Chris Scott has already speculated about what a similar scenario would look like in the A330. It's an interesting read.

The result would have been very similar, except for an earlier AP disconnect, shortly after setting the thrust levers to maximum (TOGA) thrust.


Bernd

I must have missed that. As you say - interesting.
Thanks

Jo

Flaw and hindsight
 

Two questions, the second of which is hopelessly naive but seems important to address if only to rule out:

1. Has anyone yet any confirmed example of any cavitation-induced damage to similar HP pumps on a Boeing?

2. Why is there now no presumption of a latent serious design flaw in G-YMMM and the 400 or so similar B772ER aircraft, given the tentative findings that G-YMMM was apparently operated within a normal flight envelope by competent crew with the correct spec fuel?

My initial view on the second question after the last AAIB report came out was that not grounding similar aircraft to the one with a demonstrated but inexplicable fault in which two out of two engines fail to generate sufficient power while on finals over heavily populated areas into LHR was arguably a bit like dismissing the fact that there is an elephant in your nan's living room - whether on the part of crew, SLF, the families under its flight path, manufacturer, regulators, governments, etc. As demonstrated, the flaw would seem to manifest itself at low altitude when most of us would agree that there may be no or severely limited options to extend the glide or land elsewhere.

However I have come around to the view that it is reasonable that commercial issues dictate that a plane with a latent design flaw continues to fly until testing or at least two accidents can demonstrate that the flaw exists in a way which is more than just an unfortunate and mysterious fluke, particularly given the otherwise unblemished record of a workhorse of a plane with a huge number of trouble free sectors. A probability driven analysis of a possible "catastrophic" failure repeating itself would seem to be the most rational aviation industry response consistent with the prime objective of safety as with other risk assessments associated with flying.

There does seem to be a danger though that, by giving a nugatory risk weighting to a likely repetition unless and until someone can prove the causal mechanism behind the flaw/fluke, the industry profits from continuing to fly these aircraft with the regulators' approval and such a rationale in the face of any future catastrophe is looked on with hindsight as being more cynical than sensible on the part of the various stakeholders tacitly or otherwise supporting it.

So you suggest they ground an aircraft without knowing what caused the crash? This was debated 5 months ago on these forums, and overwhelming deemed to be totally imprudent.

You cannot enitrely avoid risk anywhere, and to try to do so would set us back in the stone age. Live with it.

Ground or not to ground 777
 

I generally agree with DXZH on the question of flight certification of the aircraft which failed to operate safely within the allowable envelope. From my perspective, the TMS 777 as configured, should not be certificated in environmental conditions which resulted in the accident unless and until it can be demonstrated that such operation is safe.

The 777 is a great aircraft. Nevertheless, it failed to operate safely under conditions which resulted in the LHR accident. Given no other data, information or analysis, it seems to me that the 777 must be restricted from commercial operation which approximate and exceed the adverse atmospheric conditions experienced by BA 038,

(TMS - Type, Model, Series)

TD

To restrict or to otherwise remove the certificate of airworthiness implies that you have probable knowledge of the the degree of restriction necessary or violation of the original certificate.

That has not yet been concluded by the investigation or the regulator

It was presumed that any fleet entering service does so by carrying a degree of latent unknown risk. Historical hindsight has shown that this degree can vary over time with some risks maturing into known risks which are then addressed over time by minimizations and corrective action programs, while still newer risks develop.

The issue is not which ones are latent unknowns and which ones are known and being addressed over time, but rather the question is how much risk is being carried over what period of time.

Restrictions and/or groundings are decided on the level of risk and not just whether the total risk causes are known or unknown.

Unfortunately, everyone is waiting for someone to scream EUREKA, the problem is solved.

Interestingly, in flight shutdown events have subsided with the onset of the Northern summer?

It would be a very simple, low cost palliative to introduce a bit of heat into the tanks near the feed pipe which would ensure that ice/wax is eliminated from future possible occurrences.

Not very scientific, but IMO "prudent"

dxzh
 

Well that is an incredibly frank and honest statement.

Thinking about it, what other options do FAA/EASA have? Slice something off the environmental envelope of the plane as a precaution? Require that the engines be run up at least once during a CDA approach?

One thing that gives me hope is that there are precedents for extremely improbable failures being diagnosed years after the initial accident: e.g. UA585 and US427 ... the NTSB reports were finally adopted 10 years and 5 years after the accidents, respectively.
http://www.ntsb.gov/publictn/2001/AAR0101.pdf

Fingers crossed, but not holding my breath.

