Brake Trouble / Road Bike

Jobst Brandt wrote:

Unfortunately, bicycle repair shops, unaware of the cause, use hammers
and punches to bend the offending spring

I replied:

That technique was appropriate for some older sidepulls, models dating
from the early '70s and older, and was the only way to regulate the
centering of the caliper.

Jobst:

I don't believe it was ever appropriate. The anchor bolt on those
brakes could be rotated and the springs lubricated just the same as on
later more expensive brakes.

(I believe Jobst was referring to the center bolt, not the anchor bolt.)

Tom Ace wrote:

I had a Raleigh 3-speed with steel-caliper sidepull brakes.
The front brake bolt had a cylindrical concave surface that
mated with a convex surface on the fork. The bolt's rotation
was not adjustable. If lubrication didn't fix the centering,
the only other option I knew of was to bend the spring.

That's correct for those brakes. Those were the bikes I grew up with in
my early days of bike tinkering.

This doesn't apply just to the front brake. The rear also has a
cylindrical concavity that mates with the seatstay bridge.

Most newer sidepulls use a separate "radiussed washer" to mate the flat
back of the spring black with the fork crown or seatstay bridge. These
do permit adjustment without bending the spring, though it still
sometimes is accomplished with a hammer and punch.

Many older sidepulls, such as the typical Weinmann/Dia Compe units, had
no way to secure or rotate the spring mounting block. (This was before
Campagnolo thought of putting wrench flats on the spring block, and
before Weinmann thought of putting a 4 mm hex on the end of the center
bolt.)

When you would tighten the mounting nut on these calipers, the friction
of the threads would often tend to rotate the center bolt out of
position, even if you started with it centered.

A hammer and punch used gently on the spring, close to the central block
could rotate the centerbolt back to the correct orientation. This did
not actually involve causing the spring to yield, it was just the only
way to apply torque to the spring mounting block in the absence of
wrench flats.

For a while, special tools were available for grabbing the spring and
twisting the center bolt but these were never common, and are probably
extinct by now. The Park OBW-3 was one of these, but I don't think
these are currently available.

Sheldon "Technology Changes" Brown
+-------------------------------------------------------------+
| Want of care does us more damage than want of knowledge. |
| -- Benjamin Franklin |
+-------------------------------------------------------------+
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

Tom Ace writes:

Unfortunately, bicycle repair shops, unaware of the cause, use
hammers and punches to bend the offending spring

That technique was appropriate for some older sidepulls, models
dating from the early '70s and older, and was the only way to
regulate the centering of the caliper.

I don't believe it was ever appropriate. The anchor bolt on those
brakes could be rotated and the springs lubricated just the same as
on later more expensive brakes.

I had a Raleigh 3-speed with steel-caliper sidepull brakes. The
front brake bolt had a cylindrical concave surface that mated with a
convex surface on the fork. The bolt's rotation was not adjustable.
If lubrication didn't fix the centering, the only other option I
knew of was to bend the spring.

I don't see why that was the only available method. How did you
initially poisition the brake caliper was with pads adjacent to the
rim? Why could the centerbolt not be rotated? I don't believe it was
riveted or welded in place.

Jobst Brandt

Sheldon Brown writes:


Unfortunately, bicycle repair shops, unaware of the cause, use
hammers and punches to bend the offending spring

That technique was appropriate for some older sidepulls, models
dating from the early '70s and older, and was the only way to
regulate the centering of the caliper.

I don't believe it was ever appropriate. The anchor bolt on those
brakes could be rotated and the springs lubricated just the same
as on later more expensive brakes.

(I believe Jobst was referring to the center bolt, not the anchor bolt.)

Please explain the difference between a center bolt and anchor bolt.
I am referring to the bolt about which the caliper arms rotate and
which attaches the spring boss against the fork crown or rear brake
bridge. The ones to which I refer are interchangeable with Mafac
centerpull and Campagnolo side pull brakes.

I had a Raleigh 3-speed with steel-caliper sidepull brakes. The
front brake bolt had a cylindrical concave surface that mated with
a convex surface on the fork. The bolt's rotation was not
adjustable. If lubrication didn't fix the centering, the only
other option I knew of was to bend the spring.

That's correct for those brakes. Those were the bikes I grew up
with in my early days of bike tinkering.

This doesn't apply just to the front brake. The rear also has a
cylindrical concavity that mates with the seatstay bridge.

What is "this" in "this doesn't apply only"?

Most newer sidepulls use a separate "radiussed washer" to mate the
flat back of the spring black with the fork crown or seatstay
bridge. These do permit adjustment without bending the spring,
though it still sometimes is accomplished with a hammer and punch.

I don't understand why the center bolt could not be loosened and the
return spring rotated, the same as it was on initial installation.
Besides, if the brake was initially centered when new, clean and
lubricated, it needed only to be restored to the clean and lubricated
state to be centered. The springs did not yield in use and did not
become asymmetrically loaded.

