larssnowpharter

I do!

You have to get away from the thinking that there is a 'simple' explanation such as a mechanical failure. Mechanical failures happen. MTBFs on equipment are a known quantity by and large in the sense that we know it will happen sooner or later. Then you have to ask what mitigating measures are in place so as to prevent said failure turning into an 'event'.

I have carried out God knows how many RCAs connected with incidents in my field and only once did we - as a team - come up with 2 genuine root causes. Many, many 'causal factors', however.

Corrective actions may be broad and go outside the root cause.

But we are not disagreeing about a great deal! I accept your point about the management issues, ATCO etc. I would just like to see a better methodology used that would bring to the fore these issues as well as design, operations and the many others involved in a complex business.

vapilot2004

In my earlier comment "when in doubt, blame the crew" I should
have included this caveat:

Aircraft system operational status and performance data, when known can be subject to more than a one in five chance of error (20%), due to 1.) lack of a detailed, constant & consistent stream of raw data or 2.) applying imperfect interpretive methods - yet still be accepted as 'fact' while the crews actions are typically subject to more black & white standards and remain based on speculative, predictive behavioral factors.

Boils down to:
If the widget says "I've been told to do X", and there is a commonly accepted means for the crew to directly cause widget X outcome, barring evidence or data to the contrary, it is almost always assumed said widget was commanded to do X by crew actions alone.

As another example, in many cases, particularly on older aircraft, there is a monitoring limitation in the command/output loop that cannot be broken. Which came first, the command or the output?

It often takes additional incidents & accidents, usually where more physical evidence is found, to discover flaws in the previous investigation's assumptions. In the case of our investigating authorities and their 'final' rulings, recanting the placement of blame on crew is rarely if ever done. This has more basis in the political aspects of the investigating body rather than the 'purer' aspect of retaining scientific accuracy.

The above has nothing at all to do with the loss of AF447, it is merely offered as a reflection on how our NTSB has been known to operate in the past.

I agree 110% with this CP.

con-pilot

Okay, poor verbiage on my part. We were drilled to never guess, never assume and never pass judgment until all of the final facts and details are in. However, being human, when you come up on an accident site where someone was buzzing something and flew into a set of power lines it's pretty darn obvious what/who was at fault. The toxicology report on the body of the pilot will tell us whether the pilot was drunk, high on drugs or just plain stupid, most cases like this stupid wins.

There is a phrase called 'Designed Induced Pilot Error'. Many times it will be determined that a certain design feature will be a secondary or contributing cause of an accident. The easiest and best example of this is Beechcraft. For years Beech, for reason only known to them, had the throttle quadrant designed as follows, Prop, Throttle, Mixture, every other aircraft manufacture had/has the engine controls in this configuration, Throttle, Prop, Mixture. More than a few accident were caused by this difference by pilots whose most experience was in standard configuration aircraft. Beechcraft finally changed to the standard configuration.

Not wanting to use up any more space here, I'll be brief on the rest of your second point. You must take into consideration all of that you posted and the NTSB does. Figuring out why the crew/pilot made the error is just as important, possibly more important, when determining the cause of any given accident. All the things that you listed are researched and examined, if any of those items are found to be a factor in the cause of the accident they will be listed as a 'Contributing Factor' to the cause of the accident. Hundreds of rules, regulations and aircraft design changes have come about because of 'Contributing Factors'.

There will never be an 'accident proof' aircraft built, at least not until a long time in the future. If it is designed by man, built by man and operated by man there will be accidents. The aviation industry has made fantastic advances in safety in the last few years. Go back and study aircraft accidents from 1950 to 1970 as I have done and I'm sure you will be surprised, first on the number of airline accidents and by the reasons some of these aircraft crashed, doing things that are done daily in today's modern designed aircraft.


In closing, you must remember that there are thousands of pilots flying thousands of aircraft all over the world every day without incident. Then one aircraft, out of all those thousands, will crash. It will crash while doing the same thing all of those thousands of other aircraft were and are doing on a regular, daily basis. It is the job of the accident investigation team to discover why this crew, on this aircraft crashed when other crews and on the same type aircraft, were operating in the same or similar manner, in same airspace, nearly at the same time did not crash.