7 months and counting.
 

This incident had a cause. Period.
The consequences could have been extremely serious if the issue had occurred just a few moments earlier.
If it has happened once, it can happen again, unless something changes. Period.

The fact that the cause remains unidentified must remain as a significant concern.
IMHO and from what we have surmised sufficient data exists to implicate a combination of fuel / fuel delivery system / and environment.
Therefore I would have expected some form of interim procedure / system change to have been introduced to reduce risk until a full explanation is unravelled.

The most surprising omission IMHO is why increased fuel temp monitoring / modified temp threshold value / increased water drain off frequency (or conditions) does not appear to have been introduced, at least on this sector or with this operator.
Maybe such an action has been taken although not reported.
Such an action would be a low cost "palliative" and would also add to data collection.

IMHO I still suspect the "fuel" (i.e. including water from the CWT) in combination with other factors.
Not that it was out of spec, but that the current spec(s) / procedures are too loose in some way.
Its the only thing that was truly common to both systems.

As I understand, one of the factors in this accident was that the problem was recognized late, with very little time to react. Would it be possible (reasonable?) to change the procedures so that this kind of problem can be found earlier, for example by avoiding "long" periods at reduced thrust?

Re-Heat #1544
 

I am sorry for not making myself clear. I am not currently suggesting that the plane be grounded - quite the reverse.

However, the recovery of the aircraft largely intact and 5 months of investigation seems to make the decision not to ground the aircraft more difficult. This is because the tentative dismissal in the several months since the crash of the extraneous factors that would typically cloud the issue (eg crew performance, operation outside flight envelope, component failure, out of spec fuel, etc) makes it more likely that the flaw/fluke is a design issue.

Again, I apologise to you for not making myself clear but I strongly agree with you that a risk based assessment is the only sensible way to proceed.

Nevertheless I believe what would be imprudent conduct as you succinctly point out without evidence may become less imprudent (or even become prudent) if hard evidence seems to eliminate the other obvious possible causes, notwithstanding the causal mechanism of the flaw appears improbable and is not understood.

> Would it be possible (reasonable?) to change the procedures so that this
> kind of problem can be found earlier, for example by avoiding "long" periods
> at reduced thrust?

Perhaps but I don't think we know enough to say that whatever caused the problem can only show up after a long period at reduced thrust.

As all accidents, it had a number of causal factors. Never only one single cause.

Fuel temperature:
And:

As to fuel and water ...

... and:

If water from the centre tank had played a role during the final approach, it would have to have melted earlier during approach to be scavenged by the main tank scavenge pumps to the main tanks, and thus water would have to have been present in the main tanks. This was not the case.

(I take it that the term "significant quantities" implies such a quantity that it might exceed the operating limitation of the water scavenge system (which is effectively a fuel/water-stirrer), the fuel delivery system, the pumps, and the engines.)

And what symptom, exactly, would this "palliative" treat?

Seen another way, it would be an additional cost, with no benefit.

Still not entirely out of the question is a software problem. It is very remote, but so is every other possible scenario that has arisen so far.

If we exclude every possibility that is "very remote", we must conclude that BA 038 landed safely.

Or, according to a well-known German poet:

(A translation from Christian-Morgenstern.de:

The Impossible Fact

[...]

And he comes to the conclusion:
His mishap was an illusion,
for, he reasons pointedly,
that which must not, can not be.
)


Bernd

Directive before winter; risk and Sod's Law.
 

Warning: I'm non-professional; not crew, not engineer - just scientist guest and thanks.

We have already speculated on some form of directive/restriction on ETOPS being issued before the onset of the northern hemisphere winter.

The AAIB might not issue such a document until:
1) the northern hemisphere, summer, holiday season has passed,
2) the Beijing Olympics are over,
3) all interested parties are in agreement with the need for restrictions,
4) and, the AAIB has determined that its investigations are unlikely to find a cause and solution before the northern winter.

And, maybe, we should not be surprised if the restrictions apply to all aircraft, not just Boeing 777s. That way the authorities create a commercially level playing field; probably a prerequisite for success with point 3 above.

Much has been written recently, on this tread, about the acceptance of risk. I cannot pretend to understand the basis for the statistical arguments for the low probability of a second occurrence of the BA038 incident but I remind myself that, for example, a 1 in 100 year event can occur in any year, i.e. there could be a second occurence in 2009. That's called Sod's Law.

