The Why of the 737 Max

By now everyone has heard about the Boeing 737-800 MAX. Two of them crashed in a short time frame, and it is believed that a new automatic flight trim system is to blame. I’m not going to rehash everything floating about the news, since the Seattle Times actually does a decent enough job of it. In short, this is on both Boeing and the FAA, and both should get taken out to the woodshed over it, both criminally and civilly¹.

Now, first order of business: the why of all this.  The 737 is a fun little airliner with a rich history.  One bit about that history is that the 737-100 (the OG 737) was fitted with low-bypass turbofans.  Low-bypass engines have a rather small maximum diameter, which allowed the 737 to have a very low ground clearance and short landing gear.  When Boeing hung the larger, more modern high-bypass turbofans under the wings (the CFM-56), no one wanted to lengthen the landing gear or change the wing very much, since that would necessitate a whole lot more redesign than the (already considerable) work a new engine entails. Instead, trade-offs happened, and the nacelle was pushed forward of the wing and up, and the bottom of the nacelle was flattened, giving the engines the look of a chipmunk with stuffed cheeks².  And still, the nacelle was a mere 18″ off the ground, and the landing gear have 12″ of vertical travel.  Factor in wing flex during touch down, and if you aren’t careful, you’ll strike your engines on the deck and put your aircraft in the hanger for maintenance.

Regardless, that design has worked for a very long time.  Now, back when I was in Propulsion Aerodynamics, we were working on the re-engine of the 737, and the newer engines are even of a larger diameter, and thus to avoid significant changes to the landing gear and wing, the nacelle was hung even further forward of the wing so it could be lifted even higher up off the deck.  Problem was, as it got higher and further forward, the exhaust plume started impacting the wing aerodynamics in bad ways³.  To relieve that, the thrust of the engine was angled down ever so slightly.  Now, it’s important to remember that aircraft have three centers: a Center of Pressure, which is the point where Lift and Drag act through; an Aerodynamic Center, which can be thought of as the center of pitch; and a Neutral Point, which is where you want your center of gravity to be so the aircraft is stable. Moving the engine forward and up, and angling the thrust down changes the relationship between the Aerodynamic Center and the Neutral Point and makes the aircraft want to pitch up more than normal.

The MCAS system was designed to automatically counter that tendency by using the horizontal stabilizer more than normal for that aircraft.  See the Seattle Times link for more details about MCAS. So far, none of this is bad; it’s just engineering trade-offs. The bad comes later.

Again, I won’t go over things that the article above discusses adequately, but I will make a few points I think it misses.

Regarding Boeing acting as it’s own regulator: This is something that has been going on for a while. Engineers and technicians at Boeing receive specialized training from the FAA so they can act as FAA Reps. The deal, as it was explained to me, was that in exchange for Boeing having the FAA reps in house, Boeing was not allowed to try and pressure those reps to let things slide, as the reps are legally responsible should they fail to do their duty⁴, and they can call their FAA contacts should the feel Boeing was getting pushy. And Boeing engineers are unionized.  Keep those last bits in mind.

So, the FAA handing off duties to Boeing is not as nefarious as some make it out to be, at least not on it’s face.  Doesn’t mean it can’t fail.  As the 737-800 MAX was getting certified, Boeing’s chief rival, Airbus, was pushing it’s 737 competitor, the A320Neo, out the door as well. So there was a race to beat Airbus out the door and secure those contracts from airlines hungry for more efficient aircraft.  From what I’ve read, part of the problem here is that the FAA management drank the Boeing management Kool-Aid and decided Boeing really did need to beat Airbus.  This puts the Boeing FAA reps in a bind, because not only are they getting pressure from Boeing to push the timetable, they are also getting pressure from the FAA, who is supposed to be their 800 lbs gorilla to keep Boeing off their back.