Remember, there is very seldom a single event that will cause an accident.

grizzled

Who's on which thread?

CR2 -- I too thank you for your active involvement in this thread (and forum). I really do appreciate the difficulty of moderating these things, and I would not want to take on that task.

Though my point in answering your question about this thread may have been written tongue in cheek, I think the last two pages here show I was more correct than we all might have thought. The discourse here on methodologies for investigating incidents (in any field of activity), their strengths and weaknesses, is both highly professional and enlightening -- even to us old farts.
Like con-pilot, I started in this business a LONG time ago. At the risk of inviting a slew of posts questioning how people like us have lived so long, I will tell you that the first accident I was involved in (as an investigator) involved not composites, nor carbon fibres, nor even aluminum(ium). Twas wood and fabric.

Back to my rocker . . .
Grizz

Will Fraser

Waldo Pepper's partial inside loop?

BOAC

- post numbers adjust as deletions occur.

vapilot2004

As an aside, Permalink works as advertised.

ClippedCub

I see your point. Not only is it more complicated, but the A330 has to contend atmospheric conditions to further muddle the speculation pie.

lomapaseo

Michael Birbeck

[quote] @lomapaseo


Quote:
Of course all investigators speculate within narrow confines, "what ifs, check this check that, discard this etc."

Yes your points are valid. There is nothing wrong with forum constrained speculation and it does serve to enlighten us to ideas.

But some folks forget the limits of validity to their speculations (tested hypothisis with supporting and contradictions). Just look how long some of these theories last in post after post when there is no new information added or recognized.

I look every day for new information and find that it is rapidly drying up but the post counts for some are still going up while other more senior members have dropped off.

As far as conspiracy thorists all you have to do is see the number of posts casting aspersions on the french investigators as well as Airbus to cover up the isssues or unjustly blame the pilots.

And I guess I can't comment on what Airbus might have said since the words have not showed up in my read of the posts scattered all over this web site.

You can bet that some of the formal investigators are reading these web sites. However I don't believe that they would have a problem with the technical speculation or searching for how it works as much as they would have a problem with the finding fault with the safety related decisions before the facts are developed as contributing causes.

As always I urge any readers to read beyond the lines the posters words to find the few pearls of wisdom (there are quite a few good words here lost among the clutter)

Rollingthunder

slight thread drift

A bolt from the blue blasted a hole through the nosecone of a military aircraft visiting the local air force base this week.

The Royal New Zealand Air Force aircraft was training with Canadian Forces beneath what appeared to be benign clouds when one of the clouds suddenly began to spit hailstones.

"All of a sudden there was a massive boom and the aircraft was enveloped in light," said Wing Cmdr. Nick Olney, tactical coordinator on the flight.
The sound was quite something else."

A bolt of lightning had suddenly hit the plane, a P-3 Orion, right on the nose at around 1:30 p.m. Tuesday.

The blast punched a hole through the nose, traveled the length of the plane and then exited through a small hole punched in the tail.

The strike knocked out the aircraft's radar but the other systems were still functioning so the crew piloted it straight back to 19 Wing, where emergency personnel were waiting.

The lightning, said Olney, was completely unexpected.

"We were pretty happy with the conditions," he said. "The cloud that got us sort of came from nowhere."

Flt. Lt. Nathan McMaster was in the pilot's seat when the lightning struck right near his feet.

"I saw the flash first, I saw the bolt hit and at the same time we heard the sound," he said. "I knew it hit the nose and I felt a thump through my feet.

"A couple of the people on the flight deck with us yelped and definitely got a heck of a shock. And then it was (over) as quick as that and we were through it."

A lightning strike has been fingered as a possible contributing factor in the mysterious Air France crash near the equator that took the lives of 228 passengers.

"It's not that uncommon, but it's not common either," said Olney, who has been hit three or four times by lightning while flying. "Aircraft are designed to take this sort of impact.

"But you never know what you're going to get. You've got a massive amount of electricity going through that airplane -- it can disrupt aircraft systems. There have been occasions where the electricity has been enough to weld components together."

