Changing over to ceramic bearings.
G8RRPH
ceramic bearings? How hard would it be if one were to get the properly
sized sealed bearing to do this one their own? I used to change the
bearing on my inline skates and skateboards myself, so I think
technically it should be pretty straight forward. I have tried to find
info on the web about what size the bearings are in the bottom bracket,
the guide and tension pulleys, and of course the wheelsets, though I
think I can figure those out, it would be nice to see the specs in
writing form the manufacturer, and the Shimano website's exploded view
does not give this. any advice on if this is worth the effort, and any
info on bearing sizes would be greatly appreciated.
Thanks
Dave
RonSonic
For what they sell those bearings for they should be able to do a bit of
research for you.
Ron
Ron Ruff
G8RRPH wrote:
> I was wondering if anyone has retrofitted their bike to aftermarket
> ceramic bearings?
If you want specific info, it would help if you would offer some.
BTW... if you changed all your bearings over to ceramic you'd save a
small fraction of a percent in power. Is it worth the money and hassle
to you?
Johnny Sunset aka Tom Sherman
Replacing all the steel bearings on a bicycle with ceramic bearings
will increase the average riding speed by 0.0001487645676%!
--
Tom Sherman - Here, not there.
I am supporting cannibalism by eating more nuts.
Ted Bennett
It's not worth the effort. A properly adjusted and lubricated bearing
makes only a tiny addition to the rolling resistance of the bike and
rider in motion. A switch to ceramic will not increase your speed one
bit, nor will it make pedaling any easier.
But they will cost more.
Ted
--
Ted Bennett
richard
their ceramikc BB. It did not last nearly as long as the original...
(Sorry, no numbers to give you.)
R Brickston
wrote:
>I know of only one case - a guy blew out his FSA BB and decided to try
>their ceramikc BB. It did not last nearly as long as the original...
>(Sorry, no numbers to give you.)
I thought long life was one of the features besides the slightly lower
friction. One article claimed that if you put just one or two ceramics
in with the steels, the ceramics because they are so hard, crush any
foreign particles into a fine dust, making the whole assembly last a
lot longer.
jobst....@stanfordalumni.org
>>> I was wondering if anyone has retrofitted their bike to
>>> aftermarket ceramic bearings? How hard would it be if one were to
>>> get the properly sized sealed bearing to do this one their own? I
>>> used to change the bearing on my inline skates and skateboards
>>> myself, so I think technically it should be pretty straight
>>> forward. I have tried to find info on the web about what size the
>>> bearings are in the bottom bracket, the guide and tension pulleys,
>>> and of course the wheelsets, though I think I can figure those
>>> out, it would be nice to see the specs in writing form the
>>> manufacturer, and the Shimano website's exploded view does not
>>> give this. any advice on if this is worth the effort, and any
>>> info on bearing sizes would be greatly appreciated.
>> For what they sell those bearings for they should be able to do a
>> bit of research for you.
> Replacing all the steel bearings on a bicycle with ceramic bearings
> will increase the average riding speed by 0.0001487645676%!
... while shortening the life of the bearing races as their stress
increases from a smaller contact patch with the harder bearing balls.
And, what is the purpose of this exercise?
Jobst Brandt
Johnny Sunset aka Tom Sherman
To make a point by facetiously posting a fictional improvement with
implied accuracy to thirteen (13) decimal places. ;)
I believe all my bicycles and trike have steel bearings, unless
Shimano, Phil Wood, Ritchey and Cane Creek have been using something
else.
The crappy stock wheel bearings in my push scooter are either steel or
"pot metal".
RonSonic
<sunset...@yahoo.com> wrote:
I wouldn't have thought that much.
Ron
Johnny Sunset aka Tom Sherman
Ceramic headset bearings are especially important for speed. ;)
John Dacey
On 16 Sep 2006 10:48:17 -0700, "G8RRPH" <g8r...@gmail.com> wrote:
I see the usual roster of rbt's floccinaucinihilipilificants has
thoughtfully decided the worthiness of this issue for you.
