Engineer's take on the 737 MAX design

Status
Not open for further replies.
Since when do angle of attack indicators have a propensity to fail?

I didn't read the article because the italicized summary points posted above were frustrating enough.. but the author clearly doesn't know what an inherently unstable aircraft design is either, that would be a plane like the f-117, something that's, well, unstable. 737 is not relying on fly by wire to fly, it's a hand fly airplane and as posted above that system only activates rein certaingimes and is designed to augment the control feel so it's not overly pitch sensitive.. like the md-11 which was often blamed as well as a poor design

This honestly seems like a primarily training and input from just one angle of attack indicator issue
 
I agree, this is surprisingly well written, especially to a lay audience.

I wouldn't consider IEEE Spectrum being written for the "lay audience". This is the flagship publication of the IEEE (Institute of Electrical and Electronics Engineers), the largest professional group in the world. Laymen in terms of specializing in airplanes, perhaps, but not a lay audience as in the readership of Popular Science. Yes, I am a Life Senior Member of the IEEE.
 
Since when do angle of attack indicators have a propensity to fail?

I didn't read the article because the italicized summary points posted above were frustrating enough.. but the author clearly doesn't know what an inherently unstable aircraft design is either, that would be a plane like the f-117, something that's, well, unstable. 737 is not relying on fly by wire to fly, it's a hand fly airplane and as posted above that system only activates rein certaingimes and is designed to augment the control feel so it's not overly pitch sensitive.. like the md-11 which was often blamed as well as a poor design

This honestly seems like a primarily training and input from just one angle of attack indicator issue

Prone to failure or otherwise, it's a single point of failure in a life-critical system. What MTBF would you find acceptable in such a device?

It's the same argument I have with the word "safe". Safe just means inside a risk tolerance. Prone to failure means too failure prone for this application. I'd argue that the demonstrated MTBF for this part is sufficiently short to be prone to fail in a single-point application.

I wouldn't consider IEEE Spectrum being written for the "lay audience". This is the flagship publication of the IEEE (Institute of Electrical and Electronics Engineers), the largest professional group in the world. Laymen in terms of specializing in airplanes, perhaps, but not a lay audience as in the readership of Popular Science. Yes, I am a Life Senior Member of the IEEE.

Certainly, but I was referring to article itself. It looks aimed at the non-engineer. Maybe I'm just biased from a lifetime of reading otherwise, but that doesn't strike me as a difficult read. I think this article would work just fine in Popular Science.
 
The viewgraph engineers always sound convincing, but I have all of them reflexively on automatic "ignore." Life is too short to search for possible gold in s#itpiles even though once in a while it exists.

Regarding the thesis that it was ultimate stupidity for a single-point failure to have that effect on the airplane, of course. But none of us actually know whether that was the case or whether there was some other more complex chain of events that occurred. Whatever happened the result was preventable disaster, but I'll wait for authoritative information (like an NTSB report) before I get too spun up by what are effectively hypotheticals. Right now we mostly have press ignoramuses voraciously feeding at a trough that contains very few facts but lots of detritus contributed by said ignoramuses and being recycled by them.

It is good to remember what H. L. Mencken told us in 1917: "Explanations exist; they have existed for all time; there is always a well-known solution to every human problem — neat, plausible, and wrong."

:p I gotta say it: This guy looks like a PowerPoint engineer to me. A near-complete compendium of buzz words, lots of activities listed but few if any accomplishments, ... Technical Education: 30 YO undergraduate History degree. Patents: none Technical Publications: none.

If this resume came to me in a pile, it would have a quick trip to the wastebasket.

But to give some credit ... if the Spectrum article is a fair sample, he's a decent writer. That's a valuable skill for a PowerPoint engineer.

So, yeah, the open shirt and sunglasses shot is a bit douchey. And he's been management for as long as he's willing to list in his experience section. Wonder if I look douchey on my profile. Let me know. I can't string him up for the lack of a degree. At the time, they were still rare in com sci. I'm really the first generation where the degree was expected and he's a generation or two earlier than me.

Still not willing to eviscerate him on that for writing an article on his opinion on these disasters. Discussing the thoughts in it are a useful exercise. Denigrating the man that wrote it isn't. I don't find any value in it. What value do you find? If you don't believe the article is accurate, let's talk what's not accurate. We can discuss. I can learn something. I have already...I passed right over the dynamic/static instability thing. I realize I need to dig in to understand the difference in the definitions and why that's important to aircraft certification. In a different thread, I learned about the FAA's methodology to statistics and some of their decisions make a lot more sense to me.

But learning about Gregory Travis hasn't really furthered me in any appreciable way.
 
His 1st and 2nd points are wrong, and his 3rd point seems to be the common thought here on POA for a while. Where is the value of an article that is likely only 33% accurate?
 
Prone to failure or otherwise, it's a single point of failure in a life-critical system. What MTBF would you find acceptable in such a device?
..

It really isn’t a “life-critical” single point failure if proper procedures are followed after failure. That is why we have pilots, and will never likely have completely autonomous planes.
 
I wouldn't consider IEEE Spectrum being written for the "lay audience". This is the flagship publication of the IEEE (Institute of Electrical and Electronics Engineers), the largest professional group in the world. Laymen in terms of specializing in airplanes, perhaps, but not a lay audience as in the readership of Popular Science. Yes, I am a Life Senior Member of the IEEE.

After reading the first few paragraphs, I had little interest in continuing further. The content and author's technical knowledge, writing style, vocabulary, and grammar would be given a C- in any respectable undergraduate engineering course. I suppose that assumes the course difficulty and grading scale is similar to that which I experienced in school, which these days is doubtful.

I hope the piece is not indicative of the technical writing skills possessed by other contributors to IEEE's Spectrum.

