FYI: Broken Spoke
Luke
Two hours into my first day of a four day mini-tour, TWANG! I was
surprised: four years had passed since my last broken spoke -- and that
on a wheel I'd laced with an ornamental intertwined pattern.
The casualty was the right (drive) side spoke on a 36 spoke/3x rear
wheel built (as per the Bicycle Wheel) with 2mm straight gauge DT
stainless spokes and a Mavic Open Pro rim. Total mileage: approx.
15,000 KM. How odd: the spoke broke at the nipple rather than at the
bend. Inspection revealed that the nipple had bottomed out on the spoke
thread [1].
A week earlier an accident involving a kid and a dog upended the bike
and launched me over the bars. When the dust settled the worst damage
was to that same rear wheel: it had smashed against the curb in the
melee and was about 1cm out of true. A little twisting of the spoke
wrench and I was good to go. But the spoke that required the most
tightening was the one that ended up breaking a week following -- easy
to note because it was the right side spoke next to the valve stem.
Now even though the nipple had bottomed out, I wonder how great a role
this played in the spoke failure -- as is evident in the photo, the
nipple flats are in great shape and the post-accident truing session
did not involve any Herculean application of the spoke wrench. I'm
uncertain whether the primary cause of failure was an overtightened
spoke, a deficient spoke, or a combination of factors.
Luke
1. Click on 'Large' to view the photo at full size.
http://www.flickr.com/photo_zoom.gne?id=177134133&size=o
carl...@comcast.net
wrote:
Dear Luke,
In the Stanford spoke fatigue test for Wheelsmith about 20 years ago,
10% of spokes broke at the nipple, 90% at the elbow (8/76):
http://www.duke.edu/~hpgavin/papers/HPGavin-Wheel-Paper.pdf
Cheers,
Carl Fogel
Ron Ruff
Luke wrote:
> it had smashed against the curb in the
> melee and was about 1cm out of true. A little twisting of the spoke
> wrench and I was good to go. But the spoke that required the most
> tightening was the one that ended up breaking a week following -- easy
> to note because it was the right side spoke next to the valve stem.
I'll make a guess...
The rim was bent quite a lot from the crash... enough to take a 1cm
permanent set... and the stress on this particular spoke would have
been the greatest. It probably cracked near the threads at that time,
and quickly came apart when you rode it.
If a rim gets "knocked out of true" I consider that something must have
yielded... the rim and/or spokes, or maybe a nipple or hub flange. Just
tightening a few spokes doesn't fix things even if you can true the
wheel... you need to straighten the rim (or replace it) or replace the
spoke if that is the cause.
jobst....@stanfordalumni.org
> Two hours into my first day of a four day mini-tour, TWANG! I was
> surprised: four years had passed since my last broken spoke -- and
> that on a wheel I'd laced with an ornamental intertwined pattern.
> The casualty was the right (drive) side spoke on a 36 spoke/3x rear
> wheel built (as per the Bicycle Wheel) with 2mm straight gauge DT
> stainless spokes and a Mavic Open Pro rim. Total mileage: approx.
> 15,000 KM. How odd: the spoke broke at the nipple rather than at the
> bend. Inspection revealed that the nipple had bottomed out on the
> spoke thread [1].
The spoke shows no bend in it, so the incident with the collision had
no effect on it. Depth of engagement also had no effect. Spokes have
been advanced to a point that several brass threads in the nipple were
pressed flat and caused no problem.
What should be more apparent is that a 2mm spoke has its weakest cross
section at the thread and when fully engaged in the nipple, is the
place where cyclic stress from loaded wheel rotation is highest. A
spoke threaded from end-to-end would not suffer that fate, its stress
(and stretching) being uniform over its length.
This is a good reason to use swaged spokes and to use thin ones to
have a greater load distribution among spokes. I assume you followed
the suggestions in the book to stress relieve after tensioning?
> 1. Click on 'Large' to view the photo at full size.
http://www.flickr.com/photo_zoom.gne?id=177134133&size=o
Excellent picture!
15,000 km = 7,156,489 stress cycles. As I have mentioned, I ride a
rear wheel with spokes that have more than 230 million cycles and
still working. I don't expect them to fail now. The only ones that
failed (two) had a stick in the wheel that bent them at the nipple and
a couple where the chain dropped on them. That was about four or five
of 36 spokes. With a 6-speed cluster, todays derailleurs cannot be
adjusted to prevent over shifting into the spokes. It takes care.