Many older sidepulls, such as the typical Weinmann/Dia Compe units,
had no way to secure or rotate the spring mounting block. (This was
before Campagnolo thought of putting wrench flats on the spring
block, and before Weinmann thought of putting a 4 mm hex on the end
of the center bolt.)

I disagree, having adjusted them initially and centering them by
cleaning an lubrication. How else could one install a new brake if
its rotational position was not adjustable?

When you would tighten the mounting nut on these calipers, the
friction of the threads would often tend to rotate the center bolt
out of position, even if you started with it centered.

That takes a little skill to retain the spring in its position while
tightening the bolt. This is a common practice in machinery.

A hammer and punch used gently on the spring, close to the central
block could rotate the centerbolt back to the correct orientation.
This did not actually involve causing the spring to yield, it was
just the only way to apply torque to the spring mounting block in
the absence of wrench flats.

With a slightly looser center bolt this operation can be done without
impact and with suitable accuracy. That is the whole point of
bringing this up. I realize this does not correct the errors of
history but it was this sort of "curbstone mechanics" that I found
crude when I worked on my bicycle in grade school. I had seen better
while repairing my mothers Model-A ford.

For a while, special tools were available for grabbing the spring
and twisting the center bolt but these were never common, and are
probably extinct by now. The Park OBW-3 was one of these, but I
don't think these are currently available.

A bit of imaginative mechanics, even today, makes most of the special
tools unnecessary. I had no need for a special brake adjusting tool.

Jobst Brandt

On Tue, 26 Jul 2005 01:24:18 GMT,
wrote:

Let's assume that much about brake adjustment is understood. tension
in the cable should have little effect on centering, at least to the
amount that they might not move equally.

I have encountered one instance in which the installation of a
replacement cable assembly on a rear side-pull brake caused the brake
to fail to center reliably. The replacement cable housing was
slightly too short, so that when the brake was applied, the caliper
moved to take up slack in the housing (concealed by the ferrule)
before fully applying. When released, the caliper appeared to spring
back asymmetrically because it had been pulled off-center when
applied. Putting in a slightly longer rear housing segment fixed it.
I suspect that in some cases, a poorly-routed housing (or one that was
slightly too long) that pushed on the caliper might have a less
pronounced but similar effect.
--
Typoes are a feature, not a bug.
Some gardening required to reply via email.
Words processed in a facility that contains nuts.

wrote:

I don't see why that was the only available method. How did you
initially poisition the brake caliper was with pads adjacent to the
rim? Why could the centerbolt not be rotated?

See http://www.yellowjersey.org/ralbrake.html

You can see several parts with the cylindrical concave
surface I described, which mates with the fork (or as
Sheldon pointed out for the rear brake, the brake bridge)
in one rotational position only. If you look closely at
those pieces, you can see a groove (in the opposite flat
surface, not in the concave side) that holds the spring.

I don't know whether that piece is integral to the bolt,
but that's not the point. What matters is that the
spring is held by a piece whose rotation isn't adjustable.

Tom Ace

Tom Ace writes:

I don't see why that was the only available method. How did you
initially position the brake caliper was with pads adjacent to the
rim? Why could the center bolt not be rotated?

See http://www.yellowjersey.org/ralbrake.html

You can see several parts with the cylindrical concave surface I
described, which mates with the fork (or as Sheldon pointed out for
the rear brake, the brake bridge) in one rotational position only.
If you look closely at those pieces, you can see a groove (in the
opposite flat surface, not in the concave side) that holds the
spring.

I don't know whether that piece is integral to the bolt, but that's
not the point. What matters is that the spring is held by a piece
whose rotation isn't adjustable.

The piece is not integral with the bolt but it is registered on the
curve of the fork crown and on the diameter of some rear brake
bridges. As you see in the URL cited, the springs are symmetrical as
are the curves in the spring boss and these align centered on the
wheel if it isn't in askew.

The point is that these brakes don't even allow misadjustment when
friction draws the brake off center and cleaning and lubricating is
required. To hammer on the return spring only shifts the problem to a
slightly later time when the friction in the spring contact point
changes again.

The hammer may be a quick fix, but understanding the problem and
responding to it in kind is a better solution.

Jobst Brandt

wrote:

The hammer may be a quick fix, but understanding the problem and
responding to it in kind is a better solution.

I got started in this thread in response to your saying there
was never a need for the hammer routine with any brake.

If cleaning/lubing the Raleigh brakes doesn't center them,
what else is there left to do but to bend the spring?


Tom Ace

Tom Ace writes:

The hammer may be a quick fix, but understanding the problem and
responding to it in kind is a better solution.

I got started in this thread in response to your saying there was
never a need for the hammer routine with any brake.

If cleaning/lubing the Raleigh brakes doesn't center them, what else
is there left to do but to bend the spring?


Somehow these threads drift off into unrelated topics, probably to
defend a defenseless position that no one else knows the writer holds
until defended here.

The subject of this thread was centering a modern high quality Shimano
brake, not steel caliper Raleigh brakes of decades ago. Even those
brakes recognized the problem and made spring bosses pressed onto
knurled center bolts that were pressed in place to assist adjusting
brake position. Let's leave the hammer and punch in the historic bin.
There is no place for them on any recent aluminum alloy brake caliper.

