2 bolt vs 4 bolt stems

In article <[email hidden]>,

(PeteCresswell) said:

Per [email hidden]:

Quoted message said:

blunders such as Shimano Octalink

?

http://draco.acs.uci.edu/rbfaq/FAQ/8f.20.html

In short, Jobst has identified a failure mode in which back torque can
be a big problem, causing loosening of the crank bolts.

I run a set of these doomed cranks, but I have a few things going for
me: I'm light, I don't descend right foot forward, and they were so
cheap I won't lose any sleep if I only get a few years out of the crank
and BB.

--
Ryan Cousineau [email hidden] http://www.wiredcola.com/
"I don't want kids who are thinking about going into mathematics
to think that they have to take drugs to succeed." -Paul Erdos

Ryan Cousineau said:

Quoted message said:

Quoted message said:

...blunders such as Shimano Octalink?

http://draco.acs.uci.edu/rbfaq/FAQ/8f.20.html

Quoted message said:

In short, Jobst has identified a failure mode in which back torque
can be a big problem, causing loosening of the crank bolts.

What was sad to see is that Shimano thought the failures were caused
by insufficient spline length and came out with a second version of
the Octalink before realizing that it worked no better than the first
one. I'm not sure they understand what caused the failure, only that
the design didn't work. I've seen such things in industry a few
times, where the failure of a design was never understood and it was
left behind for another solution.

At least the new design has redundancy in that it used two pinch
bolts. Shimano may not have done that for safety but rather to get a
more elegant profile with two smaller screws instead of one large one.

Quoted message said:

I run a set of these doomed cranks, but I have a few things going
for me: I'm light, I don't descend right foot forward, and they were
so cheap I won't lose any sleep if I only get a few years out of the
crank and BB.

This has nothing to do with rider weight. Not being goofy-footed is
the main benefit. Many riders insisted that the failures were due to
faulty installation because it never happened to them. In fact it had
nothing to do with that. Most people naturally lead with their left
foot, dancing as well as on surf, snow and skate boards and when they
approach a stair.

Jobst Brandt

Per Ryan Cousineau:

Quoted message said:

I won't lose any sleep if I only get a few years out of the crank
and BB.

On one hand, after reading Jobst's explanation, it seems like as long as we
apply a torque wrench periodically there's not going to be a problem.

On the other hand, I'd definitely lose sleep over the prospect of a crank
failing under load. Been there done that with a chain lunging up a hill. It
definitely smarts - and with the chain, at least the foot stays on the pedal.

I'd think a crank arm parting under that kind of load could be catastrophic to
the rider - especially if there are moving motor vehicles in the vicinity.
--
PeteCresswell

Quoted message said:

Ryan Cousineau said:

Quoted message said:

> ...blunders such as Shimano Octalink?

http://draco.acs.uci.edu/rbfaq/FAQ/8f.20.html

Quoted message said:

In short, Jobst has identified a failure mode in which back torque
can be a big problem, causing loosening of the crank bolts.

What was sad to see is that Shimano thought the failures were caused
by insufficient spline length and came out with a second version of
the Octalink before realizing that it worked no better than the first
one. I'm not sure they understand what caused the failure, only that
the design didn't work. I've seen such things in industry a few
times, where the failure of a design was never understood and it was
left behind for another solution.

At least the new design has redundancy in that it used two pinch
bolts. Shimano may not have done that for safety but rather to get a
more elegant profile with two smaller screws instead of one large one.

Quoted message said:

I run a set of these doomed cranks, but I have a few things going
for me: I'm light, I don't descend right foot forward, and they were
so cheap I won't lose any sleep if I only get a few years out of the
crank and BB.

This has nothing to do with rider weight. Not being goofy-footed is
the main benefit. Many riders insisted that the failures were due to
faulty installation because it never happened to them. In fact it had
nothing to do with that. Most people naturally lead with their left
foot, dancing as well as on surf, snow and skate boards and when they
approach a stair.

In my off road riding I've been trying to train myself to lead with the foot
that goes down next. In other words if the trail goes left I've got my right
foot forward, if the next corner is a right turn then I'm left first. At my
otherwise inept level of bike handling it doesn't make a lot of difference
except in wear and tear on my right hip. This way both legs get the same amount
of work.

I use ISIS on my main MTB.

Ron

Pete Cresswell said:

Quoted message said:

I won't lose any sleep if I only get a few years out of the crank
and BB.

Quoted message said:

On one hand, after reading Jobst's explanation, it seems like as
long as we apply a torque wrench periodically there's not going to
be a problem.

