to anyway ).

Also see more details for this discussion on the Pilots of America web site, which seems to have quite a bit of detailed knowledge - including a picture from inside the aircraft.

Update 13May08 Attention continues officially to focus on the fuel system. There is concern the unusually low temperatures over Russia may have lead to the fuel viscosity increasing to a level that caused problems in supply the engines at a greater rate during landing. See the Daily Telegraph here. It seems strange that this problem hasn't been reported before, though that doesn't rule it out. However given the survival of the aircraft there should have been all the evidence required to discover any pilot error and that hasn't emerged as the case, I assume clearing the crew.

Update 4Sep08 The official interim air investigation report is in and ice crystals are being blamed. Apparently the same result has been reproduced by Rolls Royce. I'll be looking over the report over the next day or so ....


The Daily Pundit said...

HERF. Well spotted. Could well be. Let's hope the boffs are working out how to neutralise it.

Anonymous said...

If there was enough fuel on board, ask yourself why there was no fire. The wings are fuel tanks, and these were extensively damaged.

Also, why was the plane reported to be banking at 45 degrees, when it should have been on final approach for some time?

Man in a Shed said...

The 45' could either be due to a sudden loss of control as the power system failed before the pilot regained some control ( no idea how the backup system might work, but I assume a 777 is a fly by wire aircraft ).

Or perhaps an attempt to lose height - but that doesn't seem to have been the problem.

The good news is that with the near complete survival of the aircraft and all the crew there will be excellent evidence for the aircrash people to work with.

You just have to wonder why BA 777's are still flying right now. How can they be so confident ?

Unknown said...

Could electromagnetic interference (EMI) from security countermeasures protecting the PMs plane have caused the loss of power to the BA 777?

Elaine Scarry writes this on EMI:

Man in a Shed said...

In this case no - as the PM was in his car and close to being run over by the plane when in came in.

I suspect the EM issue is more of a concern for next generation aircraft such as the 7E7 - given their lack of a Faraday cage. ( Perhaps I'm wrong on carbon fibre and it does provide that protection or a thin metal layer is added.)

After all aircraft a regularly stuck by lightning and mostly survive.

anthonynorth said...

Your mention of how the news comes now is maybe not quite right. It seems to me that 24 hour news deals first in gossip, rather than actual news.
Hopefully, as information comes in, it turns into news, but I something think the gossip actually leads the agenda, and changes how the news is finally interpreted.

Anonymous said...

There are many oddities in this event. A sudden 45 degree bank on final would be felt by one in the cabin. Yet only an eyewitness on the ground reported this sudden banking. No passengers mentioned the banking incident. This is odd. If it did take place, then it could be a sudden loss of control due to electrical and/or power loss, or the loss of power of only 1 engine. Perhaps there was a bird strike on the port engine(#1 engine)? It is interesting to note that the plane has full right rudder applied to it as it lies on the ground....and the eyetwitness reported a suddden left banking event. Also, one passenger reports the engines making a lot of noise as it came into land which, he said, sounded more like when you take off. Given these statements, and what we know frmo the news, my guess is there was failure on 1 engine and that explains the banking, the passenger hearing an engine at full throttle (to compensate for the loss of power on 1 side) and the application of full right rudder. A bird strike could be possible.

Man in a Shed said...

Anthony, I was frustrated by the questions the interviewers were asking. Very few seemed designed to establish technical facts. Many of which were the how do your feel type, rather than what happened.(Though they were lucky to get a qualified pilot who watch the incident).

They also almost lead people into saying things - as with one passenger who eventually gave the soundbite they wanted about feeling like winning the lottery, but only after a lot of hints.

I guess my point is how the initial reporting relied almost exclusively on members of the public communicating with the BBC, rather than its reporters. Bypassing the airline and airport in almost real time.

I think that capacity will fundamentally change the way events are now recorded and reported.

BBC journos were there - just all with Gordon Brown and somewhat tied up. I think Nick Robinson managed to phone in from the plane at one point.

Man in a Shed said...

Anon - you would think that sudden single engine failure would be a design case for the aircraft. I realise power surges are hard to deal with, but some sort of emergency system obviously allowed. I think a 777 would be fly by wire.

If it was as a result of a single bird strike then it has exposed a potential serious design flaw ( it is always possible that maintenance mistakes prevent the aircraft's systems from responding to such an event as designed ). Again grounding of BA 777's would perhaps be indicated.

Its an interesting technical mystery - which thanks to the safety of the passengers and crew we can speculate openly about.

It will be interesting to see what the AAIB say tomorrow.

Anonymous said...

