Comac engineers miscalculated the forces that would be placed on the engines | Hacker News Reader

Comac engineers miscalculated the forces that would be placed on the engines

reuters.com

62 pointsinferiorhuman70 comments

The recent failures of Boeing and this story underscore how very, very, very, very difficult it is to build a modern airliner.

I'm hard pressed to come up with any other product which has as many complex interacting systems, requiring expertise in aerodynamics, materials science, mechanical engineering, electrical engineering, software, UX design among other things, with very little margin for error. Honestly they're kind of amazing.

This is sort of one of the metrics James Fallows outlined in his book, "China Airborne". In order to build a competitive airliner, it requires integration of so many discipline: metallurgy, software, aeronautics, engineering, etc. He argues that when China can build a competitive airliner, their development can be said to have to be on par with the West.

It's worth noting that Brazil[1] has been exporting aircraft for quite some time, although it's still regarded as 3rd world.

[1]: https://en.wikipedia.org/wiki/Embraer

Brasil has benefitted a lot from technology exchange with Saab. For those of you who are not familiar with the concept, the idea ,(very simplified) is that Brasil buys a lot of Saab planes, and in return gets knowledge about latest Swedish advances in the field. And Sweden, while perhaps not best in class, builds decent military planes, and I assume at least some of the knowledge is transferable to civil applications as well.

That's almost cheating given Santos-Dumont's role in the origins of aviation. That said I don't think there are any modern, high bypass turbofans coming out of Brazil. China will need those if they hope to have a fully homegrown solution.

3rd world doesn’t mean poorly developed.

It just means not 1st world (NATO countries) or 2nd (Soviet Bloc)

So for example Finland and Sweden are third world countries?

Yes.

Also Ireland, Switzerland, Austria, and almost all of the Middle East.

That was a very good book to read.

As a quick summary:

Boeing is commendable on pushing China’s aviation industry to adopt international safety standards, as well as certain key Chinese individuals that wanted China to improve themselves and be on par with international norms.

The key here, is that the international norms and safety standards were established by the FAA and American industries. But to be politically neutral and more palatable, it was rephrased as “international standards”.

But, Boeing’s goal was to improve the safety of China’s aviation industry, in order to avoid being caught up in bad press for one of their planes crashing, because of poor maintenance. And they figured by improving the safety, more people in China would fly, and thus buy more Boeing airplanes. It was a nice virtuous cycle for Boeing.

The interesting thing, is that it appears the student has now become the master, and China adopted all the international standards from Boeing, and added more of their own, that they became even stricter, and more rigorous in their safety inspections and protocols. And over the past 2 decades, their statistical safety record is better than their international peers.

And also, the other fascinating thing about the history of Boeing. Their first engineer was Wong Tsu, and he designed Boeing’s first commercial plane, the Model C water plane. This was instrumental to Boeing’s commercial success. Wong Tsu was born in China, and studied at MIT during the dawn of the airplane age. It seems the United States and China have a lot of connections going back in history.

Noah Smith has a theory that one of the best ways to develop a country that’s nearing an inflection point is to commit to building something like this. His example is a coastal African country committing to building an Aircraft Carrier. You would need to develop advanced capabilities across a dozen industries and it could realistically kickstart competition in every one of those areas as the local skills level up. It’s a provocative but plausible idea in my opinion.

I always had the imagination that the aircraft carrier wasn't that hard. The planes that can operate from the carrier. I imagined were harder.

I think you'd need a few orders of magnitude more skilled workers to build a carrier vs. building a new airframe. I think the end skills will be much more useful to a developing economy as well (since they're better suited to heavy industry vs. more technical skills when developing a plane) but I could definitely be wrong there.

key word being modern. it's easy, with today's tools, to build a plane that flies and carries passengers. it's very hard to continue pushing the bleeding-edge of fuel and cost-efficiencies.

> any other product

You can design a car with very little math. An airplane requires a lot of math. A spacecraft - everything is math, because the margins are awfully slim.

By the way, a huge reason the Wrights were successful with their 1903 Flyer and the others weren't was because they did the math, and the others were doing trial and error. Or more accurately, just error :-)

The Wrights are often dismissed as mere "bicycle mechanics", but they were anything but "mere".

>You can design a car with very little math.

I'm not trying to make it seem like car design is a higher class of work than it is, but a car designed with little math isn't going to run in any meaningful manner.

Margins/thresholds for design scope exist everywhere.

