jim beam said:Gary Young said:jim beam said:[email hidden] wrote:
> John Dacey writes:
>
>>>>> What's failing? I'm pretty sure its not the bearing balls so
>>>>> harder ones would only increase contact stress on the races.
>>>> It isn't clear to me why switching from steel to ceramic bearing
>>>> balls would meaningfully increase contact stresses. If steel
>>>> balls were consistently stressed to levels where significant
>>>> deformation occurred in bicycle hub and crank bearing applications
>>>> (assuming proper preload adjustments), I think we'd see a much
>>>> higher incidence of spalling or other failure modes. Seems like a
>>>> red herring to me.
>>> OK, let's try an example. Let's replace the bearing balls with
>>> rubber ones. Do you think the races would wear faster or slower
>>> than with steel balls, assuming the rubber balls could support the
>>> load. The modulus of elasticity of ceramic balls is higher than
>>> that of steel balls and much higher than that of rubber balls.
>> If it were actually possible to have a rubber bearing ball that had
>> the same load bearing and durability as its steel counterpart, I see
>> no reason that there'd be any difference in the wear rate of the
>> race. Unless your rubber ball were equal in those qualities
>> however, suggesting such a comparison is misleading.
> I thin I made a mistake of not explaining that the ball makes point
> contact with the circular trough race and that the elastic contact,
> called Hertzian contact is defined by the elasticity and load on the
> ball. An infinitely hard ball and race would contact only in a point,
> as ball bearings generally do when bearing no load. It is a
> geometric consideration.
>
> The more elastic the bearing ball is for a fixed steel race, the
> larger the contact area and the lower the stress. Therefore, a steel
> ball exerts less stress when loaded than a ceramic one and a rubber
> ball far less than the steel ball. Trying another example, consider
> putting a soccer ball on your chest and having someone sit on it in
> contrast to a beach ball of the same size. The beach ball would
> spread the load over a far greater area.
>
>> The fact remains that steel bearing balls aren't regularly stressed
>> to the point of elastic deformation in normal bicycle hubs and
>> bottom brackets. Therefore, the contact patch of a steel ball and a
>> ceramic ball will be essentially the same. If they're the same
>> shape and share the same load, why would a ceramic one that just
>> happens to be a little harder cause more damage to the race on which
>> it rolls?
> I don't understand why you believe that. Elasticity is a continuum
> and deforms more with greater load. In fact, bicycle bearings are
> loaded more severely than one in automobiles
are you sure???
With which sentence do you disagree? The second sentence -- "Elasticity is
a continuum and deforms more with greater load" -- is more important than
the final sentence.
i question the second - the first is obvious.
I'm guessing that what you mean is that the second sentence is obvious and
you disagree with the third. If so, that helps point up exactly where the
disagreement lies. Ultimately, I don't think it matters much to Jobst's
argument whether bicycle bearing see higher loading than those in
automobiles.
Quoted message said:
Quoted message said:I would be surprised to see you disagree with the second sentence,
since you yourself have said that bearings deform under load in
discussions of the causes of headset indexing.
eh? car bearings deform under impact too. do you work on cars much?
Whether I work on cars or not is irrelevant to this discussion since we're
talking about things that aren't evident to the naked eye. I bring up this
point only to note that this is a kind of rhetorical trick that you rely
on fairly often.
In any event, if you look again, I think you'll see that you were
misconstruing which sentence I was referring to. I don't dispute that car
bearings bear loads.
Quoted message said:
Quoted message said:Quoted message said:> and get away with it
> because they turn so slowly that the accumulate cyclic stress cycles
> many times slower than most machines that rotate at over 1000rpm.
>
>>>> Besides, wouldn't the greater asperity of steel balls over
>>>> ceramics count for something when it came to measuring wear on the
>>>> races?
>>> I don't know what you mean by asperity,
>> With respect, many here think that's a topic in which you have more
>> than a little expertise.
>>> steel bearing balls being optically smooth. I don't know what the
>>> surface smoothness of ceramic balls is but it is certainly not
>>> abrasive.
>> As smooth as a premium steel bearing ball may be, it is my
>> understanding that ceramics can be even more free of surface
>> irregularities. It strikes me that the smoother the face of the
>> ball, the less it will abrade the surface on which it rolls.