On the off chance that you elect to pursue this further however, be
aware that Shimano hubs don't use cartridge style sealed bearings.
You'll need to acquire eighteen 1/4" ceramic bearing balls (nine for
each rear cup) and twenty 3/16" bearing balls (ten for each front cup)
for most Shimano hubsets. Dura-Ace front hubs have eleven balls in
each cup, so make the adjustment if needed.
If you have a current model premium crank with external bearings
(Shimano calls them "Hollowech II"), you'll need to find ceramic
equivalents for the 6805 size sealed unit fitted in each bottom
bracket cup as standard.
You'll also find that the bearings in Shimano's derailleur pulleys
don't readily lend themselves to removal/replacement. I think there
are aftermarket replacement pulleys already fitted with ceramic
bearings that are available however. Alternately, be advised that the
plain bearing guide pulley that comes as standard equipment on many
Shimano derailleurs (e.g., Ultegra) already spins on ceramic sufaces.
-------------------------------
John Dacey
Business Cycles, Miami, Florida
Since 1983
Our catalog of track equipment: online since 1996
Phone: 305-273-4440
http://www.businesscycles.com
-------------------------------
Qui si parla Campagnolo
About 5 years ago I knew a gent that got me sets of ceramic ball
bearings in the popular 'bike' sizes..1/4, 3/16, 7/32 and
5/32...installed them into hubs, headsets, BBs....not very expensive
and geee....they didn't make any difference...c'mon people..reality
check!!!
G8RRPH
technology that is being used by the pro peloton, is popular among
remote control car and aviation enthusiasts, and is used in auto
racing. I do not believe these groups and the resources at their
disposal would use ceramic if it didn't make a difference. As prompted
by the first post, I did a more thorough search and sir, you are
correct! After googling ceramic ball bearings, one of the first sites
has everything laid out for you to pick based on your equipment,
leading me to believe I am not the only one with this odd thought.
My guess is the performance increase from using these is not going to
help as much as more training, and the cost is 2-3 times that of steel
bearings, (for one 6805 sealed cartridge, ceramic is about $60 US,
steel is about $30, and about $10 US for 20 1/4" ceramic bearings,
(thanks John)) and considering in all likelihood you will not have to
replace them, it doesn't seem like a bad investment. I know everyone
gets a chuckle when a newbie on a nice bike gets smoked by someone on a
80's era Scwinn LeTour, but I believe in using all tools and technology
at my disposal to have a great time. If I am willing and able to dump
cash into my bike, and if it gets me out more, whether from guilt or
whatever, then I am all the better.
Johnny Sunset aka Tom Sherman
G8RRPH wrote:
> I had not expected the cynicism on a new, now getting cheaper,
> technology that is being used by the pro peloton, is popular among
> remote control car and aviation enthusiasts, and is used in auto
> racing. I do not believe these groups and the resources at their
> disposal would use ceramic if it didn't make a difference.
What may make a difference in a small engine that has operating speeds
in the tens of thousands of RPM, or RC racing cars where the wheels may
spin at thousands of RPM, is not going to make a significant difference
in a bicycle. For example, a top UCI rider in a time trial will have
wheels that are turning at less than 400 RPM on the flat sections.
You could expect a similar performance improvement by drilling
lightening holes in all the cable crimps [1] on your bicycle.
[1] <http://harriscyclery.net/merchant/370/images/large/CA4104.jpg>.
[2]
[2] Yes, Ozark Bicycle, I am a shill.;)
R Brickston
<sunset...@yahoo.com> wrote:
I read somewhere that the slower the speed of the device, the more
effect less friction will have.