Jobst Brandt
jim beam
well, based on the fact that the failure was not initiated at any "new"
threading that may have been cut by bottoming, bottoming had nothing to
do with it.
the truth is, it's probably just a statistical casualty. despite what
it may suit others to tell you, stainless tell doesn't have a fatigue
endurance limit, so fatigue *cannot* be eliminated. and paradoxically,
excess "stress relieving" of spokes can initiate or activate fatigue in
notches such as those at thread roots.
bottom line, just replace and continue riding.
jim beam
> Luke Iragusa writes:
>
>
>>Two hours into my first day of a four day mini-tour, TWANG! I was
>>surprised: four years had passed since my last broken spoke -- and
>>that on a wheel I'd laced with an ornamental intertwined pattern.
>
>
>>The casualty was the right (drive) side spoke on a 36 spoke/3x rear
>>wheel built (as per the Bicycle Wheel) with 2mm straight gauge DT
>>stainless spokes and a Mavic Open Pro rim. Total mileage: approx.
>>15,000 KM. How odd: the spoke broke at the nipple rather than at the
>>bend. Inspection revealed that the nipple had bottomed out on the
>>spoke thread [1].
>
>
> The spoke shows no bend in it, so the incident with the collision had
> no effect on it. Depth of engagement also had no effect. Spokes have
> been advanced to a point that several brass threads in the nipple were
> pressed flat and caused no problem.
>
> What should be more apparent is that a 2mm spoke has its weakest cross
> section at the thread and when fully engaged in the nipple, is the
> place where cyclic stress from loaded wheel rotation is highest. A
> spoke threaded from end-to-end would not suffer that fate, its stress
> (and stretching) being uniform over its length.
rubbish! in that situation, it's /definitely/ going to fail at the
thread! what a bizarre misunderstanding.
>
> This is a good reason to use swaged spokes and to use thin ones to
> have a greater load distribution among spokes. I assume you followed
> the suggestions in the book to stress relieve after tensioning?
>
>
>>1. Click on 'Large' to view the photo at full size.
>
>
> http://www.flickr.com/photo_zoom.gne?id=177134133&size=o
>
> Excellent picture!
>
> 15,000 km = 7,156,489 stress cycles. As I have mentioned, I ride a
> rear wheel with spokes that have more than 230 million cycles and
> still working. I don't expect them to fail now. The only ones that
> failed (two) had a stick in the wheel that bent them at the nipple and
> a couple where the chain dropped on them. That was about four or five
> of 36 spokes.
yawn.
> With a 6-speed cluster, todays derailleurs cannot be
> adjusted to prevent over shifting into the spokes.
absolutely untrue. are you /really/ that ineducatably stupid or are you
having some kind of sick joke?
> It takes care.
yeah, just like checking simple facts.
carl...@comcast.net
[snip]
>With a 6-speed cluster, todays derailleurs cannot be
>adjusted to prevent over shifting into the spokes. It takes care.
>
>Jobst Brandt
Dear Jobst,
I'm puzzled.
I have no problems with overshifting my 1998 Schwinn LeTour 7-speed
Shimano, but I can't be accused of careful shifting.
I went out to the garage just now and deftly turned the wrong limit
screw, transforming my 7-speed into a 5-speed cluster that wouldn't
shift outward to the smallest two rear sprockets.
After re-adjusting my mistake, I turned the other limit screw and had
no trouble eliminating my largest rear sprocket, turning it into a
6-speed, unable to pull the chain in far enough to grab 1st gear.
Is your problem specific to Campagnolo, or have all derailleurs
changed in the last few years?
Cheers,
Carl Fogel
Tim McNamara
carl...@comcast.net wrote:
> On 29 Jun 2006 01:34:28 GMT, jobst....@stanfordalumni.org wrote:
>
> [snip]
>
> >With a 6-speed cluster, todays derailleurs cannot be adjusted to
> >prevent over shifting into the spokes. It takes care.
> >
> I'm puzzled.
>
> I went out to the garage just now and deftly turned the wrong limit
> screw, transforming my 7-speed into a 5-speed cluster that wouldn't
> shift outward to the smallest two rear sprockets.
>
> After re-adjusting my mistake, I turned the other limit screw and had
> no trouble eliminating my largest rear sprocket, turning it into a
> 6-speed, unable to pull the chain in far enough to grab 1st gear.
>
> Is your problem specific to Campagnolo, or have all derailleurs
> changed in the last few years?
