Re: What is "fretting?"

jim beam said:

and on the subject of headsets, has your "fretting" model ever managed
to explain indexing after a single blow [like a crash] yet?

Does bearing indexing necessarily have to have the same cause on every
occurence?

Jasper

Jasper Janssen said:

Quoted message said:

and on the subject of headsets, has your "fretting" model ever
managed to explain indexing after a single blow [like a crash] yet?

Quoted message said:

Does bearing indexing necessarily have to have the same cause on
every occurence?

No one has demonstrated that dimples in reasonable head bearings can
be caused by any mechanism other than fretting. It takes a massive
direct blow with a hammer on the bearing laid on an anvil to make a
visible mark on head bearing races. Beating on the underside of a
junk bicycle fork crown will not cause dimples. For those of us who
used steel cottered cranks, the exercise of pounding directly on the
BB spindle to seat cotters with a steel hammer was done often. The
absence of dimples on the spindle and cups should have raised
questions long ago on how a front tire can impart sufficient force to
cause head bearing indexing.

http://www.sheldonbrown.com/brandt/indexed-steering.html

As mentioned in the above article, the dimples are matte and oval, and
are found mainly in the fore and aft quadrant of head bearings. It is
in this area where fretting motion of a fork acts as the fork crown
oscillates with tiny in line vibration of the front axle. That there
are dimples in the upper race can also not be explained by impact.

I find this interesting because it is similar to years of people
placing the cause of spoke failure on momentary overload instead of
fatigue failures. That goes for many other failures on bicycles as
well. It should occur to riders that these failures do not occur the
first time the bicycle is subjected to a rough ride. It takes many
incidents of mostly average loading to build up cyclic failures.

[email hidden]

Jasper Janssen said:

jim beam said:

and on the subject of headsets, has your "fretting" model ever managed
to explain indexing after a single blow [like a crash] yet?

Does bearing indexing necessarily have to have the same cause on every
occurence?

Jasper

that's possibly the most astute question anyone's ever asked in the
whole headset debate - no of course not. there are two mechanisms, true
brinelling & false brinelling, what jobst calls "fretting".

true brinelling is that arising from overload, typically something like
impact. even though the bearing races are are made of very hard steel,
they still have the ability to plastically deform under load. [it was
interesting reading the recent thread on replacing cartridge bearings
for a suntour hub a few days ago - several contributors took the trouble
to warn of the dangers when fitting the replacement bearing, to avoid
loading the inner race to prevent damage.] the deformation is exactly
like that seen in brinnel hardness testing, an easily googleable term.

false brinelling is commonly associated with things like shipping damage
where a bearing is assembled but improperly greased & unloaded. the
lack of [pre]load allows relative sliding movement between the rolling
elements, which is concentrated in one spot and may eventually lead to wear.

so, it /is/ possible for a headset bearing to "fret", but for a well
lubed, properly adjusted headset, it is most unlikely. indeed, the
solution for shipping damage to bearings [the bigger ones] is to either
pre-load them or ship disassembled. my first experience of an indexed
headset was after crashing into the back of a car with a new bike. the
crash was sufficient to bend the fork slightly, but index the head set
significantly - immediately noticeable. i took it for examination at my
metallurgy lab where i was an undergrad at the time.

one caveat: if a headset /is/ brinelled with impact, it /is/ possible
for it to subsequently "fret" if it continues to be ridden. when
resting in the impact dimples, the bearing elements are no longer
preloaded as they were before, so relative sliding motion can occur.
one could therefore possibly draw the wrong conclusion about
origination, but for someone that's taken more than a cursory glance at
the subject, that confusion should not arise.

Quoted message said:

Jasper Janssen said:

Quoted message said:

and on the subject of headsets, has your "fretting" model ever
managed to explain indexing after a single blow [like a crash] yet?

Quoted message said:

Does bearing indexing necessarily have to have the same cause on
every occurence?

No one has demonstrated that dimples in reasonable head bearings can
be caused by any mechanism other than fretting. It takes a massive
direct blow with a hammer on the bearing laid on an anvil to make a
visible mark on head bearing races.

er, carl fogel beat his against a straw mat & got brinelling.

