Index Home About Blog
From: (Jobst Brandt)
Subject: Re: "False" brinells ?
Date: 28 Jan 1997 16:29:03 GMT

David G-man writes:

> Recently I read the FAQ like a well behaved newbie should.
> In the FAQ a Jobst Brandt article stated that headsets actually fret
> to failure and don't Brinell.

> I couldn't find direct evidence for the statement in the article and
> the indirect support seemed a bit weak so I asked the author how he
> had arrived at the conclusion.

[trimmed text from the FAQ]

I disagree on two points.  First, because you use the term "Brinell"
that conveys a notion as incorrect as the phrase "my chain stretched
from climbing steep hills" and second, because there is no possibility
of injury or damage from "indexed" steering head bearings.

However, dimpling is not caused by impact, but rather by lubrication
failure that occurs while riding straight ahead.  This occurs more
easily with a correctly adjusted bearing than with a loose one that
rattles and clunks.  Rattling replenishes lubricant between balls and
races that would otherwise not be present.  Off road bicycles suffer
less from this malady than road bicycles because it occurs primarily
on long straight descents where no steering motions that would
replenish lubrication occur.

If you believe it comes from hammering the balls into the races, I
suggest you try to cause some dimples by hammering on the underside of
the fork crown of a clunker bike of your choice.  Those who hammered
cotters on steel cranks will recall no such dimpling on the spindle,
even though it has a far smaller diameter than the head bearing
although the blows were more severe and direct.  No dimples were made.

Ball bearings make metal-to-metal contact only under fretting loads
(microscopic oscillations) while the bearing is not turning.  Any
perceptible steering motion will replenish lubricant from the oily
meniscus surrounding the contact patch.  Peering over the bars at the
front hub while coasting down a road at 20+ mph you will notice the
fork ends vibrating fore and aft.  This motion does not arise at the
fork end, but at the fork crown, as it bends the steer tube.  Both
head bearings rotate in fretting motion crosswise to the normal plane
of rotation, as the steer tube bends.  Dimples form in the forward and
rearward quadrant of both upper and lower bearings from this fretting.
That they also form in the upper bearing shows they are not directly
load related.

Lubrication failure from fretting causes metal to metal contact to
form microscopic welds between balls and races.  These welds
repeatedly tear material from the softer of the two causing the
elliptical milky dimples in the races.  Were these brinelling marks
(embossed through force), they would be shiny and smooth.  Various
testimonials for the durability of one bearing over another are more
likely caused by the lubricant than the design of the bearing.  The
rigidly mounted ball bearing has survived longer as a head bearing
than it should, considering its poor performance record.

Roller bearings of various designs have been tried, and it appears
that they were the ones that finally made obvious that fore and aft
motion was the culprit all along; a motion that roller bearings were
even poorer at absorbing than balls.  This recognition lead to putting
spherical seats under the rollers.  Although this stopped most of the
dimpling, these bearings did not work well because the cage of needles
tended to shift off center and drag on the housing while the conical
races also shifted causing the bearing to bind.

It appears that a solution was finally found when Shimano bought a
patent from Wilderness Trail Bikes for a ball bearing that combined
the cup and cone ball bearing with the spherical plain bearing.

Today Shimano offers these bearings model called: LX, XT, 600, STX-RC
and Dura Ace.  They have a full-complement angular-contact ball
bearing, whose races are sufficiently reentrant to snap permanently
together.  They have rubber seals that retain grease for life of the
bearing that is not exposed to weather.  The ball bearing is supported
on a spherical steel ring that forms a plain bearing against the
aluminum housing.  The plain bearing takes up the otherwise damaging
out-of-plane motion while the ball bearing does the steering.  The
bearing is only durable as long as the plain bearing remains properly

I don't see the lack of evidence or explanation in the above item that
G-man talks of.  Maybe more a more specific question would clarify
what was not not clear.

Jobst Brandt      <>

Subject: Re: Uneven Headset Bearings
Message-ID: <NTpM8.10029$>
Date: Sat, 08 Jun 2002 16:01:49 GMT

Johan Bornman writes:

>> Newer head bearings, such as the DA one I am using, have a
>> spherical face swash plate on which the angular contact cartridge
>> bearing is supported.  This takes care of alignment as well as
>> fretting dimples in the races, fore and aft skewing being taken up
>> by the swash plate.

