Re: AIRBUS vs. BOEING (cont...)

From: ("Clive Leyman")
Organization: Pucklechurch Consultants
Date:         21 Jun 95 02:56:55 
References:   1
Followups:    1
Next article
View raw article
  or MIME structure

I really did not expect Thornton to post my reply to his first query, but
since he did I think it reasonable to post my second reply. The point I
was,and am, trying to make is that there are no black and white arguments
in this debate.
>Ok, I accept that the consequences of a software bug are different, but I
>still think that there is no right or wrong solution to a given problem.
>We would probably never agree on whether AI or Boeing philosophy is
>better, but I would ask you in turn to remember that 65% of all fatal
>accidents are caused by pilot error, so it is not at all obvious that it
>is better to give the pilot the last word in all cases.
>In fact, the AI FBW system gives protection against stalling, speed
>exceedance above the authorised cruise limit, low energy (windshear)
>situations, and exceedance of structural margins. Given the number of
>times accidents have been caused by stalling and windshear events, I do
>not think those protections are an unwarranted intrusion into the
pilot's >authority.
>The exceedance of Vc is perhaps less advantageous, but hardly a real
>problem for the pilot.
>The big argument going on in this thread seems to be the limitation to
>2.5g, or 2.0g with flaps extended. I suspect that this is a pretty
>emotive subject amongst pilots, but since they anyway cannot exceed
1.69g >without stalling on the approach, there seems to be some
misunderstanding >of the real situation here.

>With respect to the A320 accidents, I have occasion to read the accident
>reports recently, as I felt it necessary to reply to someone who had
>written a magazine article purporting to show that the FBW system was
>implicated in each case.
>I tried very hard to be objective, but I have to say that I could not
>find any evidence to support his arguments.
>In one case the pilot had deliberately switched off the low energy
>protection, and had then put the aircraft into a very low energy
>situation with himself in control and got it very wrong!
>In the next incident, one pilot (under training although still carrying
>passengers) made a system selection that again disengaged the low energy
>protection, and let the aircraft get into a non-recoverable situation.
>The check pilot, who clearly understood what was happening, let him
>continue with the error until it was too late. Pilot error? or FBW
[Consideration of these two incidents, which both were caused by
deliberate or inadvertent deselection of a major protection, makes it
particularly galling to read criticisms of Airbus because the other
protections cannot be over-ridden by the crew]
>The third accident was attributed (in my view correctly) to crew
workload >combined with a misleading design of AUTOPILOT selector (NOT
FBW, but >certainly man/machine interface), together with the absence of
GPWS, >which was an airline decision not to fit. I wouldn't put this down
as >pilot error, but it certainly wasn't A FBW fault either.
>The fourth A320 accident was due to a combination of a gross overspeed
on >the approach, combined with a flooded runway which gave rise to
>aquaplaning. Again you could not honestly blame the FBW system for that.

Don't get me wrong, I am not saying the AI design is perfect, but it for
sure is not the dangerous monster that the comments in this thread seem
to suggest.

So to be objective, here are the features I don't like on the AI and
Boeing designs:-


Lack of throttle movement when under autothrottle control.

Over complication of the individual computers, and perhaps more
importantly, the complicated interaction between computers in failure


Complete reliance on electrics for signalling control demands (no
mechanical backup)

Use of a single design of computer and software to drive the FBW - the
possibilities of common mode failures must be obvious.

As I understand it,no protection against low energy (windshear) situations