Quoted message said:

Beating on the underside of a
junk bicycle fork crown will not cause dimples.

er, i beat mine against a perfectly good fork crown & got brinelling.
we both posted our results jobst. your inability to accept our evidence
is plain spite. your inability to open a book on the subject is plain
asinine.

Quoted message said:

For those of us who
used steel cottered cranks, the exercise of pounding directly on the
BB spindle to seat cotters with a steel hammer was done often. The
absence of dimples on the spindle and cups should have raised
questions long ago on how a front tire can impart sufficient force to
cause head bearing indexing.

jobst, when i turn the spindle on a brand new, fresh out of the box,
unbuilt shimano 105 hub & feel roughness in the rotation, exactly how
much aspherodicity of bearing ball is causing that? it's certainly not
visible, but very much detectable by feel. you're trying to argue that
because a dimple can't be "seen" with the naked eye on a dirty greased
up bearing, that it therefore can't exist? you're living in a twisted
myopic world of your own ridiculous design.

Quoted message said:

http://www.sheldonbrown.com/brandt/indexed-steering.html

As mentioned in the above article, the dimples are matte and oval, and
are found mainly in the fore and aft quadrant of head bearings. It is
in this area where fretting motion of a fork acts as the fork crown
oscillates with tiny in line vibration of the front axle. That there
are dimples in the upper race can also not be explained by impact.

I find this interesting because it is similar to years of people
placing the cause of spoke failure on momentary overload instead of
fatigue failures. That goes for many other failures on bicycles as
well. It should occur to riders that these failures do not occur the
first time the bicycle is subjected to a rough ride. It takes many
incidents of mostly average loading to build up cyclic failures.

[email hidden]

you can stick to the same old stories based on "selected data" as long
as you want jobst, but as long as you do, you'll never have any
credibility. go to a library. publish hard data points like your own
spoke tensions. stop fudging.

Quoted message said:

I find this interesting because it is similar to years of people
placing the cause of spoke failure on momentary overload instead of
fatigue failures.
[email hidden]

I understand journalists aren't technical writers, but I wonder what
cause might be attributed to Basso's broken spokes in this year's Giro?

http://www.cyclingnews.com/road/2005/giro05/?id=results/giro0510

Dianne who? said:

Quoted message said:

I find this interesting because it is similar to years of people
placing the cause of spoke failure on momentary overload instead of
fatigue failures.

Quoted message said:

I understand journalists aren't technical writers, but I wonder what
cause might be attributed to Basso's broken spokes in this year's
Giro?

http://www.cyclingnews.com/road/2005/giro05/?id=results/giro0510

That one may have been obvious for folks on the scene. I can imagine
that in the rough, a foreign object got in the wheel and ripped a
spoke. This was not caused by bumpy terrain.

[email hidden]

Jobst's belief that fretting occurs from long downhill rides with the
handlebars held straight is pure speculation.

In my informal survey of 12 steel bicycles with mostly campy headset -
almost all of them 30 years old, indexed headsets are highly
correlated with long head tubes, 23.5", 24.5", etc., and reynolds 531
tubing, which implies that possibly it's more likely that indexed
headsets are a result of heavier riders and/or flexion in the head
tubes.

Jobst's _explanation_ for fretting (not the mechanism, but the cause)
is imho entirely speculation.

- Don Gillies
San Diego, CA

Donald Gillies said:

Jobst's belief that fretting occurs from long downhill rides with
the handlebars held straight is pure speculation.

Quoted message said:

In my informal survey of 12 steel bicycles with mostly Campy headset
- almost all of them 30 years old, indexed headsets are highly
correlated with long head tubes, 23.5", 24.5", etc., and Reynolds
531 tubing, which implies that possibly it's more likely that
indexed headsets are a result of heavier riders and/or flexion in
the head tubes.

When you say "heavy riders" what criterion are you using. As far as I
can detect, many of the regulars who post to this group are over
230lbs. My bicycle regularly suffered from indexed steering until
Shimano introduced the spherical swivel under the head bearing as I
described. Besides, I'm' certainly around 180lbs. Do I qualify for
the heavy rider syndrome in this day of heavies in the USA?

