Indexed-headset pictures

Carl Fogel said:

Here's a badly indexed headset race:

http://img84.imageshack.us/my.php?image=250headsetdimplesve2.jpg

Quoted message said:

Here's an enlargement of the dimples:

http://img84.imageshack.us/my.php?image=250gheadsetdimplesrb9.jpg

Quoted message said:

You may be able to enlarge either picture even further by clicking
on it.

Quoted message said:

I don't know whether the magnified picture of the dimples supports
erosion by fretting or impact by denting or both.

Quoted message said:

But some posters may enjoy interpreting details that may not have been
visible before on RBT.

Quoted message said:

The headset belonged to my faithful Schwinn LeTour '98. It spent about
two miles per day coasting down a 65 mph highway on 120 psi 700x26
tires at 30-50 mph, depending on the wind. Occasionally, it drifted
over the rumble strips and buzzed like a jackhammer. Its other daily
13 miles were at about 20 mph over somewhat cracked and bumpy
pavement. No impacts or accidents were severe enough to bend the fork
back.

Quoted message said:

After years of cruel neglect, my faithful Schwinn was retired and
replaced by its twin brother (easier than maintenance).

Quoted message said:

While fiddling with my camera this evening, I wondered about headset
indexing and remembered what a shameful example I had close at hand.

Quoted message said:

After putting the bike in a vise and disassembling things, I pounded
uselessly on the top of the stem for a while with a 4-lb hammer.

Quoted message said:

Then I gave up and took more drastic measures that resembled a
partial-birth abortion. By sawing through the exposed but hopelessly
frozen stem, I extracted the stem and the intact lower race without
pounding on it.

Unfortunately this is such a poor quality head bearing that the races
are deeply eroded between the ball home positions at the dimples.
You'll find that high quality (hard races and bearing balls) show
essentially no wear between dimples that are primarily in the fore and
aft quadrant of the races, these being the only places that fret
significantly when the fork articulates about the lower bearing. Tall
bicycle have enough space between lower and upper head bearing that
the steer tube flexes enough to cause dimples in the upper bearing.

Even then, I suspect you can see less wear and dimpling in the lateral
quadrants of the races. This is hard to assess from the picture
because it is not a view looking at the whole race on axis.

Damage in the pictures is so gross that it obscures what most people
with good equipment could expect to on old style cup and cone
bearings. Unfortunately, all examples I had seem to have been
scrapped when I changed to Shimano cartridge angular contact bearings.
I think I posted some pictures a few years ago though.

Jobst Brandt

Quoted message said:

Carl Fogel said:

Here's a badly indexed headset race:

http://img84.imageshack.us/my.php?image=250headsetdimplesve2.jpg

Quoted message said:

Here's an enlargement of the dimples:

http://img84.imageshack.us/my.php?image=250gheadsetdimplesrb9.jpg

Quoted message said:

You may be able to enlarge either picture even further by clicking
on it.

Quoted message said:

I don't know whether the magnified picture of the dimples supports
erosion by fretting or impact by denting or both.

Quoted message said:

But some posters may enjoy interpreting details that may not have been
visible before on RBT.

Quoted message said:

The headset belonged to my faithful Schwinn LeTour '98. It spent about
two miles per day coasting down a 65 mph highway on 120 psi 700x26
tires at 30-50 mph, depending on the wind. Occasionally, it drifted
over the rumble strips and buzzed like a jackhammer. Its other daily
13 miles were at about 20 mph over somewhat cracked and bumpy
pavement. No impacts or accidents were severe enough to bend the fork
back.

Quoted message said:

After years of cruel neglect, my faithful Schwinn was retired and
replaced by its twin brother (easier than maintenance).

Quoted message said:

While fiddling with my camera this evening, I wondered about headset
indexing and remembered what a shameful example I had close at hand.

Quoted message said:

After putting the bike in a vise and disassembling things, I pounded
uselessly on the top of the stem for a while with a 4-lb hammer.

Quoted message said:

Then I gave up and took more drastic measures that resembled a
partial-birth abortion. By sawing through the exposed but hopelessly
frozen stem, I extracted the stem and the intact lower race without
pounding on it.

Unfortunately this is such a poor quality head bearing that the races
are deeply eroded between the ball home positions at the dimples.
You'll find that high quality (hard races and bearing balls) show
essentially no wear between dimples that are primarily in the fore and
aft quadrant of the races, these being the only places that fret
significantly when the fork articulates about the lower bearing. Tall
bicycle have enough space between lower and upper head bearing that
the steer tube flexes enough to cause dimples in the upper bearing.

Even then, I suspect you can see less wear and dimpling in the lateral
quadrants of the races. This is hard to assess from the picture
because it is not a view looking at the whole race on axis.

Damage in the pictures is so gross that it obscures what most people
with good equipment could expect to on old style cup and cone
bearings. Unfortunately, all examples I had seem to have been
scrapped when I changed to Shimano cartridge angular contact bearings.
I think I posted some pictures a few years ago though.

Jobst Brandt

Dear Jobst,

Here are the original pictures again, showing three dimples in
considerable magnification:

http://img84.imageshack.us/my.php?image=250headsetdimplesve2.jpg

http://img84.imageshack.us/my.php?image=250gheadsetdimplesrb9.jpg

And here's the whole headset, showing all dimples:

http://img170.imageshack.us/my.php?image=253wholeheadset3za6.jpg

I haven't tried to identify any quadrants with less wear, much less
where the 3 dimples in the other picture ended up.

You can measure things with pixel-level precision with a program like
Mouse-o-Meter:

http://www.freedownloadscenter.com/Utilities/Mouse_and_Keyboard_Utilities/Mouse_o_meter.html

If anyone wants to describe the position of the dimples, I expect that
treating the headset as a clock face will be the easiest method.

Again, clicking on the picture may enlarge it.

