PG...thanks for your opinion..but
 

That report will prolly be read by a wider audience than the official version and of course it is an interpretation. It has pics and good precis and is objective.

For people like me it is wunnerful because I can see the parts and understand what went on. It makes sense.

I don't think this will happen again due to the modifications made as result of this accident. That is all I seek. The outcome is postive and the world is safer..whatmore could anyone want?

Instead of swearing some crew communication could have helped to detect that the AP failed to be disconnected as it was believed.

Aerodynamic basics would state that there is no set "rule" you can derive. The higher you are, the lower the optimum flap setting would be. As you get lower the lower stall speed with more flap outweighs the drag.

It is not the AAIB's job to (re)write handling manuals: They went as far in this report to analyse the FC actions as were relevant to the accident - in short, a no notice untrained for emergency, and their actions pretty much resulted in an optimal outcome:Fundamentally this is a design/engineering accident, and little to learn from the FC pov - IMHO it is impractical to expect "drills" to be produced for every eventuality - especially since in this type of accident they would need to be "memory"...

Just my 2ps worth ;)

NoD

It's not a matter of different eventuality but basic fundamentals. Losses of thrust during initial climb requires to control:

- Safe flight path
- Optimum speed
- Configuration (drag)

There is basically no difference when this happen during an approach, there is no need for a "special" procedure. I regret that the AAIB didn't consider including few lines about it.

which will, no doubt, bring down the wrath yet again of the god-like pilots on poor old sfly. But, people, she has a point. Throughout my reading of this thread I've been amazed by the pure coolness displayed by these guys in leaving the a/p engaged until it tripped at stick-shaker. What faith in automation!

and now it turns out the F/O "omitted" to disengage it. That doesn't sound like a deliberate (and oh so cool) action.

And surely you guys out there see the point. Might it have been better to hand-fly from the moment the problem was identified? Perhaps not. Certainly the outcome could hardly have been better. But I think we'd all like to know.

Pettifogger -

The transcripts for USAIR 1549 are available. 3+28 (208 seconds) from impact to impact.

FO Skiles - Oh, !!!!!
CA Sully - Oh, yeah

After that they were too busy to make any personal comments. Sully has said that his family never crossed his mind.

I would guess that the BA038 crew had the same focus.

SFLY
95% of pilots will follow the flight director if flying manually. The flight director will be trying to keep the aircraft on the ILS, exactly what the AP was tryting to do. So why increase the workload, when trying to faultfind an emergency for which there is no drill. For the 15-20 seconds that elapsed between when the FO intended to take out the AP, and when he did, the difference in flight path is hardly worth talkingabout, if you want something pointless to contemplate, why not have a heated debate about whether they should have asked all the passengers to move to the back of the aircraft, in an attempt to reduce control drag!

Pettifogger ;

" .........the AAIB's report to be well researched, well written and comprehensive in scope. I believe that they have reached the right conclusions and have provided sufficient analysis to justify those conclusions. ......."

The report is hardly "comprehensive in scope" as it does not even begin to try to cover representative tests of the presumed actual fuel/water mixture in the plane at the time of the accident.

Likewise their "analysis" is non-existent as they totally dodge this issue.

The report screams out both a lack of rigourous testing and avoidance of the huge cost that would be involved in both undertaking such research and then applying it to ALL airframe and engine combinations.

Releasing a report with such glaring gaps is plain weird.

410,880 lbs

SoS

This is the longest, and most comprehensive, AAIB report I can recall. I guess as a taxpayer I ought to be glad that the AAIB does not have to pander to idiots who want the report to explain the movements of every component throughout their entire life, the movement of the planets and the variations in the planetary magnetic fields.

Phil Gollin
 

While the actions of the crew in the last, doubtless frenetic, seconds of the flight are both interesting and admirable - so far as I am concerned - I have to say I share Phil Gollin's concerns that 'eyes have been taken off the ball.' His well researched and knowledgable post should really have put questions of flaps raised or not, to one side. The reason those two gentlemen, plus all their crew and passengers were in that terrifying situation is because two of the most beautifully engineered machines for producing thrust, didn't. Both at the same time, give or take a few seconds.

