On Thu, 21 Feb 2013 04:58:52 -0800 (PST), thirty-six
wrote:
According to conventional wisdom, a chain should be replaced when it
stretches 1/16" out of 24 links or 1ft. *That:
* 0.0625" / 12" * 100% = 0.52%
This author recommends 0.52%
http://www.kronowit.com/bicycling/chainstretch.html
which agrees with
http://www.sheldonbrown.com/chains.html
"If the rivet is 1/16" past the mark, you should replace
the chain, but the sprockets are probably undamaged."
Again, it's uncertain if the measured deformation is permanent
(plastic) or temporary (elastic).
I'm beginning to wonder if the test is really a measurement of how
much force it takes to squeeze the lubricant out of the bushings.
31 links should be:
31 * 0.500in/link * 25.4mm/in = 393.7mm
If the chains were packed with grease or wax, the initial 10Kg load
should be less than 393.7mm. However, they're all greater, which
eliminates that theory. Oh well.
2% stretch is recommended as the limit by Wipperman (but perhaps not
strictly apply with silly small teeth on modern multi-sprocket
bikes). For 99% of the time all the chains will be below that level
of elongation taking both wear and active load into account.
It's Wippermann with 2 n's at the end. Having a similar last name
makes me rather sensitive to such creative spelling.
2% seems rather high, unless the sprocket teeth are correspondingly
worn.
Wippermann tool, which seems to be a go/no-go test, without any
measurements:
http://www.cantitoeroad.com/chain-wear-indicator-stainless-steel-wippermann
When they sponsored a chain wear test, a run-in test was performed
before making measurements. However, there's no indication that they
did that in this static test.
http://www.youtube.com/watch?v=RGjcD8xEu8o
The chains were considered to have failed at 1% elongation.
I'll not be loading the chain of my "racing" bike with a consistent
load of 150kg anytime soon.
It's more realistic to measure the load at which one can sustain (for
at least 10 minutes) with purely nasal breathing.
I would prefer to see a range of loads extending to well beyond what a
human could possibly produce so I can see where the curve is going. I
also would like to see deformation IN BOTH DIRECTIONS forming
something like the usual stress/strain hysteresis curve. That is used
to show both plastic and elastic deformation. Three data points is
not sufficient to generate a meaningful trend line or hysteresis
curve. In addition, it would be nice to be sure that they were
measuring chain stretch and not lubricant compression, which would
require a solvent bath to remove the lube before testing.
--
Jeff Liebermann
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060
http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558