Not having a six speed wheel, I can't test this. I am wondering,
though, if the range of adjustment of the limit screws on a current 9 or
10 speed derailleur is sufficient to limit spokeward motion of the
derailleur to prevent overshifting into the spokes. The inward swing
for 9/10 speeds is pretty far and the geometry of the derailleurs is
different than it was in the ancient times of 15 years ago.
Maybe Carl or jim could check this out for us, if they've got access to
a six speed bike onto which they can put a 9/10 speed derailleur. If
they haven't tried this, they have no way of verifying or refuting what
Jobst wrote.
Although I can't imagine why one would put one of these derailleurs on a
six speed bike. There are plenty of old, suitable derailleurs available
for free or dirt cheap.
--
Tim "early 80s SunTour Superbes work fine for 8 speed" McNamara
jim beam
tim, if a derailleur has limit adjusting screws, and i've never seen one
of any manufacture or vintage that doesn't, then it can be adjusted to
prevent chain fouling.
the much more interesting question, posed to yourself as someone with
more than a passing interest in human psychology, is why do some people,
believe everything told them with a certain delivery, even when simple
independent fact checking can easily and quickly be done proving that
delivery's content to be, er, "mistaken"? it absolutely fascinates me!
why is this? what's the mechanism?
i observed this in another forum recently. i attended an industry
"challenge" where two competing teams of undergrads [cal & stanford as
it happens] were posed a debate question. they had to respond with
analysis and recommendations for a solution. and interestingly for me,
the particular subject matter was one on which i am one of 5, maybe 6
subject experts, so i was in the privileged position of knowing every
detail they were trying to address. long story short, the cal team did
their homework, got all the relevant and correct information and
proposed a sensible solution in accordance with the facts. the stanford
team otoh, had serious factual problems, made wild assumptions that had
no relevance to the data they actually had bothered to collect, but
blustered on anyway. result: stanford won the popular vote. cal's
delivery was that of letting the facts speak for themselves. stanford's
was that of "assertion will prevail". and it did. so, i ask again, why
is this? what is the mechanism by which trivially exposed bullshit
seems to be so easily swallowed? and more importantly, what is the
compulsion for sociopaths to exploit it?
jobst....@stanfordalumni.org
>> With a 6-speed cluster, todays derailleurs cannot be adjusted to
>> prevent over shifting into the spokes. It takes care.
> I'm puzzled.
> I have no problems with over shifting my 1998 Schwinn LeTour 7-speed
> Shimano, but I can't be accused of careful shifting.
> I went out to the garage just now and deftly turned the wrong limit
> screw, transforming my 7-speed into a 5-speed cluster that wouldn't
> shift outward to the smallest two rear sprockets.
Do you have a 9-speed derailleur. My SunTour derailleur's limit stop
screw misses the tab it is designed to engage by a large angle
(3-speeds worth) and serves no purpose. You have a different model
that apparently won't handle 9- or 10- speeds. I ran out of those
over time as derailleur sticks ripped them apart on forest trails.
> After re-adjusting my mistake, I turned the other limit screw and
> had no trouble eliminating my largest rear sprocket, turning it into
> a 6-speed, unable to pull the chain in far enough to grab 1st gear.
Well that's the limit that doesn't work for me and hasn't for a couple
of derailleurs now.
> Is your problem specific to Campagnolo, or have all derailleurs
> changed in the last few years?
My indexed lever is attuned to Sun Tour and works well with it except
for the inner stop.
Jobst Brandt
dvt
> Carl Fogel writes:
>>> With a 6-speed cluster, todays derailleurs cannot be adjusted to
>>> prevent over shifting into the spokes. It takes care.
>> I'm puzzled.
> Do you have a 9-speed derailleur. My SunTour derailleur's limit stop
> screw misses the tab it is designed to engage by a large angle
> (3-speeds worth) and serves no purpose.
How about replacing the limit screw with a longer one?
--
Dave
dvt at psu dot edu
Everyone confesses that exertion which brings out all the powers of body
and mind is the best thing for us; but most people do all they can to
get rid of it, and as a general rule nobody does much more than
circumstances drive them to do. -Harriet Beecher Stowe, abolitionist and
novelist (1811-1896)
jobst....@stanfordalumni.org
>>>> With a 6-speed cluster, todays derailleurs cannot be adjusted to
>>>> prevent over shifting into the spokes. It takes care.
>>> I'm puzzled.
>> Do you have a 9-speed derailleur. My SunTour derailleur's limit
>> stop screw misses the tab it is designed to engage by a large angle
>> (3-speeds worth) and serves no purpose.