And, I suspect that even the most competent, and confident, investigating statistician or engineer might, on hearing that his children or grandchildren were about to fly across Siberia in January-March, suggest they take another route, or have his/her fingers crossed all the way.

Can you engineers truthfully say otherwise?

(I apologise for using emotive phrasing and for appearing to attack engineers - not so; the words are chosen to highlight our collective complaisance or, in too many cases, cynical avoidance, regarding the proper understanding of risk assessments and the probability of recurrence.)

On another matter, pilots, as I understand them, have a highly developed understanding of Sod's Law. Basically they are aware, individually and collectively, that luck can run out at any time. Therefore I would be fascinated to learn if pilots have already altered their flight management behaviour, for example, spooling up the engines on descent a couple of times - just to make sure! As I've written before, better to discover a problem with a few thousand feet to spare rather than 720.

Regards, Tanimbar

Which all falls squarely on its arse when you realise that in a business where the words cannot often be used to their full and proper extent, the AAIB really are one of the last bastions of consummate professionalism, unlike some other large organisations we could mention. "Until after the Olympics"? Are you having a laugh? This isn't the Press Office of the Labour Government we're dealing with...

dontdoit ... it's no laughing matter
 

Granted that the AAIB might be "one of the last bastions of consummate professionalism" it nevertheless does not operate in a vacuum.

I do not denigrate the integrity of the AAIB by my comment about the Beijing Olympics. If the activation of restrictions is not required until the onset of winter why not avoid false, unwarranted passenger hysteria by delaying the release of the publication until after that event. That's just common sense and would support your view of the AAIB as consummately professional.

regards, Tanimbar

dxzh wrote

You're assuming that because G-YMMM was affected it must have been particularly prone to such a failure(whatever it was).

From a statistical point of view this is not a valid assumption and the incident might actually be equally likely on other aircraft types.

We're talking tiny probabilities here and a single incident is not statistically significant.

So ithout some idea of the cause it would be difficult to justify taking action against a single aircraft type.

Well said.

You could also add that if you restrict/ground a particular aircraft type over one incident and subsequently are unable to find anything in particular that is wrong with it, how do you logically go about de-restricting/un-grounding it? Nothing has changed in the interim yet you'll have to make a 180deg. turn in terms of decision making whilst in possession of the same facts...

This also impinges on why there are no significant procedural changes in the operation of the aircraft. If you don't know what caused the problem, how can you know if your 'solution' is making things better or worse? Airlines and manufacturers are reluctant to provide 'instant fixes' as they often come back to bite them through the law of unintended consequences.

At the moment we don't know if what befell MMM was because of:

a) something unique to that airframe
b) something that could happen to RR-powered 777s
c) something that could happen to all 777s
d) something that could happen to Boeing aircraft
e) something that could happen to all commercial jets
f) etc...

Who was it that said "Sometimes the dog of fate simply lifts its leg and pisses on the leg of science." Ernie Gann?

The quote is as follows from Fate is the Hunter:

@netstruggler/FullWings
 

I see what you mean about this potentially applying to any aircraft and I am conscious that statistics can be used any which way by whoever chooses the assumptions and data to be analysed. I had not meant to be partisan in my choice of assumptions but perhaps I have introduced bias by trying to take into account the interim findings and you might be doing so if you do not take them into account?

To one person, a single incident of an uncommanded reduction in power of both engines is not, or is hardly, statistically significant in the general scheme of things and does not necessarily mean that any particular aircraft should be grounded or, as elsewhere suggested, restricted.

To another person, IF it can be tentatively assumed after investigation that contributory factors external to the aircraft itself such as crew and fuel can be discounted and factors specific to the aircraft such as the independence of the two engines is as certified, then the uncommanded reduction in power of both engines on an aircraft within seconds of each other and a failure to respond to further requests for power might be seen by many as either a highly unlikely event (if I may, equivalent to a sperm whale finding itself freefalling to earth with a bowl of petunias) or evidence suggestive of a flaw in its certificated design/performance. I accept the assumptions here are crucial.

Again I was only trying to consider, compared to many crash situations in which the airframe is not immediately recovered largely intact, the difficulty stakeholders may face in deciding not to ground or restrict aircraft (and no, I am not arguing for either) now that the initial investigations have been completed and so many of the other possible contributory factors have been tentatively discounted. As the airframe and crew survived, after a few months of investigation there seems to be a practical limit on the ability to obfuscate about the potential alternative contributory factors of the highly improbable uncommanded reduction in power of both engines, while apparently leaving those responsible for investigating, manufacturing and certificating the supply of fuel to the engines (whether in G-YMMM or other aircraft) in the invidious position of still being unable to explain the demonstrated critical failure in an otherwise highly reliable system and to make recommendations accordingly.

dxzh

I may pick at your arguments somewhat, albeit they have many sound qualities. Remove the ETOPS argument above, as the issue equally applies to all aircraft operations.