This is one of the more interesting pages of opinions I have read on this incident. There are a number of contradictions in the generally-accepted version of the event. The private pilot who watched the landing and was interviewed by the BBC said that the plane was approaching the runway from a southerly direction, which is most unusual given the runway's east-west orientation. I find it hard to reconcile the idea of a power failure with the reports from people both in and ouside the aircraft of a high level of engine noise at the time of the landing. Also, if there was a complete power and avionics failure, it's a bit of a surprise that the lights in the cabin stayed on, but I have not heard any of the passengers say that they didn't. One aviation expert this morning, looking at the pictures of the aircraft, noted that the flaps were down but the airbrakes were not deployed. Unfortunately all of these things lead me to think there may have been some sort of pilot error, or possibly a fuel emergency -- google "Gimli glider" for an amazing true story about such an incident!

Anonymous said...

don't ask me how i know, but i know.
both engines were working and responding during the incident.
as for gliding - look at the garuda flight that landed in the azores.
total fuel loss due to poor maintenance and mod incorporation. to be fair the press speculation and news reporters making quite rediculous statements beggars belief. compare the coverage now without a real clue about the cause to how much time will be given to the full report when it comes out.

Man in a Shed said...

Anon 4.24 - yes I've seen a TV film or docudrama of that incident.

Perhaps the BA 777 should have been on a glide path into Heathrow. I'm guessing here - but I would have thought you'd want aircraft just above the glide path coming in. Of course a systems disruption could lead to a loss of altitude, especially is glass cockpit instumentation is lost.

However perhaps the widely reported quote on loss of power just refers to engine power not electrical systems. I have to admit I've seen it the other way round.

Anon 4.32 - if I remember rightly the Azores problem was also a case of false hypothesis syndrome. A misdiagnosis of the problem leading to fuel transfer into a leaking tank (?). Still a long glide into the only possible airfield saved the day.

I've just seen the BA press conference which has presented the cabin crew who made a dignified statement ( it may be available here - as long as the link lasts ), but were giving no information on the nature of the emergency - claiming this was privileged information during an investigation. ( Is that really the case for an acidnet inquiry ?\ I understand when Juries are potentially involved in the future.).

I would also say that BA making such a public presentation of the crew with all their staff is a very strong sign of their support for their people.

Man in a Shed said...

Anon - I should add the cabin lights is a good point. Its the sort of question the BBC should have been asking the passengers they managed to interview.

It could be we are misinterpreting what we have heard. The reports of engine noise from observers supports your assertion that the engines were operating.

The only reason for not giving full disclosure of the incident, that I can think of that is legitimate, would be security. And that would have to be very worrying, and frankly almost impossible to keep quiet anyway.

No doubt my imagination is over clocking on the reported evidence. I'm looking forward to an explaination.

Anonymous said...

I think there's a big difference between what the AAIB are saying, and what's appearing in the media. It's all down to words really, but "failure to respond to an increase in power" does not mean "both engines have lost power". They could still be running normally. Just not increasing when asked. Normal approaches include many changes of power to keep the aircraft on its correct slope, at such a late point in the landing an inability to control thrust would be catastrophic.

Oh and FWIW, it appears *to me* from the state of the engines that they were powered when they hit dirt. There seems to be a lot of soil and mud which hit the fuse aft of the engines, which implies they were running under some thrust (albeit briefly). A fan just spooling with no power would stop pretty damn fast.

Man in a Shed said...

Anon 9.09: Yes I saw the mud on parts of the side of the aircraft. When the first pictures came out on TV it made me wonder about fire, but on closer inspection it was clearly mud ( no pun intended ).

Assuming your right I would say there is a very serrious problem. Again its odd that no grounding of similar types has occured.

As mentioned earlier witnesses report engine noise, infact greater than usual - though perhaps unusually low altitude contributed to that.

I guess the next bit of the mystery will be solved when precautionary order to check certain system are sent out.

Anonymous said...

As the world's most instant aviation expert (I have opinions on everything even when knowing absolutely nothing) I think the aircraft, when established on a normal ILS approach to 27L, with gear and flap down, had an engine flame out. The most likely cause of that is a lack of fuel to that engine and that happens when there is no fuel in the tanks on that side and the cross-feed valves are closed (which they normally are in that phase of flight).

The normal approach configuration is a high drag situation where you are on the back of the drag curve and requires thrust to maintain the 3 degree approach gradient. When making a deliberate single engine approach you use a much reduced flap setting (and a higher approach speed) because you do not have the thrust to play with. You also delay landing gear extension until the landing is assured - to reduce the drag and thrust required. If you are already in a normal two engine approach and have selected landing flap (normally when about 5 miles out) then if you have an engine failure you can be deep in the s-h-one-t with insufficient thrust available to stop you sinking below the glide slope - (hence the landing 2000 feet short of the threshold?) And you are facing directional problems as you balance thrust against the rudder effectiveness.

