If this is your first time here, why not consider subscribing to my RSS feed?
BA38 Heathrow Boeing 777 Crash: Some Thoughts And Pictures
Posted on January 17, 2008
Filed Under Personal Posts |
Watching the reports today of the Boeing 777 crash at Heathrow today, and my aviation bug kicks in and I’m trying to glean anything useful out of the BBC News 24 reports. I’m sorry but if all you have is the same few eye witnesses going ‘it looks scary’ and pumping a passenger on air with his mobile with the same ‘was it frightening?’ question four times was bad form. Facts please!
Anyway, being a glider pilot, I always joke that if I’m ever on an aircraft and there’s fish or steak on the menu (I had the Lasagna) then all I need to do is kill the engines and my gliding training kicks in. But to have a powered aircraft loose power and become a glider at 500 feet, roughly 45 seconds from landing… not nice. Not nice at all.
I’ve run some trigonometry, maths, and flightplan info to get this idea of the ground path of the event…
Or to put a bit of perspective on it for the pilots and flyers here, this would be the cockpit view at 500 feet on a 3 degree approach (standard for Heathrow).
With a car park, Hatton Cross tube station, and the A30 between them and the edge of Heathrow, it was good training that saved a lot of people today. Congrats have to go to the flight crew - it sounds like they lofted the nose and traded speed for altitude, (Update, Sunday 20th: from all the comments it’s clear this nose pull was at the last minute - push the nose down initially for best glide slope and then hop over at the end)… lengthening what ground track they could in the hope to make the edge of Heathrow. I suspect if this had been higher up, say 1000 feet, they might not have had the energy to glide in to a hard but good landing.
Good landing… one you walk away from.
Excellent landing… one where the plane can take off again.
Comments
43 Responses to “BA38 Heathrow Boeing 777 Crash: Some Thoughts And Pictures”
Leave a Reply
Latest Flickr Pics
- Ewan's Guide To Surviving Long Haul
- Is The Podcast Network Real Enough For Nick Douglas?
- The Apple iPod Shuffle Shuffle from OpenTech 2005
- Top Ten Warm Up Tracks I Use Before Podcasting or Broadcasting
- Song for Ten - Phil Spector, Doctor Who and The Christmas Invasion
- November 2008
- October 2008
- September 2008
- August 2008
- July 2008
- June 2008
- May 2008
- April 2008
- March 2008
- February 2008
- January 2008
- December 2007
-
You Can Also Find Me At...
“I suspect if this had been higher up, say 1000 feet, they might not have had the energy to glide in to a hard but good landing.”
Really???? if they had been higher they would not have been able to get there?? Don’t understand this. If they had been higher they would have bene going faster as well, so ore energy from speed and hight?
Airliners maintain a constant speed on approaches - when you hear the engines throttling it’s to change the rate of descent, so at 1000ft they would have been at the same speed as at 500 ft. My guess is by lifting the nose, and a higher angle of attack they would have been able to reduce the rat of descent, and thus lengthen the distance they could glide.
But that bleeds speed. At one point, they would have been at stall speed, and could choose to put the nose down (more speed, less left) or fall out the sky.
I can’t run those numbers in my head, but if they barely made it over the fence while trading speed… someone got a Flight Simulator handy?
Should the “not” be elided?
I suspect if this had been higher up, say 1000 feet, they might have had the energy to glide in to a hard but good landing.
He ran out of fuel No fire from ripping his gear off. No Fuel looks like it spilled? Interesting!
Joe, no, I think that by lifting the nose, they traded speed for altitude. If they were further out, they would not have been able to continue that trade.
Geoffrey, I seriously doubt it’s a fuel issue, given that the flight landed 18 minutes early, at should have been carrying at least another hour of fuel on top of that in case of diversion to Gatiwkc, Manchester or even Prestwick.
It is one possibility which will need to eliminated. You are right one hour reserves are required but it is still a human input and human error is always a part in an incident like this! The Captain will be asked this question. The sooner the better!
Via Flyer Talk…
Were I faced with a total loss of thrust at 500ft, fully configured for landing, my first response would be to pitch down to maintain the speed at Vref and not stall the aircraft short of the runway. I would shortly after expect to be alerted by my colleagues screams that we were going to land very short and then reduce the speed back to the stick shaker if necessary to ensure I cleared any big hard objects in the distant undershoot (such as Hatton Cross tube station or the Jurys Inn hotel) before easing the speed back up if possible once assured of reaching something relatively flat. If the video on the BBC is of the aircraft in question it looks very much like what they were doing. The concept of Vmd doesn’t really work in a dirty airliner, nobody really knows what it is and it’s probably too late to get to it anyway.
