Headset cheap fix - works

Clyde said:

I removed my front wheel, brake cable, and computer wire, and loosened the stem bolt (threaded
stem). Then I spun the fork around 360 degrees a few times. Put it all back together, and the
indexing is gone. I guess that the bearings got moved around just enough to make a difference.

Sometimes even just regreasing the headset works... for a short while. I tried all kinds of tricks
to cure an old headset - including using loose balls and rotating the cups. None worked in the long
run. Sorry to be negative but I bet your symptoms will return soon.

~PB

Taking the caged bearings out of the lower cup and replacing them with loose balls will do the trick
even better, as it will be less likely to "re-index". You will have to add one ball to make up the
difference.

There is no need to do this to the top bearing, as it's not under compression as much as the bottom,
therefore it has less af a tendancy to "index"..

"May you have the wind at your back. And a really low gear for the hills!"

Chris Zacho ~ "Your Friendly Neighborhood Wheelman"

Chris'Z Corner geocities.comczcornerOpen ↗

Chris Zacho The Wheelman said:

Taking the caged bearings out of the lower cup and replacing them with loose balls will do the
trick even better, as it will be less likely to "re-index". You will have to add one ball to make
up the difference.

There is no need to do this to the top bearing, as it's not under compression as much as the
bottom, therefore it has less af a tendancy to "index"..

IME the top race indexes much worse. Axial compression doesn't have all that much to do with headset
wear - it's the back-to-front flexing of the fork that wears them out.

[email hidden] (Chris Zacho "The Wheelman"😉 wrote in message
news:<[email hidden]>...

Quoted message said:

Taking the caged bearings out of the lower cup and replacing them with loose balls will do the
trick even better, as it will be less likely to "re-index". You will have to add one ball to make
up the difference

this is good advice. re: a previous poster's warning that this trick doesnt work for long:

IMHE, this works fine, in terms of takign away the old indexing. however, the headset will never be
smooth. was this was the responder was referring to? the dents havent dissapeared, they just dont
all line up with the balls at the same time.

if your priority is smooth newlike headset, solution is new headset. if $ is a problem, a tange
passage for 7 bucks or so feels like every other headset to my undiscerning hands.

if your priority is no indexing, becuase your steerer snaps dead ahead when you dont want it to, the
loose ball trick will generally fit the bill.

anthony

ant said:

Quoted message said:

Taking the caged bearings out of the lower cup and replacing them with loose balls will do the
trick even better, as it will be less likely to "re-index". You will have to add one ball to make
up the difference

this is good advice. re: a previous poster's warning that this trick doesnt work for long:

IMHE, this works fine, in terms of takign away the old indexing. however, the headset will never
be smooth. was this was the responder was referring to?

No, I don't know why, but indexing shortly returned after I replaced caged with loose balls (top &
bottom, as many as possible) - and the headset was greased and adjusted carefully after the
modification. So I do not believe it is a good enough cure for all cases (nor rotating the cups or
polishing them).

Quoted message said:

the dents havent dissapeared, they just dont all line up with the balls at the same time.

Perhaps there can be so many dents that enough balls line up (?). I should stress that this was
"light indexing" that I experienced - just enough to detect when turning bars gently with wheel off
the ground - probably not enough to make any difference on the road at normal speed but disturbing
all the same as I didn't want it to get any worse.

BTW, I now use Stronglight A9 headsets with roller bearings - which work to my satisfaction. I know
according to some that roller bearings can index in theory but I don't believe it will ever happen
with the models I'm using*, and even if it did, the tracks that the rollers run on can be replaced
easily (these cups don't wear).

* 1. Absolutely no sign of notching in my experience so far
2. I've never heard of a real case of it

~PB

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

Quoted message said:

ant said:

Quoted message said:

Taking the caged bearings out of the lower cup and replacing them with loose balls will do the
trick even better, as it will be less likely to "re-index". You will have to add one ball to
make up the difference

this is good advice. re: a previous poster's warning that this trick doesnt work for long:

IMHE, this works fine, in terms of takign away the old indexing. however, the headset will never
be smooth. was this was the responder was referring to?

No, I don't know why, but indexing shortly returned after I replaced caged with loose balls (top &
bottom, as many as possible) - and the headset was greased and adjusted carefully after the
modification. So I do not believe it is a good enough cure for all cases (nor rotating the cups or
polishing them).

Quoted message said:

the dents havent dissapeared, they just dont all line up with the balls at the same time.

Perhaps there can be so many dents that enough balls line up (?). I should stress that this was
"light indexing" that I experienced - just enough to detect when turning bars gently with wheel
off the ground - probably not enough to make any difference on the road at normal speed but
disturbing all the same as I didn't want it to get any worse.