> I'm not trying to make it seem like car design is a higher class of work than it is, but a car designed with little math isn't going to run in any meaningful manner.

I've seen a lot of custom cars built. No math was used.

Indeed, one of the few exceptions is basically a faster version of the product that goes to space.

I'm not sure it's harder. More countries have successfully shot stuff into space than have build an airliner industry (US, Canada, France, Russia, Brazil).

More so thinking of reusable spacecrafts like the Shuttle or Falcon rockets.

Space shuttle? :)

Sometimes I think we're in some weird version of a neo-Dark-Age where the ways to do everything amazing that we figured out how to do start to atrophy along the same path as those things being taken for granted. Because they're no longer considered amazing, so they're reprioritized as such, and we forget how hard they actually are to do as well as they need to be done.

We sort of seem to end up with a built-in assumption that absolutely everything can be commoditized just because we're used to having it around, not because it is actually able to be commoditized effectively.

I think this ties into the importance of living knowledge. Even the most meticulous notes must be interpreted and executed by someone, and there is no substitute for an experienced person. This has already happened with technologies like vacuum tubes. It isnt a simple matter of economics to restart production because much of the know-how must be rediscovered. In the absence of the original volume of demand that led to the first generation of tube engineers, it may be lost forever. There are some niche high power RF devices still using tubes from decades ago.

Even in something as important as nuclear weapons, the method of manufacturing a critical component was lost due to a broken chain of living knowledge:

https://en.wikipedia.org/wiki/FOGBANK

Probably why Idiocracy will not come to pass. And truly idiotic civilization will succumb to its competitors.

All the more reason to avoid a single global government. Besides never being able to escape tyranny.

Eh? This is China's first time trying to build a commercial narrow body airliner, isn't it? So its not like they used to have the capability and forgot how to do it, they're developing it for the first time.

In an ideal world, they would be able to start with all of the world's knowledge in this subject. Ie. they would start with where Boeing is now.

I think that is what OP is trying to get at.

Yes, this. Also, that even Boeing partially lost track of it along the way.

This is China's first time trying to build a commercial narrow body airliner, isn't it?

Second – the ARJ21 was the first.

Why not buy the remnants of fokker or dornier and carry on from there?

Dunno what's left of either of those two companies (or the willingness of the EU to sell to China) but neither really had anything that would compete with the A220/A320/B737.

The ARJ was an attempt at a refreshed MD-80 with tooling from McDonnell Douglas and a ton of outsourced bits and it still flopped.

Per the article COMAC is resorting to espionage, so I imagine they think going the legal route won't pan out for them.

Jonathan Blow gave a great talk on this: https://www.youtube.com/watch?v=pW-SOdj4Kkk

I stand by the assertion that we could not go back to the moon even if we wanted to right now. They had slide rules instead of computers, but a passion and attention to detail that I very rarely see in any workplace these days. Maybe they were just less distracted than we are today by our current culture of distraction.

NASA engineer here. I assure you we would be on the moon right now if it was a top priority in the Decadal Surveys. Its not. Recently, there's been a re-rush to the moon policy wise, and its been utterly delightful to watch astrobotic and several other companies step up to the plate so we dont have to. I was, just last week, working on a robotic payload for just one of those lunar landers, and my friends are working on even more ambitious ones.

They also had an effectively unlimited budget and hundreds of thousands of people to throw at the problem. You can often lay hands on some smart people under those conditions.

A lot of the parts in these planes + the automation supplied to build the planes are from companies owned by western companies. China is trying to own more parts of the process (Broetje acquisition for automation, things like Kuka and other industrial companies are getting bought up by Chinese companies).

https://spectrum.ieee.org/aerospace/aviation/chinas-comac-to...

> The engine miscalculation does not reflect a lack of theoretical understanding - China has been putting people in space for almost two decades. But it does illustrate the national aerospace manufacturer’s lack of experience in designing and building commercial aircraft.

Does this line in a Reuters article suprise anyone else as editorializing? What does COMAC's theoretical understanding have anything to do with the Long March rocket's manufacturer CALT (China Academy of Launch Vehicle Technology)?

"Someone in China could have done it, but the knowledge and experience is not as widespread as it is in other nations"

It gives context on how to interpret the two results. China can do great things, but it may not have the pipeline of engineers set up to enable widespread commercialization of those great things.