> Jobst, the bearing you're referring to is the WTB design that Shimano
> bought, right? The loose cartridge with beveled outer assembly?

I'm not sure what its designation is but it lasts more than 20000
miles with no problem for me.  As I said, the cartridge bearing is a
snap-together unit with preload on the ball contact. The swash plate,
if lubricated, absorbs most of the fretting motion and serves for
uniform bearing loading.

> I'm not convinced that it is such a good design. Not that you're
> vouching for it, but water damage on those headsets are far too
> common for such an expensive headset.  Further, in my experience,
> that particular application of plain bearing/ball bearing
> combination doesn't protect from fretting damage.  I recently
> replaced one such bearing cartridge and opened it up to find
> fore/aft fretting dimples just like in any other headset bearing.

I don't doubt that this is the failure mode, however, I have only put
one replacement cartridge in mine in about six years.  That is a cost,
up with which I can put (to not end a sentence with a preposition).
I ride through all kinds of weather and have had no water intrusion.
With the downward facing cup of this unit, it would take riding upside
down to get water intrusion... or a brisk transport on a car in the rain.

I hope you don't carry your bicycle on the roof of your car, one of
the worst gratuitous introduction of water and dirt you can make.
Just visualize squirting the head, wheel, and crank bearings with a
60mph jet of water and you should get the picture.  Even in the dry,
grease in these bearings act as flypaper for dust filtered from
passing air.

Jobst Brandt  <>  Palo Alto CA

Subject: Re: Headset cheap fix - works
Message-ID: <9JJBb.1685$>
Date: Wed, 10 Dec 2003 18:36:53 GMT

Dave VT? writes:

>> In headsets it's not an impact divot because we don't see the
>> raised material at the edge.

> Let's assume, for a moment, that it was possible to Brinell a
> headset.  After the impact there would be raised areas around the
> indentation.  Those raised areas would probably be prone to rapid
> erosion, wouldn't they?  If you turn the headset, the bearings would
> encounter a tight spot and that tight spot would seem to erode in a
> hurry.  That might explain the lack of "raised material at the
> edge."

That would be a large amount of material to wear away, and by what
method.  This assumes the upset material is softer and weaker than the
parent material and can be rolled into submission.  As I said, this
subject seems to be obscure even to tribologists.  It reduces to how
materials interact in the absence of lubricants, that in the ball
bearing are essential to its survival because the balls do not roll as
much as slide in their races.  A curved ball in a curved race must
slide and if the micro-motions are small enough that lubrication
cannot be replenished, welding must ensue.

Welding of dissimilar materials is commonly done by friction, also
known as inertia welding because it uses large flywheels to spin
mating parts as in stainless steel piston rods and hardened alloy
steel clevises (Caterpillar Tractor).  There it is obvious but in the
microscopic domain it escapes understanding for lack of Maxwell's
Demons, imagined observers small enough to enter into the microscopic
world.  Consider that a ball in a curved race skids backward at its
center and forward at the edges of contact in its rolling motion.  If
this is done in an oscillating mode, so small that the interface is
never exposed, lubrication will be driven out and welding will take
place, even if it is a carbide ball on steel.

> I can see good arguments for both sides of this case, and I'm not
> sure how to sort it out.  For example, if Brinelling were the
> explanation, one might expect the indexing to develop suddenly.

I have attempted to do the suddenly mode by striking the head sets in
the following pictures (DuraAce and Campagnolo) with a one pound
hammer with the cup, cone, and ball complement supported on concrete.
An aluminum block was used to prevent collateral damage from the sharp
hammer blows.

These are not new units, but ones that have fretting damage and have
been reused with loose balls so the spacing of the wear marks are not
uniformly spaced.  There are ones of 1/20 spacing and one with 1/22
spacing.  There are no visible Brinell dents on either bearing set.

> That's not what I have seen in my limited experience (I've been
> riding extensively for more than a decade, but I don't work at a
> bike shop).  On the other hand, if fretting were the sole culprit,
> why would misadjusted headsets seem to develop indexing more
> rapidly?

