Hacker News new | ask | show | jobs
by lmm 2591 days ago
> But the expected behavior which MCAS was to provide was not something invented only for "other 737's." Whichever plane would behave as the MAX behaves without MCAS turned on would also be rejected from certification.

No, it would have been a perfectly certifiable plane with a particular set of handling characteristics (including 777-like pitch-up). It just wouldn't be a 737, and would require its own type certificate. No-one's suggesting that large turbofan aircraft with pitch-up behaviour are inherently unsafe when flown by pilots appropriately trained and certified for them - otherwise the 777 would be grounded.

1 comments

> No-one's suggesting that large turbofan aircraft with pitch-up behaviour are inherently unsafe

It is as long it keeps 737 MAX body and engines and doesn’t have something like MCAS I.e. Inherently unsafe under conditions under which MCAS was supposed to turn on when properly functioning.

The certification requirements are the result of the clear safety goals not something invented “just so.”

MCAS was a type certificate hack.

The poster you're responding to is right; there isn't anything wrong with the behavior given the right training.

It's just that the investment in that training, and extra certification hoops to jump through would have made WhateveroModelNumberus MAX a non-starter.

It had to be a 737 to work at all.

> there isn't anything wrong with the behavior given the right training.

The behaviour without MCAS on 737 MAX is that minimal movements of pilot’s controls effectively activate what would be considered “amplification” of nose up movement, resulting in an uncontrollable plane and sure crash.

It’s definitely not something that pilots or passengers should be exposed to: being punished for approaching more dangerous position by plane forcing a deadly outcome.

Training pilots to not to move even minimally the controls in the “wrong” direction is maybe technically possible but in practice still totally wrong: It’s comparable to what Boeing told everybody before Ethiopian crash, and their attempts to blame the pilots. In reality, the pilots had almost no chance to rescue themselves and the plane.

In engineering the “positive feedback loops” (amplification of control inputs) are bad the “negative feedback loops” (correction of the input) are good.

The functioning MCAS provides a correction. The plane without MCAS amplification. Badly functioning MCAS also amplification and crash. That's why the wrong behavior was regulated, and that's why it had to be fulfilled for the certification. It’s that easy.

To convince me that 737 MAX without the "properly functioning MCAS" isn't inherently dangerous under higher angles of attack you'd have to provide some explicit proofs.

727 had about the same issue. Interactions with high lift devices would cause major problems on approach to stall.

The FAA certified it anyway. The U.K. gave it conditional certification contingent on the addition of a stick-pusher to be able to operate in U.K. airspace. See the Royal Aeronautics Society D.P. Davies Interview, specifically the 727 one.

There was quite a bit of controversy amongst test pilots at even granting the certification, seeing it as setting a precedent that would lead to a slippery slope that would culminate in less and less airworthy designs.

Nevertheless, the certification authorities accepted the argument that as long as instabilities could be countered by technological means, it would be acceptable.

Let me clarify though, that without MCAS, a responsible pilot would definitely be constrained to a much thinner envelope, but within that thinner envelope, the plane can fly just fine.

The deployment of flaps, also takes the plane out of a regime where MCAS is a factor.

So both legal, and practical precedent for it exists. Given additional training of course.

> The U.K. gave it conditional certification contingent on the addition of a stick-pusher to be able to operate in U.K. airspace.

Boeing 727 was clearly from another times: "As of July 2018, a total of 44 Boeing 727s were in commercial service" "Many airlines replaced their 727s with either the 737-800 or the Airbus A320."

> both legal, and practical precedent for it exists.

Does it? The devil is in the details. Speaking as an engineer, both the measurements of the ranges in which the changes happen and the characteristics of the responses to controls still matter. I wouldn't be surprised that it's still 737 MAX that would be "a precedent" with worse characteristics when the stall is possible (and without proper MCAS-like help) than those measured in 727.

It's the conditions under which the problems occur and the response diagrams that the regulators are supposed to verify, not the binary "has or hasn't" a problem near the stall. I'm quite sure that the technology at the time of 727 introduction was already more than capable of producing the relevant diagrams, so they can be compared. Thanks for specifying your arguments in the answer.

No problem! I'm as eager to get to the bottom of things as anyone, so I'm trying to be a careful steward of as much context as I can to keep discussions productive, and to rephrase in as many different ways as possible to increase visibility and reasonability to anyone who can help contribute more context.

>It's the conditions under which the problems occur and the response diagrams that the regulators are supposed to verify, not the binary "has or hasn't" a problem near the

Ah, I hadn't run into this tidbit before! Can you elaborate on it? I'd love to get some more detailed information if only to facilitate my own deep diving. I've been repeating the 727 simimilarity, and if there's any footwork I can do to make that more accurate, I'd be thrilled to run with it.

I do know Boeing was generally considered notorious amongst test pilots for knowing exactly how their designs would fly, so I can't imagine that those diagrams can't be found somewhere.

> To convince me that 737 MAX without the "properly functioning MCAS" isn't inherently dangerous under higher angles of attack you'd have to provide some explicit proofs.

I am legitimately curious, do you have any explicit proofs of the converse, that (as you say) "the behaviour without MCAS on 737 MAX is that minimal movements of pilot’s controls effectively activate what would be considered “amplification” of nose up movement, resulting in an uncontrollable plane and sure crash"?

The standard should be that a an unimpaired pilot properly trained in flying the aircraft will normally (practically) be in no danger of a "sure crash".

> The standard should be that a an unimpaired pilot properly trained in flying the aircraft will normally (practically) be in no danger of a "sure crash".

Exactly. And under that assumptions 737 MAX with no MCAS was never certified. Boeing didn't even want to admit that MCAS even exists to avoid even showing the flight characteristic of the plane "without MCAS." It's on Boeing to prove "737 MAX without MCAS" is safe, and up to now they did all they could to avoid that, and I expect they'll do more of that unless there is a pressure outside of both Boeing and FAA.

But there is a distinction between never getting certified because it is not safe, and never getting certified because Boeing didn't want to certify it as a new aircraft, because they wanted to sell it with the value proposition of not having to train pilots on a new type. Is that not correct?

So it seems to me that the discussion in this subthread is really around that point: could the aircraft have been certified as a new type without the horribly dangerous MCAS system in place to paper over the change in flight characteristics?

You seem to be saying no, and the other guy says yes, but you seem more to be talking past each other than offering proof of your assertions. He at least did point out (though without citing any proof, I think) that the 777 had a similar issue and was certified anyway (and I don't believe 777s have been falling out of the sky because of it). But, obviously, this is a different aircraft.