After reading some of the preliminary reports from the FDR and CVR data, there is a lot more going on in that Ethiopian crash than just MCAS trimming nose down due to faulty AOA probe. The captain's instruments seem to all be going haywire, causing all sorts of confusion as well as inputs from MCAS. At first glance it looks like the way to save the plane was to realize that he had unreliable airspeed and attitude and given control to the first officer, whose instruments seemed to be working correctly. As it was, they left the power all the way up and got going extremely fast while flailing around trying to make sense of his own instruments. Once they got going so fast it was too difficult to manually trim the stabilizer due to air loads, yet they never realized this and slowed down, they just kept accelerating all the way into the dirt. For reference, our normal climb airspeed is anywhere from 280-320 knots. At anything over 300 knots, you really start to hear it in the plane with the increased wind noise around the fuselage and especially in the cockpit. Anything other than perfectly smooth air, and all the small bumps are magnified. Yet these guys were doing 380 knots in level flight and 450 knots on the way down after they couldn't pull the nose up any more. I can't even imagine what that must have felt and sounded like in the cockpit.
Just so y'all know, we do sim training on unreliable instruments, which basically requires us to set the power at a known setting and put the nose at a known attitude so we are right in the middle of the controllability envelope until we can analyze everything and decide what's working and what's not working. The other thing that is confusing to me is why they didn't keep trimming nose up while the electric trim was still turned on. For a pilot, trimming is like breathing. You don't even think about it. If the nose is heavy you just reflexively trim back so the nose comes up. For instance, we calculate a takeoff trim setting on the ground for our takeoff weight and power setting, but if the plane isn't full and a lot of people are sitting up front, its completely normal for the yoke to require extra effort to pull back when you rotate on takeoff. Any pilot out there feels this instantly and starts trimming the nose up so the control forces are in the normal range. Again, you don't even think about it. Yet, in the report, it says a couple times that they would use the electric (normal) trim for nose up but then they stopped trimming when they got to about 2.1-2.5 units. Well normal trim range for this plane is around 5-6 units for most flight regimes, so the nose must still have been heavy as hell when they stopped trimming. And of course after they stopped trimming, the MCAS was kicking in and trimming nose down, so the whole time they are exerting a bunch of effort to keep the nose up without trimming, which is weird, and the speed keeps building up due to a combination of high power setting (which is also weird) and the nose coming down and creating a descent.
When this is all over its going to be very interesting reading. At this point it seems to be a combination of Boeing making a very bad decision on how to implement this MCAS system, and how that system contributed to making what should have been a recoverable malfunction into a mishap. But even more than that, I think what we're going to have to come to grips with is how the increasing demand for air travel is on a completely opposite curve that graphs the availability of well trained pilots. Even people who go through a complete course of approved training today are flying airplanes that do so much stuff automatically I find myself wondering how they are able to handle things when they have to take over manually. Even some of the primary training aircraft such as the Cirrus have nice autopilots and trim themselves, and people who think they are learning how to "fly" are just learning how to push buttons. Its a self-licking ice cream cone because when we have a crash, the natural inclination is to make the planes even more safe by adding more automation, but then that just contributes more to the cycle of creating pilots that don't actually fly the airplane. Not sure where it will all end, but I'm pretty sure that guys who have good flying skills are going to be in high demand for the foreseeable future. The one thing you can't buy with all the money in the world is experience. I don't want to knock those Ethiopian pilots but I'm pretty sure if they would have hit the geese out of LGA they wouldn't have ditched it safely in the Hudson like Sully.
Best Posts in Thread: 737 MAX
Hearing some rumors that in anticipation of what they think the data recorders are going to say, that there is already a software update for the plane which will certify it to fly again. The gist is that the system that automatically trims nose down if it thinks the plane is stalling can be fooled into doing the nose down trim thing if it gets bad data from the sensors such as AOA vane, pitot tube, etc. The software fix will most likely involve additional cross checks so that the trim system in question will only operate if all the air data sources are showing congruent information. In short, a system designed to help the pilot trim nose down when his airspeed gets too slow isn't inherently a bad system, but it has to be robust enough to only do that if the plane is actually slow and near stall. If the plane THINKS its slow and activates that trim while the plane is moving fast, it is no bueno. Frankly, I'm amazed that they engineered it this way in the first place. Seems like a pretty boneheaded approach to the whole thing and some really poor decisions were made if this is indeed the cause. I have no doubt the plane will enjoy a long successful service life but the start of service didn't have to be marked by these crashes.
BTW our company airplanes have over 41,000 flights and over 90,000 hours logged so far and this anomaly has not surfaced even though the data for every flight is logged and available.
