Re: Can an Airbus 320/321/330/340 perform a controlled glide?

Date:         23 Apr 97 02:58:15 
From:         Pete Mellor <pm@csr.city.ac.uk>
References:   1 2 3
Followups:    1
Next article
View raw article
  or MIME structure

In response to me:-
>>In the event of loss of all 5 computers in the Electrical Flight
>>Control System (EFCS) the crew can control pitch by moving the
>>Trimmable Horizontal Stabiliser (THS) using the pitch trim
>>wheels, and yaw (and roll to some extent) by operating the rudder
>>via the pedals.

Gary S. James <gsjames@ix.netcom.com> wrote on Tue Apr 15 22:03:25 1997:-
> I fly A-320's for a major US airline and as part of our training we
> area required to "fly" the airplane using these backup modes.  I assure
> you, that from a practical point of view the airplane is virtually
> uncontrollable in this mode and would be incapable of making anything
> remotely resembling a controlled landing.

Interesting! My understanding is that "manual back-up" is intended
solely to enable the aircraft to be controlled temporarily (and for
a short time) in cruise, to allow time for the EFCS to be restarted.

However, I believe that for certification purposes, it was required
that a landing under mechanical back-up be demonstrated. I have heard
that this was done on a test flight in Toulouse, and that the test
aircraft went through a short final approach, but did not actually
land.

I have also heard that crews are trained in "manual" landings on
simulators.

Any definitive information on the above "things I have heard" would
be welcome.

Gary added:-
> Additionally, there has been one instance where all flight computers
> were turned off momentarily in flight.  Panic ensued... and the
> aircraft nearly went out of control.

Any details about this incident? Again I have heard that on one
occasion, both ELACs were lost due to an overheating problem, but
that still leaves the three SECs, and only causes reversion to
"direct law" (i.e., the stick commands the surface positions
directly, as in conventional aircraft). No total loss of EFCS
has ever been reported.

So, what were the guys playing at in the incident you mentioned?
How did they lose all five boxes?

The buttons for each of the 5 EFCS computers are in the overhead
panel (unguarded): push once for off, and again for on.
*Accidentally* hitting five separate buttons seems an unlikely
scenario. Power failure would only do it if it were total (since
the two ELACs are on different sides regarding power supply,
and the three SECs are also spread over the two sides).

Each box (ELAC and SEC) is dual channel ("command" channel and
"monitoring" channel). Any mismatch between the outputs of the
two channels will shut down the box. (The remaining boxes then
automatically reconfigure themselves to cope with the loss.)
There is therefore the possibility of a transient (hardware
or software) failure shutting down a box, but 5 *independent*
transients on each of the boxes??

There remains the possibility that the failures were not
independent, e.g., they might have been due to similar software
bugs. Now that would be *really* interesting ...

Peter Mellor, Centre for Software Reliability, City University, Northampton
Square, London EC1V 0HB, UK. Tel: +44 (171) 477-8422, Fax: +44 (171) 477-8585
E-mail: p.mellor@csr.city.ac.uk
-----------------------------------------------------------------------------