If you doubt the shortage of fuel issue, read about the Air Canada 757 that ran out of fuel because they thought they had fuel that in fact had never been loaded. That crew also made a good landing as a glider.
BBC quote a passenger ‘The engines were roaring and then we landed and it was just banging’, which doesn’t equate with no power. Also, if all power was lost with the aircraft fully configured for landing (flaps and gear, as shown by the pix) the glide ratio would be terrible. I don’t think he would have got in from any distance more than a few hundred metres out. Sounds more like he got low and slow for some reason, and when power was reapplied it was too late - engines got spooled back?
Paul, fully aware of the Gimli Glider (Air Canada) and as I recall that was a problem betweenmetric and imperial. I’m still not buying an out of fuel situation, it would mean they were short by tonnes of weight, or deliberatly ignored options such as a landing in Frankfurt. I think we’re going to end up with some sort of loss of power issue, coupled with weather/windshear. Icing in clouds during the approach for example. In any case it’s not going to be one smoking gu, but lots of tiny holes that line up.
Richard, but that sound could easily equate to the Ram Air Turbine deploying. On engines spooling at max (as per the checklist) but not pushing out power.
The Ram Air Turbine is very small and (obviously) air driven. I can’t see that making a lot of noise. I guess a failed engine could be spooled up and not producing power, but both would not be in that state. The wind was low (25mph from the met off) so wind shear sounds unlikely as a major contribution. As you said, could be a lot of things adding up to get him behind the drag curve.
If he was flying on one engine, would not the fuel consumption be a lot higher?
How long had he been running on one engine? did that cause the fuel to run out?
Did he actually stall at 15 feet?
Anyway, it was fantastically good flying!
Sorry, only just seen this excellent discussion. Am a bit surprised not to see anyone suggesting hostile action as a cause, eg electromagnetic interference from a device on the ground under the flightpath. Or is that just paranoia on my part?
Sounds like the left engine failed causing engine yaw so the pilot counter acts that with right rudder(explains why rudder was full right on ground pics) so thats the angles of the aicraft explained but the rate of descent increased dramaticly causing the stall witch the pilot tried to pull out of and put the plane on the grass to reduce sparks from grazing tarmac igniting the fuel this is just my theory but i hope it explains the firelessness
David, that’s the paranoia talking.
Tez, again, all possible, all speculation. Right rudder could have been applied on the ground to try and effect some sort of direction. I doubt they would have deliberatly put down on grass rather than the tarmac for sparking reasons. Fire is more likley from fuel hitting hot engine parks than sparking.
Losing one engine at 2-300ft would be interesting. If the airspeed was allowed to decay while the engine out drill was executed and power was fed into the other engine, then descent rate would increase and there might not be time to get it all back before the ground arrived.
Gents, a single engine failure at 2-300 ft in a landing configured twin jet is still a non-event to a professional crew unless weight and balance, density altitide, or some other contributing factor is additive. (Other than perhaps some surface gusts, none appear to be a factor at LHR) The Vref speed margin has plenty of energy to complete an approach uneventfully from that altitude. Crews practice this in the sim frequently for bird strike and other emergency scenarios. The pilot flying on short final should have his hands on the levers, already poised for shears or go-around. At that point, with no time for a full engine out checklist, most operating procedures call for the PF to call out the emergency and advance the the levers without respect to the good engine, maintaining glide path and Vref while ruddering for directional control. Training dictates that a single engine go-around is non-option in virtually every case, so an assured landing is emphasized.
There is also a lot of discussion above that ignores the four-forces of flight in approach configuration (high drag) without power to trade. Stretching a glide with increased pitch is almost certainly a losing proposition, as induced drag will kill airspeed and bleed energy beyond the additional lift (reduced sink rate) gained. The pitch down mentioned above is the more likely input to maintain stall margin and best dirty glide distance.
The reports indicate both ground observers and passengers heard the engine(s) roaring on short final. Doesn’t sound like a complete (both) engine failure to me. How the crew managed the failure, whatever it was, and what caused the undershoot is going to come out of the FDR analysis very soon.