I joined this forum because the circumstances of BA038 were, to this uninvolved bystander engineer, quite simply; staggering. Unbelievable. And yet today the report uses the word 'probably'. Entrained ice in the fuel probably caused the engines to 'choke' simultaneously. However, unless I've missed some big numbers somewhere, it seems quite extraodinary that so little water could choke such big engines at such a critical moment in the flight.

It appears to my untrained eye that the engines were consuming fuel at varying rates - in the last twenty odd seconds - but between one and two kilograms per second. Phil Gollin's text stated that it was estimated that there was a total of 5.14 Lt of water in all the fuel in the aeroplane at take off and sucked in during the flight. It just seems too little, because if all the water in the tanks suddenly arrived at the delivery system in the last few seconds, it would still only represent about 10% of the fuel being consumed at the time.

Since entrained water and ice in fuel systems has been endemic all through the history of high altitude, high speed flight and certainly during the whole operating career of the 777 - a lovely aeroplane to my mind - why hasn't it happend before?

Perhaps it has been decided to apply the principle of Occam's Razor and accept that ice is 'probably' the cause. Although I apply the principle myself in my own job often enough, the pricking of my engineer's thumbs suggests to me that something, somewhere has been forgotten.

Roger.

Mmmayday38:
for some it is still not possible to accept that gut feeling flying may actually produce best possible results.
I still believe that your post http://www.pprune.org/5426378-post2761.html is a glamorous example of humble and professional human side of flying.

thank you!

long time ago J.Conrad wrote a book on sailors - 'Lord Jim' - from which i have learned one thing: one can not tell how he would have behaved in given situation until he would have lived through such experience himself.

hopefully the Report's recommendations might help in preventing it happening ever again...

That could very well be. BUT.....you still have an engine left for some thrust. BA038 had NOTHING.

Well done Peter and John in doing your best in such limited time. I cannot imagine being in the same position.

Additionally a fantastic job by ATC in response. Just heard the audio on BBC. Chilling stuff but well handled.

Stop criticising and let's move on.

Landroger

I recall that, from much earlier in the investigation, data on 777 engine and aircraft handling was checked for many hundreds of thousands of flights (or was that flight hours? Not sure) and the parameters found for BA038 and G-YMMM's last journey were right down at the very tail of the distribution with the lowest rates of climb and hence the lowest incremental thrust/fuel flow, coupled with a very long flight at unusually cold OAT followed by an unusual uninterrupted continuous descent into LHR with only one power application above idle which rapidly resulted in the thrust reduction seen.

While I can see Phil Gollin's point about things that have been left out of the report, it is clear that such a weight of unusualness is bound to have effects that may never have been seen before.

I also think that the volume of "ice" - by which I mean some sort of semi-frozen ice/water/fuel mixture may well exceed that theorized by a simple consideration of the effect of freezing those 5.14 litres of water, but even then the crucial pipes are quite short so a little "ice" clearly didn't need to go such a long way.

But, clearly more to be learned.

The question is, can anyone be expected to have the strength of character required to investigate further and then, if the facts show it to be necessary, convince the industry to spend what could be a very large quantity of money to correct the behaviour of every fuel system on the planet?

Occam's razor (or Ockham's razor[1]), entia non sunt multiplicanda praeter necessitatem, is the principle that "entities must not be multiplied beyond necessity" and the conclusion thereof, that the simplest explanation or strategy tends to be the best one. The principle is attributed to 14th-century English logician, theologian and Franciscan friar, William of Ockham. Occam's razor may be alternatively phrased as pluralitas non est ponenda sine necessitate ("plurality should not be posited without necessity")[2]. The principle is often expressed in Latin as the lex parsimoniae (translating to the law of parsimony, law of economy or law of succinctness). When competing hypotheses are equal in other respects, the principle recommends selection of the hypothesis that introduces the fewest assumptions and postulates the fewest entities while still sufficiently answering the question. It is in this sense that Occam's razor is usually understood. To quote Isaac Newton, "We are to admit no more causes of natural things than such as are both true and sufficient to explain their appearances. Therefore, to the same natural effects we must, so far as possible, assign the same causes."[3

More on Wiki ( from which this intro is lifted) for anyone interested.

In the modern idiom, keep it simple, stupid.

