[SystemSafety] B738M & B739M [was: Cognitive bias, safety and disruptive technology]

[Peter Bernard Ladkin]

On 2019-03-14 00:25 , Les Chambers wrote:
> 
> 
> Example - Two 737 Max 8s crash within five months of each other. Where:
> ....
> 
>                     3. Exhibiting the same porpoising  behaviour

Please be careful with this. The phenomenology of the limited data available so far indicates
differently.

Someone plotted vertical speed against time for ET-302 from the FlightRadar24 data. An eminent
aerodynamicist mentioned in private that the data seemed to indicate divergent oscillation: this is
not the phenomenon you would expect from crew countering (trying to counter) repeated stab trimming
from MCAS, which is what happened to JT 610. (I am not aware, though, of a similar plot for JT 610).

The FlightRadar24 data comes from the Mode S transponder info (ADS-B is the name of the ATC
protocol/system which uses this data). The altitude data is, at source, barometric information. It
is converted into the altitude in the International Standard Atmosphere which has that pressure,
known in aviation as pressure altitude. Pressure altitude needs to be adjusted for local QNH to get
true altitude.

There is a network run by Aerion, which I believe uses GPS data, and is a payload on the Iridium
Next satellite constellation. I don't believe Aerion is selling its service yet, but it is live and
under test. It could be that the "new information" available to the FAA is the validated Aerion data.

However, I have no idea whether the two datasets agree or not. I don't know that I have any contacts
able and willing to talk about the Aerion data.

The DFDR and CVR data are not yet available. The recorders have been found; the Ethiopians wanted
the German BFU to read them, but I understand the BFU declined, and they are going to the French BEA.

> 
> Compensating heuristic - trust your gut (born of experience) and ACT without delay.

Well, yes, in fact everybody "acts without delay". Aviation people are typically very hot on the
case, especially airlines flying that type and their crews, most of whom are as fond of retaining
their lives, or fonder, as anyone. There have been lots of very well informed people looking very
closely at events, including Southwest Airlines, which flies a lot of the type (comparatively
speaking; there aren't that many out there yet). SWA was happy continuing to fly the aircraft, and
said so, after JT 610 and also this week.

People might like to take a look at the thread on PPRuNe (under "Rumors and News"). It has about a
thousand posts so far, of which a couple of dozen contain helpful observations. Good information has
come from DaveReidUK, who is very good (and quick) at analysis (and synthesis) of FlightRadar24
data, and FCeng84, who has given succinct (and, I presume, accurate) information about the purpose
and the operation of the MCAS function.

> What was the FAA thinking?

US operations do not have a lot in common with Indonesian and Ethiopian operations and there is a
case for judging that appropriate protections against the phenomena exist in US operations.
Obviously that judgement has now changed at the regulator. Details are not yet public.

The main issue which came to light in the JT 610 aftermath was that there existed a protection
system, MCAS, which was software-based and cut in only in certain configurations and flight states,
but about which crews were not as well-informed (through training and the FCOM) as they are about
other flight assistance. The reason for the existence of MCAS is, according to FCeng84, to satisfy
an airworthiness certification requirement that, under all non-stalled flight regimes, with fixed
elevator trim, control stick forces should increase with increased AoA. At higher AoA in some
regimes, the engine nacelles on the newgen engines apparently provide additional lift, which reduced
the stick forces required to hold higher AoA. The solution was to adjust the position of the
trimmable horizontal stabiliser (THS) to render higher stick forces for elevator control as
required. This is what MCAS does in some flight regimes (I am not spelling out the exact conditions
here). There is considerable discussion on PPRuNe on the engineering merits (and demerits) of this
design.

PBL

Prof. Peter Bernard Ladkin, Bielefeld, Germany
MoreInCommon
Je suis Charlie
Tel+msg +49 (0)521 880 7319  www.rvs-bi.de





-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
URL: <https://lists.techfak.uni-bielefeld.de/mailman/private/systemsafety/attachments/20190314/fb63d69f/attachment.sig>