> How about replacing the limit screw with a longer one?
I see you don't have a clear picture of the problem. The parts of the
derailleur parallelogram move in an arc. When the sweep of those
elements is 3- or 4-speeds from the designed limit, the screw points
into empty space rather than at the limit stop it is designed to
engage. A longer screw does not change that. Besides, the screw is
long enough but is not aimed at the place where it could do its job.
This is not a situation where I have free choice of derailleurs
because it is an indexing system and levers that match derailleur
positions need to be matching. I switched to SunTour back in the days
before Campagnolo got a good slant parallelogram design.
Jobst Brandt
carl...@comcast.net
Dear Jobst,
No, my derailleur is probably designed for the 7 speed that it came
with, not a 9-speed, so that may explain things.
For $10 to $15, these 6-7 speed derailleurs (with and without hangers)
might be worth looking at. Modern cheap parts can sometimes be as good
as the best that was available years ago.
Sun Race:
Tourney:
Andrew Muzi may even have the real SunTour derailleurs that you need:
http://www.yellowjersey.org/stcass.html
Scroll to the bottom if the SunTour X1 from 1992 is what you want.
Cheers,
Carl Fogel
Luke
<jobst....@stanfordalumni.org> wrote:
>
> The spoke shows no bend in it, so the incident with the collision had
> no effect on it. Depth of engagement also had no effect. Spokes have
> been advanced to a point that several brass threads in the nipple were
> pressed flat and caused no problem.
>
> What should be more apparent is that a 2mm spoke has its weakest cross
> section at the thread and when fully engaged in the nipple, is the
> place where cyclic stress from loaded wheel rotation is highest. A
> spoke threaded from end-to-end would not suffer that fate, its stress
> (and stretching) being uniform over its length.
>
> This is a good reason to use swaged spokes and to use thin ones to
> have a greater load distribution among spokes. I assume you followed
> the suggestions in the book to stress relieve after tensioning?
Yes the wheel was stress relieved at the time of building. But isn't it
too much of a coincidence that it was the spoke located at the point of
impact (from the prior accident) that broke?
I'm inclined to believe Ron Ruff's explanation. As you've noted the
weakest cross-section of the spoke in question is at the thread, in
this case cheek by jowl to where the wheel was smashed; perhaps the
spoke developed a crack during the initial mishap and -- thinking out
loud -- couldn't the post-accident truing session further weaken it by
applying a torsional (spoke windup) force?
Thanks
Luke
carl...@comcast.net
And there are lots of NOS and lightly used SunTour rear derailleurs on
eBay:
CF
jobst....@stanfordalumni.org
If the wobble in the wheel was cured by increasing tension in that one
spoke, it is possible that it had already developed a crack and that
the increased tension was enough to make it fail. However, by that
logic, there should be other spokes in that same condition... ready to
fail. Without a bend in the spoke I don't see any reason for it to
fail from the incident.
By the way, brass does not cut threads into stainless steel spokes,
but rather the brass threads become flattened. So that is not a
source for failure.
So did you stress relieve after re-truing the wheel after the
incident? That should have either exposed a nearly broken spoke or
prevented it from getting to that condition as soon as it did.
Jobst Brandt
Luke
<jobst....@stanfordalumni.org> wrote:
> So did you stress relieve after re-truing the wheel after the
> incident? That should have either exposed a nearly broken spoke or
> prevented it from getting to that condition as soon as it did.
>
I see what you're getting at. No, that I DIDN'T do. Sloppy on my part.
Luke
dvt
> someone writes:
>
>>>>> With a 6-speed cluster, todays derailleurs cannot be adjusted to
>>>>> prevent over shifting into the spokes. It takes care.
>>>> I'm puzzled.
>>> Do you have a 9-speed derailleur. My SunTour derailleur's limit
>>> stop screw misses the tab it is designed to engage by a large angle
>>> (3-speeds worth) and serves no purpose.
>> How about replacing the limit screw with a longer one?
> I see you don't have a clear picture of the problem.
You're right. I can't see your derailer. How did you know?
> The parts of the
> derailleur parallelogram move in an arc. When the sweep of those
> elements is 3- or 4-speeds from the designed limit, the screw points
> into empty space rather than at the limit stop it is designed to
> engage. A longer screw does not change that. Besides, the screw is
> long enough but is not aimed at the place where it could do its job.