Cheers,

Carl Fogel

Carl Fogel said:

Quoted message said:

Quoted message said:

Here's a badly indexed headset race:

http://img84.imageshack.us/my.php?image=250headsetdimplesve2.jpg

Quoted message said:

Quoted message said:

Quoted message said:

Here's an enlargement of the dimples:

http://img84.imageshack.us/my.php?image=250gheadsetdimplesrb9.jpg

Quoted message said:

Quoted message said:

Quoted message said:

You may be able to enlarge either picture even further by clicking
on it.

Quoted message said:

Quoted message said:

Quoted message said:

I don't know whether the magnified picture of the dimples supports
erosion by fretting or impact by denting or both.

Quoted message said:

Quoted message said:

Quoted message said:

But some posters may enjoy interpreting details that may not have
been visible before on RBT.

Quoted message said:

Quoted message said:

Quoted message said:

The headset belonged to my faithful Schwinn LeTour '98. It spent
about two miles per day coasting down a 65 mph highway on 120 psi
700x26 tires at 30-50 mph, depending on the wind. Occasionally, it
drifted over the rumble strips and buzzed like a jackhammer. Its
other daily 13 miles were at about 20 mph over somewhat cracked
and bumpy pavement. No impacts or accidents were severe enough to
bend the fork back.

Quoted message said:

Quoted message said:

Quoted message said:

After years of cruel neglect, my faithful Schwinn was retired and
replaced by its twin brother (easier than maintenance).

Quoted message said:

Quoted message said:

Quoted message said:

While fiddling with my camera this evening, I wondered about
headset indexing and remembered what a shameful example I had
close at hand.

Quoted message said:

Quoted message said:

Quoted message said:

After putting the bike in a vise and disassembling things, I
pounded uselessly on the top of the stem for a while with a 4-lb
hammer.

Quoted message said:

Quoted message said:

Quoted message said:

Then I gave up and took more drastic measures that resembled a
partial-birth abortion. By sawing through the exposed but
hopelessly frozen stem, I extracted the stem and the intact lower
race without pounding on it.

Quoted message said:

Quoted message said:

Unfortunately this is such a poor quality head bearing that the
races are deeply eroded between the ball home positions at the
dimples. You'll find that high quality (hard races and bearing
balls) show essentially no wear between dimples that are primarily
in the fore and aft quadrant of the races, these being the only
places that fret significantly when the fork articulates about the
lower bearing. Tall bicycle have enough space between lower and
upper head bearing that the steer tube flexes enough to cause
dimples in the upper bearing.

Quoted message said:

Quoted message said:

Even then, I suspect you can see less wear and dimpling in the
lateral quadrants of the races. This is hard to assess from the
picture because it is not a view looking at the whole race on axis.

Quoted message said:

Quoted message said:

Damage in the pictures is so gross that it obscures what most
people with good equipment could expect to on old style cup and
cone bearings. Unfortunately, all examples I had seem to have been
scrapped when I changed to Shimano cartridge angular contact
bearings. I think I posted some pictures a few years ago though.

Quoted message said:

Here are the original pictures again, showing three dimples in
considerable magnification:

http://img84.imageshack.us/my.php?image=250headsetdimplesve2.jpg

http://img84.imageshack.us/my.php?image=250gheadsetdimplesrb9.jpg

Quoted message said:

And here's the whole headset, showing all dimples:

http://img170.imageshack.us/my.php?image=253wholeheadset3za6.jpg

Quoted message said:

I haven't tried to identify any quadrants with less wear, much less
where the 3 dimples in the other picture ended up.

Quoted message said:

You can measure things with pixel-level precision with a program like
Mouse-o-Meter:

http://tinyurl.com/he9ad

Quoted message said:

If anyone wants to describe the position of the dimples, I expect
that treating the headset as a clock face will be the easiest
method.

Quoted message said:

Again, clicking on the picture may enlarge it.

As I said, the deepest dimples are generally centered about the front
of head set races when the bicycle is steering straight ahead.
Therefore, when photographing it you could put that position at the
top of the picture so that 12:00 is straight up and is also the front
of the bearing.

If this were road shock induced rather than fretting the dimples would
be uniformly distributed (assuming the head tube had square ends). As
should be obvious, the forward rim of the bearing bears most of the
wheel loads and wears accordingly. The cone opposite the cup shown
will have matching marks, leaving no doubt about in use orientation.

This is more evident on a better quality bearing and if the bicycle is
used with heavy braking will have dimples in the front AND rear
quadrant. My dimpled bearings had nearly symmetrical wear in that
manner, while the side quadrants showed almost no visible wear.

Jobst Brandt

In article <[email hidden]>,

Quoted message said:

Here's a badly indexed headset race:

http://img84.imageshack.us/my.php?image=250headsetdimplesve2.jpg

Here's an enlargement of the dimples:

http://img84.imageshack.us/my.php?image=250gheadsetdimplesrb9.jpg

<snip>

Quoted message said:

After putting the bike in a vise and disassembling things, I pounded
uselessly on the top of the stem for a while with a 4-lb hammer.

Beating on the stem would not dimple the lower headset cup under any
circumstances. It'd dimple the upper headset cup and/or cone, if it was
going to dimple anything.

The photos you provide look like an excellent example of bearing wear.
The dimples are not round, as they would be if they were caused by
impact, and the wear trails between the dimples are nearly as bad as the
dimples!

I don't recall from previous threads whether the balls or the cups/cones
tend to have greater surface hardness. Anyone know?

Quoted message said:

Here's a badly indexed headset race:

http://img84.imageshack.us/my.php?image=250headsetdimplesve2.jpg

Here's an enlargement of the dimples:

http://img84.imageshack.us/my.php?image=250gheadsetdimplesrb9.jpg

You may be able to enlarge either picture even further by clicking on
it.

I don't know whether the magnified picture of the dimples supports
erosion by fretting or impact by denting or both.

But some posters may enjoy interpreting details that may not have been
visible before on RBT.

The headset belonged to my faithful Schwinn LeTour '98. It spent about
two miles per day coasting down a 65 mph highway on 120 psi 700x26
tires at 30-50 mph, depending on the wind. Occasionally, it drifted
over the rumble strips and buzzed like a jackhammer. Its other daily
13 miles were at about 20 mph over somewhat cracked and bumpy
pavement. No impacts or accidents were severe enough to bend the fork
back.