Also to be pedantic, the term rollback has historically been used to describe a different malfunction scenario so perhaps we should only use the words released to date by the investigators.

continuing in pedantic mode

There is no such thing as a fluke in statistics, as their use is only a means of communicating natural occurences (combinations)

I remind that neither the designs nor certification imply all is perfect and free from catastrophic occurences. There will allways be the rare occurences of combinations not forseen nor even recognized after the fact, that will exist in aviation for relatively short periods of time (measured in risk per flight hour).

Some risk of unknowns is acceptable even some risk of knowns is acceptable over controlled periods of time. The issue is to be certain that the level of risk is within acceptable bounds and this is done by continued dilligence and investigation of potentially related events (either in history or today's)

So in my opinion groundings or flight restrictions are not evident at this time, but continued dilligence, (investigation and understanding) is appropriate.

show me the data and I will recommend the action

With the airline (BA), Manufacturer (Boeing) and the British AAIB-CAA etc etc involved, as well as other agencies...

Being blunt and upfront literally, -- NO offence intended to any member of the crew -- no one seems to be any further forward after almost 7 months of extreme head-bashing, early mornings - long afternoons and late nights around computer simulations, test rigs, theory upon theory etc etc etc...

No one seems to agree on any particular 'reason' for this happening. Leading on from this, when do the authorities decide to call it a day - if at all?

Is this another 'unknown' phenomenon... In which case, basically case closed...

lomopaseo
 

Thanks - point taken re ETOPS, rollback and fluke.

And I agree about living with knowns and unknowns for controlled periods - we all accept a measure of risk in living our daily lives.

How do you know that?

Silverstreak, sir.

Simply because we have no further information from AAIB does not mean that AAIB has no further information.

(i) I incline to the view that AAIB does not leak- it can't see the point of it, it's not a useful exercise from their perspective.

(ii) We do not set the time frame for AAIB reportage - a process which is currently well within their standard pattern. Look at some of the previous excellent postings on this.

(iii) IMHO we need to be patient.

The frustration v trust dichotomy is never an easy one to resolve! :ok:

CW

It may well be that the AAIB does not leak, but earlier in the investigation there were "leaks" from various sources.

These have dried up - maybe because discipline has improved, maybe not.

.

M.Mouse...

OK point taken :ugh: Nothing has been done so far in the quest to find out what actually happened :mad:

You miss my point. There is a massive amount of work being undertaken. The fact that details of the results of that work have not yet been released is not an indication that 'no one seems to be any further forward'.

May I ask a couple of questions on the 777 incident?

What was the temp on the ground while the aircraft was on turn around in China?

What temp had it been exposed to, and for how long, while en route to China?

I understand that fuel "waxing" might be an issue. Fuel does not fully dewax for some time after landing. If an aircraft flies a long haul, sits on the ground in low temps and then flies long haul again the fuel wax status may be enough to prevent required fuel flow as deposits clog filters..?

Would it be desirable to alternate an aircraft between hot and cold destinations to help reduce the possibilty of waxed fuel build up?

Just thoughts..??

People get suspicious of brand new posters coming in with deep and searching questions! What was your ID before?

There was nothing unusual about the circumstances. It might have been a little colder than normal, but aeroplanes do actually operate in very cold climates all the time, and cruise for hours at extremely low OATs. Rather than simply ask, if you feel it's relevant, try a little research on your own to answer your own questions. Do you think all that wasn't examined right at the start?

Well it's nice to be made welcome, thank you Rainboe, I appreciate your rapid reply.

You obviously have all the answers, and please be aware that you have no need to label me as suspicious. Most of my days have been spent in the airline industry....

I understand from your reaction that fuel waxing is not an issue? Maybe you could direct me to the latest studies on the properties of jet fuel at low temperatures.. I would be delighted to find my own answers as you have suggested I do!!!!!!

Newbie guide?
 

Peter - there is a 'search this thread' button at the top of the forum window.

Try searching for temperature specification or other words likely to be in any post answering your question.

The extracts found by Search all have a clickable link to the actual post.