Not a nice situation - even when (as in this case) visual with the horizon and the runway.

So why run out of fuel on one side? Beijing direct is pushing it on a 777. When you arrive in UK airspace you are probably on minimum legal fuel (alternate Gatwick plus 30 minutes) - so maybe just 45 minutes planned endurance before everything goes very quiet. Of course you always have the option of refuelling enroute if below the legal minimum fuel - Paris, Brussels, Copenhagen, Amsterdam ... lots of alternates available but, unless you know of expected delays or have other operational reasons, your employer expects you to continue if the expected fuel remaining meets the requirements.

Now comes the tricky bit. You are very light and very high (41,000 ft or so) and start the descent when overhead Amsterdam. But it is a busy time of day and a queue is building so there are additional track miles as you are radar vectored across East Anglia towards south London for the 27L ILS intercept. A couple of four minute orbits in a holding pattern (maybe Clacton VOR) is not unusual and eats into the reserves big time. You, of course are flying the aircraft in the most fuel-efficient manner possible - playing at being a great big glider.

When your endurance is down to thirty minutes fuel remaining you are meant to declare a fuel emergency - but no one ever does. Well, not at that point because it happens all the time, and you can't run an ATC system with people routinely demanding priority. And there is always the argument that you can easily land at Stanstead so there was no problem anyway.

By now the remaining fuel is just sloshing around in the bottom of the two inboard main tanks. But what if you have not carefully (by cross-feeding earlier on) equalised the quantities in the two tanks? Answer - all your fuel ends up on one side only. This is one of the oldest traps in aviation - God only knows how many it has killed over the years.

And, unless the fuel crossfeed valves are open (allowing either tank to feed either engine) the engine fed by the empty tank will stop.

Looking at the actual touchdown position it would be reasonable to suggest that the engine failure occurred at about 1,500 feet and three miles from the threshold. They had no hope of making the runway and did well (or were lucky) to get it on the grass.

Amazingly, the emergency evacuation shutes seem to have all extended. They normally have a very high (25% expected) failure rate. The minor injuries were probably sustained during the evacuation scramble - heat burns on the bum are common from the friction of the slides. Ladies are recommended to stuff the emergency instruction card down their knickers but the airlines never tell you that.

That there was no fire is no surprise - there was virtually no fuel left.

The main gears seem to have been torn-off but the nose gear has remained extended.

Well ... that is my theory. Any one got a better one? jim

Anonymous said...

woops ... I should have said " at 900 feet and three miles from the runway " ... not 1,500 feet" If the engine had failed at 1,500 feet the aircraft would have been five miles from touchdown and they would have ended up somewhere in Hounslow - or, if they had been really on the ball, - they could have slammed the flaps up to an intermediate setting (lower drag) and maybe made it to the threashold on the available asymetric thrust.

Anonymous said...

The Boeing 777 is a fly-by-wire aircraft, that is with no direct physical connection between the pilots controls and the actuators that actually control the flight surfaces etc. I do believe the first 777 still had rudimentary cable links between the controls and some flight surfaces, but whether they would still have been included on 777s built in 2002 (as this one was) I don't know.

This is my theory about what happened - part of the fly-by-wire control system includes a device that aims to act against the swing in direction caused by single engine failure, by moving the rudder over to counterbalance the loss of thrust on one side. The rudder being hard over could indicate that - one engine failed (whether via fuel starvation, mechanical failure, FOD etc..), the rudder swung over to counterbalance the asymmetric thrust, the remaining engine ran up to full thrust to compensate (the aircraft would have been in high-drag configuration with gear and flaps all extended, hence needing power) but the time taken to run the remaining engine up to full thrust causes a loss in height. This is critical because the plane is already so close to the ground, so the handling pilot compensates by pulling up to attempt to gain more height, which puts the plane in an increasingly high nose-up position at low speed and full drag, which progresses into a stall, bringing the plane down short of the runway. What do you think?

This would explain the rudder position, the engine noise heard by witnesses, and satisfies some of the rumour about fuel starvation or bird strike, which would more likely affect just one engine at first.

Man in a Shed said...

Sounds reasonable to a layman.

The stalling part would fit the wild change in height and swing in direction. The stall must start on a certain part of the wing - leading to attitude changes.

Its strange there haven't been any pictures of the aircraft on approach. you would think that an airport like Heathrow would have cameras all its approaches. It wouldn't cost that much. I bet they have tonnes of CCTV in the terminals !