After years of cruel neglect, my faithful Schwinn was retired and
replaced by its twin brother (easier than maintenance).

While fiddling with my camera this evening, I wondered about headset
indexing and remembered what a shameful example I had close at hand.

After putting the bike in a vise and disassembling things, I pounded
uselessly on the top of the stem for a while with a 4-lb hammer.

Then I gave up and took more drastic measures that resembled a
partial-birth abortion. By sawing through the exposed but hopelessly
frozen stem, I extracted the stem and the intact lower race without
pounding on it.

Cheers,

Carl Fogel

Was the headset indexed when you retired it?

Or did it develop the problem while in storage - i.e. hung by front wheel?

dcg

Quoted message said:

Carl Fogel said:

Quoted message said:

> Here's a badly indexed headset race:

http://img84.imageshack.us/my.php?image=250headsetdimplesve2.jpg

Quoted message said:

Quoted message said:

> Here's an enlargement of the dimples:

http://img84.imageshack.us/my.php?image=250gheadsetdimplesrb9.jpg

Quoted message said:

Quoted message said:

> You may be able to enlarge either picture even further by clicking
> on it.

Quoted message said:

Quoted message said:

> I don't know whether the magnified picture of the dimples supports
> erosion by fretting or impact by denting or both.

Quoted message said:

Quoted message said:

> But some posters may enjoy interpreting details that may not have
> been visible before on RBT.

Quoted message said:

Quoted message said:

> The headset belonged to my faithful Schwinn LeTour '98. It spent
> about two miles per day coasting down a 65 mph highway on 120 psi
> 700x26 tires at 30-50 mph, depending on the wind. Occasionally, it
> drifted over the rumble strips and buzzed like a jackhammer. Its
> other daily 13 miles were at about 20 mph over somewhat cracked
> and bumpy pavement. No impacts or accidents were severe enough to
> bend the fork back.

Quoted message said:

Quoted message said:

> After years of cruel neglect, my faithful Schwinn was retired and
> replaced by its twin brother (easier than maintenance).

Quoted message said:

Quoted message said:

> While fiddling with my camera this evening, I wondered about
> headset indexing and remembered what a shameful example I had
> close at hand.

Quoted message said:

Quoted message said:

> After putting the bike in a vise and disassembling things, I
> pounded uselessly on the top of the stem for a while with a 4-lb
> hammer.

Quoted message said:

Quoted message said:

> Then I gave up and took more drastic measures that resembled a
> partial-birth abortion. By sawing through the exposed but
> hopelessly frozen stem, I extracted the stem and the intact lower
> race without pounding on it.

Quoted message said:

Quoted message said:

Unfortunately this is such a poor quality head bearing that the
races are deeply eroded between the ball home positions at the
dimples. You'll find that high quality (hard races and bearing
balls) show essentially no wear between dimples that are primarily
in the fore and aft quadrant of the races, these being the only
places that fret significantly when the fork articulates about the
lower bearing. Tall bicycle have enough space between lower and
upper head bearing that the steer tube flexes enough to cause
dimples in the upper bearing.

Quoted message said:

Quoted message said:

Even then, I suspect you can see less wear and dimpling in the
lateral quadrants of the races. This is hard to assess from the
picture because it is not a view looking at the whole race on axis.

Quoted message said:

Quoted message said:

Damage in the pictures is so gross that it obscures what most
people with good equipment could expect to on old style cup and
cone bearings. Unfortunately, all examples I had seem to have been
scrapped when I changed to Shimano cartridge angular contact
bearings. I think I posted some pictures a few years ago though.

Quoted message said:

Here are the original pictures again, showing three dimples in
considerable magnification:

http://img84.imageshack.us/my.php?image=250headsetdimplesve2.jpg

http://img84.imageshack.us/my.php?image=250gheadsetdimplesrb9.jpg

Quoted message said:

And here's the whole headset, showing all dimples:

http://img170.imageshack.us/my.php?image=253wholeheadset3za6.jpg

Quoted message said:

I haven't tried to identify any quadrants with less wear, much less
where the 3 dimples in the other picture ended up.

Quoted message said:

You can measure things with pixel-level precision with a program like
Mouse-o-Meter:

http://tinyurl.com/he9ad

Quoted message said:

If anyone wants to describe the position of the dimples, I expect
that treating the headset as a clock face will be the easiest
method.

Quoted message said:

Again, clicking on the picture may enlarge it.

As I said, the deepest dimples are generally centered about the front
of head set races when the bicycle is steering straight ahead.
Therefore, when photographing it you could put that position at the
top of the picture so that 12:00 is straight up and is also the front
of the bearing.

If this were road shock induced rather than fretting the dimples would
be uniformly distributed (assuming the head tube had square ends). As
should be obvious, the forward rim of the bearing bears most of the
wheel loads and wears accordingly. The cone opposite the cup shown
will have matching marks, leaving no doubt about in use orientation.

This is more evident on a better quality bearing and if the bicycle is
used with heavy braking will have dimples in the front AND rear
quadrant. My dimpled bearings had nearly symmetrical wear in that
manner, while the side quadrants showed almost no visible wear.

Jobst Brandt

Dear Jobst,

So where is the "front" in the picture of the whole bearing, which was
photographed at random?

Cheers,

Carl Fogel

Tim McNamara said:

In article <[email hidden]>,

Quoted message said:

Here's a badly indexed headset race:

http://img84.imageshack.us/my.php?image=250headsetdimplesve2.jpg

Here's an enlargement of the dimples:

http://img84.imageshack.us/my.php?image=250gheadsetdimplesrb9.jpg

<snip>

Quoted message said:

After putting the bike in a vise and disassembling things, I pounded
uselessly on the top of the stem for a while with a 4-lb hammer.

Beating on the stem would not dimple the lower headset cup under any
circumstances. It'd dimple the upper headset cup and/or cone, if it was
going to dimple anything.