BTW, I now use Stronglight A9 headsets with roller bearings - which work to my satisfaction. I
know according to some that roller bearings can index in theory but I don't believe it will ever
happen with the models I'm using*, and even if it did, the tracks that the rollers run on can be
replaced easily (these cups don't wear).

* 1. Absolutely no sign of notching in my experience so far
2. I've never heard of a real case of it

~PB

Dear Pete and others,

I'm baffled. If indexing results from fretted dimples worn into the bearing surface, how will
removing and replacing ball bearings fix anything?

Won't the same steel balls drop into the same pits, whether loose or caged?

What am I missing?

Carl Fogel

Pete Biggs said:

Quoted message said:

Quoted message said:

Taking the caged bearings out of the lower cup and replacing them with loose balls will do the
trick even better, as it will be less likely to "re-index". You will have to add one ball to
make up the difference

Quoted message said:

Quoted message said:

This is good advice. Re: a previous poster's warning that this trick doesn't work for long:

Quoted message said:

Quoted message said:

IMHE, this works fine, in terms of taking away the old indexing. However, the headset will never
be smooth. Was this was the responder was referring to?

Quoted message said:

No, I don't know why, but indexing shortly returned after I replaced caged with loose balls (top &
bottom, as many as possible) - and the headset was greased and adjusted carefully after the
modification. So I do not believe it is a good enough cure for all cases (nor rotating the cups or
polishing them).

Consider that mainly the fore and aft quadrant of the bearing need to chatter, so it is possible,
adding two balls to 20, leaves most of the dimples aligned or nearly aligned with their original
dimples, so it should be no wonder that the symptom returns rapidly, it requiring only slight wear
to get back into the groove.

Quoted message said:

Quoted message said:

The dents haven't disappeared, they just don't all line up with the balls at the same time.

Quoted message said:

Perhaps there can be so many dents that enough balls line up (?). I should stress that this was
"light indexing" that I experienced - just enough to detect when turning bars gently with wheel
off the ground - probably not enough to make any difference on the road at normal speed but
disturbing all the same as I didn't want it to get any worse.

You might review the FAQ item:

draco.acs.uci.edu8f.13.htmlOpen ↗

Quoted message said:

BTW, I now use Stronglight A9 headsets with roller bearings - which work to my satisfaction. I
know according to some that roller bearings can index in theory but I don't believe it will ever
happen with the models I'm using*, and even if it did, the tracks that the rollers run on can be
replaced easily (these cups don't wear).

Quoted message said:

1. Absolutely no sign of notching in my experience so far
2. I've never heard of a real case of it

There are other good reasons to not use roller bearings. That's why they have not been used by
the main manufacturers Campagnolo and Shimano and have not been chosen by the specialists King
and Cane Cr.

"Crane Geek" is not a great name for dyslexics to say.

Jobst Brandt [email hidden]

Carl Fogel said:

I'm baffled. If indexing results from fretted dimples worn into the bearing surface, how will
removing and replacing ball bearings fix anything?

A typical Campagnolo head bearings has 20 balls held in a cage for the purpose installing them
without tedious manual insertion of balls into a greased cup (outer race). This spacing is greater
than close-packed and will allow two additional balls to be placed in the same space if the cage
is removed.

Quoted message said:

Won't the same steel balls drop into the same pits, whether loose or caged?

By about 20/22's some of them miss the mark. As I said, this isn't a good fix and soon returns to
where it was. The problem is that the head bearing chatters and if adjusted to not chatter, it binds
when not steering straight ahead.

Quoted message said:

What am I missing?

Yes?

draco.acs.uci.edu8f.13.htmlOpen ↗

Jobst Brandt [email hidden]

[email hidden] wrote in message news:<_ANxb.8294>

Quoted message said:

Consider that mainly the fore and aft quadrant of the bearing need to chatter, so it is possible,
adding two balls to 20, leaves most of the dimples aligned or nearly aligned with their original
dimples, so it should be no wonder that the symptom returns rapidly, it requiring only slight wear
to get back into the groove.

interesting.

a cool problem solvers's product might be a new bearing retainer to 'fix' indexed headsets. an LBS
with a customer who doesnt want to pay labor and new headset cash would only need remove the old
bearings, and put the new retainer with balls in.

to avoid the problem jobst points out, the retainer could either have far fewer balls in it, like 15
instead of 20 (not ideal) or instead could have an irregular spacing of bearings, where you could
still fit 18 or so in, i would haphazardly guess.

would be pretty cheap to make. a two part production process. stamp steel plate. bend retainer
fingers in. maybe it could be worth it?

idly, anthony

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

Quoted message said:

Pete Biggs said:

Quoted message said:

> Taking the caged bearings out of the lower cup and replacing them with loose balls will do the
> trick even better, as it will be less likely to "re-index". You will have to add one ball to
> make up the difference

Quoted message said:

Quoted message said:

This is good advice. Re: a previous poster's warning that this trick doesn't work for long:

Quoted message said:

Quoted message said:

IMHE, this works fine, in terms of taking away the old indexing. However, the headset will
never be smooth. Was this was the responder was referring to?

