[Sutripta]

Originally Posted by Jeroen

Looks like the FAA is taking no chances with Angle of Attack anymore:

https://www.flyingmag.com/faa-ground...us-vision-jets

Jeroen

I think the FAA's problems run deeper than its relationship (or by the looks of it, lack of it now) with AoA sensors!

It will be interesting to see which way the FAA proceeds in the 737 Max case (and other cases in future). It will most probably get away with a slap on the wrist (done everything by the letter of the book!), but when a regulator (or trustee) fails at its job, the repercussions run for a very long time.

Regards
Sutripta

PS - one thing mentioned ove and over again in the popular/ mainstream press is that though there are two AoA sensors, only one is used in the 737Max. This I think is some oversimplification somewhere, leading to a misunderstanding. Or somebody mentioned it somewhere on the internet, got picked up by others, and landed up reinforcing each other.
Could someone with more detailed knowledge clarify.

[Sutripta]

Originally Posted by Sutripta

It will be interesting to see which way the FAA proceeds in the 737 Max case (and other cases in future).

Interesting approach. Is it normal? Out of the box thinking?
https://www.chicagotribune.com/busin...419-story.html

As an aside, what are the chances that China lands up with flight control software source code?

Regards
Sutripta

[Jeroen]

Originally Posted by Sutripta

I think the FAA's problems run deeper than its relationship (or by the looks of it, lack of it now) with AoA sensors!

True, although after something like the current Boeing AoA episode stand by for some knee jerk reactions over the years to come.

It will be very interesting to see how they amend their ways of working going forward. The whole QA/QC thinking is of course very much dependent on each party/entity in a chain doing their own QA/QC in a very transparent en well documented way.

The problem here is of course how far do you allow it to go when it comes to design matters pertaining to safety. Part of the puzzle will be how do you exercise sufficient oversight/control over a party such as Boeing. It is also a bit of a competence and skill issue.

Originally Posted by Sutripta

Interesting approach. Is it normal? Out of the box thinking?
https://www.chicagotribune.com/busin...419-story.html

As an aside, what are the chances that China lands up with flight control software source code?

Can not open this link, won’t allow it from Europe, so there is some IPR - royalty protection that is working. I am not familiar enough to say anything about how easy it is for them to get the source software. Flight Management System used to be very simple boxes from a hardware/platform point of view. Easy to get into.

But I have often wondered how much use source code really is. When I started in Telecoms many moons ago, quite some customer insisted on "source code in safe" clauses in our contracts. Even then, it would have been very doubtful what you can do with it.

Mind you, the Chinese have proven themselves to be very impressive reverse engineers in many industries.

[Sutripta]

Originally Posted by Jeroen

Can not open this link, won’t allow it from Europe,

Oh. Just says that an international panel (Australia, Brazil, Canada, China, the European Union, Japan, Indonesia, Singapore and the United Arab Emirates) will vet the approval process. But if one reads the article it seems both process and product will be vetted.

But I have often wondered how much use source code really is.

Useful? Yes and no. Since this forum is full of software guys (from coder to project manager) will let them answer it.

Mind you, the Chinese have proven themselves to be very impressive reverse engineers in many industries.

Have done reverse engineering of some pretty proprietary stuff myself! No regrets. In fact rather proud of it.

[swiftdiesel]

Check this video to see how flawed the reincarnation project of 737 has been. Such a shame that a company in aviation is infamous for its unethical ways of doing business. An industry where safety takes priority over everything else, Boeing has resorted to hook or crook methods to fill its coffers. The proxy puppet aka, FAA is nothing short of a joke. I believe global airlines should boycott this company. My heart goes out to the hundreds of lives lost and I’m also glad to see this day where this shoddy company is under scrutiny. They brought this upon themselves and rightly so! Check out the video.

http://https://www.reddit.com/r/aviation/comments/bfmjho/boeing_and_the_faa_these_two_dodgy_entities_have/?utm_source=share&utm_medium=ios_app

[RM Motorsports]

Boeing Employees Claim Shoddy 787 Dreamliner Jet Production: Report

Boeing pushed its work force to quickly turn out the Dreamliners, at times ignoring issues raised by employees, the report said.

https://www.ndtv.com/world-news/boei...report-2026227

[AlphaKilo]

This is a pretty old news on shoddy 787 workmanship. There were reports on the same some 3/4yrs back when the aircraft was brand new. This issue first came up when the batteries were heating up. Quite a lot of senior staff left the company back then just to avoid shame and not wanting to be a part of a bad incident. Thankfully, so far the 787 has been flying safely. Hope it remains the same.

[Motard_Blr]

The 787 quality issues were investigated by Al Jazeera almost 5 years ago. The entire documentary is on YouTube.

More information here: https://www.aljazeera.com/investigations/boeing787/

Watch the YouTube video:

https://www.youtube.com/watch?v=rvkEpstd9os

[itwasntme]

BOE697 MAX8 proving flight for China Southern team to test the MCAS fix?

[GTO]

Thanks to Ramanujam V V for sending this information in. Heartfelt gratitude for sharing it with other enthusiasts via this Team-BHP page!

This might be of interest to your members who are discussing the 737 Max issues.

How the Boeing 737 Max Disaster Looks to a Software Developer

[itwasntme]

Originally Posted by GTO

Thanks to Ramanujam V V for sending this information in. Heartfelt gratitude for sharing it with other enthusiasts via this Team-BHP page!