When talking about the duopoly, the article omits Sukhoi Superjet, though it seems to not be an easy alternative airliner: https://www.airportspotting.com/cityjets-sukhoi-superjets/

The comments on the page are really good. It seems the plane itself was good but simple lack of culture of fast spare parts delivery made the planes unprofitable.

I wonder if the fact that Comac is a state-owned company has anything to do with these problems. From what I have read, Chinese state-owned enterprises are inefficient and can stay in business only with major help from the government. But maybe Comac is an exception.

Minor detail: C919 somehow strikes me as a non-impressive-sounding moniker. The sound of it is somehow "off".

Does it "sound" better in Chinese?

Not sure how to explain it better.

How impressive does B747 or A380 sound? Manufacturers have largely stopped giving fancy names to airliners, and I'm sure Boeing has some regrets about calling the 787 'Dreamliner'.

919 is one of the greatest Porsches ever so some marketing dept thinks its pretty good.

see the wikipedia entry on Chinese numerology, esp. number 9.

https://en.wikipedia.org/wiki/Chinese_numerology#Nine

tl;dr - Chinese pronunciation of 9 sounds similar to the Chinese word for longevity/long life.

You'd rarely see the number 4 used since that sounds like the pronunciation for the word 'death'.

8 is another popular number, synonymous with wealth.

Go into a parking lot with many Asian drivers - you can pick many of the Chinese drivers just by looking at their license plates and checking if they have one or more 8's or 9's (and lack of 4's) in the license plate.

Applies to many number-based items - phone numbers, highrises targeted at Asians may skip the 4th and 14th floors (haven't been in such buildings with more than 20 stories, but I'd assume the pattern would extend to any floor with a 4 in it - no idea about buildings with more than 39 floors)

Once you know about it, it's hard to unsee...

OMG, that explains so many things.

Thanks!

What's a Comac?

Literally the first paragraph of the article:

> BEIJING/PARIS (Reuters) - Development of China’s C919 single-aisle plane, already at least five years behind schedule, is going slower than expected, a dozen people familiar with the program told Reuters, as the state-owned Commercial Aircraft Corporation (COMAC) struggles with a range of technical issues that have severely restricted test flights.

I don't bother reading things unless I can confirm my interest in the article. I can't do that with unknown acronyms.

I guess you never read any article about any startup then.

Basically the Chinese equivalent to Boeing or Airbus.

Very annoying when you choose to reject cookies in the optional cookie consent menu and the page refreshes to ask you again

well, if they cannot store a cookie, how would they remember that you said "no cookies" ?

There should be an option to only reject some cookies. No to tracking, yes to chocolate chip.

I'm curious of COMAC's engineers are facing similar levels of pressure to cut corners as did Boeing's engineers while working on the 737 MAX.

Does anyone know if / why we should expect COMAC to avoid repeating Boeing's institutional failures?

EDIT FOR CLARIFICATION: I'm not assuming that Boeing and COMAC face the same kinds of pressure. I'm asking if they face similar levels of pressure to have a design ready for large-scale roll-out.

Based on the very different commercial, cultural and political constraints they operate under, I would imagine any institutional failure would probably be unique to COMAC and certainly not a repeat of Boeing.

Boeing was cutting corners to try and have to avoid pilots being required to certify on 737-MAX if they have certs for previous decades of 737s.

There's literally no way that any state flight administration would allow any previous certifications to apply to such a new airplane, so it's more just growing pains of a new design.

Boeing were not cutting corner with MAX. They were trying to fit square peg in a round hole.

They tried to fit clean sheet design performance increase into age old mold. The fact that they mostly succeeded shows how amazing and talented engineers are there.

Perhaps this is just a growing pain, and they’ll eventually figure it all out. And it’s a good thing they did extensive testing and caught this now, while it was still in development. This is where you are supposed to catch these errors.

Maybe this article was purposefully released to spread lies and defame COMAC’s reputation? Because that is what western journalism is pushing out these days about China.

But the bottom line, is that COMAC didn’t have 2 of their planes fall out of the sky, like Boeing did.

> Maybe this article was purposefully released to spread lies and defame COMAC’s reputation?

I don't think so. I would guess 90+% of the people clicking on this had no idea what COMAC is. What does seem weird is that the HN headline "Comac engineers miscalculated the forces that would be placed on the engines" does support defaming COMAC narrative, while the actual Reuter's headline "China's bid to challenge Boeing and Airbus falters" reads like the dozens of articles that were written about delays the Dreamliner [1] and the A380 [2]. Compared to some of those articles COMAC is getting off easy.