I think you misunderstand. This is a problem that, in the course of
one ride, can loosen the retaining bolt far enough that the ends of
the spline in the aluminum crank shear off (the common mode of failure
for this system).

Quoted message said:

On the other hand, I'd definitely lose sleep over the prospect of a
crank failing under load. Been there done that with a chain lunging
up a hill. It definitely smarts - and with the chain, at least the
foot stays on the pedal.

I haven't experienced it but I suspect it is a mushy failure in which
the ends of the aluminum spline are sheared off.

Quoted message said:

I'd think a crank arm parting under that kind of load could be
catastrophic to the rider - especially if there are moving motor
vehicles in the vicinity.

I don't know, I've never met anyone who had such a failure. I only
saw the results in the bicycle shop and pictures here in wreck.bike.

Jobst Brandt

[email hidden] wrote in news:43ac642d$0$58089
[email hidden]:

Quoted message said:

There is no difference between number of screws that you can feel
because that would require that the clamp was loose for there to be
relative motion (something you can feel). Safety is entire
difference. Screws can fail for one reason or another, such as threads
pulling out of the aluminum or a crack in the screw itself.

Because these stems work the way they do, no single screw should alone
be be able to release the bar from the bicycle. That is why I believe
in two bolt clamping to the steer tube and four for the bar clamp.
That way there is a redundancy.

I'm trying to understand why failure of one bolt on a 2-bolt bar clamp
would automatically cause the bars to release. How do you know the bar
will release? I removed one of the bolts from my bar clamp of my
Bontrager mountain bike stem and, while the joint was a little loose, the
bars did not fall out nor did it seem like it would do so if it happened
out on a ride.

Only once did I see a bar release when the stem bolt broke. This was on a
stem which had one bolt to clamp the handlebar, and the clamp itself was
hinged at the top. The bolt itself was loaded in bending and it was made
of aluminum. The stem was subsequently recalled. It was a Trek Icon stem.

I once had a Softride stem break at the handlebar clamp. This had a
conventional attachment like a quill stem normally has. Luckily it broke
on a climb when I was seated. The attachment was so loose that I believe
the bars would have released on a descent.

As evidenced by the article at
http://nbeener.com/EASTON_TWO_BOLTS_vs_FOUR.pdf it would seem that a 4
bolt clamp has more of a potential to damage the handlebars due to
overtightening (according to the article, it's easily done), that could
cause the bars to break, so we may be trading one failure mode for
another.

--
Mike DeMicco <[email hidden]>

On 25 Dec 2005 17:05:56 GMT, Mike DeMicco <[email hidden]>

Quoted message said:

[email hidden] wrote in news:43ac642d$0$58089
[email hidden]:

Quoted message said:

There is no difference between number of screws that you can feel
because that would require that the clamp was loose for there to be
relative motion (something you can feel). Safety is entire
difference. Screws can fail for one reason or another, such as threads
pulling out of the aluminum or a crack in the screw itself.

Because these stems work the way they do, no single screw should alone
be be able to release the bar from the bicycle. That is why I believe
in two bolt clamping to the steer tube and four for the bar clamp.
That way there is a redundancy.

I'm trying to understand why failure of one bolt on a 2-bolt bar clamp
would automatically cause the bars to release. How do you know the bar
will release? I removed one of the bolts from my bar clamp of my
Bontrager mountain bike stem and, while the joint was a little loose, the
bars did not fall out nor did it seem like it would do so if it happened
out on a ride.

On a road bike it's a bit more of a problem, even a riser bar is more or less in
line with the mounting, a drop bar is cantilevered out in front of the stem and
will put you flat on your face if it's even loose. Ask how I know....

Quoted message said:

Only once did I see a bar release when the stem bolt broke. This was on a
stem which had one bolt to clamp the handlebar, and the clamp itself was
hinged at the top. The bolt itself was loaded in bending and it was made
of aluminum. The stem was subsequently recalled. It was a Trek Icon stem.

I have an Icon in a special section of my bike junk box called "stems that scare
me."

Ron

Quoted message said:

I once had a Softride stem break at the handlebar clamp. This had a
conventional attachment like a quill stem normally has. Luckily it broke
on a climb when I was seated. The attachment was so loose that I believe
the bars would have released on a descent.

As evidenced by the article at
http://nbeener.com/EASTON_TWO_BOLTS_vs_FOUR.pdf it would seem that a 4
bolt clamp has more of a potential to damage the handlebars due to
overtightening (according to the article, it's easily done), that could
cause the bars to break, so we may be trading one failure mode for
another.