You may need to get a bit more creative than simply a longer screw...
that suggestion was just a launch pad. A double nut on the end of the
longer screw would expand its lateral reach. Or perhaps you could put
the screw in backwards so the screw head serves as the stop. Or maybe
you need to add a little material to your derailer so it can reach the
screw. You could put a spacer between the derailer and the frame to move
the assembly to the right. I'm sure there are other simple things you
can do to make this work.
Of course, it would probably be easier to get a 6 or 7-speed rear
derailer. They're pretty cheap and easy to come by, as pointed out in
this thread.
Tim McNamara
jim beam <nos...@example.net> wrote:
> Tim McNamara wrote:
> > Not having a six speed wheel, I can't test this. I am wondering,
> > though, if the range of adjustment of the limit screws on a current
> > 9 or 10 speed derailleur is sufficient to limit spokeward motion of
> > the
> > derailleur to prevent overshifting into the spokes. The inward
> > swing for 9/10 speeds is pretty far and the geometry of the
> > derailleurs is different than it was in the ancient times of 15
> > years ago.
>
> tim, if a derailleur has limit adjusting screws, and i've never seen
> one of any manufacture or vintage that doesn't, then it can be
> adjusted to prevent chain fouling.
That's assuming that the limit screw has sufficient adjustment to keep
the derailleur from swinging in far enough to unship the chain into the
spokes. I wouldn't automatically assume that this is or isn't the case.
That's why I asked. I don't have a six speed bike to try it out on and
don't offhand know anyone who does.
> the much more interesting question, posed to yourself as someone with
> more than a passing interest in human psychology, is why do some
> people, believe everything told them with a certain delivery, even
> when simple independent fact checking can easily and quickly be done
> proving that delivery's content to be, er, "mistaken"? it absolutely
> fascinates me!
> why is this? what's the mechanism?
There are opposed tendencies here. One, which you are describing, is
the appeal of authority. The other is the rejection of authority. The
response is not to what has been said but to whom has said it. Either
is basically an automatic response resulting from a rigid attitude.
This explains a lot of political posturing as well (for example, why
many Republican Congresspeople support giving President Bush the line
item veto "because it's a valuable tool the President needs" when those
same people opposed giving the line item veto to President Clinton
"because it would upset the balance of power."). When someone's
authority is rated favorably, their statement are more likely to be
accepted uncritically- and of course, vice versa. There's more to it
than simply an authoritative style, however. The ideology presented by
the person has to more or less correspond to that of the audience. And
the likeability of the characters involved is also a factor.
The purpose is to avoid having to engage in the hard work of critical
thinking. This is compounded by a lack of teaching grade school and
high school students to think effectively (e.g., simply requiring a
logic class for all students). Effective thinking is harder to test and
instead we teach what can be tested for on standardized tests- facts
rather than abilities. The price is paid when these students get into
college and can't think creatively or break problems into rational,
solvable steps. (There's a book called _Effective Thinking_ by the
great Gestalt psychologist Max Wertheimer, which IMHO everyone should
read. It's out of print but can be found through Alibris and ABEBooks).
> i observed this in another forum recently. i attended an industry
> "challenge" where two competing teams of undergrads [cal & stanford
> as it happens] were posed a debate question. they had to respond
> with analysis and recommendations for a solution. and interestingly
> for me, the particular subject matter was one on which i am one of 5,
> maybe 6 subject experts, so i was in the privileged position of
> knowing every detail they were trying to address. long story short,
> the cal team did their homework, got all the relevant and correct
> information and proposed a sensible solution in accordance with the
> facts. the stanford team otoh, had serious factual problems, made
> wild assumptions that had no relevance to the data they actually had
> bothered to collect, but blustered on anyway. result: stanford won
> the popular vote. cal's delivery was that of letting the facts speak
> for themselves. stanford's was that of "assertion will prevail".
> and it did. so, i ask again, why is this? what is the mechanism by
> which trivially exposed bullshit seems to be so easily swallowed?
> and more importantly, what is the compulsion for sociopaths to
> exploit it?
People inhabiting that position get rewarded, whether they are
sociopaths or not. It's simple positive reinforcement. Easily
palatable bullshit is easier to sell than a bitter truth.
Michael Press
Luke <lucasi...@rogers.com> wrote:
Examine the fracture carefully for pattern and corrosion.
