PPRuNe Forums - View Single Post - CONCORDE ACCIDENT - PART 2
View Single Post
Old 30th Aug 2001, 19:33
  #31 (permalink)  
llamas
 
Join Date: Aug 2001
Location: Brighton, MI, USA
Posts: 37
Likes: 0
Received 0 Likes on 0 Posts
Post

Reagrding Glueball's comments about margins and redundancies in compoenents and the likelihood of their being a contributing factor to the wreck -

I've read this kind of thinking in many places and as an engineer, it bothers the heck out of me. I hope that this is not common thinking among aircrew.

Sure, engineers design margin into most things, and in the airplane business, those margins are very well-defined and specifically tested. So far, so good.

But the limit - eg the MTOW printed in the owner' manual - is not a black-and-white thing. IOW, just becauseyou are at or below that limit, does not mean that the performance of (insert part name here) will be as every bit asgood as it will for every lower loading condition. As you approach that limit, the margin designed into the part is consumed in direct proportion.

If you are at the limit - or, as it appears in this case, if you exceed the limit - you start eating margin even more rapidly, because now the part that you have stressed above its intended limit may now be asked to accomodate another failure in another area.

Bear in mind also that the analysis and testing used to set that design limit and margin of safety is only as good as the limits under which it was performed. For example, as in this case - the MTOW is intimately connected with the capacity and the margins of the tires. But did those calculations include the possibility that the wheel specer mentioned - which I understand keeps the tires from touching - was omitted? If they did not, and I doubt that they did - now you have an unknown eating away at your margin, which you have already, knowingly, largely consumed.

A single excursion to the margins is usually tolerable. It's the multiple drains upon the margins of a design which lead to catastrophic failure. Some of them may be unknown and unknowable. And it's often the "stepping" of limits which ends users up in the soup - the thinking that says 'we've operated this way before, that means we can again, and again, and again, and we will still have the same margin at this level of operation as we had at lower levels."

For a perfect case study of how this sort of thinking leads to bad outcomes, may I suggest "The Challenger Launch Decision" by (?) Vaughan. A classic example of working at and beyond the margins, and what may result from.

llater,

llamas
llamas is offline