Mike DeMicco said:

Quoted message said:

There is no difference between number of screws that you can feel
because that would require that the clamp was loose for there to be
relative motion (something you can feel). Safety is entire
difference. Screws can fail for one reason or another, such as
threads pulling out of the aluminum or a crack in the screw itself.

Quoted message said:

Quoted message said:

Because these stems work the way they do, no single screw should
alone be be able to release the bar from the bicycle. That is why
I believe in two bolt clamping to the steer tube and four for the
bar clamp. That way there is a redundancy.

Quoted message said:

I'm trying to understand why failure of one bolt on a 2-bolt bar
clamp would automatically cause the bars to release. How do you
know the bar will release? I removed one of the bolts from my bar
clamp of my Bontrager mountain bike stem and, while the joint was a
little loose, the bars did not fall out nor did it seem like it
would do so if it happened out on a ride.

First, the bar would become loose immediately and on the next pull
would bend open the cover plate that is now attached only on one end.

Quoted message said:

Only once did I see a bar release when the stem bolt broke. This
was on a stem which had one bolt to clamp the handlebar, and the
clamp itself was hinged at the top. The bolt itself was loaded in
bending and it was made of aluminum. The stem was subsequently
recalled. It was a Trek Icon stem.

Quoted message said:

I once had a Softride stem break at the handlebar clamp. This had a
conventional attachment like a quill stem normally has. Luckily it broke
on a climb when I was seated. The attachment was so loose that I believe
the bars would have released on a descent.

Quoted message said:

As evidenced by the article at

http://nbeener.com/EASTON_TWO_BOLTS_vs_FOUR.pdf

Quoted message said:

it would seem that a 4 bolt clamp has more of a potential to damage
the handlebars due to overtightening (according to the article, it's
easily done), that could cause the bars to break, so we may be
trading one failure mode for another.

I don't understand why the concept of redundancy for safety is so
obscure. The stem pictured at that URL would release the bar if one
screw stripped out or broke for some reason. Attributing the design
to fashion is silly.

As for denting the bar, the both stem and cap have a circular contour
that matches the bar diameter and will not damage the bar unless it is
too weak as a handlebar anyway. Handlebar forces are large enough to
readily approach tube deformation more sooner than clamp forces would
because they are canting forces. For this reason a bar clamp should
be at least a bar diameter wide, which two bolt clamps generally are
not.

Jobst Brandt

Mike DeMicco said:

I'm trying to understand why failure of one bolt on a 2-bolt bar clamp
would automatically cause the bars to release. How do you know the bar
will release? I removed one of the bolts from my bar clamp of my
Bontrager mountain bike stem and, while the joint was a little loose, the
bars did not fall out nor did it seem like it would do so if it happened
out on a ride.

Breakage of a single bolt would not cause the bar to "fall off" but
could permit it to rotate on the pitch axis. If you are holding the bar
itself, this is not liable to cause a loss of control, but if you're
riding on the brake lever hoods, sudden rotation of the bars could cause
your hands to lose contact, possibly causing a FDGB.

Quoted message said:

I once had a Softride stem break at the handlebar clamp. This had a
conventional attachment like a quill stem normally has. Luckily it broke
on a climb when I was seated. The attachment was so loose that I believe
the bars would have released on a descent.

This is not entirely due to "luck." The greatest stress on a bar/stem
is generally when you're climbing, because that's when you are going to
be pulling hard on it. (Landing from a jump can also be an issue. The
only time I ever broke a handlebar was going down a curb. Fortunately I
wasn't going fast, and didn't crash.

Sheldon "Ouch!" Brown
+-------------------------------------------------+
| Search the Quality Bicycle Products Catalog |
| and order on line: |
| http://sheldonbrown.com/quality |
+-------------------------------------------------+
Harris Cyclery, West Newton, Massachusetts
Phone 617-244-9772 FAX 617-244-1041
http://harriscyclery.com
Hard-to-find parts shipped Worldwide
http://captainbike.com http://sheldonbrown.com

Dans le message de news:[email hidden],
[email hidden] <[email hidden]> a
réfléchi, et puis a déclaré :

Quoted message said:

Mike DeMicco said:

Quoted message said:

There is no difference between number of screws that you can feel
because that would require that the clamp was loose for there to be
relative motion (something you can feel). Safety is entire
difference. Screws can fail for one reason or another, such as
threads pulling out of the aluminum or a crack in the screw itself.

