[TowerTalk] Guying a tower....Heresy to follow..... True statement!

[jimlux]

Doug Renwick wrote:
> Marty, I welcome your true life experience.  So here is one of the many
> problems with this reflector ... you get beat up posting your real life
> experience which trumps any theory any day. 

Nope... real life experience is an anecdote, not engineering data. It 
predicts only the past for that one case.  There's no way that anecdote 
trumps analysis, particularly if the analysis is based on test data in 
controlled circumstances.

That's the big difference between craft and engineering. With craft it's 
"do it like we've done it before, and it works ok, and we don't care 
why".  With engineering, you can predict what happens before it happens, 
so you can go beyond just duplicating something before, and more to the 
point, you don't have to worry about whether what worked before happened 
to be just about to fail.

Back in pre-engineering days (or even today, if you can't calculate with 
sufficient certainty), you can do "proof testing" (that is, test it with 
a load that is greater than you expect to see in use).  This is what we 
do with spacecraft (which are so complex that you can't calculate all 
the failure probabilities..): we test at some multiple of the maximum 
expected load or temperature or whathaveyou.


--- that said, a lot depends on the consequences of failure.  Craft may 
work just fine, if the consequences are small or unlikely or immaterial.