--
Michael Press
Michael Press
<timmcn-13A554....@news.iphouse.com>,
Tim McNamara <tim...@bitstream.net> wrote:
> The purpose is to avoid having to engage in the hard work of critical
> thinking. This is compounded by a lack of teaching grade school and
> high school students to think effectively (e.g., simply requiring a
> logic class for all students). Effective thinking is harder to test and
> instead we teach what can be tested for on standardized tests- facts
> rather than abilities. The price is paid when these students get into
> college and can't think creatively or break problems into rational,
> solvable steps. (There's a book called _Effective Thinking_ by the
> great Gestalt psychologist Max Wertheimer, which IMHO everyone should
> read. It's out of print but can be found through Alibris and ABEBooks).
Geometry is better than a class in logic. The student
learns to prove propositions. Proving them requires
connecting a series of little syllogisms into a coherent
argument. The student must critique his own arguments for
completeness, hidden assumptions, and logical coherence;
this is the difficult part. Once mastered he can apply the
skill to the arguments of others.
--
Michael Press
jobst....@stanfordalumni.org
>>> The spoke shows no bend in it, so the incident with the collision
>>> had no effect on it. Depth of engagement also had no effect.
>>> Spokes have been advanced to a point that several brass threads in
>>> the nipple were pressed flat and caused no problem.
>>> What should be more apparent is that a 2mm spoke has its weakest
>>> cross section at the thread and when fully engaged in the nipple,
>>> is the place where cyclic stress from loaded wheel rotation is
>>> highest. A spoke threaded from end-to-end would not suffer that
>>> fate, its stress (and stretching) being uniform over its length.
>>> This is a good reason to use swaged spokes and to use thin ones to
>>> have a greater load distribution among spokes. I assume you
>>> followed the suggestions in the book to stress relieve after
>>> tensioning?
>> Yes the wheel was stress relieved at the time of building. But
>> isn't it too much of a coincidence that it was the spoke located at
>> the point of impact (from the prior accident) that broke?
>> I'm inclined to believe Ron Ruff's explanation. As you've noted the
>> weakest cross-section of the spoke in question is at the thread, in
>> this case cheek by jowl to where the wheel was smashed; perhaps the
>> spoke developed a crack during the initial mishap and -- thinking
>> out loud -- couldn't the post-accident truing session further
>> weaken it by applying a torsional (spoke windup) force?
> Examine the fracture carefully for pattern and corrosion.
That the crack developed over a longer time before separation should
be apparent when inspecting the face of the fracture, especially if
ridden in rain. It will have a dirty grey area. By this you can see
how far it was cracked before failure. You ought to find waves of
crack front advance. A magnifying glass helps for viewing this.
That it separated only when the final area was about 1/3 the cross
section would show that it took many cycles from the time the crack
started until it finally said "enough" I give up!
Jobst Brandt
jim beam
which is mostly true, but the most important thing you're failing to
state which is vital for analysis purposes, and that is obvious from the
op's pic, is that there's no necking or gross plastic deformation,
therefore fatigue is almost certainly the cause of failure.
>
> That it separated only when the final area was about 1/3 the cross
> section
that's entirely presumptive jobst. it depends on the load, the
material, the rate, and a whole bunch of other stuff.
jim beam
very interesting analysis. i think i'll look for that book too. thanks
tim.
jim beam
> Carl Fogel writes:
>
>
>>>With a 6-speed cluster, todays derailleurs cannot be adjusted to
>>>prevent over shifting into the spokes. It takes care.
>
>
>>I'm puzzled.
>
>
>>I have no problems with over shifting my 1998 Schwinn LeTour 7-speed
>>Shimano, but I can't be accused of careful shifting.
>
>
>>I went out to the garage just now and deftly turned the wrong limit
>>screw, transforming my 7-speed into a 5-speed cluster that wouldn't
>>shift outward to the smallest two rear sprockets.
>
>
> Do you have a 9-speed derailleur. My SunTour derailleur's limit stop
> screw misses the tab it is designed to engage by a large angle
> (3-speeds worth) and serves no purpose. You have a different model
> that apparently won't handle 9- or 10- speeds. I ran out of those
> over time as derailleur sticks ripped them apart on forest trails.
translation: "if i can just scrape around for enough limitations on my
original statement, maybe i can..."
>
>
>>After re-adjusting my mistake, I turned the other limit screw and
>>had no trouble eliminating my largest rear sprocket, turning it into
>>a 6-speed, unable to pull the chain in far enough to grab 1st gear.
>
>
> Well that's the limit that doesn't work for me and hasn't for a couple
> of derailleurs now.
which ones? shimano adjust up just fine. see above.