Bill ATP/CFII
David Learmont of Flight magazine suggests that the most likely cause is water contamination in the fuel. It would form ice crystals when flying at high altitude and on descent would revert to water.
Once hydralic power has gone the rudder will go where the wind blows it. The CAA says that a significant amount of fuel leaked out of the damaged wings after the heavy landing. For both engines not to respond to command is very odd (very).
If you slow an aircraft to near stall speed the drag increases and you will travel less distance. Gliders have a graph that demonstrates this very well called a polar curve. Google “glider performance airpeeds” for an animated demonstration. The point being if you want to travel further dont slow down. It will be interesting to know what happened here, I have only rarely been impressed by an aircrews reaction to an emergency, an as for being “Hero’s”, well as they say, give me a break.
Any body seen this fr doc 1816?
As a retired 747 flight engineer and instructor we were always pushing the icing problems on RR engines. The 747 did experience surging and hung engines when anti ice was not selected. This AD extract (there is more) basically deletes the use of auto.
Continuous descent leaves the engines at a low power setting was there cloud early in the approach?
Did one engine surge vz reports of noise, and the other hang because of system failure in the icing selections?
Read on it almost describes the event!!
[Federal Register: January 28, 2003 (Volume 68, Number 18)]
DEPARTMENT OF TRANSPORTATION
Federal Aviation Administration
14 CFR Part 39
[Docket No. 2002-NM-318-AD; Amendment 39-13027; AD 2003-03-03]RIN 2120-AA64
Airworthiness Directives; Boeing Model 777 Series Airplanes
Equipped With Rolls-Royce Model Trent 800 Series Engines
SUMMARY: This amendment adopts a new airworthiness directive (AD) that
is applicable to certain Boeing Model 777 series airplanes. This action
requires revising the Airplane Flight Manual to specify that the engine
anti-ice must be “on” during all ground and flight operations when
icing conditions exist or are anticipated. This action is necessary to
prevent ingestion of ice that could cause shutdown of both engines
during operation in icing conditions, and result in a forced landing of
the airplane.
Google “engine icing 777″ to get the rest.
Although it may appear that the crew have been presented with a sudden, and potentially catastrophic situation, it is rather early to be feting them as ‘hero’s’. The situation that they were trying to manage, how it developed, and over how long has not been made clear, indeed, only scant information has so far been provided. A pilot has two friends, height and airspeed, and this crew had neither. Low and ‘dirty’, and without power, the aircraft was doomed,and the only option was try to prevent a stall, keep the wings level, and avoid any robust obstacles. The focus here should not be upon the act of self preservation by the first officer, but on what events went unnoticed or were mis-managed either minutes, hours or days prior to the incident. Given immediate access to an (almost) intact aircraft, the data recorders, voice recorders, and the crew themselves, accident investigators should quickly solve the mystery. Single engine failures are not rare. Near simoultaneous double engine failures are almost unheard of, unless of course, an aircraft runs out of fuel. This does sometimes happen. The incidents which are not publicised, are those where the aircraft runs dry on the runway, or whilst taxiing to the terminal! Instinct tells me that in some way, this incident was fuel related.
Hugh, one thing in the AAIB report is that the plane spilled a lot of fuel on the runway so a simple ‘fuel’ out’ isn’t going to be the answer, although contaminated fule might be one.
I also fly gliders, and (naturally) we were discussing this at the launch point yesterday (Sat 19th) as I imagine glider pilots all over the country were doing. One of our commercial pilot members was there.
If you lose power, the glide angle steepens as mentioned above. If they had been 1,000′ higher they would also have been a futher back, so the plane would have been gliding below the correct approach for longer and would have undershoot by a greater distance, failing to make the perimeter fence. If they had been 1,000′ above the correct glide slope when the engines failed they would have got a very thorough and ‘interesting’ debrief after landing.
When facing an undershoot (in anything) it’s essential to fly at the best L/D speed - the speed that gives the greatest distance flown from a given height. That’s faster than the plane’s minimum sink speed, but it’s the right speed to fly if unituitive at the time. Trying to stretch the glide by lifting the nose results in flying less distance at best, and done to extreme will result in stalling or worse. The approach to this situation in a glider (and the equivalent of a sudden power lose is flying through sink or windshear) is to keep flying at best L/D, and put the airbrakes away if they are out. If you are still undershooting, at the end you can pull up to trade speed for height, put the nose down again and hope for a not too bad landing on the field - it will be slow so the glider will tend to arrive rather than land.