The photos you provide look like an excellent example of bearing wear.
The dimples are not round, as they would be if they were caused by
impact, and the wear trails between the dimples are nearly as bad as the
dimples!

I don't recall from previous threads whether the balls or the cups/cones
tend to have greater surface hardness. Anyone know?

Dear Tim,

Bearings dent some cups, at least when banged with a hammer, squashed
in a vise, or buzzed for a few seconds with an air-hammer arrangement.

Jim Beam dents, working far too hard with a hammer:

http://groups.google.com/group/rec.bicycles.tech/msg/4aaae9db81242823

My dents, using a vise:

http://groups.google.com/group/rec.bicycles.tech/msg/834d1eb4917c2b22

My dents, using an air hammer with straw mat and wood support:

http://groups.google.com/group/rec.bicycles.tech/msg/977a3d0074b5ab4f

Cheers,

Carl Fogel

<[email hidden]> wrote in message
news:[email hidden]...

Quoted message said:

Here's a badly indexed headset race:

http://img84.imageshack.us/my.php?image=250headsetdimplesve2.jpg

Here's an enlargement of the dimples:

http://img84.imageshack.us/my.php?image=250gheadsetdimplesrb9.jpg

<snip>

Quoted message said:

Cheers,

Carl Fogel

This bearing cup appears to be case hardened rather than through
hardened. Case hardening was/is used on a lot of low quality headsets

Case hardening creates a surface layer up to .050" (1.3mm) deep that has
a hardness of around 60Rc. The advantage of case hardening is that it
provides a wear resistant surface while maintaining a tough crack
resistant steel substrate. It's frequently used for gears and shafts
where there is a lot of sliding contact.

Bicycle headsets are thrust bearings that are not subjected to much
rotational or sliding friction. Forces are concentrated in very small
areas at the tops and bottoms of a few ball bearings providing an
excellent opportunity for brinelling.

Angular race needle bearings are far better suited for this kind of
load.

This cup appears to have failed due to initial brinelling breaking
through the hardened layer then the bulk of the damage could be
attributed to "false brinelling" or caused by fretting or spalling.

http://en.wikipedia.org/wiki/False_brinelling

http://www.emersonbearing.com/fail_falsebrinell.htm

http://www.lubrizol.com/GreaseReference/glossary.asp#false

http://www.lubrizol.com/GreaseReference/glossary.asp#fretting

I managed a pro bike shop in the mid 1970's. We got tired of having to
frequently replace the lower bearing races on Campy headsets due to
"brinelling". We took over 20 new and used pro headsets to a friend's
machine shop and checked them with a Rockwell Hardness Tester. This was
before top quality Japanese components were readily available.

Campy NR & SR headsets were some of the worst coming in at <48Rc
hardness. Zeus and Stronglight S5 Super Competition headsets were all

Quoted message said:

60Rc and we saw least brinelling problems with those brands. Even with

this evidence, it was a hard sell trying to get "all Campy" bike owners
to change headset brands. Slaves to fashion - something akin to wearing
stylish but painfully uncomfortable Italian shoes!

Chas.

<[email hidden]> wrote in message
news:[email hidden]...

Quoted message said:

Quoted message said:

Carl Fogel said:

Here's a badly indexed headset race:

http://img84.imageshack.us/my.php?image=250headsetdimplesve2.jpg

Quoted message said:

Here's an enlargement of the dimples:

http://img84.imageshack.us/my.php?image=250gheadsetdimplesrb9.jpg

Quoted message said:

You may be able to enlarge either picture even further by clicking
on it.

Quoted message said:

I don't know whether the magnified picture of the dimples supports
erosion by fretting or impact by denting or both.

<snip>

The indentations appear to be equally, widely spaced indicating that the
ball bearings were held in place with a bearing cage.

I've almost always assembled headsets with loose ball bearings which in
most cases can increase the number of balls thus spreading out the load
forces.

Chas.

Tim McNamara said:

In article <[email hidden]>,

Quoted message said:

Here's a badly indexed headset race:

http://img84.imageshack.us/my.php?image=250headsetdimplesve2.jpg

Here's an enlargement of the dimples:

http://img84.imageshack.us/my.php?image=250gheadsetdimplesrb9.jpg

<snip>

Quoted message said:

After putting the bike in a vise and disassembling things, I pounded
uselessly on the top of the stem for a while with a 4-lb hammer.

Beating on the stem would not dimple the lower headset cup under any
circumstances.

not true. i've brinelled headsets in crashes and with a hammer.

Quoted message said:

It'd dimple the upper headset cup and/or cone, if it was
going to dimple anything.

The photos you provide look like an excellent example of bearing wear.
The dimples are not round, as they would be if they were caused by
impact, and the wear trails between the dimples are nearly as bad as the
dimples!

I don't recall from previous threads whether the balls or the cups/cones
tend to have greater surface hardness. Anyone know?

should be about the same, but on cheap stuff, expect the bearing balls
to be a little harder. it's a simple quality issue.

Quoted message said:

Here's a badly indexed headset race:

http://img84.imageshack.us/my.php?image=250headsetdimplesve2.jpg

Here's an enlargement of the dimples:

http://img84.imageshack.us/my.php?image=250gheadsetdimplesrb9.jpg

You may be able to enlarge either picture even further by clicking on
it.

I don't know whether the magnified picture of the dimples supports
erosion by fretting or impact by denting or both.

But some posters may enjoy interpreting details that may not have been
visible before on RBT.

The headset belonged to my faithful Schwinn LeTour '98. It spent about
two miles per day coasting down a 65 mph highway on 120 psi 700x26
tires at 30-50 mph, depending on the wind. Occasionally, it drifted
over the rumble strips and buzzed like a jackhammer. Its other daily
13 miles were at about 20 mph over somewhat cracked and bumpy
pavement. No impacts or accidents were severe enough to bend the fork
back.

After years of cruel neglect, my faithful Schwinn was retired and
replaced by its twin brother (easier than maintenance).

