Originally Posted by
Chris Scott
Thought you'd appreciate "rose-tinted spectacles" in relation to a product of "la Ville Rose"?
I didn't make that connection admittedly, but I was pretty certain you didn't mean any harm. The only reason I wanted to clearly refute the notion is because I am well aware of the system's limitations. That I don't consider it worse than anything else around at the moment is based on a lot of reading and a series of very heavy-duty modules around software engineering and reliability that made up part of my degree.
You talk of Airbus's departure from existing flight-control-system design causing "consternation... particularly in the piloting community" I think the arguments that we in the technical study groups of BALPA and IFALPA raised in the mid-1980s in opposing the non-driven throttle levers and non-interconnected sidesticks, were correct, but their ergonomic deficiencies have caused fewer problems than we predicted.
I'd slice things a little more finely than that. As an engineer I know that I'd want concerns laid out as soon as possible in order to make sure that they are dealt with. One thing that was made crystal-clear during the courses I mentioned above was that this was one of the most - if not *the* most - difficult and serious undertakings in aviation engineering of that era. The software angle alone involved heaven knows how many man-hours of testing, re-testing, hardcore statistical maths that would take me years to understand - the tools we take for granted in software metrics and design today didn't exist, so they had to be invented and built with the same level of rigour as the final product.
I'd hope that if you and your colleagues in the technology group had been privy to just how much work went into it at the time, it may have at least eased your worries slightly. The reason the flight controls and thrust levers were designed the way they were had nothing to do with an intent to take anything away from the flight crew, it was simply reasoned that the technology had become reliable enough to make that evolutionary step safe, and provide the feedback in other ways.
I take a little issue with still referring to these factors as "deficiencies" - at this stage, given that their reliability and suitability for the task has been proven I think it can be reduced to "differences".
(Of course the other big concern was the general reliability of computer systems, including Airbus's decision to use modest, tried-and-tested, commercial microchips - rather than purpose-built, mathematically-correct ones - and their decision to avoid triplex architecture. It was also widely predicted that system anomalies would be untraceable by accident investigators. Nevertheless, we didn't black-ball the inevitable.)
The Apollo programme and Concorde were of the era of purpose-built computers and logic - I suspect Airbus took their cue from the introduction of general-purpose microprocessors with the Space Shuttle, and surmised (correctly) that was the way the industry was headed.
Truth be told, with proper application of redundancy, defence-in-depth and tried-and-tested hardware - I suspect that the commercially-available equipment would provide equal or better reliability than a bespoke unit at a fraction of the cost. Of course, only time would bear that assertion out.
Two fatal accidents later brought AFS ergonomic weaknesses to light.
Which are you referring to here?
As an aside, my first ever flight was on a BCAL 1-11. I don't remember much of it however, being only three years old at the time.