The aircraft is the New Zealand air force's version of the CP-140 Aurora flown out of CFB Comox. There were 16 people on board when the lightning struck. No one was injured.

Comox Valley Echo

Loose rivets

Not to mention the DC3 then Con?


http://www.douglasdc3.com/refurb/7refurb.jpg




With half a dozen hours of commercial flying in me log book, I was nearly killed by an ex BB pilot who closed the PROPS at Rochester instead of the gas.

We lifted into the air, plonked it down at a different angle, charged up the grass field - past the passengers waiting at the hut - and headed for the trees. 'Shall I get the wheels up?' cries I. He nods, eyes staring at the leaves. We brushed the trees and fell into a gap that had been made by the highway builders. He then did it again....but this time I was ready, and put his hand on the gas.


There were leaves in the gear when back at Sen

chuks

There you have a quote that reads in so many words that it's a moot point whether "badly informed" speculation on R&N is a bad thing!

Mugging grannies for their benefits, good or bad? Umm, couldn't possibly say, really. I know the grannies complain about that but then they are always complaining. Here, Air France just seems to have no sense of humour at all about badly informed speculation but is that just the French?

We all like to bash the ignoramuses of the press for their badly informed speculation yet it's a "moot point" whether it is bad.

Well, that's me told I guess because I thought it was bad, no question about it. What is there of value in that I cannot see? Otherwise these speculators might be out there on the streets putting themselves and others at risk by walking out into traffic while speculating? Instead of that they are tucked up safe and sound in front of a nice warm monitor.

Michael Birbeck

@Chuks

With respect Chuks you have partly quoted what I said and thereby changed the intended message slightly. What I said was

I am saying that an assertion that speculatiion (i.e. from the press, PPRuNe posters, anybody for that matter) is or isn't a "good" thing is debatable. This was said in the context of a quote in another post on R&N (now deleted) that pointed out that one of the principle players in this investigation was reputed to have said (something along the lines) that speculation outside the investigation was "bad". I appreciate that you might not have seen the original post because it was deleted..

I do agree that if all speculation (from wherever it came) was proven to be bad (ill informed, tendentious, insane etc), then the case against that kind of speculation would be proven and in essence it should stop.

However (as we both agree) some speculation is well informed, either from someone close to primary investigative sources or by dint of the fact that the speculator has deep knowledge of the issues at hand.


@lomapaseo

Agree entirely.

On that note I will bow out (but will continue reading avidly) in expectation of some new proven information with respect to this tragic accident.

BOAC

If the latest reports of 'previous' on R&N are accurate, I sense an extremely turbulent ride for both Air France and Airbus. A lawyers' field day coming, I fear.

chuks

It might be that your meaning was different from your words right there on the screen to be read but I didn't distort the sense of what you wrote, just dissected it to highlight the part I found relevant.

My own opinion is that speculation (in print at least), informed or not, is usually bad. (If it were really informed then it wouldn't be speculation at all but a statement of fact! The difference between: "Joe Bloggs crashed his car. He must have been drunk..." and "Joe Bloggs crashed his car. He blew 3.5 on the breath test afterwards so that he was drunk at the time..." See the difference?)

To begin with, not to come over all stuffy but, it's really insensitive to the people directly involved in this tragedy, very much along the lines of "Aside from that, Mrs Lincoln, how did you enjoy the play?"

Even so-called informed speculation (usually nothing of the sort) inevitably brings a horde of muppets in its wake, all eager to show off that they, too, are experts in their own way.

What is the latest? Some clown looks at unused crew seats to understand yet another "death plunge." (You know he's a professional because he left out the bit about the passengers screaming in terror.) Now me personally, I have no problem with that, imagining such a horror scenario but I bet there are a lot of folks out there right now who just do not need this.

You know how people can be, especially the French, when they have lost someone from their family or just a colleague or two. In a word, sensitive!

milsabords

Could the delay for replacing Pitot tubes be explained by supply limitations?
I would think that such tubes are built to plan, based on orders from ABI and other customers to fit new a/c, plus some limited contingency for replacements.
How fast can manufacturers react to a big surge of new orders?