There have been enough 21" frames with indexed steering in the days of
such Italian frames to demonstrate that it is not only tall bicycles
that suffer this damage. However, if only tall bicycles suffered from
this, what do you propose is the mechanism?

Quoted message said:

Jobst's _explanation_ for fretting (not the mechanism, but the
cause) is IMHO entirely speculation.

Please explain how these dimples occur if it is not fretting, a common
mechanism for damaging rolling element bearings.

[email hidden]

Donald Gillies said:

Quoted message said:

Quoted message said:

Jobst's _explanation_ for fretting (not the mechanism, but the
cause) is IMHO entirely speculation.

Quoted message said:

Quoted message said:

Please explain how these dimples occur if it is not fretting, a
common mechanism for damaging rolling element bearings.

Quoted message said:

Jobst, you are not understanding my point or you would not be
complaining. i never disagreed that dimples were from "fretting".
in fact, i made a point to use "indexing" in the rest of my message
so as not get into the fretting/brinneling war.

Well that's part of the problem. You should say what side of the
fence you are on so that the discussion can progress. As you see,
with things up in the air, definitions are lacking.

Quoted message said:

but you are way off base to go beyond the type of damage and suppose
that it is caused by a certain style of riding.

I have no "style of riding" in mind but rather longer durations of
vibratory micro-motion that is caused from riding straight ahead. You
will notice that the dimples are oriented straight ahead. On top of
that, as I mentioned, automotive steering mechanisms suffered from
this endlessly leading to patents for Saginaw, Gemmer, Ross and other
steering gears that attempted to either randomize the load point or
make it insensitive to this wear.

Quoted message said:

you in fact make the same error here that you made in your book when
you claim that MA2 rims are some of the best ever made. you
confused YOUR riding style with the AVERAGE riding style and then
made value judgements based on your erroneous belief that YOU ARE
AVERAGE which you are most certainly not.

You are harping on an invalid issue. This claim is from the
popularity of the MA-2 rim here and abroad in every bicycle shop I
entered. Tourists and racers used them and before that the tubular
variety of socketed rims from Mavic. If you haven't been there you
cannot assess that widespread use.

Quoted message said:

For example, I have heard people say that you descend like a maniac,
and in your book you made a value judgement that durable rim
sidewalls are important because otherwise they wear out. I'd like
to know how many people in this country have ever experienced a
sidewall wear-out in their lifetimes vs. how many have experienced
an out of round or pretzled rim, and I think you'd find that the
latter beat the former by an order of magnitude or more.

You apparently haven't had contact with all the wheels people build
and replace rims for exactly the reason of sidewall wear-out. Today,
people cast off wheels so often that they never wear much. It's a
fashion business now, not a utility one.

Quoted message said:

In the big scheme of things sidewall wearouts are a secondary
consideration for most riders.

So what. Most riders are not the one who suffer a crash from sidewall
failure. As you may have seen we recently had a whole thread from
people who had sidewall failures.

Quoted message said:

Similarly now, you have taken one of your riding styles (I presume)
which is long downhill coasting events, perhaps in the swiss alps or
wherever, and correlated a headset "indexing" event with a ride that
had some long straight downhills, and drawn the wrong conclusion, once
again, based on your riding style, which is not average.

I think you are scraping the bottom of the barrel with this "riding
style" perspective. It doesn't wash.

[email hidden]

Quoted message said:

In short, I think that the handlebars on any bicycle rolling "straight"
down a normal road swivel constantly in tiny arcs that are considerably
greater than the width of the terminal letters of the two following words
on a 1280-pixel-wide, 12.5-inch screen line:

What do you think is the period of these oscillations? How long a time is
required for fretting to start?

--
Benjamin Lewis

Although the moon is smaller than the earth, it is farther away.

Benjamin Lewis said:

Quoted message said:

Even coasting down the smoothest, straightest road, the ends
of the handlebars constantly swivel considerably more than
0.015 inches.

How much do they swivel?