Still no public news of announcements for checks on 777s - the longer it takes the more human error looks like a factor.

Anonymous said...

Why don't you cut all the crap and wait for the AAIB to publish their findings. Anon, your'e an airline pilot are you then ?

Man in a Shed said...

Its human nature to speculate, and figuring things out is near hard wired into Engineers.

I think the days when people waited to be told what to think by authority ended some time ago - at least they did for most of us.

It is still strange that this type of aircraft continues to fly without some specific checks - given there is almost as much information as anyone could ever hope for from such an incident.

For another reason for showing a bit of doubt in official explanations I point you towards the French inquiry into the Concorde crash.

Official reports are not the truth - they are an analysis supported by available data made in the light of regularity and sometimes political constraints. The are no doubt helpful in the main and informative, but not perfect.

In my particular branch of engineering people are still discussing major incidents which happened decades ago.

There is also a key political issues concerning air traffic over southern England - especially if fuel turns out to be central to the issue.

Anonymous said...

One of the virtues of aviation is that the participants are always keen to share their experiences and discuss their operational problems. We all learn from each other's mistakes, errors and omisions. We also take all operational instructions with a pinch of the proverbial salt knowing that there is always a gap between the realistic practical and the un-realistic theoretical.

That said can I point out that the final position of a hydraulically controlled flight control surface (in this cast the rudder which is causing some interest) may, or may not, be related to the control input prior to failure? Hydraulic systems are funny beasts and where everything goes after the system pressures are removed can be anyone's guess. Furthermore, in the case of this aircraft, the systems may well have been damaged following the impact in which case one really cannot speculate as to why the rudder ended up where it is. jim

Anonymous said...

Fire (or explosion) would be just as likely if the fuel tanks were very low or empty, as fuel vapour is more flammable than actual fuel.

I'll be very surprised if this accident was not in some way atributable to pilot error.

Yes, I know it was on an ILS approach, but there are still other stages of the flight where pilot input is needed.


Anonymous said...

I am totally a layman. I only have one question. Why didn't the pilot retract the flaps a little bit to make it glide longer?

Man in a Shed said...

If the flaps are retracted then the stalling speed goes up. The aircraft would be in strong danger of dropping out of the sky in an uncontrolled manner. ( No ability to accelerate as they were out of height and out of power ).

Based on the witness reports it may be that the aircraft may have started to stall anyway.

First choice, to reduce drag, would have been to pull up the landing gear, but I doubt there was time for that.

Anonymous said...

Anon - In reply to Man in a Shed,
Quote:"You just have to wonder why BA 777's are still flying right now."
The answer to that one is simple really with the minumum cost of a 777 at $200,000,000 and a reported 357 on order and 650+ flying now worldwide, who's going to be the one to announce that maybe this aircraft isnt 100% percent yet and risk losing at least $78.5 billion.

Anonymous said...

Anon - I recall reading of an incident on the forums (I couldn't find the exact link to the report to post here, as the server was busy at the time I wrote this, sorry) of an incident in 2005 I think it was, when a BA A319 (also a "fly by wire" aircraft) had a total power loss for some 90 secs or so at 20,000+ feet. Power was restored and all was well, but the scary thing is the report concluded with some recommendations, but as yet the actual cause has not been pinpointed to date, pilot error was of course ruled out.
So why aren't all "fly by wire" aircraft grounded yet alone any of BA's aircraft ?
My personal opinion is that aircraft (in fact all modern machines and appliances) are becoming more and more technically advanced, presumably to make peoples lives easier and safer, but is this making the lives of the people who have to operate these items any easier ? more buttons to press, more procedures to remember whilst still having to remember the "good old fashioned way" just in case you have to fall back to good old fashioned manual mode. Alongside all this new technology comes more opportunity for things to go wrong, you just have to weigh that up against the benefits and decide for yourself.
Remember nothing at all is completely failsafe,just like your washing machine,TV set, or motor car things can and do go wrong (in all honesty you do get the tell tale signs of impending failure but most people can't, don't or won't act upon them), these devices aren't and probably can't be built to last forever. You just have to give them a bit of TLC and watch for and act upon signs of malfunction.
As an interesting aside some Boeings were made with full remote control flying capabilities inbuilt....Lufthansa had this stripped from their particular aircraft over concerns that "undesirables" could intercept and control the aircraft operating signals.

Anonymous said...

Anon - Here's the latest from the AAIB for anyone interested:-

Anonymous said...

A VERY intersting read here:

Anonymous said...

Please read - i think you will find it very interesting