I totally agree that certain derailers have less-than-durable springs, and you are absolutely right that the few bad ones were designed to operate overly close to the elastic limits of the spring steel that they used in those derailer's springs (I'm thinking only of certain Suntour derailers here).
I suspect also that the designers assumed that the vast majority of installations would not have the derailer fully reaching it's travel limits, but that in many cases [such as after the addition of wider freewheels, or mounting on a claw hanger (which moves the derailer outward)] the travel limits were approached perhaps too closely for those assumptions.
And yes, this would occur repeatedly in use, and even if not rapidly (like an engine's highly-stressed valve springs) fatigue sets in much faster near to the elastic limit stress level.
I still believe that it is the number of flex cycles predominately, rather than any extended static stress, which fatigue-stresses any grade of steel, so I assume that it would make no difference (to the spring) if I relieved the stress upon parking the bike or upon my first upshift the next morning. And I admit that the data I've looked at may not be able to depict (with accurate resolution) just what statistical effect that the resting-but-stressed-at-the-elastic limit condition has on the failure rate. So I'm curious now, but still not thinking that the brake calipers need to be de-tensioned each nite.
As for most derailers, the springs seem to usually out-last many of the other parts, so aside from avoiding any un-needed full-travel movement (with the limit screw backed way out) during setup or cleaning, it shouldn't matter.
I did see a customer in the shop the other day carelessly (and needlessly) fully compress a new Campagnolo Record caliper, as if for exercise, and he broke the freaking spring!
As for the OP's centurion, 1977 would be my guess, based on the component dates of the similar one I bought. I will have to look at the serial # and report back.
These bikes have very relaxed angles, btw, so are best selected from the biggest size that the rider can straddle. The effective top tube lengths are quite short and the very slack seat tube angle makes it feel even more so.
In the smallest frame sizes this is typically less true however, since the designer always tries to minimize interference between the rider's toe and the front tire.
Originally Posted by
3alarmer
Without seeming excessively argumentative, I hope you will agree that there are a wide
variety of both designs and relative qualities for rear derailleurs........the majority, in my
experience, depend on a wound coil torsion spring for cage tension. And while your example
of static strain in automotive springs may be valid, these are different springs, made to
lesser standards, by and large, and with considerably less engineering safety margin
prior to failure.
So I personally think that it is probable in the lesser quality rear derailleurs that
the fully extended position approaches and exceeds the elastic limits of these particular
springs. It is a mistake to generalize in most engineering and design applications, and
my own opinion is that this is one of them. I've seen no research on this, and doubt
any exists. I have seen a number of rear de's that have lost considerable spring tension.
My assumption is that this is a result of repeated extension past the elastic limits for
the particular steel and configuration of the spring. Would you agree?
If so, then would not logically this occur in the fully extended position whenever it
is achieved? Or in other words, why do you think rear derailleur springs weaken
and give up ?