Pax sue Boeing in DBX crash

[wiedehopf]

i wonder how many improvements are prevented because the manufacturer is afraid of lawsuits.
probably important to have the ntsb issue an suggestion or even an ADR.
in that case changing something is not an admission of guilt. could still be used in court i guess but the manufacturer has no control over it so it's not a reason not to change sth

[HPSOV L]

It's all very well in theory but unless there's a strong muscle memory associated with repetitive practice of a manoeuvre then the intuitive response under stress may not always be what you might expect. That we mentally associate thrust increase with TOGA switch operation on takeoff could short-circuit the mental process in this unpracticed configuration (coupled with a startle factor from the RAAS).
Touchy feely nonsense? Given there were two guys with normal backgrounds at the centre of this it begs the question of what proportion of a representative sample of "normal" pilots would have made the same error under the same circumstances. 1%? 20%?
The exact combination of factors happens very rarely so the data is not available; however I think there is a fatal gap in airliner auto-throttle logic (and not just in this respect) that is fixable yet remains unaddressed.
The comments about getting back to basics etc, while understandable, kind of distract from the reality that humans will always be the weak link and the systems can be improved.

[Passenger 389]

are improvements prevented by fear of being deemed an admission of guilt

On 18th Aug 2017, at 21:42, wiedehopf wrote:

i wonder how many improvements are prevented because the manufacturer is afraid of lawsuits.
probably important to have the ntsb issue an suggestion or even an ADR.
in that case changing something is not an admission of guilt.

Did some quick research. (Seemed more interesting than what I was supposed to be doing).

1. In the United States, Federal Rule of Evidence 407 addresses this issue for cases litigated in federal court:

Rule 407 – Subsequent Remedial Measures

When measures are taken that would have made an earlier injury or harm less likely to occur, evidence of the subsequent measures is not admissible to prove:

•negligence;
•culpable conduct;
•a defect in a product or its design; or
•a need for a warning or instruction.

But the court may admit this evidence for another purpose, such as impeachment or — if disputed — proving ownership, control, or the feasibility of precautionary measures. (emphasis added).


2. Important: the Federal Rules do not govern cases litigated in a state court -- and this case was filed there. (Whether it will remain in state court, or be moved to federal court, is beyond my limited knowledge.)

3. Each state enacts its own Rules of Evidence. Sometimes those rules may mirror the federal rule (here Rule 407). But a state can also choose to write its own version.

4. Adding to the confusion, State legislatures sometimes enact laws that may be determinative on an issue yet are not found within the state's formal "Rules of Evidence." Appellate court decisions can have the same affect.

In 2010, Illinois promulgated a more comprehensive formal "Rules of Evidence." The Committee that drafted it cited that as a key reason why the new version was needed:

"Currently, Illinois rules of evidence are dispersed throughout case law, statutes, and Illinois Supreme Court rules, requiring that they be researched and ascertained from a number of sources."


5. So what is the Illinois version of Federal Rule 407 (subsquent remedial measures)? Hard to say. (groan)

The Committee that promulgated the "Rules of Evidence" (adopted by the state Supreme Court) wrote in 2010 that:

"The Committee reserved Rule 407, related to subsequent remedial measures, because Appellate Court opinions are sufficiently in conflict concerning a core issue that is now under review by the Supreme Court."

In other words, the Committee left Rule 407 blank for now ("reserved.")

6. As of today, the Illinois Supreme Court's website continues to show Rule 407 as "reserved."

Would have preferred to have a clear answer for you, but .......

7. You are correct that a directive of some sort from a government body might provide a little protection.

However, if a key question in the case is the "feasibility of precautionary measures" -- then the directive (and compliance with it) might tend to show the precautionary measure was in fact feasible.

[Onceapilot]

One thing that might have influenced events is throttle handling and power setting. Maybe someone can advise on the 777-300, how often pilots will hand set TO or GA power? Without TOGA activation, is there any indication of the actual GA power setting to refer to?
I can imagine a situation where a pilot is unsure what power to manually set, especially if there is no reference to what that power setting should be?