Regarding the FAA, I have no idea what they were thinking. I suppose if I mentioned that the 737-800 MAX first flight was 01/29/16, and the first delivery was 05/22/17, one could compare those dates with other events occurring in the executive branch and draw some conclusions…

Regarding Boeing, I have a pretty good idea.  IMHO, Boeing has serious management problems.  First off, there is a culture of near-worship of executives and senior management.  Lots of massaging of bad news and ass-kissing and the like.  I mean, it’s bad.  Along with that there is a lot of ‘managing up’ going on, where senior managers and executives spend more time focusing on the politics of their positions, and how good they look to the people above them and how those people view them, rather than focusing on what the people below them are doing, or saying.  I’ve heard of multiple incidents where mid-level leadership was pushing the 1st and 2nd line managers to get things done faster, and those managers and their senior employees had to push back with the safety of the aircraft (i.e. if you make us do this faster, we won’t have time to check our work, and you put the safety of the aircraft at risk).  Tack onto that the mantra of shareholder value over customer value.⁵

Anyway, both organizations have to own this.

By the way, the bit about the FAA Union being critical of Boeing having FAA duties, I’d take that with a grain of salt.  It’s a smaller Union wanting to boost it’s membership by essentially poaching people from the larger Boeing Union (SPEEA).

Now, onto my final point, the pilots.  This is not victim blaming, per se, since I know those pilots did everything they knew how to keep those planes flying.  The key here is “everything they knew how”.  This is the issue; pilots are in demand across the world.  It takes a lot of time and money to train a pilot, even if you get someone like a former military pilot, who already has the basics down.  That is time and money the airlines don’t want to spend, and not a ton of people want to invest in⁶ since the working hours of a commercial pilot can be hard on families⁷.  Places like the US and Europe can afford to shell out the time and money to competently train pilots.  Not everyone can, but everyone the world over expects airplanes to fly them about.  This means that less affluent countries are in a bind. They need pilots, but they can’t attract them from places like the US, Japan, and Europe, so they have to train them locally, and too often, they teach their pilots to fly the computer rather than how to fly the plane.

I mean, a modern airliner can almost fly itself.  The pilot really just needs to taxi it out to the runway and wait for permission from the tower.  The plane can, for all intents and purposes, take off on it’s own, climb to altitude, cruise to it’s destination by way point, and if conditions are good, land itself, whereupon the pilot taxis to the gate.  Of course, all of that assumes good conditions.  No bad weather, no unexpected air or ground traffic, no avians in the inlets, no faulty sensors to confuse the automatic systems, that kind of stuff.

So pilots get trained to basically let the plane fly itself, and if not enough attention is paid to what to do when the plane can’t fly itself. Now, lots of American pilots have encountered a dodgy MCAS, and they just shut the system off and logged the anomaly.  I even saw a report that the Lion Air plane that crashed, on a recent previous flight, had problems with the MCAS, and the the pilots were at a loss.  Luckily for them, a more experienced pilot was on board as a deadhead and told them how to turn the MCAS off.  So the issue was known, and how to correct it was known, but it wasn’t known to everyone, and if you are largely trained to let the plane fly itself, then not knowing is a problem.

And it’s not like modern countries don’t suffer the same issue to some extent.

Ergo, at the end of this tragic chain is the fact that accurate documentation matters, adequate training matters, and no one can take short cuts to training pilots.  And autopilots, while incredibly impressive, are not actually intelligent.  They are only as good as their inputs and software allow.  Only the pilots are intelligent, provided they are given the correct training and information.

^[1] I’m not entirely confident any individuals will see the wrong end of an indictment, nor will the FAA actually get in trouble, but monies will be paid, and it’s going to hurt.

^[2] The cheeks are actually stuffed with the engine auxiliary equipment: fuel pumps, oil pumps, generators, etc.

^[3] Remember, the air over the wing is supposed to move faster than the air under it.

^[4] AFAIK, they are still Boeing employees, and thus don’t enjoy the immunity government employees might.

^[5] E.g. the fact that Boeing did not include a warning light (that the critical sensors were not in agreement) as a standard option in the cockpit; it was a add-on the airlines had to opt-in and pay for.  I mean, WTF?!

^[6] Figure the cost to become a commercial airline pilot is easily in the ballpark of $100K.

^[7] My cousin flies for Southwest, his schedule is nuts, and he has 3 kids.

Photo by Gordon Werner