Quoted message said:

No, I don't know why, but indexing shortly returned after I replaced caged with loose balls (top
& bottom, as many as possible) - and the headset was greased and adjusted carefully after the
modification. So I do not believe it is a good enough cure for all cases (nor rotating the cups
or polishing them).

Consider that mainly the fore and aft quadrant of the bearing need to chatter, so it is possible,
adding two balls to 20, leaves most of the dimples aligned or nearly aligned with their original
dimples, so it should be no wonder that the symptom returns rapidly, it requiring only slight wear
to get back into the groove.

Quoted message said:

Quoted message said:

The dents haven't disappeared, they just don't all line up with the balls at the same time.

Quoted message said:

Perhaps there can be so many dents that enough balls line up (?). I should stress that this was
"light indexing" that I experienced - just enough to detect when turning bars gently with wheel
off the ground - probably not enough to make any difference on the road at normal speed but
disturbing all the same as I didn't want it to get any worse.

You might review the FAQ item:

draco.acs.uci.edu8f.13.htmlOpen ↗

Quoted message said:

BTW, I now use Stronglight A9 headsets with roller bearings - which work to my satisfaction. I
know according to some that roller bearings can index in theory but I don't believe it will ever
happen with the models I'm using*, and even if it did, the tracks that the rollers run on can be
replaced easily (these cups don't wear).

Quoted message said:

1. Absolutely no sign of notching in my experience so far
2. I've never heard of a real case of it

There are other good reasons to not use roller bearings. That's why they have not been used by the
main manufacturers Campagnolo and Shimano and have not been chosen by the specialists King and
Cane Cr.

On the other hand, enough people have been buying Stronglight roller bearing headsets for what,
about two decades, that they have not felt market pressure to change. I recently bought an A9
myself, not because I think there is anything wrong with ball bearing headsets, or anything
necessarily superior about roller bearings, but because I wanted a French headset for my French
bike, and there is no practical evidence that I shouldn't go with a Stronglight. Replaceable
bearings and bearing tracks do seem to make the risk pretty low, though. Would you say that there is
an economic incentive to both bike shops and manufacturers to use a design that requires replacing
the whole headset instead of just the bearing set when the cups are damaged? And they can say to you
while they're installing the new one, "See what happens when you don't take care of your headset?"

JP

Quoted message said:

There are other good reasons to not use roller bearings. That's why they have not been used by the
main manufacturers Campagnolo and Shimano and have not been chosen by the specialists King and
Cane Cr.

the primary reason from what i know is economics, not mechanics.

rollers cost significantly more than balls and require a different race machining setup. balls
already last very well - there's no reason to bother "upgrading".

jb

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

Quoted message said:

Carl Fogel said:

I'm baffled. If indexing results from fretted dimples worn into the bearing surface, how will
removing and replacing ball bearings fix anything?

A typical Campagnolo head bearings has 20 balls held in a cage for the purpose installing them
without tedious manual insertion of balls into a greased cup (outer race). This spacing is greater
than close-packed and will allow two additional balls to be placed in the same space if the cage
is removed.

Quoted message said:

Won't the same steel balls drop into the same pits, whether loose or caged?

By about 20/22's some of them miss the mark. As I said, this isn't a good fix and soon returns to
where it was. The problem is that the head bearing chatters and if adjusted to not chatter, it
binds when not steering straight ahead.

Quoted message said:

What am I missing?

Yes?

draco.acs.uci.edu8f.13.htmlOpen ↗

Jobst Brandt [email hidden]

Dear Jobst,

Ignorance is bliss, since it makes things so interesting. I know that a caged-ball-bearing assembly
is much easier to install, but I thought that its real virtue was that the hard steel balls rotated
against the soft cage metal instead of wearing against each other.

That is, a dozen ball-bearings kept from touching each other by a soft metal cage are generally
better than a baker's dozen of the same ball-bearings uncaged and stuffed into the same space, all
rubbing against each other, even though the extra ball might seem like a good idea.

Is this just another of my misconceptions?

Carl Fogel

Pete Biggs said:

BTW, I now use Stronglight A9 headsets with roller bearings - which work to my satisfaction.

As they did for me for many years. Light, cheap, what's not to like?!

Quoted message said:

I know according to some that roller bearings can index in theory but I don't believe it will ever
happen with the models I'm using*,

One of the A9s finally indexed on our tandem. Photo here:
geocities.comindex.htmlOpen ↗

Quoted message said:

and even if it did, the tracks that the rollers run on can be replaced easily (these cups
don't wear).