R Brickston
<pe...@vecchios.com> wrote:
I think that the technology is a little different now. Here an excerpt
from a Cycling News article: (Disclaimer, I'm not promoting ceramics
and don't have a clue. Maybe it's just FSA marketing bs. If they just
gotta have those low spoke count botique wheels, maybe they "deserve"
the ceramics as well)
http://www.cyclingnews.com/sponsors/fsa/2006/2006.php
FSA's Claudio Marra explained to Cyclingnews that "our testing shows
that ceramic ball bearings are 20% more rigid, 10X rounder and 60%
lighter than standard steel ball bearings, which reduces friction. We
have found that using ceramic ball bearings can provide speed
increases of up to 4%, or a forty meters distance gained over every
kilometer."
carl...@comcast.net
<rb20170REMOVE.yahoo.com@> wrote:
Dear R.,
That might explain why the maximum effect of the improved bearings is
noticed when the bicycle is standing still. :)
Here's a note on the kind of poorly done testing that finds amazing
advantages to such bearings:
***
>http://www.rouesartisanales.com/6-categorie-94458.html#english
>
>In the test, the same bicycle and rider rolled down a 3.25% grade (the
>degree sign is a misprint) twenty times, ten with ordinary wheel
>bearings, ten with ceramic bearings, and found that the ceramic
>bearing runs were 3.8% faster over a roughly 55-second, 500 meter
>course, 47.3 kmh top speed versus 45.4 kmh.
>
>Before I spent hundreds of dollars on ceramic wheel bearings, I'd want
>to see a drum spin-down test.
>
>If the wind on the half-kilometer hill changed 1.9 kmh after the first
>ten runs, which would take half an hour to do, then that wind change
>would account for the difference--and a 1.2 mph wind speed difference
>is pretty much undetectable.
>
>Then there's the double-blind problem. How did the bicycle start
>rolling? A tiny difference in the initial push makes a surprising
>difference in coasting tests. And how closely did the rider duplicate
>his tuck from the start to the finish, knowing that he was looking for
>improvements on the ceramic bearings?
http://groups.google.com/group/rec.bicycles.misc/msg/68be537fcb5913ce
***
To put that "test" in perspective, here's a speed calculator:
http://www.kreuzotter.de/english/espeed.htm
For the default rider on the drops to reach the 45.4 kmh of the
ordinary bearings on a level surface, it predicts 422 watts are
needed.
Reaching the 47.3 kmh claimed for the ceramic bearings requires 452
watts.
That's a 50 watt difference that the "test" claims to find in improved
bearings--as much as five 10-watt bicycle-light generators.
Cheers,
Carl Fogel
R Brickston
I think a racer would want to see results from an independent and
scientific test before making a decision. Rooting around on this
subject I saw a bottle of 100 1/4's for $600.
carl...@comcast.net
<rb20170REMOVE.yahoo.com@> wrote:
Dear R.,
Note the weasel-wording of "up to 4%"--it's very hard to come up with
plausible figures.
In 2006, the Tour leaders averaged about 27 mph. If they went 4%
faster, they'd have been going 28.08 mph.
Here's a speed calculator:
http://www.kreuzotter.de/english/espeed.htm
For the default rider on the drops, 27 mph takes 375 watts.
(The Tour riders do a lot of drafting to get around this kind of raw
power requirement.)
To go 28.08 mph (4% faster) takes 417 watts, 11.2% more power.
(Wind drag is a beast--as speed rises, even tiny speed increases
require larger and larger power increases.)
It's unlikely that ordinary wheel and transmission bearings are
wasting 37 watts of power that the ceramic bearings could save.
After all, a good chain and set of gears will be 95% efficient and
lose less than 20 watts at the 375~400 watt level.
Cheers,
Carl Fogel
John Dacey
On 17 Sep 2006 08:52:55 -0700, "G8RRPH" <g8r...@gmail.com> wrote:
>My guess is the performance increase from using these is not going to
>help as much as more training, and the cost is 2-3 times that of steel
>bearings, (for one 6805 sealed cartridge, ceramic is about $60 US,
>steel is about $30, and about $10 US for 20 1/4" ceramic bearings,
>(thanks John))
Just a footnote to my earlier reply about bearing size: Dura-Ace model
7800 rear hubs use fourteen 3/16" balls in each cup. All their other
rear models have nine 1/4" balls per side as noted previously.
