Chain skipping on large chain ring.

Aside to OP: the index on your inner chainring shows it to be oriented
144 degrees from the recommended position. Although you probably won't
notice any performance difference, you might advance the position of
the 'ring two holes to the right. Just because.

Aha! I wondered what that nub was... I'll line it up properly after I
remove the rings for the photograph...

--
Mike (Xyzzy)

That would be handy, actually. Then we can see the wear on all the
teeth much more clearly.

http://www.teamendorphin.com/jpg/53.jpg
http://www.teamendorphin.com/jpg/39.jpg

I found out last night the spider on the crankset is a regular 130mm
BCD deal, so I will most likely just order new chain rings... I really
don't want to destroy my new chain or my cassettes... My guess is
Cannondale may have tried a little too hard to shave weight on these
rings and either didn't use enough material or used a material that is
too soft...

The chain ring bolts are pretty wild too... They thread into the 39,
not into little nuts like most I have seen...

If I continue using the 39 for a few rides do you think I will harm
anything? It will take me a week or so to order and receive new
rings... I don't have a backup bike to ride in the meantime... I'll
most likely put 300 or so miles on the bike before the new stuff comes
in...

Finally, apparently I have a lot of choices for replacements... Are
there any particular ones I should consider? So far I have seen Campy
rings, FSA rings and TA rings... I'll most likely get 53/39 sizes
although I am very tempted to get 52/42, since I could use a tiny bit
more top end on the small ring... Going from a 39 to a 42 would give me
one more gear higher across the board...

BTW, I do know I must match the rings for shifting purposes... IOW, the
53 for the 39 is different than the 53 for the 42...

--
Mike (Xyzzy)

Marvin _Meredith writes:

PS - If it would help with the diagnosis, I can easily remove the
chain rings and photograph them on a piece of white paper... Let me
know if this will help and I will do it...

That would be handy, actually. Then we can see the wear on all the
teeth much more clearly.

Another point I forgot to mention in my mammoth last post: if you do
ride with even a slightly worn chain, the wear is worst at the
points where the chain gets the highest tension. If the cranks are
horizontal, this is usually at the top and bottom of the chainring.
Hence you can have some teeth absolutely perfect and some
significantly worn, which can cause havoc trying to detect wear.

The highest stressed sprocket teeth on a driving sprocket (chainwheel)
are the first ones to engage the chain. Thereafter, they make no load
contact with the teeth and roll idly through the root of the tooth
profile before climbing over the back side of the teeth if in contact
at all. That contact depends on the tension given by the derailleur.
Crank position has little to do with where the chain bears most
heavily, although that affects how much force a rider can put into the
chainwheel.

With a new (or properly worn) chainring and chain this doesn't happen,
because the load is taken evenly on all teeth. Even a small stretch
means the majority of the load is taken by the first five or so teeth,
the rest of the chain/chainring going slack. You can see this by
putting some load on the pedals with a relatively worn chain and then
trying to lift the chain off the ring at various points.

I think you are not assessing the tooth profile of the sprocket shown at:

http://home.comcast.net/~carlfogel/download/zzz_006.jpg

The root of the tooth gaps is relatively unchanged from new and shows
how far the contact pint has shifted up the pressure face of the teeth
by the load pocket near the end of the hook. The tooth nearest the 50
was centered on that number and the attachment bolt hole. Its apex is
now substantially shifted by wear. Although severely worn this
sprocket still accepts a new chain without problem and climbs the
steepest hills as when new. I have others like this.

Just as an aside, the difference in wear between heavily and lightly
loaded chainring teeth is quite a good indicator of riding style - big
difference = stomper, small difference = spinner. I regularly impress
easily impressed customers with this particular sleight of hand :-).

I think you should explain how you believe this comes to pass. My
worn chainring has about 100,000 miles on it, that is, it was in use
for about 10 years and looks much like many others I have seen that
were not changed out by equipment fanatics.