Our tame commercial pilot assures us all this is true for commercial planes as well, though best L/D speeds are a lot higher and best glide angle is a lot steeper.
Hopefully this is exactly what the PIC did - flew at best L/D, pulled up at the end to hop over the boundary fence and stalled onto the ground. Time and the AAIB report will show all - the trace showing speed, attitude, height and so on from the data logger will be interesting.
And thankfully the worst injury was a broken leg, whereas had it undershoot further it would have been much, much more serious.
Here is an alternative !
Two Giant Panda’s were hiding in the landing gear. When the landing gear was dropped the hungry Panda’s eat some “bamboo” which turned out to be the cables responsible for the control systems. Now if they find black and white fur underneath the crash then this theory is on track. (Apols)
if all the passengers had jumped up at the same time the plane would have made it to the runway! then all the pilot had to do was wait for a stall at about 2 metres altitude and drop the plane at zero knots.
basic, pure basic physics!
oh, almost forgot, it is documented that a 767 had both engines shut down at altitude by the pilot. something about all those switches and dials looking the same!
http://pages.slu.edu/student/paisd/research/dragproj.pdf
The above link is to an article that does a reasonable job of calculating the drag polar of the 777 ‘clean’. As you might expect the best L/D (about 23:1 is achieved at way faster than the 160 kts approach speed. Clean the thing gets close to 20:1 at 160 kts but is on the loosers side of the polar as far as the effects of loosing airspeed are concerned with the polar getting significantly below 20:1 once the speed drops below 160 kts. The energy deficit at the point of touchdown is significant - about 220MJ - calculated from the difference in height and speed at the actual touchdown (zero ft and
No fire despite fuel spillage is a true miracle … a spark is all it takes, but with engines out they probably cooled sufficiently to prevent them being a risk as it was very cold and air would pass through at up to 140mph, say.
But my main comment was Ewan you said the aircraft up and down the throttle to change altitude on approach. Strictly this is not so … when descending on the Instrument Landing System (ILS) comprising a Localizer and Glide Slope the aircraft would be, as you say, bleeding off any last excessive speed BUT as wind conditions slow slightly the aircraft or tailwind speeds up the aircraft, it leaves the actualy glideslope.
As you are already close to stall speed on approach you cannot correct for falling “behind” or under the glideslope by lifting the nose as this will loose speed and potentiall cause a stall. Instead you power up to push the aircraft a little further forward for the same descent rate to rejoin the glideslope. For a craft overflying theslope you throttle back a touch for the same reasons but reverse operation.
Way too early to be calling these pilots heros. Double engine failure seems impossible, throttle request not reaching engines - eletronic? - again unlikley. Have I missed any mention of communications to tower? They must have said SOMETHING - and it should have been juicy enough for the media to pick-up on. AND, no panic over other 777s’ My favorite is a fuel-related issue.
I am afraid I am going for the “777″ has a software problem, which by the way has showed itself on several near incidents in the past. Also this has happened on a few airbuses.
I will put some money on software problem causing command conflicts to the engines.
BA 777 Heathrow
1. No post crash fire
2. Wing root area punctured by LH MLG.
3. At time of arrival at Heathrow, the fuel tanks should have had fixed plus variable plus diversion plus unused block fuel. Presumably this quantity would have been distributed between more than one tank.
4. The aircraft would have had to have been defueled as one of the first post crash actions, certainly before shifting the aircraft. There does not appear to be any evidence of defueling tankers at the crash scene.
5. If the crew discovered a fuel problem prior to the crash, they may have been committed to feeding both engines off one tank and if this became empty, both engines could have quit almost simultaneously.
But the engines did not quit, as per the official report both engines spooled down to just above flight idle speed within seconds of each other. This for me is not a fuel problem but a command fault in the autopilot/throttle system communication with the turbines.
If you had contamination of the fuel, you would get a “hunting” effort just like you get in an internal combustion piston engine, which means you would start getting power fluctuations but not an immediate spool down to idle on both engines.
It will be particularly interesting to see the final report and its findings. At this stage it is certainly engrosing to speculate and excercise one’s mind but that’s all it is: speculation.
David G (January 18) has suggested hostile action as a cause, e.g. electromagnetic interference from a device on the ground under the flightpath.
I understand that wherever the PM is travelling, certain procedures (e.g. electromagnetic interference) are instituted by UK agencies in the area around the PM to help prevent terrorist devices from being remotely triggered by, for example, mobile phone.