Quoted message said:

(At a modest 30 mph, a bike covers 44 feet of
road surface per second--a 0.1 degree change of course
amounts to only 0.92 inches in 44 feet. Few people can shoot
1-inch groups with .22 target rife at 50 feet from a
standing position.)

Why is this a good comparison? You don't steer with your hands when
coasting downhill, and you don't even give indication in your rifle example
of how fast the angle is changing, which is of course important in the case
of coasting on a bike.

why is the rate of angle change important?

Benjamin Lewis said:

jim beam said:

Benjamin Lewis said:

[email hidden] wrote:

>Even coasting down the smoothest, straightest road, the ends
>of the handlebars constantly swivel considerably more than
>0.015 inches.

How much do they swivel?

>(At a modest 30 mph, a bike covers 44 feet of
>road surface per second--a 0.1 degree change of course
>amounts to only 0.92 inches in 44 feet. Few people can shoot
>1-inch groups with .22 target rife at 50 feet from a
>standing position.)

Why is this a good comparison? You don't steer with your hands when
coasting downhill, and you don't even give indication in your rifle
example of how fast the angle is changing, which is of course important
in the case of coasting on a bike.

why is the rate of angle change important?

Because it might be slower than the length of time it takes fretting to
start.

what length of time "might" that be? i don't know, but i'd be surprised
if it's less than the lube replenishment period in a system like this.

jim beam said:

Benjamin Lewis said:

jim beam said:

Benjamin Lewis wrote:
> Why is this a good comparison? You don't steer with your hands when
> coasting downhill, and you don't even give indication in your rifle
> example of how fast the angle is changing, which is of course
> important in the case of coasting on a bike.

why is the rate of angle change important?

Because it might be slower than the length of time it takes fretting to
start.

what length of time "might" that be? i don't know, but i'd be surprised
if it's less than the lube replenishment period in a system like this.

I have no idea how long either of these times are, which is why I was
asking. I wouldn't be at all surprised if it were less.

--
Benjamin Lewis

Although the moon is smaller than the earth, it is farther away.

Benjamin Lewis said:

jim beam said:

Benjamin Lewis said:

jim beam wrote:
> Benjamin Lewis wrote:
>> Why is this a good comparison? You don't steer with your hands when
>> coasting downhill, and you don't even give indication in your rifle
>> example of how fast the angle is changing, which is of course
>> important in the case of coasting on a bike.
>
> why is the rate of angle change important?
Because it might be slower than the length of time it takes fretting to
start.

what length of time "might" that be? i don't know, but i'd be surprised
if it's less than the lube replenishment period in a system like this.

I have no idea how long either of these times are, which is why I was
asking. I wouldn't be at all surprised if it were less.

Dear Benjamin,

Asking about both times sounds sensible.

The time for replenishing the squashed-out oil is probably
the time for a single roll in any direction of one tenth of
one degree, corresponding to the length of time for the
outside of the handlebar to move 0.015 inches. That is, a
single roll to a new patch of oily surface is all that's
needed.

Figuring out how long it takes a small bearing to move less
than 0.01 degrees would require first knowing how the
bearing's circumference and dividing that by 3600.

A 1/4" bearing has a circumference of 0.7854 inches, so 0.1
degrees of movement (1/3600th) means a roll of 0.000218
inches.

A 3/16" bearing has a cirumference of 0.5890 inches, so 0.1
degrees of movement (1/3600th) means a roll of 0.000163
inches.

A 5/32" bearing has a circumference of 0.4909 inches, so 0.1
degrees of movement (1/3600th) means a roll of 0.000136
inches.

(For comparison, a human hair is about 0.001000 inches, five
times the size of the largest movement above.)

Of course, another factor is how long it takes for the ball
to penetrate the protective oil layer. Given the viscosity
of oil, it seems likely that this will be longer than the
ball will sit still and move less than 0.0002 inches, just
from ordinary handlebar vibration due to road buzz.

That leads to the related question of how fast the
handlebars are moving back and forth. The accelerometer
tests from Specialized measure vibrations between the axle
and the handlebars up to at least 600 Hz:

http://www.specialized.com/OA_MEDIA/pdf/Witchcraft.pdf

Copies mailed to me from Specialized were clear enough to
see that the horizontal scale on those two lower graphs runs
0, 100, 200, 300, 400, 500, 600 Hz, while the vertical scale
is log, going down 1, 0.1, 0.01, 0.001, 0.0001.