* 1. Absolutely no sign of notching in my experience so far
2. I've never heard of a real case of it

See above.

Quoted message said:

One of the A9s finally indexed on our tandem. Photo here:
geocities.comindex.htmlOpen ↗

that's an interesting pic!

maybe i'm just seeing what i want to see, but it appears that you have classic brinelling
going on there.

i.e._ / \ \_/ < crater "lip" ______/\ /\______ \_/

there appears to be material deformed _up_ out of the crater forming a lip as well as that indented
down. text book brinell! not what the "lubrication failure" folks will want to see.

Quoted message said:

Quoted message said:

ant said:

>Taking the caged bearings out of the lower cup and replacing them with loose balls will do the
>trick even better, as it will be less likely to "re-index".

-snip-

Quoted message said:

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

Quoted message said:

No, I don't know why, but indexing shortly returned after I replaced caged with loose balls (top &
bottom, as many as possible) - and the headset was greased and adjusted carefully after the
modification.

-snip- Carl Fogel said:

I'm baffled. If indexing results from fretted dimples worn into the bearing surface, how will
removing and replacing ball bearings fix anything? Won't the same steel balls drop into the same
pits, whether loose or caged? What am I missing?

You're not. Index headsets index less severely when the cups are removed and reinstalled - the balls
center on a dimple on one side but fall between divots on the other race. In theory, tossing
retainers and using loose balls (changing the ball count) can possibly prevent the bearings from
aligning in more than a few dimples. Simply spinning the headset can only give momentary relief -
the balls will find their homes again quickly.

--
Andrew Muzi www.yellowjersey.org Open every day since 1 April, 1971

A Muzi <[email hidden]> wrote:

: But headsets are so darned cheap. They cost,what, $20~50 for ten or fifteen years? Why "fix" 'em?

: LBS phenomenon I've noticed - a guy who will think nothing of $150+ for new rubber, cables and
: tape every season [censored] on and on about an index headset every 10+ years. Go figure
: --
: Andrew Muzi www.yellowjersey.org Open every day since 1 April, 1971

Andrew, My bikes seem to have indexed headsets. I can feel the indexing when the bike is in the
stand. But I cannot feel it when I am riding. Is this one of those things that you can just ignore?

--------------------------------
Bob Masse' [email hidden]
--------------------------------

jim beam said:

Quoted message said:

One of the A9s finally indexed on our tandem. Photo here:
geocities.comindex.htmlOpen ↗

that's an interesting pic!

maybe i'm just seeing what i want to see, but it appears that you have classic brinelling going
on there.

i.e._ / \ \_/ < crater "lip" ______/\ /\______ \_/

there appears to be material deformed _up_ out of the crater forming a lip as well as that indented
down. text book brinell! not what the "lubrication failure" folks will want to see.

I wish I'd saved the parts. They've long since gone in the trash...
:-(

ant said:

a cool problem solvers's product might be a new bearing retainer to 'fix' indexed headsets.

-snip- Been done , sort of. An outfit named TefSeal (Tulsa? Ok City?) used to make a seal for normal
1039 headsets and a nylon torus to replace the 2100 retainer. You could get another few (sticky)
seasons on an indexed headset with a pair of those for five bucks.

But headsets are so darned cheap. They cost,what, $20~50 for ten or fifteen years? Why "fix" 'em?

LBS phenomenon I've noticed - a guy who will think nothing of $150+ for new rubber, cables and tape
every season [censored] on and on about an index headset every 10+ years. Go figure
--
Andrew Muzi www.yellowjersey.org Open every day since 1 April, 1971

Quoted message said:

A Muzi <[email hidden]> wrote:

: But headsets are so darned cheap. They cost,what, $20~50 for ten or fifteen years? Why
: "fix" 'em?

: LBS phenomenon I've noticed - a guy who will think nothing of $150+ for new rubber, cables and
: tape every season [censored] on and on about an index headset every 10+ years. Go figure

Quoted message said:

My bikes seem to have indexed headsets. I can feel the indexing when the bike is in the stand. But
I cannot feel it when I am riding. Is this one of those things that you can just ignore?

Sure. Ignoring it or changing it won't make your bike any faster. If it isn't annoying, it is not a
problem IMHO.

But next time you are in the mood to deal with this, why not run a set of gauges across that
frame/fork? Might show a problem. Be sure to check the column with a straightedge, too.

Or buy a headset with installation from a trusted shop with the express expectation that you want at
least, say, six years from it. Review the history with the mechanic and maybe he can point up the
problem / suggest a solution. This is s solvable problem, not really expensive.

--
Andrew Muzi www.yellowjersey.org Open every day since 1 April, 1971