The PM was at LHR when BA038 landed. Could such interference have been used with the collateral effect of knocking out the onboard computers on BA038?
S. Embee - I do not believe this is acredible explanation - Why would no other aircraft at LHR have been affected in a similar manner at the same time (i.e. subject to apparent software-driven malfunctions or otherwise uncommanded events)? Okay, BA038 wa sthe only aircraft at that point in teh sky, but there would have bene numerous aircraft with engines running at the airport at that same moment. Or, would you suggest any such similar events these have been “covered up” by the authorities - presumably as part of a wider conspiracy.
Sorry, a software glitch somewhere in the engine management system or fuel flow system seems most likely to me, i.e. somewhere in the maze of electronics a software “instruction” got lost or mistranslated by the next step in the chain.
Lets wait and see what AAIB investigation uncovers.
When I first heard of this incident I thought “both” engines??? Then I thought..long flight, auto pilot auto throttle approach, windy conditions, could there have been some windshear?
Had the engines been spooled back to idle…then you might need 8 or 9 secs to get full power, by which time it could be too late. (Remember the Air France Airbus that went into the trees near
Basel? You could hear those engines starting to spool up just as they began eating leaves.
However, the FADEC system must also be suspect. How the aviation industry got away with this I don´t know…NO MANUAL OVERIDE…unbelievable.
All in the interests of economy/profit…not safety! I understand this system has had glitches in the past..Maybe this is what one of your correspondents might be alluding to with reference to ice and fuel, after all, the sensors that feed the FADEC system could have been fooled by the presence of contaminated fuel. When you fly to and from the “boonies” this is the risk you take.
I hope the flight recorder is sufficiently sophisticated to cover ALL the important parameters, otherwise our “heroes” may be hung out to dry by the operator. Much cheaper & easier than having to instal a manual backup system.
My thinking is tied to the PM being at the airport. I am in electronics and know little to nothing about aircraft so for whatever it is worth i suggest if you want to see what a strong radio signal does to a computer take a cell phone in use and bring it near your Computer.
I just had a friend that wanted me to listen to the audio from his speaker and as soon as his cell phone got close the sound broke up so bad that you could not understand it.
Do a google search about cars not starting around the empire state building and when towed away 5 blocks they are fine it makes the thought that this was the cause sound not so strange.
If it was caused by some kind of jaming i am sure they will find something else to blame it on.
The fact that they have not grounded all 777 tells me they know the cause and are just trying to figure out what to blame that is not to big to do to all 777’s
About 5 seconds before the crashed plane came to a rest , the UK media were waving a figure @ Boeing as the source for the crash. Those evil Americans ..AGAIN!!!!
I kept hearing and reading “engine trouble” , “Boeing” in the same sentence.
What I kept seeing in the crashed plane engines were two really big R’s as in RR…
Hmmmmmmmmmmmmm I thought to myself …Hmmmmmmmmmm
I guess if one was not a pompous UK type..cough ..cough ..
one might care to look up the number of times a 777 in commercial flight and use has encountered problems with those mighty RR engines while in flight.
If you take the C and L off the word clASS in the phrase “ClASS society ” well..you get the idea…
“Cheers”
When I first heard about this crash landing at Heathrow,and when the pilots described the non response of ‘both’ engines to manual throttle I immediately suspected that some software or electromech problem had surfaced. This has to be an unprecedented incident where a large passenger jet almost crashed,due to an electronic glitch. You’d wonder why the whole fleet of Boeing777’s was’nt immediately grounded pending a full investigation. Remember all these systems are supposed to be thoroughly tested and in my mind this whole incident puts big question marks over the whole certification process of computer controlled aircraft. There are over 600 of these aircrft flying currently. If there is an inherent controls problem that’s a lot of lives put at risk.
[…] at Heathrow who decided that the best way to boost confidence is to taxi past the remains of the BA38 777 that crashed on approach recently […]
http://www.theregister.co.uk/2008/09/04/heathrow_777_verdict/
September 4, 2008
The Air Accidents Investigation Branch (AAIB) has concluded that the 17 January crash-landing of a Boeing 777 at Heathrow was probably caused by “ice within the fuel feed system” which restricted flow to the engines.
Related stories:
http://www.theregister.co.uk/2008/05/13/aaib_777_update/
Heathrow 777 crash: Siberian cold to blame? (May 13, 2008)