- Like x 8
As a matter of fact, keeping it simple is what has driven a lot of decision to make the Max instead of a new plane. At SWA, the entire operation is built on simplicity. Unlike other airlines with up to a dozen different types of airplanes, when you have only one type then every pilot can fly every plane, every FA and mechanic is familiar with every plane, the ground equipment, tools, jetways, etc all work with every plane, the dispatching and planning software works with every plane, etc. It is a huge cost advantage, which lets the company sell cheaper tickets and still make a profit.
This is the real world of trying to run a business. Trying to make passengers understand the constraints is frustrating. Every time a flight is delayed or canceled, its the same old questions. Why don't they have a spare aircraft? Why aren't their spare crew available? Why don't we have more runways or more de-ice trucks or more baggage handlers or mechanics or spare parts? The answer is we could have all that stuff but the price of your ticket would double, and then no one would fly and there would be no airline.
Now of course that being said, the trick to running a reliable and profitable operation is balancing those factors with the inevitable breakdowns, weather events, etc that occur every day. You can't plan an operation with the idea that every plane will be used, every seat will be full, no one will call in sick, etc. Because of course the first time any of those things happen the whole house of cards comes down. Frankly I find it amazing that after a big event like the blizzard they just had in Denver, just how quickly an airline can get its operation back to some sense of normalcy. Its a massive undertaking to move that many pieces of a puzzle around on a daily basis, really mind-blowing if you think about it.
I know I've strayed from the topic of the Max but yes they could have refused to build a new model 737 and started from scratch with a completely new design, but I just wanted to point out that its not a simple from an execution standpoint as some might be led to believe. There is a massive cost for such a decision. I don't know how long SWA will be an entirely 737 fleet but it'll probably be at least until I retire in 14 years.
- Like x 5
1) Boeing did not disclose to airlines that they introduced the MCAS system into the MAX family. It is not part of the training manual, sim training, nowhere to be found. Boeing figured there was no need to share information about the MCAS with the airlines because it would essentially operate in the background and didn't require intervention from pilots. Inexperienced pilots or not, if you are unaware of a new system, it's pretty damn hard to troubleshoot an issue when it arises.
2) For a fact, the Angle of Attack (AoA) vane was feeding incorrect data to the onboard computer of the Lion Air flight. This enabled the MCAS as the system "believed" the aircraft was approaching a stall even though this wasn't the case. So imagine yourself as a pilot, receiving incorrect AoA data and your aircraft automatically dives from the input of a system that you aren't aware of. The Ethiopian pilots reported to ATC that they were receiving erroneous speed readings which could've also initiated the MCAS, again as the system "thought" it was nearing a stall.
Experienced pilots or not, Boeing completely schit the bed by introducing the MCAS without disclosing or explaining the system to the airlines.
Ya wanna know what's really scary about the Ethiopian crash? That the First Officer had a whopping 200 hours of pilot time. Basically a student that shouldn't be allowed to fly even 3 people around for money, much less 160 people. You can't even get the license required to APPLY for that job in the states without 1500 hours, and with that the only place that's going to hire you is a tiny puddle jumper outfit.
Heck, the F18, which is so old they are retiring it, isn't even controllable at all without computers running. Using computers to augment the flight controls is neither new or a design flaw.
They're probably going to figure this out really soon when the get the box data. Let 'em figure it out. Make me guess, sounds like crappy software and training issues. Problem is, that at this point it is a wild guess. It could very well be two entirely unrelated sequences of events. Old airospace engineer here who hasn't designed anything in years.
Wait till Boeing stock tanks a little more then buy. The 737 Max is not the de Havilland Comet.
Probably the best article about the situation I've read.
Flawed analysis, failed oversight: How Boeing, FAA certified the suspect 737 MAX flight control system
- Like x 2
- Facepalm x 1
1. Don't like shooting my mouth off when I don't have the facts.
2. Read an earlier post I made about cascading events (and see Detroit's post above).
3. It's a plausible explanation BUT all the facts aren't in yet.
4. Focusing on software for a moment. If all of his post is close to the truth, he's got one thing wrong. Of course software was A PROBLEM. I don't view the job of a systems engineer as separate from software design. This would be a Harvard case study where the system engineer/software designer fails to understand, pay attention to, or communicate it's impact on the entire system. And consequently the guy coding and other people involved have no clue. The bigger and more integrated the system, the more critical it is to talk to one another. And that's why we do not outsource software development. Something always get's lost in the wash. What you get is not what you need or in this scenario what is absolutely critical.
5. We still don't know for sure. And I'm still having a hard time believing they dropped the ball like this.