Originally Posted by
infrequentflyer789
I think that's one possibility / interpretation, but back in June at
theaircurrent.com Jon Ostrower suggested that it took too long because the test pilot "
found the electric trim switches on the pilot’s yoke unresponsive". If that is correct and the cause of delay is overwhelmed FCC (as is also stated in that article) then the implication is that manual electric trim doesn't override autopilot (with relays or switches) but rather manual trim is a request to HAL which then decides
(in software) if it will allow you to override its trim command.
I will state very clearly:
I struggle to believe that interpretation and the implication, because I can't see how you could do it that way without triplex redundancy and voting etc. - but I may very well be wrong. I await more definitive information with interest.
While that was how the test failure was initially reported, later reporting refuted this particular detail. A subsequent article discussed how an intentionally induced fault (caused by "flipping bits" on the processing chip) created a runaway stab trim situation with the A/P engaged. It was said that this fault had nothing to do with MCAS and was only theoretical as it had never been seen in actual operations. The evaluating pilot judged that it took too long to intervene and complete the runaway stab procedure.
I'm reading a bit between the lines here because the report did not go into a lot of detail. However, a stab runaway with the A/P engaged would likely first present as a "Stab Out of Trim" annunciation. This light illuminates when the A/P is holding too much elevator pressure and
usually warns that the A/P trim is not keeping up with stab trim inputs. This occasionally happens with rapid speed and configuration changes that require more than the A/P trim rate. The first step of the Stab Out of Trim NNC directs the crew to see if the stab is actually moving. If so, then their instructions are to "Continue normal operations" and
exit the checklist. If the stab isn't moving (A/P not working), they are directed to disconnect the A/P and trim with the Main Electric trim.
If the underlying problem, however, was not a case of A/P stab trim getting behind, but rather a runaway trim, then the time to run this checklist ( see the annunciation, designate the flying pilot, pull out the manual, turn to the correct page, read the execute the steps, put the manual away) are all seconds delaying the response to a runaway trim. If the flaps are extended and the A/P is trimming at the high rate (.27 degrees/sec), this is not a trivial amount of time. If the actual malfunction is a trim runaway, then the crew may not realize it until the A/P disconnects in an already highly out of trim state.
Instead of exiting the checklist, this non-normal
should direct the crew to monitor the direction of the trim for reasonableness (trim up if slowing, trim down if accelerating) and be prepared to intervene if runaway trim is suspected. BTW, this issue would affect the 737NG as well as the MAX.