Also, did the tubes on the 5 or 6 a/c which experienced icing problems come from a serie which could exhibit technical defects which might have slipped through quality insurance?

Michael Birbeck

You are absolutely right. We owe it to ourselves, this industry and to the public to be cautious in our opinions, written or otherwise.

arcniz

Everyone with two working frontal lobes comprehends that aviation is a rather vulnerable, small, and narrow business with a lot of complexity, a lot of business risk, and close connection between the fortunes of airlines, flight crew, other crew, airframe manufacturers, etc.

When eye-opening aviation accidents occur, free discussion about "possible causes" continues only until some theme or thread of causality, true or not, comes into focus as the likely mechanism. The spin-meisters on various corporate staffs and in certain public agencies seem to appreciate the relief of pressure from the press that evolves after a public "theory" of cause takes over and closes off most speculation about OTHER concepts of cause and effect.

One virtue of WILD SPECULATION - especially by people who are somewhat informed as to how things work - is that it surfaces information for discussion that one would never otherwise be likely to encounter. Two examples here:

a) Somewhere up in the R&N thread is a reference to a PDF about the AA-587 crash, written by a very legalistic and argumentative German Professor who has obviously been retained as an expert witness, asserting that everyone should know any pilot can immediately crash a heavy passenger aircraft by horsing around with the rudder pedals for a few SECONDS, so as to overstress and remove the tail. He claims that this proves Airbus has no fault, because, Airbus is no worse than any other any other aircraft manufacturer in this regard.

b) In the discussion of automated envelope modes on Airbus, a number of seemingly informed technicians and pilots seem to agree it is a FEATURE and not a bug that the protective modes can be stripped away summarily and without any effort toward graceful degradation when quality issues arise regarding the sensor airdata and/or function of various subsystems. In the scenario visualised for AF447, this behavior, alone, could largely account for loss of control of the aircraft in the turbulence around tropical TWX - at night. One moment you're flying a warm-fuzzy all-electronic aircraft at the edge of various envelopes, and the next moment you're in a snakepit with no reliable operational data, flying a completely different aircraft past the edges of multiple envelopes.

At one level all this commentary seems to make sense - we have little doubt that the engineers and builders and crews do and have done their best to deal with complex and difficult circumstances, given the resources and constraints with which they labour.

At another level, you have to ask yourself, why don't modern aircraft have rudder systems that are smart enough to prevent the vertical stabilizer from being overloaded under any circumstance short of collision? Could we design and build "smart tails" that would take this problem out of the equation for good? Heck yes!

Same general concept seems to apply for the all-or-nothing approach of going from compulsively autonomous operation to "ït's all yours, fella" on short notice. Surely, given the costs and risks involved in aircraft hull losses, a little more clever automation of the basic airbourne controls might be worthwhile. Can we not afford a fully inertial mass & motion guidance system which could be integrated with the enviro sensors to provide a truly independent opinion when ambient air sensors and systems are briefly or durably unreliable?

These are two examples that come forward in the context of "wild speculation". Perhaps there is more than the fine details of this sad circumstance to suggest the need for a broader and deeper "reality check" regarding some of the more slippery assumptions about evolving uses of automation in large civil aircraft?

Michael Birbeck

Having just read the link to Professor Peter Ladlin's site posted by mod PPRuNe Towers on R&N, I am fascinated to see that the professor has some comments to make on Bayesian inference on his site.

Bayesian techniques are almost certainly informing the search for the A300's FDR and CVR. I look forward to reading more on this in R&N.

Alongside techniques like root cause analysis, probability analysis is also a useful tool in the accident investigator's box.

Bayesian Belief Nets (Something for the Tech Forum I suppose).

@ArcniZ

As you imply Arcniz, there is sometimes little but semantics between a feature and a bug in sofware design philosophy and implementation..

lomapaseo

arcniz

I thought that I agreed with you about good and bad speculation, but your examples confused me.

I thought that the first one was fact while the last one was dreams

I'm becoming convinced now that good vs bad speculation is only in the readers mind.

At any rate we can't stop thoughts can we?