PPRuNe Forums - View Single Post - Boeing 737 Max Recertification Testing - Finally.
Old 9th Oct 2022, 01:17
  #772 (permalink)  
fdr
 
Join Date: Jun 2001
Location: 3rd Rock, #29B
Posts: 2,956
Received 861 Likes on 257 Posts
What I say here will annoy almost all drivers, that's OK.

We seem to have devolved into a compliance evidence based system that is obviously one way to do something, but borrowing shamelessly from The Princess Bride, where Inigo Montoya says:

You Keep Using That Word, I Do Not Think It Means What You Think It Means
The average driver can hop from a DC-9 to a B-777, put the pennies in the slot, get noise going, and follow a checklist, grab data and go aviate, following the jepp procedure, anywhere in the world. When they get a surprise, they will respond with "[expletive]".... and a few seconds later, confirm the blue is on the top, they are still flying, and ask for "...checklist please...." Pretty much the rest of the mess before during and after is lip gloss.

We don't bingle from a lack of knowledge normally, although on occasions, the drivers and their load have met their demise with no comprehension of what was happening. And that is actually the point; We lose planes for a failure of situational awareness in most events. 441; full back stick? what plane has anyone flown that will not desist the art of flying with full back stick for a couple of minutes? all the rest, 427, 585. Very few cases would insider knowledge massively alter the outcome, and in those cases, it was information that was not included as a matter of course in training programs. UAL 232, the guys are coping with a catastrophic event, and manage to keep the blue up top through both competency and a modicum of good fortune. Fantastic job, the last phugoid nearly turned it to custard, but their clean living came to the fore. UAL 881, the drivers had severe compound emergencies, and simplified the stuff to the point of being in a survival mode, and they got the bird back. Stunningly well composed team. Clipper Young America; RWY 01L, approach lights. The hole they dug was recovered from by using their inherent skills at the most simplistic level, to get the mortally wounded aircraft back onto the deck. XL888T, went out of their box ticking process into "there be dragons" and under perceived time pressure did an ad-hoc test point without proper set up, and then ran out of airspeed, elevator authority, and altitude without comprehending the condition they were in, "USE MANUAL TRIM", an advanced, directive lost in the clamor of all of the bells and whistles that they unleashed all because of an AOA probe that was frozen (sound familiar).

The surviving critical events are most frequently characterized by rational loadshedding of the cues and noise, and reverting to the simplest basics. We teach this, but the difference is getting to happen in real world while dealing with the sudden stress of a critical event.

Assuming that an advanced alerting system will make planes safer is unfortunately not supported by the facts in evidence.

Is flying different types that much of a complication? Maybe, maybe not. I checked my log back a ways, when I was doing some testing, I did 12 different fixed wing and a helicopter flights in the same day, and the only issue was it was hot and a number of the planes had no aircon. Point is simple, push/pull, chat, blue on top etc doesn't change between types. The emergency drills for the all start the same, blue on top; "checklist please..." Everything else is the same. The same thing will kill you in a C-150 or a PA-18-150 as will ruin your retirement in a B777-300ER, although to be fair, PT-22's were designed altered to kill students on base turn.... (thanks to Hap Arnold's input into design, good leader though). Losing a primary control is time critical, and that is where we seem to get discombobulated most, and our adherence to bureaucratic processes seems to be limiting. There are classes of aircraft that will behave in manners that are common, as that is how they are designed. There are some outliers, but a pitch issue in a B737 is similar to one in a B747, or an A320 if in alternate control laws. The aerodynamics are pretty agnostic. We are running through the UPRT stuff now, which is in essence a great step forward for aviation but we have added another level of bureaucracy, and my concern is that when trying to apply Ed DeBono's simplicity concepts to what remains a simple set of problems, we are adding more trees to the forest.

How bad is the problem? How many Amazon Prime customers missed their delivery after a B767-300F plants itself at a steep angle into the bay east of Houston? We did the same sort of stuff on the runway at Dubai with a B777 that apparently didn't have the skids added to the fuselage, the A320 that did the carnival 1 arrival into Bahrain, the B738 that put a neat hole into the ground proximate to Rostov runway. These crews got outside of their comfort zone, and did not have enough time or capacity to recover their SA and avoid trying to nudge the earth into a new orbit.

The large planes more or less (MD11's being an outlier) behave in a similar manner. Even T tailed vs conventional, until provoked they are similar (F100 being suitably odd, for reasons of design).

The problem with advanced alerting systems is, even the simplest get overlooked and the crew get outside of their safe space, and have issues determining they have an SA issue, and then have inadequate time to recover the SA loss and recover the aircraft to a safe state.

Having better alerting is fine, but our problems don't stem from the alerting, they arise from how we cope with SA loss in a stressful situation, and that doesn't have a simple solution from within § 25.1322. XL888T had a cockpit-full of state of the art alerting.... and that was in fact part of the problem.

Are there solutions? Sure, stop flying removes the risk, otherwise, the crews need to be able to comprehend their world in real time under levels of stress, and that is not an alerting function, it is a matter of simplicity such that the reduced bandwidth of the driver is not compromised. The recovery thereafter is a formality.

IMHO

[Before the bricks start flying, I can only talk from my experience, and that includes evaluation of over 3500 serious events that had 98% of cases where an SA loss was a primary factor in the event; from many major accident investigations where the crew have little comprehension of their dynamics until the moments at the end of the CVR/FDR recordings. In 110+ types that I've flown, the closest to a mean streak I have encountered in a plane was from the revised layout of the PT-22 following Hap's demand, which turned a sweet handling plane into a little rascal. Still really pretty though, and the engine note of a 3 stack exhaust Kinner radial puts a smile on your face...]
fdr is online now