This assertion is painfully incorrect:

It’s as simple as that. The most effective way to make an engine use less fuel per unit of power produced is to make it larger. That’s why the Lycoming O-360 engine in my Cessna has pistons the size of dinner plates. That’s why marine diesel engines stand three stories tall. And that’s why Boeing wanted to put the huge CFM International LEAP engine in its latest version of the 737.

Then there is this:

MCAS is implemented in the flight management computer, even at times when the autopilot is turned off, when the pilots think they are flying the plane. In a fight between the flight management computer and human pilots over who is in charge, the computer will bite humans until they give up and (literally) die.

The author reveals his lack of knowledge; the MCAS system is not "implemented in the flight management computer", whatever that's supposed to mean. It is only active when the autopilot is off. The second sentence is ridiculous and untrue.

An unfounded and odious claim:

Finally, there’s the need to keep the very existence of the MCAS system on the hush-hush lest someone say, “Hey, this isn’t your father’s 737,” and bank accounts start to suffer.

This is complete nonsense:

The flight management computer is a computer. What that means is that it’s not full of aluminum bits, cables, fuel lines, or all the other accoutrements of aviation. It’s full of lines of code. And that’s where things get dangerous.

Those lines of code were no doubt created by people at the direction of managers. Neither such coders nor their managers are as in touch with the particular culture and mores of the aviation world as much as the people who are down on the factory floor, riveting wings on, designing control yokes, and fitting landing gears. Those people have decades of institutional memory about what has worked in the past and what has not worked. Software people do not.

The above statements are so juvenile and unsound they strain the limits of credulity.

This person has no engineering skill. He writes like a recently hired news reporter from a downmarket television station.
 
Last edited:
So, yeah, the open shirt and sunglasses shot is a bit douchey. And he's been management for as long as he's willing to list in his experience section. Wonder if I look douchey on my profile. Let me know. I can't string him up for the lack of a degree. At the time, they were still rare in com sci. I'm really the first generation where the degree was expected and he's a generation or two earlier than me.

Still not willing to eviscerate him on that for writing an article on his opinion on these disasters. Discussing the thoughts in it are a useful exercise. Denigrating the man that wrote it isn't. I don't find any value in it. What value do you find? If you don't believe the article is accurate, let's talk what's not accurate. We can discuss. I can learn something. I have already...I passed right over the dynamic/static instability thing. I realize I need to dig in to understand the difference in the definitions and why that's important to aircraft certification. In a different thread, I learned about the FAA's methodology to statistics and some of their decisions make a lot more sense to me.

But learning about Gregory Travis hasn't really furthered me in any appreciable way.

Thanks for pointing this out. I enjoy this site but sometimes it’s filled with people who have never walked into a room without thinking they are the smartest person in it. That’s frustrating to me and this thread is an example of how that can play out often on this site.

I’m about as far from an engineer as one could get but I found the article to contain a lot of interesting points. The one I found most interesting was the potential economic impact the lack of expansion the 737 model seems to have left for Boeing. They can’t make it much longer( without adding in some new landing gear) can’t seem to make it much more powerful or fuel efficient based on this latest issue with the MAX, and this article highlighted the reasons for that very well.

I know Boeing has not announced anything about the 737 line. My hunch is they will continue to design systems to work with this current airframe but should that be cause for concern? It seems like maximizing profits at the expense of designing a new airframe is driving Boeing right now. Can these new systems efficiently keep the flying public safe remains to be played out. I know Boeing’s stock has done remarkably well since the incident but I’m not so sure people aren’t seeing this as a possible situation where “max capacity” for the 737 line may be playing out.
 
... I can't string him up for the lack of a degree. At the time, they were still rare in com sci. I'm really the first generation where the degree was expected and he's a generation or two earlier than me. ...
Well Donald Knuth released "The Art of Computer Programming" in 1968 as an undergraduate textbook. Travis was an undergraduate twenty years later. By that time, we certainly were expecting CS or EE degrees of fresh graduates, though the pre-1970 generation of old bears did not have them. In 1989 a degree in History would not have passed "Go" applying for a technical job.

...Still not willing to eviscerate him on that for writing an article on his opinion on these disasters. ...
Nope, neither am I. It's a free country and there is a lot of stupider opinion -piece stuff out there.

Discussing the thoughts in it are a useful exercise. Denigrating the man that wrote it isn't. I don't find any value in it. What value do you find? ... ...
I didn't "denigrate" him in my initial posts. If my conclusion that he is a PowerPoint engineer is correct I don't know whether that is denigration or not. But is has been my understanding that people post their information to LinkedIn precisely because they want people to judge them on it. Is there any other possible reason?

... If you don't believe the article is accurate, let's talk what's not accurate. We can discuss. ...
No interest. As I have said, I don't waste time on material from unqualified sources. That includes the news media, incidentally. I think you may be especially interested in the article because you are in agreement with at least some of it. https://en.wikipedia.org/wiki/Confirmation_bias
 
Well Donald Knuth released "The Art of Computer Programming" in 1968 as an undergraduate textbook. Travis was an undergraduate twenty years later. By that time, we certainly were expecting CS or EE degrees of fresh graduates, though the pre-1970 generation of old bears did not have them. In 1989 a degree in History would not have passed "Go" applying for a technical job.

Nope, neither am I. It's a free country and there is a lot of stupider opinion -piece stuff out there.

I didn't "denigrate" him in my initial posts. If my conclusion that he is a PowerPoint engineer is correct I don't know whether that is denigration or not. But is has been my understanding that people post their information to LinkedIn precisely because they want people to judge them on it. Is there any other possible reason?