(Regrettably, there was no further improvement in
legibility, but a new scanner and a related set of graphs
may prove to be of interest.)

Anyway, those are some of the figures that I can see, though
they're not enough to answer the question directly. However,
the fretting theory itself seems to insist that ordinary
riding is sufficient to stop the balls from penetrating the
oil, micro-welding to the race, and tearing loose. So the
fretting theory itself predicts that the steering motion
from merely turning the pedals on the flats is enough to
make a difference--which suggests a very slow penetration
time.

The jiggling of a rider's hands on the bars while coasting
down a normal road at 30 mph is far more than the tenth of
less than one degree predicted to be enough for
replenishment. The vibrations are measured in hundreds of
Hertz (cycles per second), so they're orders of magnitude
faster than any pedal cadence measure in rpm.

I still waver on the matter and hope to hear more, but I
have to side with Jim Beam on which time is likely to be
faster. Until someone comes up with an explanation that goes
against these kinds of figures, I expect the headset balls
to roll back and forth 0.1 degrees on a normal road far more
rapidly than they penetrate the oil--if they didn't, they'd
fret themselves during any kind of riding, not just downhill
coasting.

Carl Fogel

Jobst Brandt wrote (about headset race dimples)

Quoted message said:

That there
are dimples in the upper race can also not be explained by impact.

What about the cases where only the lower race has dimples?
I remember that around 1980, one of the mail-order houses
(probably Bikecology or Nashbar (originally Bike Warehouse))
included in their catalog a replacement bottom half of a
headset, for riders who had detented steering due to dimples
in the lower race of their Campi headsets. I know that the
marketing of a part by no means guarantees that it solves
a real problem--but around that time, my Campi Record road
headset (#1039) developed dimples in the bottom half only,
and I doubt I was the only rider with that experience.
I replaced the bottom half with Stronglight roller bearing
parts and never had the problem thereafter.

Tom Ace

Donald Gillies said:

For example, [...] in your [Jobst'] book you made a value judgement that
durable rim sidewalls are important because otherwise they wear out.
I'd like to know how many people in this country have ever experienced
a sidewall wear-out in their lifetimes vs. how many have experienced
an out of round or pretzled rim, and I think you'd find that the
latter beat the former by an order of magnitude or more. In the big
scheme of things sidewall wearouts are a secondary consideration for
most riders.

Ahh, but Don, you're in sunny (? maybe not this year) Southern
California. I'm in wet Western Oregon, and I've had a sidewall wearout,
as has a friend when he lived here.

I'm sure what you say is true for low-rain areas; when I lived in a
low-rain area, I would have agreed with you. No longer, I've learned
the hard way. Fortunately I was moving slowly when the sidewall peeled
off and jammed in the brakes, locking the rear wheel.

Now I pay attention to the conditions of the rim sidewalls on bikes that
get ridden lots in the rain.

Mark Janeba

Quoted message said:

The time for replenishing the squashed-out oil is probably the time for a
single roll in any direction of one tenth of one degree, corresponding to
the length of time for the outside of the handlebar to move 0.015
inches. That is, a single roll to a new patch of oily surface is all
that's needed.

Figuring out how long it takes a small bearing to move less than 0.01
degrees would require first knowing how the bearing's circumference and
dividing that by 3600.

[...]

Quoted message said:

Of course, another factor is how long it takes for the ball to penetrate
the protective oil layer. Given the viscosity of oil, it seems likely
that this will be longer than the ball will sit still and move less than
0.0002 inches, just from ordinary handlebar vibration due to road buzz.

That leads to the related question of how fast the handlebars are moving
back and forth. The accelerometer tests from Specialized measure
vibrations between the axle and the handlebars up to at least 600 Hz:

http://www.specialized.com/OA_MEDIA/pdf/Witchcraft.pdf

Do these vibrations cause horizontal steering movement? How much?

[...]