Jobst Brandt

Michael Vang writes:

That would be handy, actually. Then we can see the wear on all the
teeth much more clearly.

http://www.teamendorphin.com/jpg/53.jpg
http://www.teamendorphin.com/jpg/39.jpg

Aha, as I suspected. brand new chainrings with no appreciable wear on
the tooth load faces. The light anodizing wear-off on the sides has
no effect on chain retention. You'll notice that the tooth contours
are identical and smoothly uniform on both chainwheels. Worn
chainwheels display wear pockets where the loaded chain bears. As was
mentioned, these are greatest at the top and bottom of the CW with
respect to cranks being horizontal. There is no such asymmetry
visible here.

Whose manufacture are the CW or are they Cannondale. I find the ramp
style teeth they have, a result of misunderstanding of how these
things work, a typical occurrence for the bicycle industry. You won't
find this in any other chain applications. It is both a waste of
design effort and material.

I found out last night the spider on the crankset is a regular 130mm
BCD deal, so I will most likely just order new chain rings... I
really don't want to destroy my new chain or my cassettes... My
guess is Cannondale may have tried a little too hard to shave weight
on these rings and either didn't use enough material or used a
material that is too soft...

I doubt it. There are many thousand of these in service. There is
another cause for your skipping chain and it is not material hardness
or sprocket shape. I don't see what the cause is but it is not the
large CW which has the same tooth profile as the smaller one that
apparently works fine.

The chain ring bolts are pretty wild too... They thread into the 39,
not into little nuts like most I have seen...

If I continue using the 39 for a few rides do you think I will harm
anything? It will take me a week or so to order and receive new
rings... I don't have a backup bike to ride in the meantime... I'll
most likely put 300 or so miles on the bike before the new stuff
comes in...

What could it possibly do? Chains are expendable and easily replaced.
I cannot imagine how an aluminum CW could damage a chain.

Finally, apparently I have a lot of choices for replacements... Are
there any particular ones I should consider? So far I have seen
Campy rings, FSA rings and TA rings... I'll most likely get 53/39
sizes although I am very tempted to get 52/42, since I could use a
tiny bit more top end on the small ring... Going from a 39 to a 42
would give me one more gear higher across the board...

BTW, I do know I must match the rings for shifting purposes... IOW,
the 53 for the 39 is different than the 53 for the 42...

You don't have to synchronize the sprockets for shifting purposes.
These things have been shifted for more than 50 years without any
reference to one another and do so as fast as the shift lever can be
moved.

Jobst Brandt

wrote:
Marvin _Meredith writes:

PS - If it would help with the diagnosis, I can easily remove the
chain rings and photograph them on a piece of white paper... Let me
know if this will help and I will do it...

That would be handy, actually. Then we can see the wear on all the
teeth much more clearly.

Another point I forgot to mention in my mammoth last post: if you do
ride with even a slightly worn chain, the wear is worst at the
points where the chain gets the highest tension. If the cranks are
horizontal, this is usually at the top and bottom of the chainring.
Hence you can have some teeth absolutely perfect and some
significantly worn, which can cause havoc trying to detect wear.

The highest stressed sprocket teeth on a driving sprocket (chainwheel)
are the first ones to engage the chain. Thereafter, they make no load
contact with the teeth and roll idly through the root of the tooth
profile before climbing over the back side of the teeth if in contact
at all. That contact depends on the tension given by the derailleur.
Crank position has little to do with where the chain bears most
heavily, although that affects how much force a rider can put into the
chainwheel.

Okay, that's what I meant, the force on the crank changes which changes
the amount of wear. So you can (and indeed usually do) have a
difference in wear between teeth at different points on the ring.

Your ring doesn't particularly show this, which, again, I suspect is
down to a brand-new chain spreading the load over more teeth.

With a new (or properly worn) chainring and chain this doesn't happen,
because the load is taken evenly on all teeth. Even a small stretch
means the majority of the load is taken by the first five or so teeth,
the rest of the chain/chainring going slack. You can see this by
putting some load on the pedals with a relatively worn chain and then
trying to lift the chain off the ring at various points.