No doubt. But when sliding along on the grass? Have a look at this at 2.08 to 2.16
BBC News - BA crash captain: 'I thought it was my time to die'
This clip also contains Capt. P Burkhill's recollection of his flap/drag decision and observed consequences.

As a humble SLF and one-time glider pilot,

Having read this thread and various offshoots, and having just listened to the ATC recording, which highlighted how incredibly calm and measured the responses were from each of the people/groups involved, may I just say how much I was impressed by the actions of all those involved in this incident. Truly professional !!

There are those who post here who will always try to seek blame and unfairly denigrate individuals, regardless of the facts. To them I would say **** ***

CS

phil gollin

Your last statement desrves a rebuttal

It's not the job of the investigating agency to research and develop all possible contributors in future accidents. It is generally considered that an investigation report has done a fair job if its recommendations are timely and sound enough to make it unlikley that the accident will be repeated before the designer-regulator has addressed the issues.

Sometimes it's important to be aware of the fine line between design intent, regulation and after-the-fact discovery. Lessons Learned

Feathers McGraw
 

I accept that BA038's flight profile might have been unusual, or rather an unusual concatenation of ordinary events, the chance pattern of which placed them at an extreme point on the 'bell curve'. But even the extreme position on the curve does not preclude the possibilty of bilateral 'engine failure' if say, just one extraordinary item was removed and thus something like this ought to have happened before?

I don't know just how sophisticated any tests have been, but it is not beyond the wit of man and an imaginative test rig, to reproduce the accident conditions routinely. Some aircraft components - although not a whole engine I wouldn't have thought - a quantity of appropriate fuel and reliable means of lowering the temperature of tanks, pumps, pipes and control systems. Unless and until someone has done this and reproduced the fault - where reproduce is the scientific key - it will not be possible to say we have the answer to BA038.

I think Phil Gollin has said as much, but with greater scientific weight than I could possibly aspire to. It seems that the fault condition could be achieved - intermittently - but only by forcing certain parameters well outside those thought to obtain in G-YMMM. One of which was to artificially increase the proportion of water by direct injection near to suspected critical components. (Italics my speculation.)

If this localised increase in water density was the only way to trigger the fault condition, then the fact that - according to Phil Gollin's text - ice tends to form on the bottom - coldest - part of the tank (and was extremely difficult to dislodge) suggests that this would keep the bulk of the available water away from anything vital. And even if it did in one tank, enough to choke its engine, why did it happen in the other at the exactly the same time?

All I say is, again in my opinion, that this accident was extraordinary from the start, but after all this time the answer appears to put it in the category of 'the perfect crime' and only just short of it being caused by an Alien Energy Beam.

Roger.

Landroger

I discard "holes in cheese", and Occam is more like a group vote than a tedious research into cause and effect. If I could call up the spirit of Pinkman I would welcome his input relative to water/fuel/ice.

I think you envision a puddle of Water freezing at the bottom of a tank or an elbow in the transit pipe. There is always water in fuel, that is a given. In the least amount perhaps 7ppm. Depending on piping architecture, tank pressure and boost pumps, the system is engineered to preclude clogs of solid water Ice getting to the engines. From the beginning here, from Boeings chief safety pilot to posters on this thread, a theory of migratory slurry came about that was theorised to have broken off the walls of piping to flow with the current to end up at the FOHE face, where it plugged fuel flow after several seconds of full chat, a matter of collected data on the record. GE versus Pratt, Chinese versus US or Brit fuel, bowser this, sumping that, or not, and the procuring cause of the problem was?????? UNKNOWN. The theory was not duplicable, and the mandated fix involves a little closer shave to the face for the fuel tubes in the HE cannister. Evidently, Boeing and FAA are confident the fix will prevent further occurrence of quiet Trent syndrome. Since nothing in the report explains how this happens, one may ask really?

bear

You aren't the only one.

That shouldn't be a surprise, it wasn't to me. Since the early interim reports it's been made clear that there was some sort of obstruction in the fuel system, that disappeared before investigation. Ice / wax / slush typically doesn't stick around to be investigated, and can form and behave "randomly" making reproducible experiments practically impossible.

Realistically, "probably" was always as close as we were going to get.

