Behold the difference between the Federal Aviation Administration (FAA) and the National Transportation Safety Board (NTSB). The US House of Representatives Transportation & Infrastructure Committee‘s Aviation subcommittee held a hearing on May 15 focused on the 737 MAX certification process, crashes, grounding and other associated events. The FAA was represented by Acting Administrator Daniel Elwell while Chairman Robert L. Sumwalt represented the NTSB. Not surprisingly, Elwell took most of the heat during the hearing, but the way in which he responded, compared to the responses of Sumwalt, are telling in many ways.
The independent NTSB is focused on reviewing incidents, determining the root cause of those incidents, and making recommendations that will prevent those incidents from occurring again.
The agency is spectacularly conservative when it comes to assigning blame or even intimating that there are specific bits of blame to be cast before an investigation is complete. In the testimony today, prodded by Missouri’s Representative Sam Graves, Elwell showed little of that discretion.
Instead, “speaking as a pilot” he chose to call out the performance of the Lion Air and Ethiopian Airlines pilots as contributing factors in their respective crashes.
What concerns me about the data from the FDR is the apparent lack of recognition of the runaway stab trim. Runaway stab trim is taught in the earliest stages and it is so important that you don’t pull out a checklist; it is memorized and you are tested on it all the time. You turn off those stab trim motors. In the Lion Air accident it is significant that even though the aircraft was pitching against the pilot’s commands – that’s a classic stab trim – the stab trim motors in 13 minutes were never turned off. In the Ethiopian Air flight they did turn them off although they didn’t adhere to the emergency AD we put out. Subsequently, about a minute before the end of the flight they turned them back on. – FAA Acting Administrator Daniel Elwell
Representative Graves was happy to join in the blame game and was not nearly as subdued in his comments as Elwell. The xenophobia in this rant is astounding.
Grounding and ungrounding the MAX
Elwell offered some useful insight on the initial Airworthiness Directive and the specifics it included:
He also talked about the grounding and suggests that the USA was the first (then admits really second) country to react based on data rather than instinct. He does leave out the part where Aireon suggests it provided the same data to the FAA the day before it was provided to Canada. The discussion around “refining” the data remains very confusing as Aireon previously stated it only ever delivered the same set of raw data; nothing was further processed by its team.
That data-driven approach is critical to the success of the agency, according to Elwell.
What will it take to end the grounding? Elwell describes part of the process in this clip from near the end of his testimony.
Same question, VERY different answers
In two portions of the testimony the two witnesses were given essentially the same question to answer. The way in which they answered and the information they chose to share is telling. The NTSB comes across as singularly focused on aviation safety; the FAA, not so much.
One of those topics is flight deck cameras, something the NTSB has had on its list of recommendations for a while now but which never makes it past the FAA. Elwell’s explanation of that phenomenon is, well, phenomenal. And not in the good way.
For their closing remarks Sumwalt chose to address the pilot training issue that was mentioned so many times during the prior testimony. It was a coherent and thoughtful response to the prior hours of commentary.
We’ve heard questions about pilot training and the different standards around the world. If a manufacturer is going to sell airplanes around the globe then it is important that pilots operating those planes in those parts of the globe know how to operate them. To say that the US standards are very good and that this might be a problem in other parts of the globe is not the answer. The airplane has to be trained to the lowest common denominator.
Elwell’s statement was far more generic and without much significance.
Does any of it matter? Hard to tell at this point. But the difference is incredible to see.
A favor to ask while you're here...
Did you enjoy the content? Or learn something useful? Or generally just think this is the type of story you'd like to see more of? Consider supporting the site through a donation (any amount helps). It helps keep me independent and avoiding the credit card schlock.
As a non US viewer, I found this hearing dispiriting.
Boeing have clearly decided their ‘line to take’ designed to minimise any legal ramifications; protecting the compnay comes first. It was also clear their lobbying efforts were paying dividends.
The line – which has been pursued from the outset – of blaming the poor piloting skills in other countries (and the subliminal patriotism behind bigging up the inherent superiority of American pilots) was unpleasant to behold.