[tdracer]

There is a 'target' bug on the EICAS N1 display, and there is also an 'N1 Command' bug. Manual power set is done by moving the throttles such that the Command bug and the target bug pretty much line up. While the engines are spooling up there is also what we call a 'command sector' - an arc between the N1 actual and the N1 command. Manually setting N1 command to the target is pretty close to trivial (especially on the 777 which has what I believe to be the best throttle Boeing's ever done).
Sadly, if someone never does a manual power set, they likely don't know or have forgotten about it...

[galaxy flyer]

As a pilot, not an eminent engineer (), move the throttles until I start getting the performance I expect or need. One exception, C-5 AAR, doing so could result in significant damage, but the engineer's tone usually gave a clue.

[hec7or]

setting TOGA thrust without TOGA actuation will result in a no FD raw data GA, also undesirable.

[EMIT]

Simple

OAP

The thrust lever position for a Go Around is pretty simple - full forward until it is stopped by the forward limit of the slot it moves in. Rather hard to forget that one.

[Dan Winterland]

That's the Airbus philosophy and it works very well. Personally, I never liked this aspect of the Boeing autothrust system on the 744. I recall being told about it in training, but I suspect the lesson could easily be forgotten in the situation of a long landing with the RAAS going off.


I don't like RAAS either. My airline didn't fit it. It was considered to be too much of a distraction.

[FlyingStone]

Unless flap lever is at zero. Then it becomes very funny...

[RAT 5]

will result in a no FD raw data GA, also undesirable.

Undesirable, maybe; difficult for a properly trained pilot, no.

The thrust lever position for a Go Around is pretty simple - full forward until it is stopped by the forward limit of the slot it moves in. Rather hard to forget that one.

It was the case on the Boeings that I flew that GA thrust was a 'straight arm' not to the stops. It is also the case that full power is not necessary on an all engine GA. However, if made within kissing distance of the ground I can understand why guys might want all they can get, initially.
On B777 FBW is the power/pitch couple taken care of by more automatics and much of the pitch trim removed? On steam driven a/c it could be quite a handful at full power to keep the nose from being over enthusiastic as you pull back and power up.

[Onceapilot]

tdracer
Thank you for your reply. Can you just confirm what the target bug would display in a non TOGA situation that they had? Would there be a bug in that circumstance? Would it be a weight-related GA power? How do the pilots access the full thrust bug info, is the max always bugged? Thanks

[Onceapilot]

RAT 5
Yes, weight related GA is commonly used. However, full thrust is also handy to have bugged somewhere, in case of engine failure and GA.

[alf5071h]

Dan, #69, agree, and a similar dislike for RAAS, but why, yet there are some very good bits.
One aspect is that the ‘SmartRunway’ feature may be less accurate than pilot judgement, creating cognitive dissidence,- ‘do I believe this or not’ feeling, but of course it is installed to catch those conditions when the computation may be better than the human, but how might we know.
One question in this accident is if the alert was appropriate for the situation, i.e. proportional to the actual distance available and aircraft performance.

Another problem is that the ‘SmartRunway’ appears to be certificated as ‘advisory’, or at least that's the title of the system.
How can a regulator or airframe manufacturer approve a system which offers ‘advice’, requiring a high level alerting, but the advice given might not be the best option. The risk in this is passed to the operator, who has a choice for some of the functions and distances alerted, and in operation, there could be a choice for the pilot. Apparently none in this accident, follow the SOP, but what if the trigger conditions for the SOP are inappropriate. The crew would blamed for not following the alert, but also blamed for not ignoring it.

Then there is the interaction with the aircraft, TOGA. Who holds the responsibility for appropriate operation. The pilots should know but perhaps don't, the operator should but didn't think about any adverse reaction, the system manufacturer (RAAS), but is not his aircraft (and it only advisory), or the airframe manufacturer who ‘must’ know about TOGA, but didn't or did not have to know about add-ons (not my job), or didn't fully consider the interaction with an add-on installation.
What responsibility does the regulator have in this, certification, operational approval, do they talk to each other; problems of supplementary type certification (STC) ?

“Responsibility lies with those who could act but do not, it lies with those who could learn but do not and for those who evaluate it, can add to their capacity to make interventions which might make all our lives the safer.” Phillip Capper - ‘Systems safety in the wake of the Cave Creek disaster

[Sailvi767]

Interesting thread and comments. Perhaps the throttles should be designed so if you want power you simply push the throttles up! Oh wait, they do that already!

