Also of interest, it reads like there is a logic error in the software.
In the current design, the system engages for 10 seconds at a time, with five-second pauses in between. Under conditions similar to the Lion Air flight, three engagements over just 40 seconds, including pauses, would send the plane into an unrecoverable dive, the two people involved in the testing said.
We knew this shortly after the Lion crash when Boeing fessed up about MCAS. Boeing's original 'reminder' said something to the effect (can't find exact wording just now) "left unchecked, MCAS can render the plane uncontrollable" (may have said unrecoverable).
The article is pointing out that just three full cycles of MCAS activity (10s trim down, 5s pause, 10s trim down, 5s pause, 10s trim down) without
any pilot counter action will leave the plane unrecoverable.
It was shocking when we found out about it. It's still absolutely shocking that Boeing would design such a system, have it rely on a single sensor, have it certified and not tell pilots about it.
There are lots of aircraft where 40 uncorrected seconds of a 'runaway' action at certain points in the flight envelope can cause an accident. If you know how to react 40 seconds is a long time, if you don't it is a very short time. Pilots have a lot less than 40 seconds to make a decision on what to do during an engine failure on take-off. Keep in mind to counter the action of the MCAS requires one of the following
[list=]Manual activation of nose-up trip every 5 second period countering any nose down trim for the entire flight (not something a pilot wants to do)
Engaging the trim cutouts and manually trimming the aircraft[/list]
Now pilots who were well versed in the stab trim runaway scenario would be able to either. However, it is increasingly obvious pilots are not well versed in the overall procedure as in most prior cases or runaway it could be stopped by either disconnecting the autopilot or pulling on the yoke in the opposite direction. Therefore, they only practice the first few steps in the overall process and are likely to forget the ultimate mode of engaging the stabiliser cutouts. If you remember all of these no problem; however, if you never get to the third action you will tend to forget that part.
While I think not making use of the second AoA vane in the MCAS system is very poor design it isn't, by itself, a catastrophic failure point. There should be enough potential checks that it shouldn't be an issue. Unfortunately, these were also removed over the years in the way Boing/airlines trained and airlines flew their NGs. This is where mentioning the MCAS system to pilots, specifically that it takes using the cutouts might have helped. Unfortunately, unless the cutout part was regularly practiced it might have had no effect. The Lion Air pilots never tried the cutout even though it has been in the FCTM for a number of years.
This is the inherent problem with any 'differences' training, it has to overcome all of the preconceived biases/shortcuts that pilots use from the prior models. These accidents bear a surprising similarity to British Midland Flight 92. The system changed, even with the differences training, the pilots reverted back to what they had done over and over. Note: I don't think having the AoA disagree light on LNI610 would have solved the problem. However, having it for the prior flight might have helped. The reason why is that the prior crews reported IAS disagree, but the MTC logs had AoA disagree, this caused confusion that helped lead to the problem not being properly addressed.