No interest. As I have said, I don't waste time on material from unqualified sources. That includes the news media, incidentally. I think you may be especially interested in the article because you are in agreement with at least some of it. https://en.wikipedia.org/wiki/Confirmation_bias

I almost certainly have confirmation bias on this one. I noted it in a previous post, #31: "Confirmation bias is definitely possible." Beating back confirmation bias is generally done by debating the points. Attacking the author only gets folks circling the wagons in two camps. If you're in the group to whom the article speaks, like me, an attack on the author feels like an attack on yourself. That's probably why your original comment stood out for me. It meant I had to recognize that there was confirmation bias present in myself and fight the emotional tide the comment created. At the risk of sounding like I'm patting myself on the back, that's a tough thing to do.

This is my point about ad hominem: It rarely gets anyone closer to the truth. Maybe that wasn't your aim, but why post in the thread at all if it wasn't?

I'll admit that I want desperately to understand what the hell happened. It's both fascinating and terrifying. I'm at a point in my career where my responsibility has extended beyond just writing good code to actually understanding and governing the processes that produce it and ensure it. Most interesting here, for me, is why do people who know better not do better? It's my current struggle: I know several places where quality is lacking in my organization. I can identify it, I can communicate it, I can evangelize it, but I'm only marginally effective at facing down the institutional inertia that keeps it in place. I need to do better. I'm hoping to take lessons from what I learn about this disaster.
 
...I'll admit that I want desperately to understand what the hell happened. It's both fascinating and terrifying. I'm at a point in my career where my responsibility has extended beyond just writing good code to actually understanding and governing the processes that produce it and ensure it. Most interesting here, for me, is why do people who know better not do better? It's my current struggle: I know several places where quality is lacking in my organization. I can identify it, I can communicate it, I can evangelize it, but I'm only marginally effective at facing down the institutional inertia that keeps it in place. I need to do better. I'm hoping to take lessons from what I learn about this disaster.
Good for you. But ... patience. patience. patience. Gather all the facts (i.e. NTSB, etc.) before forming hypotheses. Premature conclusions are the enemy of accurate ones. Maybe resolve yourself to not read, not to discuss, and not to hypothesize for six months. Then stick your head up and see if all the important information is available and decide whether it's time to try for that understanding.
 
Good for you. But ... patience. patience. patience. Gather all the facts (i.e. NTSB, etc.) before forming hypotheses. Premature conclusions are the enemy of accurate ones. Maybe resolve yourself to not read, not to discuss, and not to hypothesize for six months. Then stick your head up and see if all the important information is available and decide whether it's time to try for that understanding.

I understand the concern. The earliest info is often just plain wrong. The stuff that follows is more accurate, but generally incomplete. Only with careful investigation and consideration does the most complete story come out.

We each have different ways of getting there. For me, this iteration through each round of information is how I get there. Parts of the story come into better focus with time, but I find it difficult to go directly to the conclusion and absorb the impact directly. There's stuff along the way that informs the context of the final, dry facts. This exercise is instructive for me: form the model, see how well it plays out as more info comes. Adjust the model. Rinse and repeat. I don't have conclusions yet, just a hypothetical model.

And...it runs into another weakness I have. I tend to just put statements out without surrounding them with my level of confidence. I'm used to being surrounded by others with strong opinions and strong voices. They openly disagree when something doesn't match. I expect that, and it's been a useful tool for getting to the right place. Unfortunately, two things have happened. First, the advent of electronic communication has taken some of the nuance from something like that, making something that is a strong conjecture just sound like a firmly held belief. The second is, as I've gotten more senior, there are now many, particularly juniors, who won't disagree openly for a variety of reasons. I'm trying to adjust how I do things, but years of doing this successfully makes it a surprisingly hard habit to break.

Long story short, this interaction is really comfortable to me. But you may have the impression I have a hard position on what happened. I do not. I have a model and I work from it. It's open to adjustment.
 
The airframe, the hardware, should get it right the first time and not need a lot of added bells and whistles to fly predictably. This has been an aviation canon from the day the Wright brothers first flew at Kitty Hawk.

Um, I think aviation has been pretty much exactly the opposite of that throughout history. Think rudders, ailerons, flaps, trim tabs, yaw dampers, vortex generators, modal suppression. Lots of "bells and whistles"

After reading the article and the comments it seems the author at least is convinced that the 737 MAX is fatally flawed and that when the AOA inevitably fails there is absolutely nothing that can be done to save the aircraft. Obviously it's more complicated than that and likely that if the crews in the two fatal crashes knew what we all know now they both would have recovered and returned safely to the airport.

Sorry but this article is not particularly good in my opinion.
 
Hi,

Gregory Travis here. A friend forwarded a link to this discussion.

First, Ron it's fantastic to see you here. I am looking at my copy of Kitplane Construction on my bookshelf right now :)

Second, I'd be happy to discuss the points in my article here. I am not perfect, by a long shot, and it's more important for me to be correct than to be right.

Third, re: Powerpoint. Guilty as charged. Though I've read Tufte, I still can't resist those darn bulletpoints.

Fourth, re: Note. It did exist, there is a Facebook page dedicated to it. And I wrote it in 1977, when I was 13 years old, in COMPASS (assembly language for the CDC 6000 series of computers). Hopefully that un-does some of the douchery that my third point, above, did. My two heroes are Seymour Cray and Dennis Richie, who taught me the most valuable lesson ever taught: Keep It Simple, Stupid.

(I realize that nothing can un-do the douchery of the sunglasses and open shirt)

Fifth, an earlier poster hit the nail on the head. The points about aerodynamics (my passion) and even the points about software design (my profession) are not the story. They are there to make the story accessible and understandable for the public. The story is a human story, namely (to quote the Talking Heads) "How did we get here?" This is not an engineering text. It is a murder-mystery.

Sincerely,

Greg
 
Last edited:
In 1989 a degree in History would not have passed "Go" applying for a technical job.

Unfortunately, I think that's a large reason why there are 300+ bodies at the bottom of the ocean and scattered across the desert.
 