I think you are not assessing the tooth profile of the sprocket shown at:

http://home.comcast.net/~carlfogel/download/zzz_006.jpg

The root of the tooth gaps is relatively unchanged from new and shows
how far the contact pint has shifted up the pressure face of the teeth
by the load pocket near the end of the hook. The tooth nearest the 50
was centered on that number and the attachment bolt hole. Its apex is
now substantially shifted by wear. Although severely worn this
sprocket still accepts a new chain without problem and climbs the
steepest hills as when new. I have others like this.

As I've said, your chainrings are atypical - I see *no* rings like that
at the LBS where I work and many, many with sharper ramps like the one
of Carl Fogel's I keep dragging up as a convenient example. I believe
this is related to the amount of care taken replacing worn chains.

This might be a long example, so bear with me:

Assume we have a new chain and chainwheel. As the chain rolls on to
the chainwheel, each roller will come into contact at a certain point.
If the chain is the correct pitch, it will contact at the same point as
the previous roller, which should be the base of the tooth profile, and
all is well.

Now the chain wears faster than the chainwheel, so in due time we have
a worn chain and new chainwheel. Now when each roller engages it is
slightly further from the previous roller than it should be, so it
starts taking load and wearing some way up the tooth profile.

Assume we don't replace either, and continue riding. The worn chain on
the worn chainring is now running steadily higher on the tooth profile,
and as it wears there will be less and less teeth taking any load so
the wear gets faster. The profile of the tooth is gaining a relatively
straight ramp until it wears right to the top of the tooth, sharpening
the top of the tooth quite nicely and slipping under any load at all.

If, on the other hand, the chainring never gets used with an
appreciably worn chain, the wear stays low on the tooth profile. It
still moves up slightly as shown by your extreme example, but by no
means as fast as with a worn chain.

So, in conclusion, the wear moves up the tooth faster with a worn
chain, and the wear is faster with a worn chain.

Just as an aside, the difference in wear between heavily and lightly
loaded chainring teeth is quite a good indicator of riding style - big
difference = stomper, small difference = spinner. I regularly impress
easily impressed customers with this particular sleight of hand :-).

I think you should explain how you believe this comes to pass.

It's very simple, more load at a given point on the chainwheel equals
more wear. Worn chains don't spread the load anywhere near as evenly
between teeth and therefore wear chainrings faster. I don't see why
this is contentious.

My
worn chainring has about 100,000 miles on it, that is, it was in use
for about 10 years and looks much like many others I have seen that
were not changed out by equipment fanatics.

I'll say it again, if someone is careful about changing their chain
they can indeed expect their chainrings to last for ludicrous numbers
of miles. If they aren't, well, 10,000 miles is a closer par for the
course.

Whose manufacture are the CW or are they Cannondale. I find the ramp
style teeth they have, a result of misunderstanding of how these
things work, a typical occurrence for the bicycle industry. You won't
find this in any other chain applications. It is both a waste of
design effort and material.

They are made by Cannondale... BTW, what do you mean by "ramp style
teeth?"...

I am glad to hear they are not worn out... I will save my money then,
and ride the crap out of these...

I'm still not sure why I had this problem to begin with, but I'm happy
with the outcome and I learned a bit in the process...

Thanks to all for the help!

--
Mike (Xyzzy)

Marvin Meredith writes:

PS - If it would help with the diagnosis, I can easily remove the
chain rings and photograph them on a piece of white paper... Let
me know if this will help and I will do it...

That would be handy, actually. Then we can see the wear on all
the teeth much more clearly.

Another point I forgot to mention in my mammoth last post: if you
do ride with even a slightly worn chain, the wear is worst at the
points where the chain gets the highest tension. If the cranks
are horizontal, this is usually at the top and bottom of the
chainring. Hence you can have some teeth absolutely perfect and
some significantly worn, which can cause havoc trying to detect
wear.