While fiddling with my camera this evening, I wondered about headset
indexing and remembered what a shameful example I had close at hand.

After putting the bike in a vise and disassembling things, I pounded
uselessly on the top of the stem for a while with a 4-lb hammer.

Then I gave up and took more drastic measures that resembled a
partial-birth abortion. By sawing through the exposed but hopelessly
frozen stem, I extracted the stem and the intact lower race without
pounding on it.

Cheers,

Carl Fogel

you are much emboldened by the world of digital photography carl!

i'm waiting for someone to show similar pics of sealed cartridge headset
bearings showing the same features. i suspect i may be waiting a while
since the seals keep the lube in and the grit out thus removing two
major failure factors.

* * Chas said:

<[email hidden]> wrote in message
news:[email hidden]...

Quoted message said:

Here's a badly indexed headset race:

http://img84.imageshack.us/my.php?image=250headsetdimplesve2.jpg

Here's an enlargement of the dimples:

http://img84.imageshack.us/my.php?image=250gheadsetdimplesrb9.jpg

<snip>

Quoted message said:

Cheers,

Carl Fogel

This bearing cup appears to be case hardened rather than through
hardened. Case hardening was/is used on a lot of low quality headsets

Case hardening creates a surface layer up to .050" (1.3mm) deep that has
a hardness of around 60Rc. The advantage of case hardening is that it
provides a wear resistant surface while maintaining a tough crack
resistant steel substrate. It's frequently used for gears and shafts
where there is a lot of sliding contact.

Bicycle headsets are thrust bearings that are not subjected to much
rotational or sliding friction. Forces are concentrated in very small
areas at the tops and bottoms of a few ball bearings providing an
excellent opportunity for brinelling.

Angular race needle bearings are far better suited for this kind of
load.

This cup appears to have failed due to initial brinelling breaking
through the hardened layer then the bulk of the damage could be
attributed to "false brinelling" or caused by fretting or spalling.

absolutely.

Quoted message said:

http://en.wikipedia.org/wiki/False_brinelling

http://www.emersonbearing.com/fail_falsebrinell.htm

http://www.lubrizol.com/GreaseReference/glossary.asp#false

http://www.lubrizol.com/GreaseReference/glossary.asp#fretting

I managed a pro bike shop in the mid 1970's. We got tired of having to
frequently replace the lower bearing races on Campy headsets due to
"brinelling". We took over 20 new and used pro headsets to a friend's
machine shop and checked them with a Rockwell Hardness Tester. This was
before top quality Japanese components were readily available.

Campy NR & SR headsets were some of the worst coming in at <48Rc
hardness. Zeus and Stronglight S5 Super Competition headsets were all

Quoted message said:

60Rc and we saw least brinelling problems with those brands. Even with

this evidence, it was a hard sell trying to get "all Campy" bike owners
to change headset brands. Slaves to fashion - something akin to wearing
stylish but painfully uncomfortable Italian shoes!

Chas.

awesome post - testing, comparison - excellent.

Carl Fogel said:

Quoted message said:

Quoted message said:

>> Here's a badly indexed headset race:

http://img84.imageshack.us/my.php?image=250headsetdimplesve2.jpg

Quoted message said:

Quoted message said:

Quoted message said:

>> Here's an enlargement of the dimples:

http://img84.imageshack.us/my.php?image=250gheadsetdimplesrb9.jpg

Quoted message said:

Quoted message said:

Quoted message said:

>> You may be able to enlarge either picture even further by clicking
>> on it.

Quoted message said:

Quoted message said:

Quoted message said:

>> I don't know whether the magnified picture of the dimples supports
>> erosion by fretting or impact by denting or both.

Quoted message said:

Quoted message said:

Quoted message said:

>> But some posters may enjoy interpreting details that may not have
>> been visible before on RBT.

Quoted message said:

Quoted message said:

Quoted message said:

>> The headset belonged to my faithful Schwinn LeTour '98. It spent
>> about two miles per day coasting down a 65 mph highway on 120 psi
>> 700x26 tires at 30-50 mph, depending on the wind. Occasionally, it
>> drifted over the rumble strips and buzzed like a jackhammer. Its
>> other daily 13 miles were at about 20 mph over somewhat cracked
>> and bumpy pavement. No impacts or accidents were severe enough to
>> bend the fork back.

Quoted message said:

Quoted message said:

Quoted message said:

>> After years of cruel neglect, my faithful Schwinn was retired and
>> replaced by its twin brother (easier than maintenance).

Quoted message said:

Quoted message said:

Quoted message said:

>> While fiddling with my camera this evening, I wondered about
>> headset indexing and remembered what a shameful example I had
>> close at hand.

Quoted message said:

Quoted message said:

Quoted message said:

>> After putting the bike in a vise and disassembling things, I
>> pounded uselessly on the top of the stem for a while with a 4-lb
>> hammer.

Quoted message said:

Quoted message said:

Quoted message said:

>> Then I gave up and took more drastic measures that resembled a
>> partial-birth abortion. By sawing through the exposed but
>> hopelessly frozen stem, I extracted the stem and the intact lower
>> race without pounding on it.

Quoted message said:

Quoted message said:

Quoted message said:

> Unfortunately this is such a poor quality head bearing that the
> races are deeply eroded between the ball home positions at the
> dimples. You'll find that high quality (hard races and bearing
> balls) show essentially no wear between dimples that are primarily
> in the fore and aft quadrant of the races, these being the only
> places that fret significantly when the fork articulates about the
> lower bearing. Tall bicycle have enough space between lower and
> upper head bearing that the steer tube flexes enough to cause
> dimples in the upper bearing.

Quoted message said:

Quoted message said:

Quoted message said:

> Even then, I suspect you can see less wear and dimpling in the
> lateral quadrants of the races. This is hard to assess from the
> picture because it is not a view looking at the whole race on axis.

Quoted message said:

Quoted message said:

Quoted message said:

> Damage in the pictures is so gross that it obscures what most
> people with good equipment could expect to on old style cup and
> cone bearings. Unfortunately, all examples I had seem to have been
> scrapped when I changed to Shimano cartridge angular contact
> bearings. I think I posted some pictures a few years ago though.