Each post has a poster. If the post makes sense, click on the poster's pseudonym, then follow through to the poster's profile, and to a list of other posts by the same poster.

Note the post numbers of anything interesting, then when you refer to them in your post, your readers can find and assess the posts for themselves.

Newbies get mixed welcomes here - probably for good historic reasons.

Welcome!

Many thanks RightBase

Your welcome and guidance very much appreciated.

330 versus 777
 

I do think a 330 would have done very well with hopefully no casualty as well.
But my view is slightly different from Chris Scott view

By manually fire walling the thrust levers, Go Around Auto Flight logic is triggered.
FD bars command SRS (Speed Reference System) in this case probably VAPP (VLS + 5kt)
As speed was already around VLS or even below and as bizarre we might think FD will command aircraft to pitch down (waiting for thrust increase …)
Note: AP is still ON unless manually disconnected or sidestick manipulated.
Now aircraft is below G/S but still at VAPP.
Houses are getting closer … time to refuse ground contact by pulling the sidestick (AP now disconnects)
Speed is bleeding AOA increasing to AlphaPROT … (maybe AlphaMAX but I don’t think AlphaMAX is achievable below 100 feet …) STALL protection remains active.
Less drag in the early portion and better ground effect, aircraft touches down somewhere in the grass in a nose high attitude.
V/S at touchdown depends on pilot fortune in the overall process.

Would be nice to validate (or not) this scenario on a 330 sim if ever able to simulate the thrust deficit with THR LVR in TOGA gate.



BA038 touched down in a very flat attitude witch means it had stalled already.
Combination of soft terrain + landing gear impact absorption made things very comfortable for passengers at least.

Too bad AAIB still keeps very secretive on aircraft performances as well as on front crew actions …

To Peter Brown
 

I understand very well your concern about the "temperature history" of BA38. http://www.pprune.org/forums/images/...ons/icon14.gif

I follow this thread since post # 1 and I didn't get documented data about ground and flight temperatures the aircraft met during the flights and the days before.

We know that there was no "fuel draining" before the flight to London and that there were some fuel ice messages.

We know also that the flight was operated well higher and colder than initially planned, and that the crew didn't ask for descent into warmer temperatures over eastern Europe while other aircraft did. That's about all ...

I just hope the investigators are looking carefully into weather records.

In my past career, I got my share of temperature related problems with fuel, fuel systems and other aircraft systems as well, all in duly certificated aircrafts and components. And yes, at some points, we had to switch aircraft on cold sectors.

Welcome! http://www.pprune.org/forums/images/...cons/icon7.gif

Bis47:
However, water draining checks were performed just prior to refuelling for departure to Beijing on January 15th (see AAIB special bulletin S1-2008, page 5) and could you please explain when those "fuel ice messages" were presented to the crew? These are not mentioned in the AAIB reports.


Thanks and regards,
Green-dot

@Bis47, I guess your "fuel ice messages" reference is to the unverified report on pilotsofamerica.com forum (see UPDATE on the LHR (Heathrow) 777 incident - Pilots of America Message Board) on 30 January 2008 (and picked up by blogs elsewhere) as to the theories being investigated by the AAIB. One theory the unverified report mentions as being investigated is that:

"Ice in the fuel somehow limiting the fuel flow to the engines. A maintenance message indicating excessive water in the center tank was set during taxi on the two previous flight legs, although it cleared itself both times."

For what it is worth I note that the unverified report was introduced by the words "Aspects that the FAA believes the investigation is concentrating on are: [...]" but also that the alleged fact re maintenance messages is not mentioned in any AAIB statement published to date (whether Statement of accident, Initial Report, Initial Report Update, S1-Report or S3-Report).

Assuming ice in the fuel was being investigated at the time, it seems less likely to be relevant now given the subsequent analysis and AAIB statement in its S-3 Report that: "The fuel has been tested extensively; it is of good quality, in many respects exceeding the appropriate specification,and shows no evidence of contamination or excessive water".

@Peter Brown, you might also have a look at articles re flight in polar conditions at Aero 16 - Polar Route Operations and on fuel temperature data on polar flights at http://www.fsinfo.org/FSI-journals/4q_2000.pdf.

There is no fuel ice message on the B777.
There is a water in fuel message. However this is never presented to the crew. It is hidden away at the bottom of the fuel qty maint page which few people ever look at unless you have a fuel qty problem. However it is stored in EICAS NVM so can be seen by the investigators.

Exactly.:ok:
That is why i wondered how Bis47 came to such a conclusion.:confused:



Green-dot