Quoted message said:

Anyway, those are some of the figures that I can see, though they're not
enough to answer the question directly. However, the fretting theory
itself seems to insist that ordinary riding is sufficient to stop the
balls from penetrating the oil, micro-welding to the race, and tearing
loose. So the fretting theory itself predicts that the steering motion
from merely turning the pedals on the flats is enough to make a
difference--which suggests a very slow penetration time.

The jiggling of a rider's hands on the bars while coasting down a normal
road at 30 mph is far more than the tenth of less than one degree
predicted to be enough for replenishment. The vibrations are measured in
hundreds of Hertz (cycles per second), so they're orders of magnitude
faster than any pedal cadence measure in rpm.

I still waver on the matter and hope to hear more, but I have to side
with Jim Beam on which time is likely to be faster. Until someone comes
up with an explanation that goes against these kinds of figures, I expect
the headset balls to roll back and forth 0.1 degrees on a normal road far
more rapidly than they penetrate the oil--if they didn't, they'd fret
themselves during any kind of riding, not just downhill coasting.

I'm not sure what you're suggesting. Are you arguing that fretting does
not happen (in spite of pretty clear evidence, IMO), or merely that it is
not more likely to happen during downhill coasting than in other types of
riding?

My intuitions are that a) fretting is a somewhat marginal event whether
downhill coasting or not, or we'd see it occur much more rapidly, and that
b) it is most likely to occur during periods of minimal steering, and
downhill coasting seems to me to be the best candidate.

--
Benjamin Lewis

Although the moon is smaller than the earth, it is farther away.

Carl Fogel said:

Quoted message said:

Please explain how these dimples occur if it is not fretting, a
common mechanism for damaging rolling element bearings.

Quoted message said:

While searching for posts about the number of bearings in a headset,
I stumbled across a 1997 post from you about dimpled headsets that
included this intriguing calculation:

Quoted message said:

"It is not at all a bumpy descent problem and not one of pushing
aside grease, but rather one of penetrating the boundary layer
between ball and race. The bearing is not lubricated by the grease
as such, but rather the oil that the grease exudes. This is a film
no thicker than a few micro-inches (10^-6). It takes a ball
rotation of less than a degree to replenish this film, a steering
motion of 1/10 of that, the ratio of ball to bearing diameter being
about 1:10."

http://groups-beta.google.com/group/rec.bicycles.tech/msg/2d541ce44398f645

Quoted message said:

The explanation makes sense, but I was startled by how tiny a
handlebar movement seemed to be needed to prevent headset dimpling
by fretting--less than 1/10 of less than one degree.

Quoted message said:

Curious, I slapped a tape measure across my touring bike's
handlebars and found that its modest antlers have a 17 inch
spread--8.5 inches to either side of the headset.

Quoted message said:

A tenth of a degree on a 17-inch diameter circle is about 0.015
inches, less than the thickness of a sheet of copier paper.

Quoted message said:

Even coasting down the smoothest, straightest road, the ends of the
handlebars constantly swivel considerably more than 0.015
inches. (At a modest 30 mph, a bike covers 44 feet of road surface
per second--a 0.1 degree change of course amounts to only 0.92
inches in 44 feet. Few people can shoot 1-inch groups with .22
target rife at 50 feet from a standing position.)

It isn't smoothness of surface but speed and not having a disturbing
grip on the bars. Classically, tucked in coasting down HWY1 to Santa
Cruz at 50+mph with wind causes no steering motions that you can see.
I have the hands on the stem and take pleasure in riding ON the 3"
wide edge stripe on the side of the road. Meanwhile the fork crown is
vibrating and because it is a Shimano pre-loaded angular contact
bearing, these motions are taken up by the spherical seat under the
ball bearing, something inherited from needle bearing headsets that
indexed worse than ball bearings.

Quoted message said:

It sounds as if the very road vibration that the fretting theory of
dimpled headsets relies on to micro-weld and then tear apart the
balls and races would be ample to replenish the protective oil film.

The tear apart occurs in the range of ball fretting rotations, there
is no excursion outside of the elastic contact zone of the ball. That
is why the bearing fails.