That they fail more rapidly is not determined.  Just as so many claims
that shimmy is caused by a bent frame, unbalanced wheels, or
misadjusted bearings.  These are straws in the wind and are not proven.

> I have had a headset fail after less than 2 years' use after I had a
> stem slip on a threadless system.  I'm fairly certain the indexing
> developed rather rapidly, certainly within 1000 miles and probably
> even faster than that.

You have no control experiment to sow that there is a correlation.

Jobst Brandt

Subject: Re: 1" versus 1 1/8" Headset???
Message-ID: <%6KBb.1689$>
Date: Wed, 10 Dec 2003 19:04:27 GMT

Ryan Bates writes:

> Hi, wondering what the performance differences are between a 1" and
> a 1 1/8" headtube?  seems to me like the larger diameter would weigh
> more and create a greater frontal area?

This fits into the other subject of head bearing dimpling failures.
These are not so much a bearing problem that needs solving, but rather
a fork problem.  If the steer tube didn't flex when the fork vibrates
fore and aft under load, its bearings would last substantially longer.

The larger steer tube prevents flexing, the fore and aft rocking of
the head bearing that is not designed to move in that direction
especially at such small excursions.  That is the advantage of the
larger diameter.  The reason for the one inch steerer is that it is
strong enough to do the job reliably for a long time.  What was
overlooked is that it also leads to head bearing failure.  Bearing
designers have devised ways to prevent such damage even with a one
inch steerer.  Road bicycles will probably stay with the smaller size.
At least I hope so.

Jobst Brandt

Subject: Re: Headset cheap fix - works
Message-ID: <6G4Cb.1930$>
Date: Thu, 11 Dec 2003 20:43:46 GMT

Carl Fogel writes:

> We're probably banging our heads against a question that can only be
> settled by having someone with a microscope examine a lot of real
> examples and figure out how to tell fretting pits from Brinelling
> dimples. (The raised lips of Brinelling are apparently so small in
> tiny dimples in hard surfaces that it's not the easy tie-breaker
> that I hoped for.)

As I said, proof lies in the matte finish of fretting dimples as well
as their occurrence in an lower AND upper head bearing of a bicycle
with a separation of about 260mm.  It takes only a little deductive
reasoning to realize that these are not caused by impact.

That marks cannot be made on a new bicycle by riding off a high curb
or down a set of stairs ought to be an hint.  This is a time consuming
process that removes microscopic bits over millions of vibratory
oscillations.  I wrote the FAQ item on fretting after hearing many
sage assessments of peoples dimpled head bearings.  It's not as big a
subject as spoke failure and wheel strength but it is obscure, as I
said, even to tribologists.  Fretting is an uncommon phenomenon to
most mechanical engineers.  You should have seen the disbelief that
erupted when this concept was proposed here on  We're
seeing some of that again now.

> Still, when all you have is an air tool and a helpless headset...

Leave the jackhammer out of this.

Jobst Brandt

Subject: Re: Headset cheap fix - works
Message-ID: <cTaCb.1996$>
Date: Fri, 12 Dec 2003 03:47:20 GMT

Dave VT?n writes:

>> As I said, proof lies in the matte finish of fretting dimples as
>> well as their occurrence in an lower AND upper head bearing of a
>> bicycle with a separation of about 260mm.  It takes only a little
>> deductive reasoning to realize that these are not caused by impact.

> That the indents occur in the upper bearing is, in my opinion, the
> most convincing argument that fretting is the primary cause of
> indexing.

> Based on the other observations in this thread, I'm still not
> convinced that indentation plays *no* part in this game. The other
> observations to which I refer are Anthony's description of an
> asymmetrically indexed headset, Jim beam and Carl Fogel's ability to
> dent headsets using common household tools, and anecdotal evidence
> (from myself and others) of headsets failing prematurely due to poor
> adjustment. None of these observations have the authority of
> scientific proof, but they still seem to raise some questions.

