pyotr filipivich wrote:

The cheap/bad stuff wore out years ago.

The interesting question is, how much of this bad old stuff was there?

There is also the issue of "over engineering". My understanding
is that now adays we can get the numbers crunched closer to optimal,
and the result is that rated capacity tends to be closer to the actual
capacity than in "the old days". E.g. you could run something 120% of
'max' because of the "margin". Or rather you ran it at 80% of
theoretical capacity. Nowadays, you run stuff at 95% of theoretical
capacity, so you can only over load it by 4%.

This is very true. You can build a product which works satisfactorily
when new with a fairly small factor of safety. But abuse or overload the
product and it'll fail, whereas a product with a larger factor of safety
may not. Also, a product with a larger factor of safety will often wear
more slowly under normal use. I think reduced factors of safety are
quite common because computers allow more accurate modelling. In a way,
it is one of the downsides of more advanced modelling techniques.

There is a story that Henry Ford sent engineers out to the junk
yards to find out what ports were not broken on junked Model Ts. Turns
out that it was the king pins. So he figured they were "overbuilt"
and reduced their heft. Saved on production costs (fifty cents time 2
times N units per year= real money), but people are saying "they don't
make them like that anymore".

I wonder if that story is true. There is also a story that Joe Bamford
(of JCB excavator fame) would send out his engineers whenever he heard
that a JCB excavator was broken down nearby. That way he could see what
was broken and do better next time.

Best wishes,

Chris