It reminds me of the banking crisis, where they were seen as ‘too big to fail’ and were therefore defended even when their role in precipitating the crisis was clear toa ll.
Well said. One repercussion that Boeing is overlooking by victim shaming the airlines that crashed and trying to share the blame with the FAA is the loss of business that will result. If a foreign buyer can’t trust Boeing and can’t trust the FAA, Airbus starts to look a whole lot better.
And Boeing tried to play the bird strike card. An engineering issue should not be fixed by software.
Seth Miller says
Sometimes the software is part of the engineering. In this case they are definitely intertwined. Not an excuse, but both hardware and software choices appear to have been bad in this case.
david innes says
There aint been an airliner which did not depend on software to achieve the required functionality. Without software we would be stuck at the early 707 level of technology
CHARLES S COLEMAN says
Absolutely agree agree with the software comments made by Seth Miller! Further, it is obvious Boeing is doing everything it can to protect itself from all the liability caused by these two Boeing aircraft accidents.
I see it as the FAA having given Boeing too much control to oversee itself and the lack of “real proper” oversight may have led Boeing to get a bit lazy. Thus we have seen the results. Remember, Boeing has told us that there were options carrier could purchase (extra cost) and it appears most chose not to purchase them.
Also, after the first accident, Boeing failed to immediately put all of its cards on the table; rather just sat passively by (perhaps as a one off accident) until the second accident and then it had to do something.
Look at how long this has been and this has yet to be fully resolved.
I believe both Boeing and the FAA share in these tragedies that should have never happened.
Thanks for watching C-SPAN so the rest of us don’t have to 🙂
In all seriousness, I’m not sure how I feel about the pilot training issue. If the US clears a plane to fly then it would seem that every other nation that flies that plane have similar certification processes; is this not currently the case? At the bare minimum it would seem that an airline should have processes in place to train pilots to safely and correctly fly the plane type. The NTSB comments that Boeing needs to train to the lowest common denominator seems short sighted if the rest of the world assumes training based on US standards; and wouldn’t that then push the training blame squarely on the FAA rather than Boeing? I don’t know much of the whole story but it seems that either the FAA needs to demand better training standards (that would thus apply worldwide) OR any country/airline that buys a plane needs to thoroughly examine training and needs before allowing the plane to fly. And Boeing should have, it seems, made more notice of the change to the MAX operation vs past 737 models.
Seth Miller says
There is something of a chicken-and-egg game being played, I suppose. On the one hand, I’m not qualified to be a pilot so training to my level is clearly not the right choice. But what determines the minimal qualifications as a pilot? And as a pilot for a certain aircraft type? If, as was suggested by the Administrator and Congressman, being able to hand-fly a 737 is a minimal necessary component then shouldn’t that be the training standard applied? Shouldn’t the simulators and other training curriculum test against those skills? Maybe they already do, but it does not appear that way, at least not right now.
I understand Elwell’s point that it doesn’t matter why (i.e. MCAS/AoA issues) the plane pitched nose-down so much as being able to quickly and correctly respond to that situation. At the same time, however, perhaps not designing (and certifying) planes that can decide to erroneously pitch down based on a single input should also be at least considered. That’s a blame game that really, really doesn’t look good for Boeing or the FAA and I think that’s what Sumwalt was getting at with his comment at the end there.
It should be noted that Mr. Elwell has been acting head of the FAA since Jan 6, 2018. The President just got around to naming Stephen Dickson to be permanent head of the FAA in March, almost 17 months after the previous director stepped down. Mr. Dickson is currently going through Senate confirmation hearings.
John Eogerson says
Wow. The only xenophobia is see in this article is from the author. It’s a fact most foreign countries have more lax training requirements for pilots. To make bigoted claims against the Boeing employees for pointing this out is a form of xenophobia.
It’s a fact?
Want to share your evidence for that?
Seth Miller says
That’s not what xenophobia means. Also, I didn’t think the Congressman or the Acting Administrator was an active Boeing employee, but I could be wrong there. Are we basing that on PAC contributions or more traditional payroll?