The highest stressed sprocket teeth on a driving sprocket
(chainwheel) are the first ones to engage the chain. Thereafter,
they make no load contact with the teeth and roll idly through the
root of the tooth profile before climbing over the back side of the
teeth if in contact at all. That contact depends on the tension
given by the derailleur. Crank position has little to do with
where the chain bears most heavily, although that affects how much
force a rider can put into the chainwheel.

Okay, that's what I meant, the force on the crank changes which
changes the amount of wear. So you can (and indeed usually do) have
a difference in wear between teeth at different points on the ring.
Your ring doesn't particularly show this, which, again, I suspect is
down to a brand-new chain spreading the load over more teeth.

It doesn't because I rotate it to spread the wear. Unfortunately it
is a 5-pin system instead of something divisible by four, so I can
only approximate. If we saw the who;e CW, I think you could see some
teeth are less worn than the average.

With a new (or properly worn) chainring and chain this doesn't
happen, because the load is taken evenly on all teeth. Even a
small stretch means the majority of the load is taken by the first
five or so teeth, the rest of the chain/chainring going slack.
You can see this by putting some load on the pedals with a
relatively worn chain and then trying to lift the chain off the
ring at various points.

I think you are not assessing the tooth profile of the sprocket
shown at:

http://home.comcast.net/~carlfogel/download/zzz_006.jpg

The root of the tooth gaps is relatively unchanged from new and
shows how far the contact pint has shifted up the pressure face of
the teeth by the load pocket near the end of the hook. The tooth
nearest the 50 was centered on that number and the attachment bolt
hole. Its apex is now substantially shifted by wear. Although
severely worn this sprocket still accepts a new chain without
problem and climbs the steepest hills as when new. I have others
like this.

As I've said, your chainrings are atypical - I see *no* rings like
that at the LBS where I work and many, many with sharper ramps like
the one of Carl Fogel's I keep dragging up as a convenient example.
I believe this is related to the amount of care taken replacing worn
chains.

Maybe in your circle of riders, but for mine it is typical. We ride
bike to ride and don't mind riding on less than new components.
Carl's CW has no teeth. Those are only stubs of teeth that could not
drive a new chain. I also have stacks of rims with merely 0.5mm wall
on brake surfaces, also atypical in general, people always being on
the latest fad wheels.

This might be a long example, so bear with me:

Assume we have a new chain and chainwheel. As the chain rolls on to
the chainwheel, each roller will come into contact at a certain
point. If the chain is the correct pitch, it will contact at the
same point as the previous roller, which should be the base of the
tooth profile, and all is well.

Now the chain wears faster than the chainwheel, so in due time we
have a worn chain and new chainwheel. Now when each roller engages
it is slightly further from the previous roller than it should be,
so it starts taking load and wearing some way up the tooth profile.

That assumption of wear rate is immaterial because the pitch of the CW
remains constant while its pressure pocket in its teeth increases in
height above the base circle as chain pitch increases from wear.

Assume we don't replace either, and continue riding. The worn chain on
the worn chainring is now running steadily higher on the tooth profile,
and as it wears there will be less and less teeth taking any load so
the wear gets faster. The profile of the tooth is gaining a relatively
straight ramp until it wears right to the top of the tooth, sharpening
the top of the tooth quite nicely and slipping under any load at all.

From elastic considerations only the first few teeth of the CW bear
any significant load. You can test this by lifting the chain off the
CW starting at its exit end.

If, on the other hand, the chainring never gets used with an
appreciably worn chain, the wear stays low on the tooth profile. It
still moves up slightly as shown by your extreme example, but by no
means as fast as with a worn chain.

So, in conclusion, the wear moves up the tooth faster with a worn
chain, and the wear is faster with a worn chain.

Yes??? and now what?

Just as an aside, the difference in wear between heavily and
lightly loaded chainring teeth is quite a good indicator of riding
style - big difference = stomper, small difference = spinner. I
regularly impress easily impressed customers with this particular
sleight of hand :-).