Gregory Travis here
Thanks for coming here to post, that's respectable! Out of curiosity, was this read: https://seekingalpha.com/instablog/...930-boeing-737-max-8-crashes-case-pilot-error

I appreciate the effort to make this something digestible to the public, events like this are challenging because there is a lot of nuance with how the public receives information, the final findings aren't out yet, and most people's skillset when it comes to aviation, engineering, medicine, etc., is very limited.. so attempting to break something down to the masses is difficult

If you'll entertain me the opportunity to share my feedback, maybe you can help better educate me as well, since I can't help but feel like there's a clear intent based on the 3 premis below that ther OP wrote to paint a very particular picture of an inherently unstable and cheaply built aircraft without regard to the bigger picture of prior safety bulletins, the pilot's role in actively flying the aircraft, etc. Again, thanks for posting here..

"So Boeing produced a dynamically unstable airframe, the 737 Max. That is big strike No. 1."
-I think this is a dangerous opinion to share with a lay person or general public. "Dynamically unstable" implies the aircraft doesn't have any inherent stability, as in, if the controls are let go of it will come apart into some doom's day spiral event. In reality, the plane is trimmable and hand flyable without computers, it's not a FBW plane, and the MCAS is not active at all times. As posted by others, and I think well known at this point, MCAS was introduced to activate at certain flight regimes to mimic the feel of prior 737 models to avoid having to give it a new type certificate. There's a big difference in "this 737 will fly differently than other 737" VS "this 737 is dynamically unstable and can't be flown without it" - MCAS was installed due to the former

"Boeing then tried to mask the 737’s dynamic instability with a software system. Big strike No. 2."
This makes it seem like a group of comic book villains sat around a conference room.. "mask" implies that there was intentional deception for a defective product. The reality of it is that they added the software system to avoid requiring a new type certificate by making the airplane feel more similar to other 737. The bigger engines, etc., meant that it would feel different during high AoA and power increases, etc., so they added a control augmentation system. Boeing obviously has the ability to certify and get new type ratings, part of the reason they "had" to keep this rating was as a response to customer fleet requirement. Commonality is not some "cheap" excuse way to certify planes, the 757/767 share a type, and was a big driver with Airbus. So implying that Boeing was intentionally deceiving the poor innocent customer and cheeping out on a type rating is just wrong

"Finally, the software relied on systems known for their propensity to fail (angle-of-attack indicators) and did not appear to include even rudimentary provisions to cross-check the outputs of the angle-of-attack sensor against other sensors, or even the other angle-of-attack sensor. Big strike No. 3"
Point A: Is there some evidence to back up AoA sensors' propensity to fail? How often do they fail relative to other AC items? Genuine question, but I've never heard of the AoA as some particularly vulnerable item, at least not more so than a pitot or static port

Point B: There are many *very* rudimentary provisions to flying the aircraft that don't rely on AoA.. such as being aware of what the plane is doing, observing the pitch, power, redundant airspeed and artificial horizon indicators, etc. I would think in most cockpits seeing an uncommanded trim activation would at least get the attention of the crew, especially if they're unfamiliar with MCAS and would not be expecting it. This implies that the aircraft's entire fate was dependent on one junkie part that the pilots could do nothing about. In reality, they let the trim wheel spin full nose down at which point the plane was unflyable given the altitude they had left.


Please don't take my points personally, and I thank you for coming on this forum so I can ask these directly from you. But why not just explain what happened, vs assign specific blame? The biggest issue here was not training new pilots properly on the MCAS system, its failure modes, etc., and having the software rely on just the one AoA input (this part I totally agree with you). I do still believe that a well trained and active crew who stays ahead of the aircraft (vs software systems operator) would have seen that trim wheel going nuts and disconnected it well before things became unrecoverable at the altitude they were at. We also don't know the actual cause yet.. there are several elements that disable MCAS, like AP being on, flaps engagement, etc. So we still have a lot of unknowns here, and also a lot of people who claiming causes as facts
 
"So Boeing produced a dynamically unstable airframe, the 737 Max. That is big strike No. 1."

First, my understanding of the issue -- gleaned both from public reports as well as private correspondence is this:
1. The mass of the engines (the LEAP engines are about 1500 pounds heavier, per engine, than the CFM56s they replaced) coupled with their location forward of the longitudinal center of gravity means that they impart a large inertial moment. Meaning once the airplane starts to rotate in pitch, the mass of the engines resists a change in rotation. Once it starts to rotate, it wants to keep rotating. To me, and I am just a software engineer, that sounds dynamic.
2. The engine nacelles (cowlings) generate lift. That lift is a function of the angle of attack of the cowling. The more the angle of attack, the more lift. The more lift, the more aerodynamic moment. Meaning, once the nacelles start to generate lift, they tend to rotate the aircraft to a higher angle of attack, which causes them to generate more lift. Again, to me that sounds dynamic, not static.
3. Thrust changes cause also a pitch up. I cannot go to why they cause a pitch up other than to note that pitch up with increasing thrust is common (my Cessna does it). This is perhaps, to me, the only non-dynamic part of things because the last thing I would want as a pilot would be an aircraft that pitches up in response to a thrust decrease (i.e. pitches down in response to a thrust increase) as that situation, again to me, would be dynamic.

By dynamic here I mostly mean to imply "makes the problem get worse."

I believe that the 737 MAX is statically stable. By that, I mean that if flying in trim and a pilot removes her hands from the controls, the airplane will continue to fly stably. It will seek its trim speed.

"Boeing then tried to mask the 737’s dynamic instability with a software system. Big strike No. 2."

My best answer here is something I wrote in response to an email query. Here is the thread:

Hello, Mr. Travis. Some friends of mine and I have been having an argument over a couple of sentences in your article, namely this:

"Boeing then tried to mask the 737’s dynamic instability with a software system. Big strike No. 2."