The radius was there to offer initial centering on similarly curved
fork crowns. Unfortunately many fork crowns had a flat mounting
surface, something I ran into right away in those days. Anyway, if
the brake subsequently did not center, it had probably already been
misshapen by hammer and punch. At that point, lubricating and
cleaning has little effect. This was a self perpetuating routine.

Jobst Brandt

Quoth Jobst Brandt:

Unfortunately, bicycle repair shops, unaware of the cause, use
hammers and punches to bend the offending spring

I plied again:

That technique was appropriate for some older sidepulls, models
dating from the early '70s and older, and was the only way to
regulate the centering of the caliper.

Jobst:

I don't believe it was ever appropriate. The anchor bolt on those
brakes could be rotated and the springs lubricated just the same as
on later more expensive brakes.

Tom Ace took the words right out of my mouth:

I had a Raleigh 3-speed with steel-caliper sidepull brakes. The
front brake bolt had a cylindrical concave surface that mated with a
convex surface on the fork. The bolt's rotation was not adjustable.
If lubrication didn't fix the centering, the only other option I
knew of was to bend the spring.


I don't see why that was the only available method. How did you
initially poisition the brake caliper was with pads adjacent to the
rim? Why could the centerbolt not be rotated? I don't believe it was
riveted or welded in place.

The block on the centerbolt that had the groove for the middle of the
spring also had a concave groove on the other end.

The front one had a large radius to match the curvature of the front of
the fork crown.

The rear one had a smaller radius, transversely oriented groove that fit
snugly against the seatstay bridge.

These parts could not be rotated because of the curved interface to the
frame/fork.

Bending the spring was the _only_ way to adjust the centering.

Sheldon "http://sheldonbrown.com/english" Brown
+----------------------------------------+
| Yes Britain set the world ablaze, |
| In good King George's glorious days! |
| --W.S. Gilbert |
+----------------------------------------+
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

Jobst Brandt wrote:

Unfortunately, bicycle repair shops, unaware of the cause, use
hammers and punches to bend the offending spring

I replied:

That technique was appropriate for some older sidepulls, models
dating from the early '70s and older, and was the only way to
regulate the centering of the caliper.

Jobst:

I don't believe it was ever appropriate. The anchor bolt on those
brakes could be rotated and the springs lubricated just the same
as on later more expensive brakes.


(I believe Jobst was referring to the center bolt, not the anchor bolt.)


Please explain the difference between a center bolt and anchor bolt.
I am referring to the bolt about which the caliper arms rotate and
which attaches the spring boss against the fork crown or rear brake
bridge.

That is the "center bolt." In the industry, the term "anchor bolt"
generally refers to the bolt that secures the plain end of a brake or
shift cable to the caliper/cantilever or derailer.

See: http://sheldonbrown.com/gloss_an-z.html#anchor

The ones to which I refer are interchangeable with Mafac
centerpull and Campagnolo side pull brakes.

Those are not interchangeable.

Mafac and other centerpulls use a semi-standard M6 bolt, though usually
with a "low profile" head to improve arm clearance.

Campagnolo and most other single-pivot sidepulls use a double-ended
center bolt. It has a "block" in the middle with a groove that secures
the spring, and has male threads protruding from both ends. One of the
male threads attaches it to the frame/fork. The other male thread holds
the arms and associated washers in place, using an acorn nut and a jam nut.

Shimano and some other single-pivot brakes use a bolt with a head on one
end, but a dual thread. There's a 7 mm thread that screws into the
spring block and an associated jam nut, then it necks down to the
standard 6 mm size where it fits through the frame/fork.


Many older sidepulls, such as the typical Weinmann/Dia Compe units,
had no way to secure or rotate the spring mounting block. (This was
before Campagnolo thought of putting wrench flats on the spring
block, and before Weinmann thought of putting a 4 mm hex on the end
of the center bolt.)

Jobst:

I disagree, having adjusted them initially and centering them by
cleaning an lubrication. How else could one install a new brake if
its rotational position was not adjustable?


When you would tighten the mounting nut on these calipers, the
friction of the threads would often tend to rotate the center bolt
out of position, even if you started with it centered.


That takes a little skill to retain the spring in its position while
tightening the bolt. This is a common practice in machinery.

But the spring is a _spring_! If you hold the brake by one or both of
the caliper arms, the spring deflects due to torque on the center bolt,
and the spring block gets misaligned.

A hammer and punch used gently on the spring, close to the central
block could rotate the centerbolt back to the correct orientation.
This did not actually involve causing the spring to yield, it was
just the only way to apply torque to the spring mounting block in
the absence of wrench flats.

With a slightly looser center bolt this operation can be done without
impact and with suitable accuracy.

I prefer to get the center bolt tightened securely.

Sheldon "It Takes All Kinds" Brown
+----------------------------------------+
| All theory, dear friend, is grey, |
| but the golden tree of actual life |
| springs ever green. --Goethe |
+----------------------------------------+
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