What you may have missed is that the suspect is not water ice, in fact I think the AAIB may have confused many by referring to it throughout as "ice". It is actually a slush made up of fuel and water, at above the waxing point of the fuel but below the freezing point of water. It isn't clear that this is a previously known phenomenon - it isn't a well researched state of fuel, if it was known about at all.

The reported tests showed both that a spoonful of water could create enough of this "ice" to block the FOHE, and that larger quantities of this "ice" (more than enough) could acrete on pipework over time at low fuel flow rates.

AAIB: "Data mining showed that the accident flight was unique amongst 175,000 flights..."

That was a big data excercise, and found no other comparable flights, so we have no idea how many flights you would need before you would be likely to see a repeat of these circumstances. Quite possibly more than the total 777 flights to date.

To apply Occams Razor, you need to have other hypotheses that also fit the evidence. I see none. Up to now I would have said that the spar-valves were still a suspect on the info we had, but I believe I see in this report the evidence on which the AAIB have ruled that out (not as clearly presented as I would like, but I think it is there).

What I do see in this report, in much more detail than we had before, is lots of elimination of other hypotheses before ending up with the "ice" as (only) probable cause, ie. "When you have eliminated the impossible...". That isn't Occams Razor - in fact it is more the reverse of it.

Agreed and worth noting that some 18 months ago the AAIB recommended that the regulatory authorities consider, as phil gollin says: "undertaking such research and then applying it to ALL airframe and engine combinations."

That ball has been in the regulators court for 18 months now, and it is out of order to blame the AAIB if the regulators haven't done anything with it.

BA038
 

The CBC here in Canada are reporting that the crew of this flight have been off duty since the accident and the Captain "is looking for work".
:confused:
Do I conclude from this that;

A. BA severed the crew
B. The Canadian media are just as good at reporting aviation stories as everyone else
C. This crew have decided to "moonlight" as industry saviours
D. Fiction is stranger than fact

Willie :ugh:

Don't know about the F/O's, but the captain resigned last August 2009 and if he is looking for work, then we'd all be doing a good deed in closing that thread that is in no way helpfull.

Aside from the ice stuff; there's some interesting theoretical questions that come as a result of this:

-does a three degree glideslope have enough of a safety margin for an approach vector. Would a higher slope with a lower throttle setting (hence less difference should thrust fail) be safer with regard to a loss of power on finals, or is the requirement for a better go around response more pressing (presumeably, it's faster to go around with a lower glideslope)

-Are runway safety margins big enough; should there be some sort of retardation surface at the end of runways in the event of such an incident or is grass good enough? What if it had been frozen?

-How the hell did a guy make it back on board to retrieve stuff? I know the slide at the rear right door was almost horizontal from pics I remember seeing but when did he get a chance?

-Good boeing quality control I see; with plastic ice scrapers being found in the tank; and FOD having a possible effect on the water scavenging in the tanks.

Yes, of course... But the aerodynamic principles are the same. The concern is to get rid of the parasite drag. On a heavy jet with the landing gear down and full landing flaps and no thrust you decent like a stone at VREF :rolleyes: Or you bleed off your speed down to the speed shaker in no time if you maintain a 3 deg G/S, like the crew of the TA B738 in Amsterdam experienced with their throttles closed at idle.

The difference in stall speed and hence available lift is not much at the later stages of the flap settings - but the difference in drag is enormous. Which is handy to have, if you got your engines running. It will require a rather high power setting, which gives you a much faster throttle response.

So IMHO even if they couldn't maintain the speed and raised the AoA to get more lift and therefor increased induced drag, they drag was still less then the parasite drag you have with full flaps. I think the report also mentioned, that if they had maintained the configuration, they would have touched down a bit earlier. Which seems to support my thoughts above.

This isn't correct, as you can see on the following illustration (B787), the later stages of flaps are having a lower impact on the L/D performances than the earlier. This means lift benefits are more predominant over drag in this stages than in the earlier.

http://www.lissys.demon.co.uk/samp1/lspb.png

While L/D improvements resulting from flap retraction are visible on this illustration, you have to keep in mind that this is valid provided that the aircraft flies the appropriate Vref. In other words, compare the little L/D improvements that you obtain through flap retraction with the major L/D reduction caused by flying at the minimum speed and you will find out what was the most efficient way to control gliding performances. The lack of speed control caused greater performance deterioration than the flaps retraction could improve.