Some of us maintain by the use of the word "mask" here, you mean that Boeing is "compensating" for the new instability, and they believe that there is nothing wrong with using software in this manner. Others of us believe you are using the word as in to hide or disguise and it is the action of hiding that is the problem, not necessarily the use of software per se. Can you settle this for us? What was your intent here?
To which I replied:

Much more the latter — it is the action of hiding MCAS that is the root of the ethical problem. As your friends note, compensating for an aerodynamic issue with some kind of machinery is fairly common practice. The use of yaw dampers on swept wing jets being the best example I can think of

Yaw dampers are a necessary evil (but evil still as they add complexity and thus fAilure modes). But no one ever deliberately tried to pretend they didn’t exist
"Finally, the software relied on systems known for their propensity to fail (angle-of-attack indicators) and did not appear to include even rudimentary provisions to cross-check the outputs of the angle-of-attack sensor against other sensors, or even the other angle-of-attack sensor. Big strike No. 3"

Rhetorically I would argue that there are two destroyed aircraft and over 300 destroyed lives that would, if they could, take exception to the notion that AOA sensor failures are rare. Less emotionally, I would argue that irrespective of the failure rate, it is beyond foolish to design a system that a) relied on only a single sensor output to b) make configuration changes to the aircraft that render the aircraft uncontrollable.

I am glad you mentioned the static port. A static port is nothing more than a hole. My lowly Cessna has a static port. It also has an alternate (backup) static port, which I can switch to with a control in the cockpit. Yes, my Cessna has a backup hole in case the primary hole fails. That should give some idea of how reliable something needs to be before it doesn't need a backup, in aviation. It's something like 100% reliable, maybe a little more.

Point B: There are many *very* rudimentary provisions to flying the aircraft that don't rely on AoA.. such as being aware of what the plane is doing, observing the pitch, power, redundant airspeed and artificial horizon indicators, etc. I would think in most cockpits seeing an uncommanded trim activation would at least get the attention of the crew, especially if they're unfamiliar with MCAS and would not be expecting it.

My understanding, based on my own experience plus that of talking with commercial pilots, is that AoA is pretty much a cosmetic indicator in commercial aircraft. Nobody flying a commercial airliner gives a **** about AoA. They, like me, fly by attitude (pitch) and airspeed. AoA is important in military fighting, but unless you want to take that 737 into a dogfight where you need to maneuver as close as you can to the stall, it doesn't matter.

Regarding pilots seeing an uncommanded trim activation -- speculation along this line (as well as the speculation regarding pilot error in the article you cite at the top) really gets to me. There is the supposition that a) The pilots are at fault with b) Because they are not skilled enough (Boeing's marketing department does not want people telling airlines that only their most skilled (expensive) pilots should fly the 737 and c) They are not skilled enough because they are brown. In other words, there is a NASTY bit of implicit racism going on here that is entirely unfounded.

In high stress, lift-threatening situations, decision making ability and rational thought go RIGHT OUT OF THE WINDOW. How do I know? I've been hijacked, at gunpoint, while flying a plane (I'd put a link to the story on Medium here, but PilotsOfAmerica says I don't have enough hours to include links in posts, yet ) and it made my brain turn to mush. I don't think that any pilot, of any training, could have known that hitting the stabilizer cutout switches in the event of an MCAS malfunction was a thing to do. Why? First, because an MCAS malfunction does not present as a classic trim runaway. Second, because the pilots were not even told that MCAS existed, or under what circumstances it might malfunction and what that malfunction would look like.
 
I don't think that any pilot, of any training, could have known that hitting the stabilizer cutout switches in the event of an MCAS malfunction was a thing to do. Why? First, because an MCAS malfunction does not present as a classic trim runaway. Second, because the pilots were not even told that MCAS existed, or under what circumstances it might malfunction and what that malfunction would look like.

Unless I mis read, a jump seat pilot on a previous flight did in-fact use the trim cut out with success, so at least one pilot knew to take that corrective action.
 
I’m a software engineer, well at least according to my job title. I do not hold an engineering degree, nor have I passed a FE exam.

Would I assume I know the first damn thing about aeronautical engineering? Hell no, I’m no more qualified then the next guy on the street.

Should “engineering” be a part of my job title given what I mentioned? Who knows. I spend years building complex systems that are liable for billions of dollars of other people’s money. Someone with a masters degree or equivalent experience is an entry level position on my engineering teams. Complex enough to justify calling it engineering? Probably. However there is zero relation to aero engineering.

Any software engineer that tries to imply his opinion on the Boeing situation is more valid then anyone else’s is pretty ridiculous.

Lots of software engineers ask me for my opinion on the Boeing situation...they do that because they know I’m a pilot. Not because I’m a software engineer.

So really, what we have here, is just an opinion of a pilot. No more valid then the opinion of any other pilot on this board.
 
Last edited:
First, my understanding of the issue -- gleaned both from public reports as well as private correspondence is this:
1. The mass of the engines (the LEAP engines are about 1500 pounds heavier, per engine, than the CFM56s they replaced) coupled with their location forward of the longitudinal center of gravity means that they impart a large inertial moment. Meaning once the airplane starts to rotate in pitch, the mass of the engines resists a change in rotation. Once it starts to rotate, it wants to keep rotating. To me, and I am just a software engineer, that sounds dynamic.
2. The engine nacelles (cowlings) generate lift. That lift is a function of the angle of attack of the cowling. The more the angle of attack, the more lift. The more lift, the more aerodynamic moment. Meaning, once the nacelles start to generate lift, they tend to rotate the aircraft to a higher angle of attack, which causes them to generate more lift. Again, to me that sounds dynamic, not static.
3. Thrust changes cause also a pitch up. I cannot go to why they cause a pitch up other than to note that pitch up with increasing thrust is common (my Cessna does it). This is perhaps, to me, the only non-dynamic part of things because the last thing I would want as a pilot would be an aircraft that pitches up in response to a thrust decrease (i.e. pitches down in response to a thrust increase) as that situation, again to me, would be dynamic.