Quoted message said:

Quoted message said:

Quoted message said:

Here are the original pictures again, showing three dimples in
considerable magnification:

http://img84.imageshack.us/my.php?image=250headsetdimplesve2.jpg

http://img84.imageshack.us/my.php?image=250gheadsetdimplesrb9.jpg

Quoted message said:

Quoted message said:

Quoted message said:

And here's the whole headset, showing all dimples:

http://img170.imageshack.us/my.php?image=253wholeheadset3za6.jpg

Quoted message said:

Quoted message said:

Quoted message said:

I haven't tried to identify any quadrants with less wear, much less
where the 3 dimples in the other picture ended up.

Quoted message said:

Quoted message said:

Quoted message said:

You can measure things with pixel-level precision with a program like
Mouse-o-Meter:

http://tinyurl.com/he9ad

Quoted message said:

Quoted message said:

Quoted message said:

If anyone wants to describe the position of the dimples, I expect
that treating the headset as a clock face will be the easiest
method.

Quoted message said:

Quoted message said:

Quoted message said:

Again, clicking on the picture may enlarge it.

Quoted message said:

Quoted message said:

As I said, the deepest dimples are generally centered about the front
of head set races when the bicycle is steering straight ahead.
Therefore, when photographing it you could put that position at the
top of the picture so that 12:00 is straight up and is also the front
of the bearing.

Quoted message said:

Quoted message said:

If this were road shock induced rather than fretting the dimples would
be uniformly distributed (assuming the head tube had square ends). As
should be obvious, the forward rim of the bearing bears most of the
wheel loads and wears accordingly. The cone opposite the cup shown
will have matching marks, leaving no doubt about in use orientation.

Quoted message said:

Quoted message said:

This is more evident on a better quality bearing and if the bicycle is
used with heavy braking will have dimples in the front AND rear
quadrant. My dimpled bearings had nearly symmetrical wear in that
manner, while the side quadrants showed almost no visible wear.

Quoted message said:

So where is the "front" in the picture of the whole bearing, which was
photographed at random?

As I said, there where the dimples are most pronounced. With lighting
is not ideal for seeing them, it seems to me that straight ahead is at
11:30. You can see that better than I, hand held with good lighting.

Jobst Brandt

jim beam said:

Tim McNamara said:

In article <[email hidden]>,

Quoted message said:

Here's a badly indexed headset race:

http://img84.imageshack.us/my.php?image=250headsetdimplesve2.jpg

Here's an enlargement of the dimples:

http://img84.imageshack.us/my.php?image=250gheadsetdimplesrb9.jpg

<snip>

Quoted message said:

After putting the bike in a vise and disassembling things, I pounded
uselessly on the top of the stem for a while with a 4-lb hammer.

Beating on the stem would not dimple the lower headset cup under any
circumstances.

not true. i've brinelled headsets in crashes and with a hammer.

Quoted message said:

It'd dimple the upper headset cup and/or cone, if it was
going to dimple anything.

[snip]

Dear Jim,

I think that Tim's point is that whomping down on the the stem of a
bicycle frame held off the floor in vise shouldn't affect the lower
headset cup--the impact should force the bearing assembly apart, not
together.

The situation in the vise, where nothing supports the fork except the
corroded stem, leaving no pressure on the lower bearings, is different
from the normal situation, where the tire pressing on the ground can
compress the bearings.

Cheers,

Carl Fogel

jim beam said:

Quoted message said:

Here's a badly indexed headset race:

http://img84.imageshack.us/my.php?image=250headsetdimplesve2.jpg

Here's an enlargement of the dimples:

http://img84.imageshack.us/my.php?image=250gheadsetdimplesrb9.jpg

You may be able to enlarge either picture even further by clicking on
it.

I don't know whether the magnified picture of the dimples supports
erosion by fretting or impact by denting or both.

But some posters may enjoy interpreting details that may not have been
visible before on RBT.

The headset belonged to my faithful Schwinn LeTour '98. It spent about
two miles per day coasting down a 65 mph highway on 120 psi 700x26
tires at 30-50 mph, depending on the wind. Occasionally, it drifted
over the rumble strips and buzzed like a jackhammer. Its other daily
13 miles were at about 20 mph over somewhat cracked and bumpy
pavement. No impacts or accidents were severe enough to bend the fork
back.

After years of cruel neglect, my faithful Schwinn was retired and
replaced by its twin brother (easier than maintenance).

While fiddling with my camera this evening, I wondered about headset
indexing and remembered what a shameful example I had close at hand.

After putting the bike in a vise and disassembling things, I pounded
uselessly on the top of the stem for a while with a 4-lb hammer.

Then I gave up and took more drastic measures that resembled a
partial-birth abortion. By sawing through the exposed but hopelessly
frozen stem, I extracted the stem and the intact lower race without
pounding on it.

Cheers,

Carl Fogel

you are much emboldened by the world of digital photography carl!

i'm waiting for someone to show similar pics of sealed cartridge headset
bearings showing the same features. i suspect i may be waiting a while
since the seals keep the lube in and the grit out thus removing two
major failure factors.

Dear Jim,

There's something enchanting about scrolling around a 20" high image
of a headset bearing.

It gives me the hope that someone can point to the details.

I was tempted to experiment with the editor and see if I could sneak a
faint smiley face into one of the dimples, but my courage failed me.

Still, it's amazing what a modern $140 camera and computer can do,
even in the hands of the most naive amateur.

When I went to make a movie of the Park tension gauge stubbornly
deflating in synch with the tire, I had a vague notion that the camera
could only capture 3 seconds of motion and didn't realize that it
would have an audio track. It turned out that the camera captured the
sound of the air hissing out of the tire and would have happily
continued recording for twenty minutes.