[hec7or]

RAT5

will result in a no FD raw data GA, also undesirable.

Undesirable, maybe; difficult for a properly trained pilot, no.

I think you have been a bit of a lightweight in your response here, a properly trained pilot will use his knowledge of SOPs as dictated by the SFI/TRE cadre, as opposed to a properly experienced pilot who will fly the aircraft comfortably using raw data.

Here's the issue, press TOGA and nothing happens, oh dear, advance the thrust levers to GA thrust, brilliant, now what? No FD! Something's wrong, but what I've got a full script of SOP call outs to make, but no prompts from the FMA. What do I do? No Flight freeze and no one to ask, only a first officer who is just as whacked out as I am after a long sector.

In the 21st century, we are scared to fly outside the SOP box due to the brainwashing from the SOP brigade and we have noise sensitive tracks to follow and altitude constraints to observe, so raw data is a brave option.

Until the OPC/LPC includes TOGA switch failure in the tick box, and an SOP is written, this will be a recurrent (no pun) challenge.

BTW, I've recently flown a raw data takeoff following TOGA switch failure and also a GA from platform alt where TOGA doesn't engage and can confirm that it's much harder than you'd imagine.

and... Flown gliders, singles, single engine jets, RAF, TRI/TRE

[Herod]

There is always a lot of talk about loss of flying skills. Yes, the accountants like to see automatic flight where possible, since it saves money. However, automatics should be considered aids. Maybe airlines should encourage manual flight below, say, 1000' in good conditions. In this case, had PF been using manual control, including manual throttle, might this have been prevented? Caveat: I last flew an airliner in 2004.

[CDRW]

And no one remembers this!!
Similar ??

[Ian W]

Originally Posted by Herod

There is always a lot of talk about loss of flying skills. Yes, the accountants like to see automatic flight where possible, since it saves money. However, automatics should be considered aids. Maybe airlines should encourage manual flight below, say, 1000' in good conditions. In this case, had PF been using manual control, including manual throttle, might this have been prevented? Caveat: I last flew an airliner in 2004.

It is actually a little worse than that. The certification of many of the automatics is based on them being considered 'aids'/'support tools'. This makes their testing and certification easier because failure modes can be expected to be dealt with by the flight crew. So the software is built with a design that when things get too difficult (the otherwise case ) it hands the decision to the crew as a bag of bolts for the sky-gods to take over.
Thanks to the actions of the beancounters the and senior pilot/managers that scrutinize FOQA and the SOP writers, the 'aids' and their 'advice' are required to be treated as infallible. But they were not designed to be perfect they are intended to assist and reduce workload.
This is a major systems analysis failure by the designers and a missed 'cheese hole' by the regulators.
Also, IFF, the systems are not for sky-gods but instead are designed to be perfect and not fail as the current senior pilots/managers/beancounters wish them to be; then there is no need for sky-gods, indeed no need for even cruise-pilots - as the 'automatics' now do it all themselves and do not have an otherwise case. That is where the attitude always follow the automatics, always obey RAAS, always use TOGA, always follow SOPs is leading. The day is not far off where even remote pilots will be unnecessary - because of the attitude to automation, exhibited by suing a manufacturer because the crew treated advice as an edict and were unable to carry out a simple go-around without the assistance of automatics that were limited in their operation to cope with possible crew error, without that limitation on TOGA to avoid human error the automatics would have worked alone RAAS could have triggered TOGA and the FMS fly the aircraft.

I (unfortunately) can remember when elevators all had operators, when trains always had drivers and when there were no automated cars. This court action will hasten the demise of the job of pilot.

(And before anyone asks - yes I would fly in an unpiloted aircraft - the one I am in is currently on full automatic operation over the East Atlantic and the automatics will be in control until a few hundred feet on finals in Atlanta, and that could easily become an autoland.)

[Capn Bloggs]

This court action will hasten the demise of the job of pilot.

What it should hasten is more sim time for exploring/practising weirdo stuff so we can better cope with a out-of-left-field event such as this.