Re: Design Philosophy -Reply
Stephen Stough (stough_at_earthlink.net)
Tue, 25 Mar 1997 17:38:57 +0000
Geoffrey McCarthy wrote:
>
> I do have a serious thought on this: there can be no doubt that
> differences in levels of automation, and differences in tasks
> automated exist among manufacturers and airlines. More important, I
> believe, is the underlying understanding of the role and needs of the
> pilot who is in "control". It is my opinion that a certain European
> manufacturer has made a fundamental, fatal error in isolating the
> pilot from essential aircraft feedback by making the (side) stick not
> proportional, and disconnecting the throttles from movement by the
> autopilot. What is the jet doing? What is the real power setting?
> Moreover, it could be argued that the bank limit (55deg, I beleive)
> is detrimental; someday a pilot will need to go beyond it to save the
> jet and HAL won't let him. I conclude that an excess of accidents in
> the life of this airframe has already occurred because the crew
> couldn't cope with the automation and no basic, "manual flying" mode
> is available. To be impartial, a noted safety researcher, Dr. C.
> Billings, who has intensively studied automation accidents, does not
> agree with my inferences. I should also state that I have some time
> with fly-by-wire (F-16) which has built in limiters to keep the
> pointy end forward and prevent stalls and departures (at least one
> accident might have not occurred without them). There is no doubt
> that automation will continue simply because the technology exists
> and because of the assumption that it lowers workload and increases
> safety. It has however grown without much consideration of the
> human. Opinions my own. Geff McCarthy MD
Geoff, I think what you are trying to say is that the Airbus is a killer
due to the level of control authority and lack of feedback provided to
the pilot. As a technologist, you may find it surprising that I agree
with you. Every piece of added anything (more people, more software,
more components) will tend to increase the catastrophic failure rate.
That is why engineers in companies like mine (a large aerospace
manufacturer) spend so many years trying to develop requirements and
test prototype designs with users. You comments above, as simple as
they may seem to you, help those of us in the technology biz better
understand what we are designing and if we should be doing it at all.
---- Steve Stough