After Ethiopian 302 crashed on takeoff from Addis Ababa, I thought the decision to ground the airplane was hasty. But now after reading the entire preliminary report that came out last week, I’ve changed my tune. Now that we have more complete information, I would agree that grounding the airplane is the right thing to do.
I would recommend a read through the to get a full sense of what is known, but I’m going to focus on what I found so concerning here.
There’s nothing overly surprising about how the flight started. The take-off roll was normal, but once the airplane took flight, things started to go wrong instantly.
The Angle of Attack (AOA) sensors started showing extremely different numbers. The right sensor was correct showing the airplane at around 15 degrees to the air flow, but the left sensor dropped to 11.1 degrees and then spiked as high as 74.5 degrees. That data is obviously incorrect, but Boeing’s problematic MCAS system — a perfect example of awful design — only looks at the one sensor, and that is the sensor it used. Thanks to that bad data, the pilots were about to be thrust into a nearly-impossible fight for their lives, despite Boeing saying it should have been easily over-ridden.
Less than two minutes after departure, the flaps were retracted and the auto-pilot disengaged. (It’s unclear from the report why the latter happened.) That paved the way for MCAS to do its thing for nine long seconds. It saw the false angle-of-attack and changed the pitch trim from 4.6 to 2.1 to push the nose down. That may not mean anything to you, but put it this way. That stopped the airplane from climbing and put it into a slight descent. Once the MCAS stopped, the pilots recovered and pointed the nose up.
A mere 11 seconds after MCAS stopped pushing the nose down, it tried to do it again. This time pitch trim went down to 0.4 and the airplane descended more quickly.
Only 15 seconds later, the pilots did what they were trained to do in a situation like this, just as Boeing said after the Lion Air accident last year.
At 05:40:35, the First-Officer called out “stab trim cut-out” two times. Captain agreed and First Officer confirmed stab trim cut-out.
This should have prevented the MCAS from forcing the nose down again, and indeed, it did.
At 05:40:41… a third instance of [Aircraft Nose Down] automatic trim command occurred without any corresponding motion of the stabilizer, which is consistent with the stabilizer trim cutout switches were in the “cutout” position
The MCAS tried to push the nose down again, still based on the faulty sensor data, but the operative word is “tried.” This time, with the automatic trim disabled, the MCAS efforts did nothing. Good news, right? Well, yes and no.
The pilots at this point were trying to stabilize the airplane and continue climbing. But while they were doing that, the airplane started going faster and faster. Though the two airspeed indicators showed different numbers, the lowest one showed them accelerating to 340 kts. I don’t know Ethiopian rules, but in the US airplanes are supposed to be under 250 kts when they are below 10,000 feet. At those high speeds, manually trimming the airplane is exceedingly difficult thanks to the airflow over the surface.
The pilots continued to try to get the airplane to pitch up, and the Captain even asked the First Officer for help. One minute after the trim cutout was flipped, the First Officer said that he could not manually trim the airplane despite multiple attempts.
While we don’t know from the preliminary report what exactly happened in the next minute and a half, it appears that the auto-trim was somehow reactivated. Presumably the pilots, realizing that they weren’t able to manually trim the airplane, tried to flip the auto-trim back on. The cynic in me says that Ethiopia left this detail out of the preliminary report because it would have implicated the pilots in having done something wrong, but of course, we don’t know that. All we know is that the simple solution of flipping the trim cutouts did not save this airplane. It left the pilots with an airplane that was way out of trim and going far too fast. Boeing’s easy fix was anything but.
Looking at the flight data recorder (FDR) data shown on page 26 of the report, at this point it looks like the airplane was relatively level but that’s not good enough when you’re at such a low altitude.
Below you’ll see the end of the flight. In the top chart, look at the black line as showing the airplane’s altitude. Then look at the pink spikes at the bottom which show the pilot efforts to pitch the nose up. It didn’t have a big impact, and the airplane remained level. At this point, the auto-trim went back on, and you can see what happened in turquoise at the bottom. That’s where the MCAS kicked in and pushed the nose down, putting the airplane on its final descent into the ground.
The preliminary report confirms as much.
At 05:43:20… an [Aircraft Nose Down] automatic trim command occurred and the stabilizer moved in the [Aircraft Nose Down] direction from 2.3 to 1.0 unit in approximately 5 seconds. The aircraft began pitching nose down. Additional simultaneous aft column force was applied, but the nose down pitch continues, eventually reaching 40° nose down. The stabilizer position varied between 1.1 and 0.8 units for the remainder of the recording.
Now, had the auto-trim not been reactivated, could the airplane have been saved? If the data is right and it was flying mostly level AND if there were no obstacles that required the airplane to climb to clear them, then yes, they could theoretically have pulled back on the throttles, slowed the airplane down, and returned to the field. But with all these problems going on, that very well could have required some super-human piloting efforts. No perfectly-flyable airplane should ever put its pilots in that position.
You have two pilots overwhelmed by a host of problems right after takeoff that were caused by the airplane itself. Had the cutout returned the airplane to normal operations and they could have manually trimmed it and continued on their way, then fine, this works. That’s what Boeing said should have happened, and it’s what we’ve all been operating under. But the manual trim didn’t work, and the pilots were forced to try to fix a complex issue with little room between them and the ground.
To me, this goes back to the shockingly poor design of the MCAS system. I talked about this a couple weeks ago, and now I find it even more unconscionable that this system was allowed to fly on an airplane. Boeing designed this poorly and the FAA’s lack of oversight failed to stop it. In the short run, the software fix that Boeing is putting together will solve the problem. Then at least these airplanes can get back in the air. But in the long run? It’s going to be a very rough ride for Boeing and the FAA.