[1] https://www.google.com/search?q=boeing+787+dreamliner+delays

[2] https://www.google.com/search?q=airbus+a380+delays

What does seem weird is that the HN headline "Comac engineers miscalculated the forces that would be placed on the engines" does support defaming COMAC narrative, while the actual Reuter's headline "China's bid to challenge Boeing and Airbus falters"

I picked that because having to redo the nacelles was new news to me. Most of the rest of the article covers previously published problems.

FWIW the 787 was hardly treated with kid gloves. It wasn't just random delays, it was things like hamfisted outsourcing, scrapping frames assembled with hardware from Home Depot, major customers refusing to take delivery from a specific factory, battery fires, and major engine problems.

> But the bottom line, is that COMAC didn’t have 2 of their planes fall out of the sky, like Boeing did.

Customers are not flying the COMAC C919 - it is not in passenger service.

I agree with you, I've seen articles like this on hacker news for the past 5 years about how China is failing and China is unable to produce what North America is capable of. And the interesting thing is yes, at first they fail. Why should we expect them to get it right the first time? You learn with practice, and mistakes, and testing. I guarantee eventually they will have a functioning aircraft. This is normal process, with a narrative western spin.

Why should we expect them to get it right the first time?

This isn't their first time (the ARJ21 was). Will COMAC succeed where Mitsubishi (Japan), Bombardier (Canada), Sukhoi (Russia), and Lockheed (USA) have failed? Maybe. But China's set a deadline of 2025 to have their own homegrown product in service and it's doubtful that specific goal will be met.

The western aerospace industry is hardly betting against China here. I think Airbus was expecting COMAC to be competitive by 2020. Hell, a huge part of the reason that the 737 MAX exists today is that the Boeing CEO thought China was their biggest threat.

Bombardier (Canada) failed because they had produced a better plane, and boeing made efforts to make it impossible to sell in America. So they've teamed up with Airbus to be able to access the market.

They said that about the Japanese. I remember around 1990 sitting as some of my fellow engineers said the same about Hyundai, they'll never be able to build good cars. I took the other end of the bet. That a country with enough resources with commitment to make the engineering and manufacturing cultural changes will eventually succeed.

Change since thing is Hyundai builds decent cars and a lot of other heavy industrial goods. And the United States has regressed because we believe the free market will provide while the third generation post war management class loots everything not nailed down.

Yeah but COMAC has maybe 2 commercial jets that have entered service, Boeing has shuttled millions of people around the world safely for decades. We're riding an aviation news scare wave right now, so every burp and fart from the industry is getting reported. That said, there's no question that COMAC doesn't have the experience of Boeing or Airbus in designing production airliners. It's like trying to throw a bunch of money at a new software team to replace Windows 10, and then get Big Corp to start replacing their machines with that.

This is exactly why it’s an embarrassment for Boeing.

The fact that they have over 100 years of experience doing this, but yet, they made an elementary mistake. Two of their planes fell out of the sky!

And at first, they blamed it on mediocre 3rd world country pilots.

Then, the truth finally came out. Their design for the 737 MAX was junk. It was fatally flawed. It’s a flying death trap!

They put on a larger sized engine to get more fuel efficiency. But moved the nacelle forward, which changed the flight characteristics, causing the plane to pitch its nose up. Which they tried to fix, by having 1 faulty sensor detect the balance, and to correct it with software to automatically adjust the flaps. But the sensor was wrong, and it wasn’t redundant, and the software over-corrected. And what’s the result? The planes fall out of the sky.

They paper-bagged one problem with another, and another. Then they lied and sold it to the public, like it was the safest plane ever made.

> And at first, they blamed it on mediocre 3rd world country pilots.

> Then, the truth finally came out. Their design for the 737 MAX was junk. It was fatally flawed. It’s a flying death trap!

Well, both are true.

The Indonesian investigators share blame between their pilots, mechanics and Boeing. Insiders expected this.

And Boeing forgot they were in the aerospace business.

I think the problems from Boeing mark a new era when you cannot rely anymore on a previous good record. The new structure how FAA approves airliners proven to be insufficient, for both the Dreamliner & MCAS with 737. If Boeing continues going down this path than it won't matter how long and how many passengers they shuttled.

There's figuring out what went wrong, and then there's figuring out why nobody thought anything was wrong. At least, nobody with the authority to do something about it.

Reply on news.ycombinator.com