Hacker News Reader

Comment by anon-sre-srm | Hacker News Reader

They will try to MacGyver some shit together last minute.

The irony is the 707 replaced then De Havilland Comet 4, which is much like the 737 MAX today. The difference being Boeing is a strategic defense contractor that is "too big to fail". There are enormous pressures on the FAA to let Boeing slide with a stern warning again.

View on HN

Eh, I'm not sure that's a great analogy; the Comet 1 was the _really_ problematic plane.

The DeHavilland Comet was the first practical civil jet airliner ever produced. It literally invented the systems that were onboard. The 737 is a 60 year old design with a stellar safety record up to this point. It's not comparable at all. The only reason those planes fell apart is rotten management.

You're off your rocker. Comet's square windows were fundamentally unsound. 737 NG contains Ducommun garbage structural components deemed "airworthy", going back between 1996 and 2004. 737 Classic != NG != MAX. No amount of MCAS or engineering exceptions will make it so, or make it safe.

> Despite findings of the Cohen Inquiry, a number of myths have evolved around the cause of the Comet 1's accidents. Most commonly quoted are the 'square' passenger windows. While the report noted that stress around fuselage cut-outs, emergency exits and windows was found to be much higher than expected due to DeHavilland's assumptions and testing methods[122] the passenger windows shape has been commonly misunderstood and cited as a cause of the fuselage failure. In fact the mention of 'windows' in the Cohen report's conclusion, refers specifically to the origin point of failure in the ADF Antenna cut-out 'windows', located above the cockpit, not passenger windows.[123] The shape of the passenger windows were not indicated in any failure mode detailed in the accident report and were not viewed as a contributing factor. A number of other pressurised airliners of the period including the Boeing 377 Stratocruiser, Douglas DC-7, and DC-8 had larger more 'square' windows than the Comet 1 and experienced no such failures.[124] In fact, the Comet 1's window general shape resembles a slightly larger Boeing 737 window mounted horizontally. They are rectangular not square, have rounded corners and are within 5% of the radius of the Boeing 737 windows and virtually identical to modern airliners.[124] Paul Withey, Professor of Casting at the University of Birmingham School of Metallurgy states in a video presentation delivered in 2019, analysing all available data that: "The fact that DeHavilland put oval windows into later marks, is not because of any 'squareness' of the windows that caused failure."[125] "DeHavilland went to oval windows on the subsequent Marks because it was easier to Redux them in (use adhesive) – nothing to do with the stress concentration and it's purely to remove rivets." (from the structure)[126]

https://en.wikipedia.org/wiki/De_Havilland_Comet#Square_wind...

The Comet 4 didn't _have_ those, tho (its windows were round, precisely because of the incorrect perception that the Comet 1's window shape was problematic).

It was an antenna opening in the roof, not one of the - by the way perfectly fine - windows that blew.

> The 737 is a 60 year old design with a stellar safety record

It isn't. Or at least the max isn't. It's a big change from the original 737, a change that makes the type IMHO unairworthy beyond repair.

But that is not even the biggest issue, its the fact that boeing thinks they should withold essential information (such as the existence of an MCAS) from the pilots, who ultimaltely bear the responsibility of the safety of hundreds of passagers. Even now, in the light of what happened, apparently boeing still hasn't changed policy.

It was pressure from airlines to not have to retrain pilots, because that costs them money and whatnot. And in almost all cases when it comes to Boeing, the 737, and further issues, these airlines are also involved.

At this stage of the game, I seriously have to ask: is retaining the type rating actually a money-saver for the carriers?

I mean, airlines obtain new types all the time and pilots get trained to fly them. If it were so onerous, selling new aircraft would be practically impossible. (Yes, I know many Airbus share a type classification.)

It feels like there's an historical truth that's now no longer the case, especially when you consider all the associated compromises and, now, the resulting safety & engineering concerns.

Yes. Generally speaking, you want your pilots as proficient as possible on a single type. They will need many hundreds of hours on the new type before they are considered seasoned. A lot of disasters in the last 30 years have been amplified by pilots with low hours on the type they were flying. It takes a lot of training to memorize checklists, switches, levers, time to get muscle memory for various emergency situations. It’s a lot of time in the simulators. Lots of airlines also fly only one type, so they are majorly impacted by needing training for their entire pilot crew. It’s their own fault for choosing that but still. Combine this with the fact that airlines are razor thin margins and always seem a couple bad quarters away from bankruptcy and you get what we have today.

That's what everyone says, but I don't think the numbers are there. And, to be frank, most of the accidents you infer, are with non-US pilots, who's culture & training just doesn't match up with what the States has (hey, we invented the airplane).

[disclaimer: former chief dispatcher for a large regional US airline]

The problem isn't in retraining a pilot to fly the new plane. The problem is that this pilot is now no longer interchangeable with legacy 737 pilots, and cannot fly a legacy 737.

This means that if you have a hardware issue, you can't swap the plane without also swapping the flight crew.

There is a thing called equipment change. We use it all the time. At airline scale a captain (and crew) is worth an airplane. (And, come to think about it, it's a 1:1 relationship, so that follows Noether's Lemma, I think).

Don't overvalue the airship in the equation. (And I've done this job. I know what levers we can pull and how often it comes up.) Show me your data.

Reply on news.ycombinator.com