> If I am willing and able to dump
>cash into my bike, and if it gets me out more, whether from guilt or
>whatever, then I am all the better.
I concur.
Johnny Sunset aka Tom Sherman
Maybe as a percentage, but overall?
To use some hypothetical numbers, which has the greater overall effect:
a 1% reduction in friction on a wheel turning 200 RPM or a 0.1%
reduction on an engine at 40,000 RPM [1]?
[1] Not unheard of for RC aircraft and car engines.
Bruce Gilbert
"G8RRPH" <g8r...@gmail.com> wrote in message
news:1158508375.7...@h48g2000cwc.googlegroups.com...
One of my friends runs an industrial bearing supply down here in Houston. He
has a word of caution about mixing ceramic with metal races. According to
him, the ceramic will wear at a slower rate than will the metal causing the
race to wear faster than the bearings. He said that you want the bearings to
wear out faster than the races. That way you replace bearings instead of
hubs... food for thought.
Bruce
>
R Brickston
Yeah, but the claim is made that (supposedly) the ceramics last 5 to
20x over metal. Can the hardened races keep up?
Ron Ruff
R Brickston wrote:
> FSA's Claudio Marra explained to Cyclingnews that "our testing shows
> that ceramic ball bearings are 20% more rigid, 10X rounder and 60%
> lighter than standard steel ball bearings, which reduces friction. We
> have found that using ceramic ball bearings can provide speed
> increases of up to 4%, or a forty meters distance gained over every
> kilometer."
I thought that was funny nice all the *real* studies show very little
loss even with regular bearings... something like <1% in power. That
would be more like 0.4% in speed... but only if the ceramics reduced
the friction to *zero*... which they don't. If we are generous and say
they cut the friction in half (0.5% power), then it'd be more like 0.2%
in speed... so now you can TT at 28.056mph instead of 28.000mph.
This is something only top pros or riders with too much money need to
be concerned about.
John Forrest Tomlinson
>and the cost is 2-3 times that of steel
>bearings,
That's less a difference than I'd have thought.
--
JT
****************************
Remove "remove" to reply
Visit http://www.jt10000.com
****************************
John Dacey
On Sun, 17 Sep 2006 23:40:04 GMT, R Brickston
<rb20170REMOVE.yahoo.com@> wrote:
>Yeah, but the claim is made that (supposedly) the ceramics last 5 to
>20x over metal. Can the hardened races keep up?
That ceramic bearing balls are harder than steel cones and cups
doesn't mean that they'll cause those races to deteriorate any sooner
than they would using premium (e.g., grade 25) steel balls. In fact,
to the extent that ceramic balls are generally regarded to have
smoother surfaces, better sphericity and better corrosion resistance
than their steel counterparts, hub race and cone life seems likely to
be longer when fitted with ceramics.
R Brickston
<usenet...@jt10000.com> wrote:
>On 17 Sep 2006 08:52:55 -0700, "G8RRPH" <g8r...@gmail.com> wrote:
>
>>and the cost is 2-3 times that of steel
>>bearings,
>
>That's less a difference than I'd have thought.
This can't be right, 1/4 ceramics by themselves are $600 per 100. How
many pennys each are they in steel?
frkr...@gmail.com
John Dacey wrote:
>
>
> On 17 Sep 2006 08:52:55 -0700, "G8RRPH" <g8r...@gmail.com> wrote:
>
>
> > If I am willing and able to dump
> >cash into my bike, and if it gets me out more, whether from guilt or
> >whatever, then I am all the better.
>
> I concur.
