Originally posted by Ted Bennett
David Reuteler <[email protected]> wrote:
> anyone else thinking of running biopace rings on their fixed?
Yes, me. I got the idea from Sheldon Brown. It seemed so nutty that I had to try it, and guess what,
I love it.
As others have noted, the chain tension variation is very small but not zero, because of the varying
angle the chain makes with the chainstay. This small variation is swamped by the variation caused by
various eccentricities of sprockets.
Since the two sprockets are of unequal size, there is a periodicity of the chain tightness and it is
much longer than one crank revolution. That's why (and this will be more grist for the mental
processes of Carl Fogel) that the position of the rear axle whould be set only after observation of
a number of crank revolutions.
If your bike can coast, you need to FIX it.
--
Ted Bennett Portland OR
Dear Ted,
You may be answering the question that I
just logged in to ask, but I'm not sure. I
think that your "periodicity" means how
often the tightest position of the front gear
coincides with the tighest position of the
rear gear.
Some browsing showed me that all these
fixed-gear fanatics agree that their chain
tension varies noticeably as they spin
their rear wheels by hand.
It could be something explained in
"Extraordinary Popular Delusions and
the Madness of Crowds," but I tend to
believe Sheldon Brown and Andrew
Muzi when they agree about basic
bicycle maintenance.
But what still bothers me is that everyone
keeps blaiming the imperfectly concentric
gears, bottom bracket, bearings, cones,
cups, hub, and axle. It's cumulative, or so
I'm told, and adds up to enough to affect
the chain tension noticeably.
So how come the back half of this wobbly
mess doesn't affect the wheel rim? You
know, that thing sticking out on the end
of some long spokes attached to the hub,
cones, bearings, and axles? The rim that
they true to the thickness of a couple of
sheets of paper?
Are the gears really that badly mounted
and poorly machined?
I'm looking at a nice new 53-tooth Shimano
chain-ring. It looks pretty well machined me.
Maybe there are deliberate irregularities
machined into it to help shifting, but it
doesn't look too sloppy to me.
I'm hoping that some of the fixed-gear folk
who own bike shops and aren't mechanical
klutzes like me will take a few moments, slip
their chains off, and check the irregularity of
their front and rear gears. Wrap a piece of
soft wire around the frame and pretend to
true a small wheel.
With no chain mounted, how much slop is
there in those gear teeth? Any variation
should be at its worst at the tips of the teeth,
reflecting all accumulated slop of bearings and
mountings and so forth.
If there's enough slop in the observed path
of the gear teeth to account for the tightening
and loosening of the chain, then I'll be convinced
that--
Well, that I've got another reason to like
derailleurs.
But if the observed slop is a lot less than
everyone seems to expect, then I'm going
to ask whether the real culprit is the
chain, which no one mentions as a
suspect, but which has far more moving
parts than the rest of the drive train put
together and which engages individual
teeth in ways that are not well addressed
by "imperfectly round."
I'd do the testing myself, except that
a) I have no fixie and b) none of you
seem dumb enough to believe figures
from someone who measures a few times,
cuts, cuts a little more, and then goes
back to the lumber yard to buy another
board.
Are wobbly gear teeth what really causes
the chain to tighten and slacken? If so,
Ted's "periodicity" between two gears will
be fairly tricky, but it shouldn't coincide with
the length of the chain.
Or is the variation really due to the the chain
and the way that it engages the individual gear
teeth? If so, the "periodicity" might be a fairly
regular pattern corresponding to chain length.
This isn't quite as good as a front-wheel
speed-slapping contest, but I do hope to
hear from fixed-gear riders, boasting about
how much or how little their gears wobble.
Of course, there may be a thread or page
somewhere that already lists the average
out-of-roundness of sample gears, but that
would be almost as nice.
Carl Fogel