How the Boeing 737 Max Disaster Looks to a Software Developer

Read it a couple of days back, thanks for sharing here. Scary stuff. I hope they fix and test everything very well before they put them back into service. Could have happened so easily to any IN carrier too. Just pure luck it did not!

I looked through my flight records and realized that my wife and me flew the Max 8 12 times before it was withdrawn from service.

[Sutripta]

Originally Posted by GTO

Thanks to Ramanujam V V for sending this information in. Heartfelt gratitude for sharing it with other enthusiasts via this Team-BHP page!



How the Boeing 737 Max Disaster Looks to a Software Developer

At least some information (true/ false/ qualifiers yet to be determined) on the use of a single AoA sensor.

For people going through the article, (which really does not have anything new, except for the single AoA stuff, and that the MCAS/ AutoPilot physically overpowering/ tiring out the pilots) do go through the comments.

Regards
Sutripta

[Jeroen]

Originally Posted by Sutripta

At least some information (true/ false/ qualifiers yet to be determined) on the use of a single AoA sensor.

The 737 Max has two AoA sensors, and they provide input continuously to various parts of the flight management/control system. However, as far as I can tell the MCAS is triggered by the reading of only one AoA sensor. Apparently there is some simple routine that alters which sensor it uses. I think I read somewhere it alters Left/right every flight or so.

Whereas the pilots have some control over which AoA sensors power their flight instruments, as far as I know they do not control with AoA sensor controls the MCAS.

So the pilot's instrument, e.g. the primary flight display will each have it’s own AoA, independent from one another. But the MCAS uses only one input at the time.

There were some interesting functional diagram shown on PPRune, but that thread has become so long, it is almost impossible to find it.

https://www.pprune.org/rumours-news/...wn-africa.html

It is one of the main criticism on the Boeing design. A single point of failure on an essential part of the flight controls.

[GoBlue]

Originally Posted by GTO

Thanks to Ramanujam V V for sending this information in. Heartfelt gratitude for sharing it with other enthusiasts via this Team-BHP page!

How the Boeing 737 Max Disaster Looks to a Software Developer

Quoting one of the comments out of the several that are resulting in a fiery discussion on this article. This comment brings out a very different perspective on the issue.

----Comment Start------
One of the thoughts that occurred to me when I started looking at this (I am not a pilot or an aerospace engineer -- I have sometimes worked with distributed control software for databases) was why there had never been a parallel problem with far-more-heavily automated Airbus planes. As I started to look into this I became rather alarmed at the direction of the 737-MAX control systems. I believe that some of the problems are actually even worse than described in the article so I am adding my thoughts here as well. The closest you have is Air France 447 but that was caused by pilots overwhelmed by warnings and failing to follow protocol rather than failing systems themselves. Airbus has never had a malfunctioning anti-stall system actively bring down an airplane. So why not? Why is this a specific problem with the MAX fleet and not the A320's, A330's etc?

The Airbus A320 was built to be an automation-centric airplane. As such it follows a large number of standard rules for making sure that the automation is safe. These include gracefully degrading the automation (alternate law and such), and even disabling envelope protection when the plane is too far out of the envelope. Also all control systems in these aircraft are triply redundant and rely on consensus to ensure that they are all functioning correctly. In this regard, if an angle of attack sensor is malfunctioning, the system notices that the two other sensors agree and use them instead. This isn't perfect as a failure in two sensors can still cause unexpected inputs but it is far better than having only two systems to consult. Yes, the A320 has triple redundancy on sensor input.

What this means is that the hardware design of the Airbus aircraft is built to support safe automation and control software. Any single component that fails can be identified by software, and input can automatically take that into account. This is not new. The Space Shuttle had triple redundancy in computer control systems also. Without triple redundancy, you can't do consensus-based automation.

The 737 never needed this because the plane was designed to have the ultimate control in the hands of the pilots. If a sensor fails, humans can triage, decide what to believe, or if all else fails look for visual inputs. But now with the MCAS for the MAX series, you have a critical system that operates silently behind the scenes, and has final authority over an airplane where the control systems were never designed for that level of decision-making. You don't have triple redundancy on the inputs. There is no requirement that an airplane can be competently flown by the pilot with this system disabled. This is also engineering malpractice, and yet another giant strike against the airplane. Not only is the fix to a hardware problem implemented in software, but it is implemented on a platform that was not even designed to support that level of automation.

As I have looked at the MAX series and these crashes, I have gone from "this is a major event in aerospace history but the plane is probably very safe" to "all these planes need to be recalled and the entire line discontinued." I no longer believe these planes to be safe by contemporary standards.
------Comment End------

[vharihar]

Originally Posted by Sutripta

At least some information (true/ false/ qualifiers yet to be determined) on the use of a single AoA sensor.

For people going through the article, (which really does not have anything new, except for the single AoA stuff, and that the MCAS/ AutoPilot physically overpowering/ tiring out the pilots) do go through the comments.

Regards
Sutripta

Very interesting to read the comments at the IEEE Spectrum article!

Sometime ago when I had suggested putting multiple AOA sensors and having the system take a majority vote, some folks had ridiculed the idea. Looks like that is precisely what Airbus does. Uses 3 AoA sensors.