Comment by hooby | Hacker News Reader

Comment by hooby | Hacker News Reader

Parent Full thread

Engines got a lot larger over time, more fuel efficient, more reliable, etc.

On a 737 those engines cannot be placed in an ideal position, because that would mean they'd clip into the ground.

The engines are being put above and I think in front of (might be wrong there) where they should be for maximal flight stability.

The original 737 was self-correcting - meaning that if for example the aircraft stalled, the resulting forces would cause the nose to dip down, pulling the plane out of the stall automatically.

The newer versions no longer have that inherent self-correcting ability - and instead use computer software to take over controls and pull the nose down, in case of a stall.

A new airplane design could be made to accommodate much larger, modern engines, while also having the stability and inherent self-correcting ability of the original 737.

But that would require changes to the overall shape and form of the plane - which will require complete recertification. By sticking to the old design, they avoid that.

ELI5! thanks so much!

so the crashes were because the new software was confusing to the pilots and did something unintuitive? and now they want to do another change to how the airplane behaves (interacts with) the pilot?

Afaik, the crashes were caused by the software mistakenly believing the plane was in a stall due to a single faulty sensor and pulling down the nose - while the pilots had absolutely no clue what was going on (because they never were told the software could do this), and had no way to disable/counteract the software.

I believe the updated software now uses both sensors and only acts if both sensors agree (which is actually standard for such systems and should have been done right from the start) - and can be disabled by the pilots, who are now aware of it's function.

But please don't trust my word... I'm no aviation export by a long stretch.

For a more qualified take on this, done by an actual pilot and pilot trainer, I recommend checking out "Mentour Pilot" on Youtube. He's got a number of videos on the 737 Max and the reported problems at Boeing.

> and had no way to disable/counteract the software.

Both pilots pulling up on their yokes couldn't override the system?

Another armchair pilot here, also recommending Mentour Pilot's channel, but hopefully I can summarize it well enough.

In this case no, because the way the MCAS system operated was to actuate the stabilizer trim which changes the angle of the entire huge tail plane (the horizontal part of the tail). For the most part the yoke controls the elevator (a smaller flap on the trailing edge of the tail plane), so the pitch control authority available in the yoke movement is not enough to override the adjustment to the tail plane trim MCAS was making.

Once MCAS moved the trim far enough, you couldn't raise the nose with the yoke alone. To correct it, pilots needed to deactivate the trim adjustment motor ("stab trim cut-out" you might have heard about) and manually re-trim the tail plane by folding out a literal crank and spinning a wheel by their legs in the cockpit[1]. The concept of doing this was not new with the MAX (similar procedures around trim system malfunctions existed), but what was new was that MCAS existed in the first place, that it could repeatedly activate if the erroneous inputs were still present, and that it could move the trim quite far before pilots realized what was happening. As I understand it that is pretty different than a traditional runaway trim malfunction for which they'd had training for. I believe that looks more like a stuck switch making a constant movement of the trim until you hit the cut-out. If you cut it back in, either it immediately starts moving again and you know you have a problem or it doesn't and leaves you wondering.

[1](https://www.trbimg.com/img-5d0ba28b/turbine/ct-biz-wp-boeing...)

MCAS uses trim rather than elevator? That seems insane, though I suppose trim has more authority than elevator...but yeah...in the time it takes for the pilot to notice the trim has moved itself to some insane level, they've been nose down for so long that they couldn't possibly recover.

Wasn’t this anti-stall software exactly what caused those two fatal crashes a few years back?

How hard is it to do a complete recertification. Given that these planes cost over $100 million each it seems like that probably wouldn't be a huge cost compared to the revenue that a new model of plane would generate.

It's not just the plane that needs recertification.

All the pilots will need to get a type-rating on the new plane type as well, before they are allowed to fly it.

And this is why focusing exclusively on Boeing in the 737 MAX controversy is, IMO, quite short-sighted.

It's not just Boeing's greed. Pretty much the entire industry has been happy to look the other way and enable their nonsense for years, since it meant they wouldn't need to invest extra in training their pilots.

Maybe the discussion should be about how much changes regulation should allow being done to a plane, before it must become a new type.

The 737 Max is like double the size of the original 737 - roughly twice the passenger seats, more than twice the weight, roughly 1.5 times the length... How much larger can a plane get, while still keeping the same certification?

Mmmm, I think there's some confusion here. The 737 MAX has its own certification, as has every generation of the 737 (Original, Classic, Next Generation and now MAX). However, if a pilot already has another certification of the same type then they are only required to take a differences course to operate other variants (which comes out to a few simulator sessions and a couple of check rides), as opposed to earning an initial type rating from scratch (which can require over a month of instruction).

Also it's entirely normal for a variant of a vehicle to have significant physical dimension differences. They're considered variants because their [chassis, airframes, hulls, etc] follow the same design and assembly, and because they handle in a similar way; they don't need to basically the same vehicle in a different coat of paint.

I guess a somewhat decent comparison is...Apple makes a type of computer (the MacBook Pro) and releases different variants (and within that, versions) of it. A fully specced out 2023 16-inch MacBook Pro and its entry-level 13-inch variant are both fundamentally the same type of computer even though the former is significantly bigger and beefier. Now, if Apple tried to shove a whole RTX 4090 graphics card into a MacBook Pro's characteristic svelte and relatively vent-less chassis we would say "no, that's stupid. Make a different type of computer if you want to do that". But that doesn't and shouldn't mean that a 16-inch laptop is a different type of laptop from a 13-inch one.

(As an aside, your numbers are a bit off, probably because you're comparing the wrong things. The "original 737" is two planes/variants, not one, and the "737 MAX" is similarly four planes not one)

Exactly, and that means that airlines like Southwest wouldn't even consider it.

One of the ways that low-cost regional airlines save money is by standardizing on a single aircraft. For Southwest, that's the 737. Every single pilot, inspector and mechanic in their entire organization only needs to be certified on a single airplane, they only need to have one parts supplier, only one set of tools, etc. - it's enormous cost savings.

Reply on news.ycombinator.com