By dynamic here I mostly mean to imply "makes the problem get worse."

I believe that the 737 MAX is statically stable. By that, I mean that if flying in trim and a pilot removes her hands from the controls, the airplane will continue to fly stably. It will seek its trim speed.

And here is part of the problem--you are using standard engineering terms like "dynamic stability" with your own made-up definitions! What you describe as "static stability" is pretty much how us engineers define and use "dynamic stability." You must use the right words, or your meaning will never be communicated. The term "PowerPoint engineer" meant "someone with little knowledge or expertise, who survives in the corporate world by giving presentations of questionable origins and suspicious conclusions not based on actual knowledge" [at least that was how I understood the term to have been used], and was not a compliment--it was an around-the-bush way of saying that you don't know what you're talking about.

A dynamically stable plane in flight will continue on its merry way not only when the controls are released, but if they are released and bumped or if the plane hits turbulence, it will find it's way back to where it had been, going where it had been going. A dynamically unstable plane will diverge from its path when the controls are released or turbulence is encountered, and if the controls are nudged the rate of divergence will increase. So using these basic engineering terms properly, there is no disagreement that the 737MAX is a dynamically stable airplane. Although the words in your article say otherwise, this is what you wrote above. Shame on you! Go back to Freshman Statics and read the durn book, you silly programmer masquerading as an engineer!

This is why it is important to review the author's credentials, and why the negative comments above are appropriate and not an ad hominem attack. The author is not writing what he means, because he doesn't know the definitions of the words he is using. He is unqualified to address the subject, but he sure sounds good to himself and probably makes slick presentations (which I learned to do while earning my Master of Engineering degree well after my BSME).
 
Unless I mis read, a jump seat pilot on a previous flight did in-fact use the trim cut out with success, so at least one pilot knew to take that corrective action.

Now we're debating whether or not a 737 MAX should require a three-person crew.
 
A dynamically stable plane in flight will continue on its merry way not only when the controls are released, but if they are released and bumped or if the plane hits turbulence, it will find it's way back to where it had been, going where it had been going. A dynamically unstable plane will diverge from its path when the controls are released or turbulence is encountered, and if the controls are nudged the rate of divergence will increase.

You've just described a 737 MAX with MCAS inoperative.

Your objection reminds me of someone someone said either here or elsewhere. They took exception that I said that "[stability] was aviation canon since the days of the Wright brothers at Kitty Hawk." Their objection was that I should have said "Santos Dumont" since it is (convincingly) argued that he proceeded the Wright brothers.

Perhaps 85% of my audience knows who the Wright brothers are. Maybe 5% knows who Santos Dumont is. The selective use of examples, so as to effectively communicate the concept to your audience, is a hallmark of a successful writer.
 
Now we're debating whether or not a 737 MAX should require a three-person crew.

Nope, responding to your comment that no pilot of any training would use the trim cut out in that situation. Fact is one did, he knew how to deal with a run away trim condition.
 
Nope, responding to your comment that no pilot of any training would use the trim cut out in that situation. Fact is one did, he knew how to deal with a run away trim condition.

Now you're putting words in my mouth. I explicitly said that pilots were not only unaware of what to do in the case of an MCAS failure (i.e. untrained), they were unaware that MCAS even existed. Not even Boeing's own test pilots knew that MCAS existed.
 
Although they knew about Run away trim procedures and from my Understanding they haven’t changed since the 737’s inception. That’s what they had, doesn’t matter what caused it.
 
Last edited:
First, my understanding of the issue
Unfortunately, the premise of your core understanding is flawed. The MAX can fly quite happily and stabile through it's entire certified flight envelop. Period. This has been discussed and verified on multiple platforms. The sole reason the MCAS was introduced was because at certain flight attitudes and airspeeds the cockpit flight control feed back forces did not meet the linear requirements of FAR Part 25. Nothing more. Those are the facts.

Half the issue is whether a flight crew is able to handle a failure of the MCAS or one of its associated systems. The one thing missing from most articles on this subject is that the MCAS worked as designed in both the accidents. It took the data it received and increased the control forces. Nothing more.

The other half of the issue is whether the current status of crew training and ability is at a level equal to the level of the current state of automation in most modern aircraft. There have been discussions on these issues for years before the MAX accidents. Lack of manual flying training and tne preference to follow the magenta line have dominated most previous accident discussions. The problem is Boeing, Airbus, or name whatever manufacturer you want, has zero authority on the training/ability of flight crews. That falls solely to the operator and the CAA.

So when the Lion jumpseater suggested to use an existing emergency procedure or the Lion 610 captain who maintained control through through 21 MCAS activations only to have the SIC lose control after 4 activations, or the fact that Ethiopian Airlines did not formally incorporate any of the the flight safety bulletins into their MAX crew training/documentation after the Lion accident, does not support your premise at all that it was a Boeing or aircraft issue.

And while you prefer to use your Cessna as an analogy to the flight chararistics of a MAX, I wonder how a pilot with 361 total flight hours (EA 302 SIC) would handle a major emergency like an engine failure at take off? Not to mention what that same pilot might be able to handle on a MAX with an erroneous AoA sensor, stick shaker, and various other emergency procedures that require 2 minds and 4 hands.

While I respect your opinion, I think it it may help if you include all the facts when writing to the general public.
 
