Re: Trim Activated by the Autopilot

Date:         18 Jan 2000 05:24:31 
From:         Pete Mellor <pm@csr.city.ac.uk>
References:   1
Followups:    1
Next article
View raw article
  or MIME structure

On 8 Jan 2000, John M. Hunt wrote:

Dear John,
> This brings up another of my pet gripes.  Why can't the Flight
> Management System simply announce in a very bossy voice "The aircraft
> is seriously out of longitudinal trim"

Your suggestion is very similar to the reaction of one of my colleagues
after I had delivered a seminar to the Centre on the Nagoya crash. His
suggestion was that a permanently visible display would have done the job.
In fact, there *is* a permanently visible indication of the stabiliser
position on the A300, but it is in the form of a mechanical pointer
attached to the trim wheels on the central console, requiring the crew to
look down and slightly aft to view it. (It is not illuminated.)

An audible warning of the type you suggest would (IMHO) be feasible and
highly specific in terms of the warning information given to the crew.

The movement of the THS by the use of the electrical trim by one of the
crew (activated by a thumb switch on the control wheel) is annunciated by
a "whooler" (a totally specific "owl-hoot" sound, completely unlike any
other audible annunciation in the cockpit) if the THS movement persists
for more than a second or so. According to the Nagoya accident report,
this was suppressed in the case of THS movement by the AP at the request
of the CAA on the grounds that it would occur frequently during "normal"
automatic go-around and distract the crew! The notion that a warning could
be given in precisely those circumstances where a dangerous out-of-trim
configuration is developing (and that this would be easily detectable by a
sophisticated automatic flight system) does not appear to have occurred
either to the certificators or to the certificatees.

Amazingly, the joint efforts of a number of expert people, all acting with
the best of intentions, has resulted in a botched design with a critical
feature interaction and no annunciation of the dangerous configuration to
which it can give rise.

I remain steadfast in my conviction (pace PBL) that this constitutes a
critical design fault, and that it was only a matter of time before it
caused a fatality.

Pete Mellor