> Bearing balls run with an oil film between race and ball,
not at bicycle rpm's they don't. especially not at bb rpm's.
http://tribology-abc.com/calculators/image/e12_3a.gif from
http://tribology-abc.com/calculators/e12_3.htm
> contact is
> made by impurities of external grit.
no, contact is made by asperities between the ball and race. there is
no elasto-hydrodynamic separation for most of a bicycle bearing's
existence. see above. grit or contamination can add to that, but
primary contact is from the elements themselves.
It's not clear to me why, on your account, oil or grease is of any
benefit at all at "bicycle rpm's." The references I've glanced at on
"boundary lubrication" indicate that while it is not an ideal form of
lubrication, it is nonetheless still a kind of lubrication and results
in the interposition of oil between moving parts. The oil may be fairly
tightly bonded to metal, and thus not perfectly fluid, but it's still
there, or so I gather. Is anything more than that required by Jobst's
account?
eh? gary, with respect, i have no idea where you draw an inference that
lubrication is pointless. jobst wants to imply that bearings are
hydrodynamically separated.
Where does Jobst imply that bearings are hydrodynamically separated?
Please be specific.
If oil doesn't create hydrodynamic separation at bicycle rpm's, then what
does it do that benefits bicycle bearings? My understanding is that it
creates a different, less optimal, form of separation between the ball and
race called boundary lubrication. As I said before, it's not clear to me
that you take into account boundary lubrication. You imply that if there's
no hydrodynamic separation, then the only alternative is full
metal-on-metal contact.
I you do recognize that boundary lubrication takes place in slow-moving
bearings, then why doesn't that satisfy Jobst's claim that the parts are
separated by a film of oil?
Quoted message said:truth is, they're not on bikes because
their speed isn't high enough. stick to the point.
I hope it's clearer to you now that I have been sticking to the point.
Quoted message said:
Quoted message said:Headsets generally rotate less and at lesser speeds than hubs or bottom
brackets and yet you've frequently said that the type of metal-to-metal
fretting that Jobst points to as the source of indexing should not
occur in a well-lubricated headset. How do you jibe that position with
what you've written here?
read what i wrote above!
Let me try to be more explicit about the contradiction I see. On the one
hand, you've often claimed, during the course of discussions of
headset damage, that lubrication is sufficient to protect bearings from
metal-on-metal contact EVEN WHEN THERE'S LITTLE MOVEMENT OF THE BEARING
(e.g., when a car is being transported on a railcar).
On the other hand, you now suggest that there's no oil separating
balls and races except at high speeds.
Is there any way I could possibly make the contradiction -- or tension, if
you will -- any more explicit than that?
Quoted message said:Quoted message said:Quoted message said:> Beside that,, bearing races are
> usually 100x more rough than bearing balls.
are you sure? there's not 100x between 1 micron and 0.25 micron, the
optical "smooth" transition.
>>> Back to the original question: What is failing?
>> We are failing to get an answer to the simple question of whether
>> ceramic bearing balls offer any better or worse performance for
>> bicycles. The matter is too often deflected by ascetic issues
>> regarding their cost before resolution on the technical merits.
> This whole subject arose because bearings were failing. I ask again,
> what part of the bearing is failing?
>
> As for ceramic bearing balls, they offer no advantages and give
> marginally poorer wear life because they cause higher race stresses.
too much of a generalization. hybrids can have longer lives in
certain conditions.
Do those conditions resemble anything seen in a bicycle?
what would be /your/ criteria gary? there's lots to choose from. tell
me what's bugging you about an overly generalist statement followed by a
generalist response.
Here's the point -- for purposes of the present discussion, it doesn't
matter whether some hybrids are superior under certain conditions if those
conditions don't obtain in bicycle bearings.
As a general matter, I think you could put your knowledge to better use
in this newsgroup if you would pay more attention to the relevance of the
facts you deploy. Knowing obscure facts about outliers and border
conditions is obviously a good thing, but reciting those facts when
they're not relevant to the discussion merely muddies the waters.
Surely that's not your intention.
Quoted message said:
Quoted message said:Quoted message said:[very much more expensive] all-ceramics can have even
greater lives than that.