3. Thrust changes cause also a pitch up. I cannot go to why they cause a pitch up other than to note that pitch up with increasing thrust is common (my Cessna does it). This is perhaps, to me, the only non-dynamic part of things because the last thing I would want as a pilot would be an aircraft that pitches up in response to a thrust decrease (i.e. pitches down in response to a thrust increase) as that situation, again to me, would be dynamic.

Every plane with underwing engines produces a nose-up moment when you add power. In jets, the moment is surprisingly strong. Add full power at low speeds and you need over 50% of elevator deflection just to keep the nose from pitching up.
The 737 is stable, MCAS is designed to operate in parts of the envelope you would never experience in normal operation.
Completely different reason though than why your Cessna does it.
 
First, my understanding of the issue -- gleaned both from public reports as well as private correspondence is this:
1. The mass of the engines (the LEAP engines are about 1500 pounds heavier, per engine, than the CFM56s they replaced) coupled with their location forward of the longitudinal center of gravity means that they impart a large inertial moment. Meaning once the airplane starts to rotate in pitch, the mass of the engines resists a change in rotation. Once it starts to rotate, it wants to keep rotating. To me, and I am just a software engineer, that sounds dynamic.
2. The engine nacelles (cowlings) generate lift. That lift is a function of the angle of attack of the cowling. The more the angle of attack, the more lift. The more lift, the more aerodynamic moment. Meaning, once the nacelles start to generate lift, they tend to rotate the aircraft to a higher angle of attack, which causes them to generate more lift. Again, to me that sounds dynamic, not static.
3. Thrust changes cause also a pitch up. I cannot go to why they cause a pitch up other than to note that pitch up with increasing thrust is common (my Cessna does it). This is perhaps, to me, the only non-dynamic part of things because the last thing I would want as a pilot would be an aircraft that pitches up in response to a thrust decrease (i.e. pitches down in response to a thrust increase) as that situation, again to me, would be dynamic.

By dynamic here I mostly mean to imply "makes the problem get worse."

I believe that the 737 MAX is statically stable. By that, I mean that if flying in trim and a pilot removes her hands from the controls, the airplane will continue to fly stably. It will seek its trim speed.



My best answer here is something I wrote in response to an email query. Here is the thread:

Hello, Mr. Travis. Some friends of mine and I have been having an argument over a couple of sentences in your article, namely this:

"Boeing then tried to mask the 737’s dynamic instability with a software system. Big strike No. 2."

Some of us maintain by the use of the word "mask" here, you mean that Boeing is "compensating" for the new instability, and they believe that there is nothing wrong with using software in this manner. Others of us believe you are using the word as in to hide or disguise and it is the action of hiding that is the problem, not necessarily the use of software per se. Can you settle this for us? What was your intent here?
To which I replied:

Much more the latter — it is the action of hiding MCAS that is the root of the ethical problem. As your friends note, compensating for an aerodynamic issue with some kind of machinery is fairly common practice. The use of yaw dampers on swept wing jets being the best example I can think of

Yaw dampers are a necessary evil (but evil still as they add complexity and thus fAilure modes). But no one ever deliberately tried to pretend they didn’t exist


Rhetorically I would argue that there are two destroyed aircraft and over 300 destroyed lives that would, if they could, take exception to the notion that AOA sensor failures are rare. Less emotionally, I would argue that irrespective of the failure rate, it is beyond foolish to design a system that a) relied on only a single sensor output to b) make configuration changes to the aircraft that render the aircraft uncontrollable.

I am glad you mentioned the static port. A static port is nothing more than a hole. My lowly Cessna has a static port. It also has an alternate (backup) static port, which I can switch to with a control in the cockpit. Yes, my Cessna has a backup hole in case the primary hole fails. That should give some idea of how reliable something needs to be before it doesn't need a backup, in aviation. It's something like 100% reliable, maybe a little more.



My understanding, based on my own experience plus that of talking with commercial pilots, is that AoA is pretty much a cosmetic indicator in commercial aircraft. Nobody flying a commercial airliner gives a **** about AoA. They, like me, fly by attitude (pitch) and airspeed. AoA is important in military fighting, but unless you want to take that 737 into a dogfight where you need to maneuver as close as you can to the stall, it doesn't matter.

Regarding pilots seeing an uncommanded trim activation -- speculation along this line (as well as the speculation regarding pilot error in the article you cite at the top) really gets to me. There is the supposition that a) The pilots are at fault with b) Because they are not skilled enough (Boeing's marketing department does not want people telling airlines that only their most skilled (expensive) pilots should fly the 737 and c) They are not skilled enough because they are brown. In other words, there is a NASTY bit of implicit racism going on here that is entirely unfounded.

In high stress, lift-threatening situations, decision making ability and rational thought go RIGHT OUT OF THE WINDOW. How do I know? I've been hijacked, at gunpoint, while flying a plane (I'd put a link to the story on Medium here, but PilotsOfAmerica says I don't have enough hours to include links in posts, yet ) and it made my brain turn to mush. I don't think that any pilot, of any training, could have known that hitting the stabilizer cutout switches in the event of an MCAS malfunction was a thing to do. Why? First, because an MCAS malfunction does not present as a classic trim runaway. Second, because the pilots were not even told that MCAS existed, or under what circumstances it might malfunction and what that malfunction would look like.
Yeah you have some good ideas but are not really sure and put it out there like you have solid understanding

Hence me throwing the BS penalty flag on the field of play. I haven’t picked it up yet.
 
You've just described a 737 MAX with MCAS inoperative.

No, he described a 737 MAX in an extremely small sliver of its flight envelope - a situation that should never be encountered during normal operations. I've flown the MAX-8 without MCAS activated for hundreds of hours, and it's a more stable, better flying airplane than the -800.
 
Last edited:
Now we're debating whether or not a 737 MAX should require a three-person crew.