Quoted message said:

Quoted message said:

Because these stems work the way they do, no single screw should
alone be be able to release the bar from the bicycle. That is why
I believe in two bolt clamping to the steer tube and four for the
bar clamp. That way there is a redundancy.

Quoted message said:

I'm trying to understand why failure of one bolt on a 2-bolt bar
clamp would automatically cause the bars to release. How do you
know the bar will release? I removed one of the bolts from my bar
clamp of my Bontrager mountain bike stem and, while the joint was a
little loose, the bars did not fall out nor did it seem like it
would do so if it happened out on a ride.

First, the bar would become loose immediately and on the next pull
would bend open the cover plate that is now attached only on one end.

Quoted message said:

Only once did I see a bar release when the stem bolt broke. This
was on a stem which had one bolt to clamp the handlebar, and the
clamp itself was hinged at the top. The bolt itself was loaded in
bending and it was made of aluminum. The stem was subsequently
recalled. It was a Trek Icon stem.

Quoted message said:

I once had a Softride stem break at the handlebar clamp. This had a
conventional attachment like a quill stem normally has. Luckily it
broke on a climb when I was seated. The attachment was so loose
that I believe the bars would have released on a descent.

Quoted message said:

As evidenced by the article at

http://nbeener.com/EASTON_TWO_BOLTS_vs_FOUR.pdf

Quoted message said:

it would seem that a 4 bolt clamp has more of a potential to damage
the handlebars due to overtightening (according to the article, it's
easily done), that could cause the bars to break, so we may be
trading one failure mode for another.

I don't understand why the concept of redundancy for safety is so
obscure. The stem pictured at that URL would release the bar if one
screw stripped out or broke for some reason. Attributing the design
to fashion is silly.

As for denting the bar, the both stem and cap have a circular contour
that matches the bar diameter and will not damage the bar unless it is
too weak as a handlebar anyway. Handlebar forces are large enough to
readily approach tube deformation more sooner than clamp forces would
because they are canting forces. For this reason a bar clamp should
be at least a bar diameter wide, which two bolt clamps generally are
not.

Jobst Brandt

I *completely* agree with you, except for more sooner. I don't see a good
reason for having split caps, in contrast with a single cap.
--
Sandy
--
S'endormir au volant, c'est très dangereux.
S'endormir à vélo, c'est très rare.
S'endormir à pied, c'est très con.
- Geluck, P.

Quoted message said:

I think you misunderstand. This is a problem that, in the course of
one ride, can loosen the retaining bolt far enough that the ends of
the spline in the aluminum crank shear off (the common mode of failure
for this system).

Perhaps they were not installed with the correct torque?

I have been running XT octalink v2 cranks (M751 and M752) for five
years on my MTBs. I pop them off routinely to change the rings and
when I replace the BB (every year or so as preventative maintenance).

I always install the cranks using a torque wrench, and I always
recheck after the first few rides. Only once do I recall a crank bolt
that loosened slightly after the install (I don't remember if that was
within a few rides of a re-install or just during normal use).

I frequent one of the best shops in my area, and am always surprised
when they "laugh off" the need to use a torque wrench. For that
reason, I try to never let them get near my stems...

Phil who? said:

Quoted message said:

I think you misunderstand. This is a problem that, in the course
of one ride, can loosen the retaining bolt far enough that the ends
of the spline in the aluminum crank shear off (the common mode of
failure for this system).

Quoted message said:

Perhaps they were not installed with the correct torque?

Aaargh! You seem to have missed the earlier part of this thread or
any of the similar ones that preceded it. The failure mode occurs
with "goofy footed" riders, ones who stand on their pedals with the
right foot forward. You are apparently not one of these and should
have no problem. That is why Shimano couldn't discover what caused
the failures. They did not recognize this effect even though it had
been discussed on this forum often.

Quoted message said:

I have been running XT Octalink v2 cranks (M751 and M752) for five
years on my MTBs. I pop them off routinely to change the rings and
when I replace the BB (every year or so as preventative
maintenance).

It makes no difference how tight you make the retaining bolt. It will
unscrew if you stand on your pedals when coasting downhill or jumping
over bumps.

Quoted message said:

I always install the cranks using a torque wrench, and I always
recheck after the first few rides. Only once do I recall a crank
bolt that loosened slightly after the install (I don't remember if
that was within a few rides of a re-install or just during normal
use).