Cheers,

Carl Fogel

Charles who? said:

Quoted message said:

Here's a badly indexed headset race:

http://img84.imageshack.us/my.php?image=250headsetdimplesve2.jpg

Quoted message said:

Quoted message said:

Here's an enlargement of the dimples:

http://img84.imageshack.us/my.php?image=250gheadsetdimplesrb9.jpg

Quoted message said:

This bearing cup appears to be case hardened rather than through
hardened. Case hardening was/is used on a lot of low quality
headsets

Quoted message said:

Case hardening creates a surface layer up to .050" (1.3mm) deep that
has a hardness of around 60Rc. The advantage of case hardening is
that it provides a wear resistant surface while maintaining a tough
crack resistant steel substrate. It's frequently used for gears and
shafts where there is a lot of sliding contact.

Quoted message said:

Bicycle headsets are thrust bearings that are not subjected to much
rotational or sliding friction. Forces are concentrated in very
small areas at the tops and bottoms of a few ball bearings providing
an excellent opportunity for Brinelling.

Quoted message said:

Angular race needle bearings are far better suited for this kind of
load.

I have not found that to be true, having a needle bearing headset that
dimpled sooner than any 20 ball Campagnolo bearing. The problem is
fretting and that can't be solved by more contact such a s a roller
instead of a ball. These bearings, both roller and ball bearings work
best for rotating loads that keep a reasonable oil film between
rolling elements and their races.

A head bearing does not rotate sufficiently on a road bicycle when
costing down long descents where there are essentially no steering
motions while the fork vibrates fore and aft. It is the fore and aft
motion that causes fretting damage. Preloaded angular contact
cartridge bearings as used by Shimano, supported on a spherical plain
bearing that can tilt fore and aft, solve this problem.

Quoted message said:

This cup appears to have failed due to initial Brinelling breaking
through the hardened layer then the bulk of the damage could be
attributed to "false Brinelling" or caused by fretting or spalling.

Make up your mind, was it fretting or spalling. They are definitely
not similar.

http://en.wikipedia.org/wiki/False_brinelling

This article beats around the bush and does not understand fretting
damage to rolling element bearings or in press fits, where it occurs
between shafts and sleeves mounted on them.

I think that is better explained in the FAQ item:

http://www.sheldonbrown.com/brandt/fretting.html

http://www.emersonbearing.com/fail_falsebrinell.htm

http://www.lubrizol.com/GreaseReference/glossary.asp#false

http://www.lubrizol.com/GreaseReference/glossary.asp#fretting

Quoted message said:

I managed a pro bike shop in the mid 1970's. We got tired of having
to frequently replace the lower bearing races on Campy headsets due
to "Brinelling". We took over 20 new and used pro headsets to a
friend's machine shop and checked them with a Rockwell Hardness
Tester. This was before top quality Japanese components were
readily available.

Quoted message said:

Campy NR & SR headsets were some of the worst coming in at <48Rc
hardness. Zeus and Stronglight S5 Super Competition headsets were
all >60Rc and we saw least Brinelling problems with those brands.
Even with this evidence, it was a hard sell trying to get "all
Campy" bike owners to change headset brands. Slaves to fashion -
something akin to wearing stylish but painfully uncomfortable
Italian shoes!

BMW suffers from fretting the most among cars that I hear passing as I
ride. I was first exposed to this problem by an article from GM
Laboratories that investigate this phenomenon in the days (1950's)
when their cars were shipped farther and more securely restrained in
their RR cars than they are today. The failure of pinion gear
bearings in differentials, the one with the highest speed when
running, were fretting themselves to death in shipment. This damage
is audible as a rushing (hissing) sound coming from the rear end.

Since then I observed sport car transmissions with special ratios
gears pressed onto the main and countershafts that all had rouge in
their bores or bearing sleeves and on the shafts onto which they had
tight press fits.

Obviously a shaft under torque twists and that twist is not
transmitted ideally to the part pressed onto it. My first
introduction to this was before I understood metal elasticity as a
child watching failed Ford taper-cone rear axles being replaced. They
where they were pressed into the brake drum and retained with a large
castle nut. Later I saw that VW spline rear axles were always full of
rouge and chirped when used in reverse forcefully.

Jobst Brandt

Charles who? said:

Quoted message said:

Quoted message said:

> Here's a badly indexed headset race:

http://img84.imageshack.us/my.php?image=250headsetdimplesve2.jpg

Quoted message said:

Quoted message said:

Quoted message said:

> Here's an enlargement of the dimples:

http://img84.imageshack.us/my.php?image=250gheadsetdimplesrb9.jpg

Quoted message said:

Quoted message said:

Quoted message said:

> You may be able to enlarge either picture even further by
> clicking on it.

Quoted message said:

Quoted message said:

Quoted message said:

> I don't know whether the magnified picture of the dimples
> supports erosion by fretting or impact by denting or both.

Quoted message said:

<snip>

Quoted message said:

The indentations appear to be equally, widely spaced indicating that
the ball bearings were held in place with a bearing cage.

Quoted message said:

I've almost always assembled headsets with loose ball bearings which
in most cases can increase the number of balls thus spreading out
the load forces.

How many more? Not to worry, they make their own dimples and home in
on them, especially the one in the front and rear. Fretting is the
problem rather than load distribution. It doesn't take a large load
for this to occur, so more balls is not a solution. You'll note that
light weight riders also had these problems.

Jobst Brandt

[snip]

Emails have complained about the free image hosting site bogging down,
so here are the three original pictures at that site, plus duplicates
at another site (two have border material removed to squeeze under a
1mb upload limit).

