The crash of Ethiopian Airlines Flight 302 at Addis Ababa has never been a complete mystery. The accident, which claimed the life of all 157 onboard, seemed from the beginning likely to be related to the MCAS system that Boeing installed on the 737 Max to help it pass its flying qualities tests with the FAA and achieve certification quickly, allowing Boeing to bring the new 737 version to market to compete against the Airbus A320neo. And on the heels of the MCAS-related Lion Air 610 disaster in late October, the Ethiopian mishap had the telltale signs of an automation failure from the start. Now new evidence and theories are coming together to likely give investigators a clear picture of what happened and why things happened as they did in that short, horrifying flight of ET 302. That answer is related to facts both common to all airplanes and specific to the 737, and not just the Max.
New Details Emerge From Data Recorder
While many analysts are still discussing this as an accident caused by MCAS, that might or might not be the case depending on how you interpret the evidence. As with the Lion Air 610 crash, which is suspected to have been caused by a faulty angle of attack sensor, which triggered the MCAS to lower the nose, thinking the plane was on the verge of stalling. With Ethiopian Airlines 302 the AOA vane is also suspected as the initial complication. But in this case, there are signs that it broke off altogether before the accident, based both on the flight recorder data and the accident site investigation. It would not be the first instance of an AOA vane breaking off. Bird strikes have been known to take them out.
That is what Peter Lemme, a former Boeing engineer, writes in his blog, satcom.guru, going into some detail about the possibilities, developing in the process a very credible scenario for the accident chain of events.
Regardless of whether the vane broke off or not, it was disabled, based on the recorder data. The faulty sensor data it supplied the AOA would have activated MCAS, which would have then trimmed nose down, aggressively so, causing the nose to drop, to which the pilot flying, the captain according to the FDR data, continually trimmed against MCAS, which would re-engage after a few seconds. On two occasions the crew engaged the autopilot, only for it to disengage after a short time. Early in the flight the captain reduced angle of attack and also commanded flaps up. The airplane was accelerating rapidly and the crew was already struggling with control. And to complicate matters immensely, the left side stick shaker started going off early on and continued throughout the flight.
There were several questionable calls the crew made during the short and tragic flight. But several of these are part of basic flying skill. In an airplane that was accelerating toward disaster, the first thing a pilot should do is pull back on the power, but the accident pilots kept the engines spooled up to max continuous thrust, causing the overspeed horn to go off, which it continued doing throughout the flight.
One very troubling report is that the crew turned off the electric trim, which deactivates the MCAS functions, only to turn it back on again, all as they were fighting to gain control of the plane. Some observers have weighed in that the reactivation of electric trim was inexplicable, but it’s not. There are, in fact, almost always compelling reasons that pilots take action during an emergency, even when those reasons reflect a misunderstanding of the systems, resulting in intensifying the emergency.
In the case of the crew of ET 302, evidence suggests that the reason they reactivated electric trim is that after deactivating it earlier, they found that they couldn’t manually trim the plane. So they went back to electric trim, which, while not solving the problem, at least did something, and might buy them additional time.
But why they couldn’t manually trim the plane very likely had everything to do with airspeed control. The pilots, remember, left in full power, and as MCAS trimmed the plane nose down and it accelerated to well over its max allowable speed, the aerodynamic forces quickly multiplied, giving the manual trim a much heavier load to work against. Add to that the fact that MCAS had trimmed the plane nose down and it’s very likely the manual trim felt as though it were welded in place.
By slowing down might the crew been able to manually trim and thereby arrest the lethal descent? Maybe, maybe not. It was the right thing to do regardless, but the 737 (and every jet) can get to a flight condition at very high speeds where large trim corrections cause an unrecoverable pitch over. That appears to have happened with Ethiopian 302. Its descent angle as it dove toward the ground was a horrifying 40 degrees nose down. During that descent the captain and first officer both pulled back on their control columns, almost certainly with all their might, but there was nothing they could have done to prevent disaster by that time.
In concluding his analysis in a recent blog post, Lemme sums up both the temptation to blame the crew for their actions or inactions and why that is the wrong approach to take.
Human Factors is a science to understand what prevented a pilot from taking the expected action. It is easy to pontificate from a distance. Even the flight crews lost in these accidents would agree what they should have done in hindsight. The challenge is understanding why they made the choices they made and what would have directed them to a better outcome.
It’s an important goal, to work toward understanding what the crew could have done to affect a better outcome in part because the actual outcome could not have been any worse.