As I said, I pounded on the DuraAce and Record head bearings with a
one pound hammer on solid concrete backing.  There is no visible
damage other than the fretting dimples of old, the ones for which the
bearings were removed from service.  You can see these on Carl's web

This whole matter came to my attention when I first started riding a
good bicycle and discovered head bearing chatter after a thousand
miles or so.  I was sure it could not be impact because there was none
of note.  At least nothing that I heard make a sharp clunk that would
occur with an impact sufficient to dent steel.

Coincidentally, at that time, I read of GM cars, that in those days
were all built in Detroit, that they arrives in California with damage
to their differential pinion bearings and some even had pinion tooth
damage.  It was found that these cars were shipped as usual except that
they were tethered tighter than they had in the past and could not
roll fore and aft when the train braked or took up slack.  The
continual rocking in the train caused the damage to bearings that were
not even in service yet.

This all made sense to me and I investigated the head bearing on that
basis finding on closer inspection that I also had dimples in the
practically unloaded upper bearing.  I also had such failures in
Stronglight needle bearings but they were out of the question anyway
because they skewed and bound up if adjusted with no slack.  I think
looking at this realistically, there is not sufficient impact in
normal riding to produce even the slightest dimple.  Arguments for
that have been presented by several contributors in this thread.

Jobst Brandt

Subject: Re: Question - Ball vs Needle Bearing Headsets?
Message-ID: <qTDc9.17173$>
Date: Mon, 02 Sep 2002 06:51:02 GMT

David White writes:

> I have read some archived stuff on this (mostly Jobst) but come away
> from it all still unsure of the following:

> (1) Are needle bearing headsets more long-lived than ball bearing
> ones?

They remain a marginal product because they perform more poorly than
ball bearings for this task that requires a bit of spherical motion in
addition to rotary.  My experience is that they bind but to make up
for that, they still develop dimples.

> (2) Among ball bearing headsets: do smaller balls offer any
> advantage over larger ones?

The larger the balls (a complement in which balls are not separated by
spaces) the larger the balls the stronger the bearing until there are
only three balls.  However, the shaft size limits this to much smaller
and more balls.  Taking it to the limit, infinitely small balls make a
zero area line contact between races, three giant balls have an
enormous contact area in comparison.  There is a practical number in
between and I think Campagnolo has that in their unit.  Shimano has
other constraints, being a cartridge bearing that make their balls
smaller but with a preloaded angular contact bearing it is far better
than the 20 ball Campagnolo bearing in my experience.

> (3) Is the old presumption that having more loose ball bearings is
> better than fewer (in a cage) still considered true?

In use they aren't really "loose" there always being several balls in
contact so that the neighbors are pinned, so to speak.  Gravity makes
them slide to the low side initially.  Later their dimples, although
small, "call them home" when returning to straight ahead after a rotary
excursion.  They make their dimples just as well as caged balls.

> (4) Are cartridge bearing headsets (King, Shimano, etc) using balls
> or needles?


Jobst Brandt  <>  Palo Alto CA

Subject: Re: No keyed washer on headset
Message-ID: <haegc.6335$>
Date: Sat, 17 Apr 2004 17:53:17 GMT

Chris Zacho writes:

> First of all, the keyed washer is not a lockwasher. It's to keep the
> cup from turning when you tighten the lockNUT.

> If you have two spanners, one for the cup, one for the locknut, it
> is not necessary. Which most cyclists who do their own repairs, all
> shops and professional team mechanics have.

> An un-keyed washer will work fine.  If your headset is properly
> tightened down, it won't loosen up.

I have taken the washer out if there was enough threads in the nut or
filed the tab off to prevent thread damage.  The tab washer is a poor
idea and does not work as imagined by its inventor.

What is less apparent in many instances, is that the "lock nut" is
the principal load bearer and not just a locking device.  The bearing
cup, that has plenty of threads in it that do not directly bear the
load, while the lock nut with only a few threads, takes the brunt of
any impact load of the bearing.  The steer tube is not such a bad one
but wheel bearing lock nuts are much worse off, with only a thread and
a half, they are mostly unloaded when the QR is tightened.  I don't
think bicycle parts designers give these concepts any thought.

These are the hidden snags that cause problems in, for instance,a the
recent disc brake discussion.

Jobst Brandt

Index Home About Blog