The three originals:

Here's a badly indexed headset race:
http://img84.imageshack.us/my.php?image=250headsetdimplesve2.jpg

Here's an enlargement of the dimples:
http://img84.imageshack.us/my.php?image=250gheadsetdimplesrb9.jpg

Here's the headset dead-on:
http://img170.imageshack.us/my.php?image=253wholeheadset3za6.jpg

And here they are at another site:

Small angled view of most of the headset, trimmed to squeeze under a
1mb upload limit:
http://server5.theimagehosting.com/image.php?img=250headset%20dimplestrim.jpg

The three dimples from that picture, enlarged:
http://server5.theimagehosting.com/image.php?img=250gheadsetdimples.jpg

The headset dead-on, trimmed to squeeze under a 1mb upload limit:
http://server5.theimagehosting.com/image.php?img=253wholeheadset3trim.jpg

CF

Quoted message said:

Charles who? said:

Quoted message said:

Here's a badly indexed headset race:

http://img84.imageshack.us/my.php?image=250headsetdimplesve2.jpg

Quoted message said:

Quoted message said:

Here's an enlargement of the dimples:

http://img84.imageshack.us/my.php?image=250gheadsetdimplesrb9.jpg

Quoted message said:

This bearing cup appears to be case hardened rather than through
hardened. Case hardening was/is used on a lot of low quality
headsets

Quoted message said:

Case hardening creates a surface layer up to .050" (1.3mm) deep that
has a hardness of around 60Rc. The advantage of case hardening is
that it provides a wear resistant surface while maintaining a tough
crack resistant steel substrate. It's frequently used for gears and
shafts where there is a lot of sliding contact.

Quoted message said:

Bicycle headsets are thrust bearings that are not subjected to much
rotational or sliding friction. Forces are concentrated in very
small areas at the tops and bottoms of a few ball bearings providing
an excellent opportunity for Brinelling.

Quoted message said:

Angular race needle bearings are far better suited for this kind of
load.

I have not found that to be true, having a needle bearing headset that
dimpled sooner than any 20 ball Campagnolo bearing. The problem is
fretting and that can't be solved by more contact such a s a roller
instead of a ball. These bearings, both roller and ball bearings work
best for rotating loads that keep a reasonable oil film between
rolling elements and their races.

A head bearing does not rotate sufficiently on a road bicycle when
costing down long descents where there are essentially no steering
motions while the fork vibrates fore and aft. It is the fore and aft
motion that causes fretting damage. Preloaded angular contact
cartridge bearings as used by Shimano, supported on a spherical plain
bearing that can tilt fore and aft, solve this problem.

Quoted message said:

This cup appears to have failed due to initial Brinelling breaking
through the hardened layer then the bulk of the damage could be
attributed to "false Brinelling" or caused by fretting or spalling.

Make up your mind, was it fretting or spalling. They are definitely
not similar.

http://en.wikipedia.org/wiki/False_brinelling

This article beats around the bush and does not understand fretting
damage to rolling element bearings or in press fits, where it occurs
between shafts and sleeves mounted on them.

jobst, you're either sadly confused or deliberately trying to mislead.
fretting between the race and its housing is ABSOLUTELY NOT any damned
thing to do with the false or true brinelling evidenced between a
bearing ball and the race.

Quoted message said:

I think that is better explained in the FAQ item:

http://www.sheldonbrown.com/brandt/fretting.html

your faq's are /full/ of errors on a multitude of subjects. your
suppositions about bearings and their failure modes are prime examples.
remember the time you tried to b.s. about rolling element bearing
spalling and cited a journal bearing as an example?

Quoted message said:

http://www.emersonbearing.com/fail_falsebrinell.htm

http://www.lubrizol.com/GreaseReference/glossary.asp#false

http://www.lubrizol.com/GreaseReference/glossary.asp#fretting

so what? both failure modes exist, but it's you that seems to have the
problem differentiating between them.

Quoted message said:

Quoted message said:

I managed a pro bike shop in the mid 1970's. We got tired of having
to frequently replace the lower bearing races on Campy headsets due
to "Brinelling". We took over 20 new and used pro headsets to a
friend's machine shop and checked them with a Rockwell Hardness
Tester. This was before top quality Japanese components were
readily available.

Quoted message said:

Campy NR & SR headsets were some of the worst coming in at <48Rc
hardness. Zeus and Stronglight S5 Super Competition headsets were
all >60Rc and we saw least Brinelling problems with those brands.
Even with this evidence, it was a hard sell trying to get "all
Campy" bike owners to change headset brands. Slaves to fashion -
something akin to wearing stylish but painfully uncomfortable
Italian shoes!

BMW suffers from fretting the most among cars that I hear passing as I
ride. I was first exposed to this problem by an article from GM
Laboratories that investigate this phenomenon in the days (1950's)
when their cars were shipped farther and more securely restrained in
their RR cars than they are today. The failure of pinion gear
bearings in differentials, the one with the highest speed when
running, were fretting themselves to death in shipment. This damage
is audible as a rushing (hissing) sound coming from the rear end.

rubbish. false brinelling was a problem in the 50's due to [censored] steel
and inferior lubricants. vehicles are shipped in exactly the same way
all over the globe half a century later, and there are no more bearing
failures. it doesn't say much for your credibility if you can't be
bothered to update your information. but then again, you wouldn't want
to have facts get in the way of a good criticism would you - your other
misplaced bleatings about anodized rims spring to mind.

Quoted message said:

Since then I observed sport car transmissions with special ratios
gears pressed onto the main and countershafts that all had rouge in
their bores or bearing sleeves and on the shafts onto which they had
tight press fits.

those are not bearings!!! sheesh. just because fretting occurs doesn't
mean it's responsible for the ills of the world, especially not your
apparent confusion between the noise made by a bad bearing and a car's
tire scrub.

Quoted message said:

Obviously a shaft under torque twists and that twist is not
transmitted ideally to the part pressed onto it. My first
introduction to this was before I understood metal elasticity as a
child watching failed Ford taper-cone rear axles being replaced. They
where they were pressed into the brake drum and retained with a large
castle nut. Later I saw that VW spline rear axles were always full of
rouge and chirped when used in reverse forcefully.

so freakin' what? qr shafts fret inside their hollow axles too. it
makes no difference to their operation or longevity.

this post is typical jobstian nonsense. a small nugget of trivial truth
is the hook, but you follow it with supposition and error because you
can't be bothered to get current information.

and let's not overlook the fact that you're trying to trivialize the
posting of a guy that actually did hardness testing on bearing failures
- something you've never done. but then again, maybe we should be glad
you didn't do a dye penetrant test to determine hardness either.