Boeing may be now well and truly fucked | Hacker News Reader

Boeing may be now well and truly fucked

havecoffeewillwrite.com

24 pointsneverminder34 comments

Open articleView on HN

I've gone from admiring Boeing all my life to completely not trusting anything today's Boeing says after this scandal.

It's very likely that they are overrun by a bunch of MBA-ridden suits rather than engineers (not claiming all engineers are benevolent but that cultural shift always results in an erosion of quality), like most companies of its type. And no good decisions will be made.

I have always said (although without any supporting data), that the safest time to fly is probably after a plane crash because most airlines are probably double and triple checking their planes to be safe.

But that's not true of the 737-MAX. You can't fix what they've screwed up. Because the fundamental design change was borne out of greed, and not good engineering. I'll continue to avoid the 737-MAX like I'm sure many others will.

the 737max is not really any more unsafe than any other fly-by-wire aircraft - aerodynamically unstable aircraft are not really unusual, and this one isnt even really unstable in 99% of its flight envelope, just one specific place - and its not even unstable, it just handles differently than other 737's.

I agree with you, Boeing should have done a clean sheet redesign of the aircraft for this segment, not patch up the 737 for another decade of production - but I cant really tell them the reasons they made this choice are bad - airlines like and want more 737's, a completely new type creates a bunch of extra costs on all parts - so a choice was made, and now Boeing will live with the consequences.

We all will avoid the 373 MAX luckily, because it's blocked by every Airline right now.

It's crazy how anything in the universe grows and decays.. even when said thing is somehow sentient and knows about this

This is just a copy-paste of https://news.ycombinator.com/item?id=19758373 (which is the original source and on a site that isn't down)

I don't care what Boeing does, I won't fly the Max going forward unless it gets a triple-redundant redesign for MCAS. Yep, triple-redundant can also fail, but I expect Boeing to over-design the fix to raise consumer confidence.

The entire article is blogspam for this article:

https://nader.org/2019/04/04/boeings-homicide-will-give-way-...

Boeing still has Outperform rating from Leeham and others. It will be interesting who will be right: random bloggers or financial "analysts"

https://leehamnews.com/2019/04/24/initial-analyst-reaction-t...

> Using only one operating sensor (Airbus A320neo has three sensors)

I thought the boeing plane only needed the system because their engines were positioned so far forward, and that the airbus plane didn't have that problem, so shouldn't airbus not need the stall system?

Many planes have angle of attack sensors. Normally it is just a warning system. Boeing used theirs as input for their control loop of the MCAS system.

And most have redundant systems.

Airbus and Boeing take different approaches towards flight control. Airbus will override pilot inputs to keep the plane from stalling or otherwise flying out of the envelope. Boeing allows pilots to override the plane even if the plane thinks it's a dangerous maneuver. Airbus's system needs to be more reliable since a malfunctioning sensor will cause the plane to crash itself.

Wasn't the problem that Boeing requires pilots to keep overriding the system in the case of a failure, while not training them to do so, or even mentioning the fact in the documentation?

Yes, but the theory was that since the pilot could override the system it was less critical compared with Airbus where the systems controlled by the sensors are considered critical for flight. That's why Boeing used 1 while Airbus uses 3.

I see, though for the airbus I'd have thought a warning system would be better to implement than more complicated to implement control system.

Do you know exactly how fly-by-wire Airbus planes are? In an emergency, if the avionics computer goes crazy and starts making the plane do barrel rolls, can the pilot override it and take manual control?

>Do you know exactly how fly-by-wire Airbus planes are?

In normal operation it takes the input from the pilot and decides how to accomplish what it thinks the pilot wants. For example, if the pilot pulls all the way back the plane will translate that into the maximum safe climb rate. You can think of it as the pilot expressing a desired state and the plane figuring out how to get there.

>In an emergency, if the avionics computer goes crazy and starts making the plane do barrel rolls, can the pilot override it and take manual control?

AFAIK there's no way to disable the flight control. The computer is supposed to be reliable enough to not do crazy stuff and it accounts for sensor errors. Based on what sensors are malfunctioning different safety features are disabled. For example, if it can't get a reliable angle of attack reading the stall prevention is disabled.

Airbus systems have alternate law mode which kicks in when computers, sensors fail, read about Air France crash

I think this is true. Also, in the Air Asia crash when pilot disengaged autopilot by mistake using a reset switch changing flight mode to manual.

As far as I know, if the autopilot can't make reliable decisions, it rings alarm bells and disengages fly-by-wire. Looks to me like quite elegant design.

Most aircraft have the stall sensors for warnings, right down to small private planes.

The 737 Max 8 is different because the MCAS system actually overrides the pilot's input, rather than just sounding a warning to alert the pilot to a potentially dangerous situation.

Consumer confidence in the DC-10 disappeared after its closely-spaced and related fatal crashes. This led to the plane being relegated to cargo duties and, eventually, to McDonnell Douglas's financial failure and sale to Boeing.

The TL;DR? If people don't want to fly the MAX, there's nothing Boeing can do to fix it. They're better off abandoning the design and doing the clean sheet plane they should have done from the beginning.

Don't know why you're getting the downvotes.

Thanks for the TL;DR - as the site didn't work for me. Also, the analysis is spot on.

Here's Nader's original:

https://nader.org/2019/04/04/boeings-homicide-will-give-way-...

http://web.archive.org/web/20190426150157/http://havecoffeew...

I've read elsewhere that there aren't a whole lot of choices since competitors have limited capacity to ramp up? Also, some choices are also built by Boeing. (Such as other 737's.)

Other 737s aren't inheritantly unstable airframes that need software hacks to keep them flying.

Mind you, the entire MCAS issue would be mitigated, if the plane were re-certified.

The MAX8 is only really unstable in one specific set of circumstances where every other 737 would be stable. It doesn't need MCAS to be safe let alone fly. It needs MCAS so that pilots don't need to be trained (expensive) on the differences between the MAX8 and every other 737.

That one specific set of circumstances is that when the plane starts stalling during take-off, it wants to stall more.

Pilots need to be trained on that difference, because even with the software patches, MCAS will not correctly operate all the time.

That's not really accurate - the DC-10 was moved out of passenger operations because of operating costs - TriJets (DC-10/MD-11/L-1011) soldiered on in passenger roles until the mid-90's when better fuel performance from twin-jets and natural age based attrition moved them into freight roles.

ETOPS allowing dual engined aircraft to do routes that previously needed >2 engines sealed their fate.

Unfortunately, from what I understand, it’s more like MCD bought Boeing with Boeing’s money. Officially, Boeing bought MCD, but MCD executives, leadership, and design process replaced a lot of their Boeing redundancies.

Site doesn't work for me

same,

- google cache has no entry

- printfriendly fails on the same error

- can't try archive.is right now

Workaround: https://web.archive.org/web/20190426150157/http://havecoffee...

Reply on news.ycombinator.com