Quoted message said:

Was your 1997 calculation wrong about the necessary steering motion?
Or am I missing something crucial about the fretting theory?

No.

[email hidden]

Carl Fogel said:

Quoted message said:

Quoted message said:

Even coasting down the smoothest, straightest road, the ends of
the handlebars constantly swivel considerably more than 0.015
inches.

Quoted message said:

Quoted message said:

How much do they swivel?

Quoted message said:

Quoted message said:

Quoted message said:

(At a modest 30 mph, a bike covers 44 feet of road surface per
second--a 0.1 degree change of course amounts to only 0.92 inches
in 44 feet. Few people can shoot 1-inch groups with .22 target
rife at 50 feet from a standing position.)

Quoted message said:

Quoted message said:

Why is this a good comparison? You don't steer with your hands
when coasting downhill, and you don't even give indication in your
rifle example of how fast the angle is changing, which is of course
important in the case of coasting on a bike.

Quoted message said:

It seems unlikely that any rider can lean on a pair of 17-inch-wide
handlebars and coast downhill on an ordinary paved road without the
ends of the handlebars swivelling constantly back and forth more
than 0.015 inches.

I don't know where you get these dimensions but bumps on pavement
surfaces do not induce steering. At high speeds such steering motions
are essentially impossible unless you are striking your bars with your
fist. What I see here is that you are trying to nullify the
definition and common occurrence of fretting that was dug up by
posters to this subject. The definitions are there, accept it. As I
mentioned, automotive steering gears were classic places where such
problem existed.

Quoted message said:

That's roughly 1/64th of an inch, the thickness of a fifteen
human hairs or a sheet of paper. The bumps on a well paved
road are considerably larger than that, and the effect is
magnified at ends of the bars. How steady can any rider hold
the handlebars against the buzz of road vibration?

I don't know where you are measuring, with angles and distances mixed.
By getting diverted into minutia, you are losing track of the issue.

Quoted message said:

Riders are unconsciously making endless small corrections, (even
when riding no-hands) much larger than a tenth of less than one
degree of steering motion, which is the 0.015 inch movement said to
be sufficient to replenish the protective oil layer under a headset
bearing.

That's what gets repeated often but these people apparently never
descend at speeds over 30mph where you have a hard time guessing
where any steering corrections are being made. As I said, I often
coast downhill riding in the width of the 3" wide edge stripe just for
the hell o9f it, just as I usually cut between pairs of Botts dots on
double yellow line marked roads. That's a slot of 2" taken at an
angle.

Quoted message said:

One easy way to demonstrate this is to try to hold one side of your
front tire within an inch of a highway stripe for 44 feet at 30 mph
(one second).

I do that for a 1/4 mile at least but that doesn't mean you need to do
that to avoid steering motions. Most people take a random azimuth to
road striping when coasting on empty roads and follow an equally
straight course, only on that is not parallel to any prescribed line.

Quoted message said:

A More difficult demonstration is to find a 30 mph roll-out through
a convenient puddle at the bottom of a hill after a rain. The
tracks on the dry pavement past the puddle will vary rather more
than an inch to both sides side of the initial line every 44 feet.

I don't get it. What is this about and what does it prove?

Quoted message said:

(For those who lack surveyor's equipment, a peep-sight rifle is a
handy tool, but pick a deserted road--bystanders tend be easily
alarmed. Because our muscles are shakier than we like to think,
sighting steadily even on an inch-wide target at 50 feet with a
24-inch sight radius (easier than a 17-inch handlebar) is so hard
that you'll quickly realize why surveyors use tripod rests and
cross-haired telescopic sights.)

How does the NRA keep getting into bicycle steering gears?

Quoted message said:

Closer at hand, pick up your keyboard, which is conveniently about
the width of touring handlebars. Brace its middle against a corner
of your desk, extending your arms in roughly a bicycling posture.
Do the ends of the keyboard move more than the thickness of a sheet
of paper?

This sounds like some office exerciser routine. I've heard other
things one should pick up and do lifting motions.

As I said, you don't steer when you descend fast, It's mainly no-hands
control that takes over even with hands on the bars.

[email hidden]