Beyond that, I’m taking the word of the many, many pilots not on Boeing’s payroll who all suggest that this is a bad design. So do many independent engineers. No system that is designed to assist in maintaining control of the plane should have a set of circumstances under which it creates a impossible or nearly impossible set of control conditions for the aircraft. Especially not from only a single data input.
I am not claiming – and never have claimed – that the pilots acted perfectly in these incidents. But starting and finishing with the blame squarely on them given an aircraft with a control system designed so badly that the planes have been grounded for months while Boeing works to rebuild it is all sorts of wrong.
david innes says
There are several drivers to my perception.
1) The desire for a Common Type rating, in spite of significant system changes, hence IMHO the limited differences training – less training= less cost to customers…=product more attractive Talking about pilot skills varying is just a touch racist, but they should design sysytems, manual and training for the actual demonstrated skill levels (and language dfferences)
2) Protecting the company. From the timeline I reckon they knew what had happened after crash 1 and hence the document they produced- No SW changes, no manuals changed, no training changed but the bit of paper, which cost very little gives them a level of “we told you so”
3) The fact that the updated MCAS SW was available so quickly might lead you to think they were already aware and had the fix, but decided to withhold offering it, since an AD must be paid for by the supplier, not the airline(s)
Can anyone see a common thread?
Its not how I would have done things in the FAR 25 airliner programs on which I was Lead “DER” for all avionics /electronic aspects of flight control.
But then again, we designed these systems using multi channel architecture from the beginning, so no single sensor faults could cause …. and I sleep well at night with 2500 ish airliner flying around, with NO FARE PAYING PASSENGERS killed or injured due to flight control failures in millions of flight hours..
I feel charges of corporate homicide waiting in the wings…
BTW we had the equal but opposite issue with engines on one of my airliner projects.
Tail mounted engines, moved aft with a fuselage stretch, so the nacelles added to the effective fixed area of the tailplane (Horizontal stabiliser) making the aircraft more stable. We had to make the elevators bigger!!
On another program, to meet elevator feel force rules, due to downwash changes we added a function to put in a “step” of trim, with Flaps… but it was a “single shot” and required two computers to agree. One computing the change, and asking for trim, and the other doing the same calculation but then only enabling trim for that one shot.. (as long as AP was not engaged) But note the need for two computers to agree AND the single shot, so single faults could not cause a hazard, and the trim system did not fight the AP
Same for mach trim… I could go on
Seth Miller says
A desire to avoid sim time for the pilot transition was (and is) a clear factor in the design decisions made. There’s a huge cost factor associated with that. Reportedly tying the system to both AoA vanes would require the sim, but I’m not entirely clear on that. Still a poor choice to design a system that has control over aircraft movement without the built-in redundancy that aviation historically has been famous for.
CHARLES S COLEMAN says
Edgar Numrich says
What is increasingly clear from these testimonies: “It’d the other guys’ fault.” The “system” of aircraft and aviation safety is both corrupted and broken. I suspect it’s been that way for some time (just as with the big banks). And certainly not ironic these crashes occurred during the Trump presidency: The blueprint is on everything the president touches,.
Seth Miller says
FAA approval of the MAX came in March 2017 after a process that ran nearly entirely under the Obama administration. I’d be careful about accusations/assertions that either political party is culpable here.
david innes says
While the philosophy may echo the orange one, you cant allocate this catalogue of debatable decisions to him (I never thought I would defend the great Donald, but we must avoid distracting these major issues with political bias)
Doug Lynch says
The F-16 was one of the early fly-by-wire aircraft. It has a great flight control system back up: an ejection seat. I know that with current designs, hardline controls are no longer in place, but without them, the pilot never has full flight control authority.
Ferris Bueller says
Former Boeing engineer here. It was an unconscionable design decision to have a redundant AoA input and not divert to a correct reading when one failed, or at least alert the PIC to mismatched readings. As for your claim that Boeing had the SW patch completed and just didn’t release it (due to the cost of an AD), I greatly doubt that. The SW patch / logic change is a relatively simple one for flight controls geeks, and probably took all of a week to resolve. Regarding training to the lowest common denominator, well, the pilot community should tread carefully there. A decent AI program may already be available that meets that requirement.