It's exactly like trying to clear an obstacle during a climb by retracting the landing gear while the aircraft's speed is far from Vx. It only reduces the "negative" effects.

L/D
 

I admit this isn't my line of work but doesn't a given configuration give a given L/D ratio which defines the best flight path angle (best in terms of range) and the crew have to find an angle of attack which achieves the flight path angle with the speed following from there ?

For each config you have an optimum L/D angle of attack. Deviating from this value will reduce the L/D ratio.

I agree that on a 3 degree slope an a/c is always going to be short of energy if it has a complete power loss.

So why not fly a bit faster?

What, for pitys sake, is the point of flying at Vref+5, 5 miles (or more) out?

Is it SO hard to slow down when near to the runway?

The extra 'handling speed' is useful too.

Because a stable approach is a fundamental tenant of a safe operation of a large commercial airliner. Accurate speed control is absolutely vital to putting the aeroplane down in the right spot and getting it safely stopped.

It is? Why? Are you suggesting Boeing Airbus et al need extra speed to handle better? Whatever better is...

Why are you posting on a subject and thread you clearly have absolutely no knowledge of?

L337

I have 10000 hours in commercial ops.

Where I work you have to be stable by 500' but they teach people to fly Ref+5 for miles before that.


Do you know what 'handling speed' is? It's not type specific.

As for flying a bit faster: I think most folk could and still be stable at 500'

Maybe you'd struggle on your type? I know not. Depends on the type. And the pilot.;)

I cannot believe for one second that you have ever flown a heavy jet. And whilst we are willy waving. I am a 747-400 Captain, I have well over 20,000 hours of which 7000 are on 747s.

I sure would struggle to be safe if I flew as you suggest.

Well, there you go then. Don't carry any extra speed. I'd hate you to have an accident!;)

An extra 5-10 kts would likely have got the 038 on to the runway, is all.

My point is simply that Ref+5 is great at 500' -and you need to make it so, because nowadays you're not trusted to slow down any later.

But why at 4 miles? I mean -are you going to land out there?

(Biggest I've flown is MTOW 48000kg. Small to a Jumbo.:))

Just a wee edit to add that we can be at Vref+19 at 500' and still be officially 'stable'.

L337 , well, I have to say you are very fortunate to always fly to destinations where there is no, or very lax ,speed control.
At LHR/LGW/STN and many many major European airports the requirement is 160 to 4.(indeed there are some that ask 180 to 4 which is exaggerating just a little)
So, if you are able :rolleyes: 25kts to be bled off, in the case of this 777 , in the last 4mls.
Well, in my mere 16500hrs I seem to have observed most of my colleagues remaining safe whilst complying with that, what are you missing ?

Bearfoil
 

I happen to be looking at a S Korean Jet A-1 CofQ as we speak. FP quoted is -51C. This particular one had bacterial contamination as it happens but that's nowt to do (it seems) with BA038.

Come to think of it, one of my colleagues dealt with a major particulate contamination (again, nowt to do it seems with BA038) on an Ulsan jet cargo late last year, I'll check on quoted FP when he returns to the UK next week.

I never for one moment suggested that I do or would ignore speed controls. I said: "a stable approach is a fundamental tenant of a safe operation of a large commercial airliner. Accurate speed control is absolutely vital to putting the aeroplane down in the right spot and getting it safely stopped."

The bulk of aircraft accidents happen in the landing phase. The major cause of landing accidents can be traced back to rushed approaches, and unstable approaches. 4 miles is 1200'. With planning it is possible to be stable at 1000, and definitely stable at 500'.

Carrying unnecessary excess speed is one cause of a unstable approach.

Recent studies show that, most CFIT accidents occur within 8nms of the airfield, and a very large proportion of these are on the centreline within 3nms.

1000' is about the last point to have energy management issues resolved in order to ensure maximum capacity for the last 3nms of the approach to monitor and manage the most exposed part of the flight.

Can anyone explain why the incident happened when it did, just before the runway? Why did it not happen earlier in the flight? To me, this is just too much of a bit of luck to have been pure happenstance. It suggests human error, possibly pilot error. But what?