Quoted message said:

I frequent one of the best shops in my area, and am always surprised
when they "laugh off" the need to use a torque wrench. For that
reason, I try to never let them get near my stems...

Oh how nice of you, but they are right about that. Torque has no
effect on loosening of the retaining bolt. To be repetitive:

I think you misunderstand.

Jobst Brandt

Quoted message said:

Aaargh! You seem to have missed the earlier part of this thread or
any of the similar ones that preceded it. The failure mode occurs
with "goofy footed" riders, ones who stand on their pedals with the
right foot forward. You are apparently not one of these and should
have no problem. That is why Shimano couldn't discover what caused
the failures. They did not recognize this effect even though it had
been discussed on this forum often.

It makes no difference how tight you make the retaining bolt. It will
unscrew if you stand on your pedals when coasting downhill or jumping
over bumps.

Oh how nice of you, but they are right about that. Torque has no
effect on loosening of the retaining bolt. To be repetitive:

I think you misunderstand.

Thanks for taking the time to post the explanation again. Maybe I
did/do misunderstand, but I did read all of the posts in this thread
as well as your comments in the FAQ (link provided by Ryan Cousineau
above). In the FAQ, you describe a reverse torque effect (that I
interpret as a backwards ratcheting effect) that only occurs when the
rider stands on the pedals with right foot forward.

Well, I did mention in my post that I never had an octalink v2 crank
bolt loosen significantly during a ride, but I failed to mention that
I always stand on the pedals when riding over technical terrain, and
usually with my right foot forward (i dunno why my right foot likes to
be forward, it just does). I think that makes me "goofy footed". I've
been bouncing down the trails this way for five years now...

Ok, so if proper torque has no effect, and I am not sure that I
believe that yet, but if so, then the only other thing I can think of
is that maybe my light weight (around 125 lbs) limits the extent of
reverse ratcheting that I can produce?

Phil who? said:

Quoted message said:

Aaargh! You seem to have missed the earlier part of this thread or
any of the similar ones that preceded it. The failure mode occurs
with "goofy footed" riders, ones who stand on their pedals with the
right foot forward. You are apparently not one of these and should
have no problem. That is why Shimano couldn't discover what caused
the failures. They did not recognize this effect even though it
had been discussed on this forum often.

Quoted message said:

Quoted message said:

It makes no difference how tight you make the retaining bolt. It
will unscrew if you stand on your pedals when coasting downhill or
jumping over bumps.

Quoted message said:

Quoted message said:

Oh how nice of you, but they are right about that. Torque has no
effect on loosening of the retaining bolt. To be repetitive:

Quoted message said:

Quoted message said:

I think you misunderstand.

Quoted message said:

Thanks for taking the time to post the explanation again. Maybe I
did/do misunderstand, but I did read all of the posts in this thread
as well as your comments in the FAQ (link provided by Ryan Cousineau
above). In the FAQ, you describe a reverse torque effect (that I
interpret as a backward ratcheting effect) that only occurs when the
rider stands on the pedals with right foot forward.

Quoted message said:

Well, I did mention in my post that I never had an Octalink v2 crank
bolt loosen significantly during a ride, but I failed to mention
that I always stand on the pedals when riding over technical
terrain, and usually with my right foot forward (i dunno why my
right foot likes to be forward, it just does). I think that makes
me "goofy footed". I've been bouncing down the trails this way for
five years now...

Quoted message said:

OK, so if proper torque has no effect, and I am not sure that I
believe that yet, but if so, then the only other thing I can think
of is that maybe my light weight (around 125 lbs) limits the extent
of reverse ratcheting that I can produce?

I find that interesting, because the spline has backlash. In use,
while the retaining screw is backing out, cranks show no looseness
because they are on a slight taper on the tubular spindle. Friction
will not prevent the that backlash from being taken up when reverse
torque occurs from standing on the left crank in the rearward
position. Similarly pedal attachments move with every hard pedaling
stroke even tough the pedal has been tightened as much as is possible.
Left pedals have left hand threads because they always move.

That your crank, with right foot forward riding, did not loosen means
either that you tightened it often enough (it creeps only a fraction
of a degree for many load changes) at which time you should have
noticed that the retaining bolt could be tightened. I cannot imagine
operating these cranks for any longer period, right foot forward
standing, without unscrewing one or both the retaining screws.

In any event, the Octalink failed often enough under these
circumstances that Shimano first tried to fix it by a lengthening the
spline, not having understood the cause of failure, and then finally
giving up on the concept.

Jobst Brandt