I think you should explain how you believe this comes to pass.

It's very simple, more load at a given point on the chainwheel equals
more wear. Worn chains don't spread the load anywhere near as evenly
between teeth and therefore wear chainrings faster. I don't see why
this is contentious.

It isn't, but you seem to suggest one should ride the lowest gear
possible for the benefit of bicycle components, modifying one's riding
to something other than what might come naturally? I believe the
bicycle is the riders servant, not the converse. Besides, this does
not affect the problem at hand.

My worn chainring has about 100,000 miles on it, that is, it was in
use for about 10 years and looks much like many others I have seen
that were not changed out by equipment fanatics.

I'll say it again, if someone is careful about changing their chain
they can indeed expect their chainrings to last for ludicrous numbers
of miles. If they aren't, well, 10,000 miles is a closer par for the
course.

So what is wrong with riding a CW for 10x as long by replacing the chain
at 0.5-1.0% chain wear?

Let's get back to the failure analysis at hand. I find it mysterious.
In fact every time I read of chains jumping over CW's in this forum, I
wonder how it is done, not having had this occur on my bicycle or
those of my riding companions since I got my first derailleur bicycle
in the 1950's.

Jobst Brandt

Michael Vang writes:

Whose manufacture are the CW or are they Cannondale. I find the
ramp style teeth they have, a result of misunderstanding of how
these things work, a typical occurrence for the bicycle industry.
You won't find this in any other chain applications. It is both a
waste of design effort and material.

They are made by Cannondale... BTW, what do you mean by "ramp style
teeth?"...

The teeth are not symmetrically shaped having a back side that has a
lower slope than the functioning front side. It gives the tooth
profile that ocean wave look. You notice that the teeth are
reminiscent of a circular saw blade. Neither Shimano, Campagnolo, nor
other mainstream component manufacturers have such tooth profiles
although some used maimed teeth at some ideal derailing spots.

http://www.wallacemachine.com/w4_whatwedo.htm

I am glad to hear they are not worn out... I will save my money then,
and ride the crap out of these...

Well, we're still not to the bottom of the phenomenon.

Jobst Brandt

wrote:

Well, we're still not to the bottom of the phenomenon.


Maybe we are getting close...

The OP wrote in his original post "The big ring, however, has a bit of
hooking action going on when pedalling forwards, and if I pedal
backwards the chain rides up the teeth every once in a while and then
plops back down where it is supposed to be."

So, except for the "bit of hooking action", which I'd like to know more
about, the problem is when pedaling backwards. Since the teeth on the
large ring are cut in such an extreme way, isn't it possible that it
simply doesn't work well going backwards?

The chain is good, and the chainring isn't worn. Either the chainring
is defective, or it wasn't designed to go backward in the first place
(or both). I'm more interested in how it is performing in the forward
direction: What does "hooking action" mean? Is noisier in the large
ring vs. the small, and how would you describe the sound? When you
pedal backward, does it ride up in all the gears, or only when
cross-chained?

-Ron

The chain is good, and the chainring isn't worn. Either the chainring
is defective, or it wasn't designed to go backward in the first place
(or both). I'm more interested in how it is performing in the forward
direction: What does "hooking action" mean? Is noisier in the large
ring vs. the small, and how would you describe the sound? When you
pedal backward, does it ride up in all the gears, or only when
cross-chained?

When I pedalled it forward, without load (The bike was upside down) the
chain, instead of travelling straight back to the lower pulley, arced
up a bit right behind the large chain ring... The arc was maybe 3/4
inch...

Now that I have rode it a little the hooking action is mostly gone It
arcs up maybe 1/16 inch... There is still a very slight riding up when
pedalling backwards...

The old chain never rode up when pedalling backwards...

I emailed Park Tool about my CC-2 being possibly miscalibrated and they
offered to take a look at it, so I mailed it to them today... I still
think the fact the CC-2 will not fit into the chain is a bit weird...

--
Mike (Xyzzy)