This comment is ridiculous, and reinforces my opinion you don't have the qualifications to opine on anything regarding engineering, statistics, aircrew training, human behavior, and all the other things you presented as facts in your silly article.
 
The sole reason the MCAS was introduced was because at certain flight attitudes and airspeeds the cockpit flight control feed back forces did not meet the linear requirements of FAR Part 25.

Reference, please. The specific section of part 25 that goes to your statement would be fine.

The one thing missing from most articles on this subject is that the MCAS worked as designed in both the accidents. It took the data it received and increased the control forces.

Reference, please. In particular that MCAS "[increases] the control forces." I thought it drove the stabilizer jackscrew through the electric trim motor.

Every plane with underwing engines produces a nose-up moment when you add power.

Yet not every plane with underwing engines requires MCAS. In fact, no 737 prior to the MAX required MCAS. What is different with the 737 MAX that it requires what no other 737 needed?
 
Reference, please. In particular that MCAS "[increases] the control forces." I thought it drove the stabilizer jackscrew through the electric trim motor.

You don't know that altering trim or stabilizer incidence increases or decreases control forces felt by the pilot?
 
Yet not every plane with underwing engines requires MCAS. In fact, no 737 prior to the MAX required MCAS. What is different with the 737 MAX that it requires what no other 737 needed?

Larger engines mounted further forward, and due to the single type rating requirements, the designers had a very limited toolbox in use to fix it.
It's all about control force requirements for Part 25 standards. Movement of the stab was required to make the forces linear.
"(a) Each control and control system must operate with the ease, smoothness, and positiveness appropriate to its function." for example. There are other quotes from Part 25 as well.
 
Last edited:
It's all about control force requirements for Part 25 standards.

Cite the relevant sections of Part 25, please.

Movement of the stab was required to make the forces linear.

On the 737, control forces, particularly elevator control forces, are created by the elevator feel computer. Not the stabilizer trim.

Also please cite the relevant sections of part 25 that require to make the forces linear. I'm also curious how the A320 escaped from that mythical section, given that it has no control forces whatsoever. I suppose an argument that zero force throughout the envelope and throughout control movement is linear could be made.
 
Cite the relevant sections of Part 25, please.



On the 737, control forces, particularly elevator control forces, are created by the elevator feel computer. Not the stabilizer trim.

Also please cite the relevant sections of part 25 that require to make the forces linear. I'm also curious how the A320 escaped from that mythical section, given that it has no control forces whatsoever. I suppose an argument that zero force throughout the envelope and throughout control movement is linear could be made.

I just did.

You're saying stab trim position is not an input for the elevator control feel computer? Don't forget the single type rating limitations here. Can't redesign too much.

You also need to read about augmentation systems (A320 being the case you refer to) and how A320 control forces feel in alternate, abnormal alternate and direct law, and why that is.
25.671/672 for example to start with.
 
Also please cite the relevant sections of part 25 that require to make the forces linear. I'm also curious how the A320 escaped from that mythical section, given that it has no control forces whatsoever. I suppose an argument that zero force throughout the envelope and throughout control movement is linear could be made.
CFR14 25.173 requires control force gradient of not less than 1 lb per 6 knots speed change from initial trim. It does not have to be linear but it does specify a linear gradient that must be met. As for the A320, do you know for a fact that part 25 was the cert basis, with no waivers or variances?

Regarding 'stability as canon', bare airframe (unaugmented through any means) stability has not been 'canon' for decades. There are many ways of making an airplane feel stable or more (or less) stable - mechanical, aerodynamic, analog, digital,... and they've been in use since not long after the Wright bros (or pick your figurehead) began flying. Ever fly a GA airplane with a bobweight? Downspring? Yaw damper? Servo/Anti-servo tab? I suspect a lot of people here have, and whether they realized or not these are all devices to change *apparent* stability without actually changing the aerodynamic stability.

Nauga,
WDTSFAL
 
CFR14 25.173 requires control force gradient of not less than 1 lb per 6 knots speed change from initial trim. It does not have to be linear but it does specify a linear gradient that must be met. As for the A320, do you know for a fact that part 25 was the cert basis, with no waivers or variances?

Regarding 'stability as canon', bare airframe (unaugmented through any means) stability has not been 'canon' for decades. There are many ways of making an airplane feel stable or more (or less) stable - mechanical, aerodynamic, analog, digital,... and they've been in use since not long after the Wright bros (or pick your figurehead) began flying. Ever fly a GA airplane with a bobweight? Downspring? Yaw damper? Servo/Anti-servo tab? I suspect a lot of people here have, and whether they realized or not these are all devices to change *apparent* stability without actually changing the aerodynamic stability.

Nauga,
WDTSFAL

The A320 has no "feel" in normal law, but alternate and below, it has artificial control forces in the sidestick. I understand this is to fulfill .671 and .672 and .173 etc.

I only have 10-ish hours in a sim in a 'bus, but when it kicks off Normal law, the stick forces completely change.
 
I just did.

I've read all of part 25. Nowhere is a linearity requirement for control forces specified. So, no, you've not.

how A320 control forces feel in alternate, abnormal alternate and direct law

They feel the same. There is no artificial feel in the A320 control. There are a set of centering springs and dampers, that's it, and they behave identically irrespective of the control laws in effect.
 
I've read all of part 25. Nowhere is a linearity requirement for control forces specified. So, no, you've not.



They feel the same. There is no artificial feel in the A320 control. There are a set of centering springs and dampers, that's it, and they behave identically irrespective of the control laws in effect.

How much time you've got in an A320? You're saying that with AP on, the stick still only has "springs and dampers" and feels exactly the same? There's much more happening in that sidestick than you seem to understand.

Also, A320 is not applicable. You'll find in the Federal Register that there are plenty of "Special Conditions" for the A3xx.
 
Last edited:
Status
Not open for further replies.
Back
Top