Then you'll both want to invest in the new self-adhesive racing stripes
I've just developed! My own tests show that their carefully calibrated
roughness can cause them, when properly applied, to energize boundary
layer flow and reduce air drag of vertical bicycle tubes by up to 2%.
And they're red, which everyone knows is important for psychological
boost, plus opponent intimidation!
The beneficial effects are guaranteed to be present, even when
undetectable, just like ceramic bearings!
Cost is only $99.99 per bike. Not available in stores. Just send your
credit card info, bank account numbers, social security number, all
passwords, and mother's maiden name to:
www.racers_buy_it_so_it_MUST_work.com
- Frank Krygowski
John Dacey
The fractional performance enhancements offered by lighter weight
bikes, aero wheels and components that lower rolling resistance are
often of little import to riders outside venues of competition. That
you may not find the cost of exotic componentry justified for the kind
of riding you do doesn't mean that all riders will share the way you
structure priorities for yourself. I find it curiously inconsistent
that you lobby so vociferously for the right to choose whether to wear
a helmet,yet are so ready to hold up to ridicule those who don't
conform to your values in the instant matter.
Maybe somebody should seek legislation to make ceramic bearings
mandatory. . .
****************************
A picture I took:
http://www.shutterbugs.biz/cgi-bin/blackandwhite.pl?potd_date=2006-09-18
David L. Johnson
>>>and the cost is 2-3 times that of steel
>>>bearings,
>>
>>That's less a difference than I'd have thought.
>
> This can't be right, 1/4 ceramics by themselves are $600 per 100. How
> many pennys each are they in steel?
5 to 10. That is, $5 to $10 per hundred. We are talking 100 times the
price of steel.
Bruce Jackson
> Yeah, but the claim is made that (supposedly) the ceramics last 5 to
> 20x over metal. Can the hardened races keep up?
You must be talking about hybrid ceramic bearings (ceramic balls in
steel races).
The high moot bearings are full ceramic. In these both the balls and
races are ceramic.
FWIW, I still have the same set of steel 6001 bearings in my front
Bullseye hub that I had 20 years ago. I overhaul them every now and
then and they seem to last forever. If ceramic bearings last 5-20x as
long than I should be able to hand my hubs down to my great grand
children if I upgraded them to ceramic.
I've never seen a full ceramic bearing with seals. If I could find a
sealed version I might actually invest a couple of c notes in buying a
pair just for the heck of it.
dvt
> Maybe somebody should seek legislation to make ceramic bearings
> mandatory. . .
Imagine the bandwidth required to discuss that!
> A picture I took:
> http://www.shutterbugs.biz/cgi-bin/blackandwhite.pl?potd_date=2006-09-18
Congrats! Nice photo.
You named your dog Keirin?
--
Dave
dvt at psu dot edu
Everyone confesses that exertion which brings out all the powers of body
and mind is the best thing for us; but most people do all they can to
get rid of it, and as a general rule nobody does much more than
circumstances drive them to do. -Harriet Beecher Stowe, abolitionist and
novelist (1811-1896)
G8RRPH
FWIW, I was at Boca bearings dot com, they have:
1/4" silicon nitride (Si3N4) bearings, 5 for $5 US (grade 5)
1/4" alumina or zircon, 10 for $4 US (grade 10)
5/16" Si3N4, $9 for 5, Al and Zr, $7 for 10
The 6805 hybrid ceramic balls in steel races, with seals, $119 for a
pair.
I am sure you could find better prices, but the cost does not seem to
be that much relative to what a new set of mid to high level wheels
will set you back.
Dave
G8RRPH
John Everett
> I was wondering if anyone has retrofitted their bike to aftermarket
>ceramic bearings? How hard would it be if one were to get the properly
>sized sealed bearing to do this one their own? I used to change the
>bearing on my inline skates and skateboards myself, so I think
>technically it should be pretty straight forward. I have tried to find
>info on the web about what size the bearings are in the bottom bracket,
>the guide and tension pulleys, and of course the wheelsets, though I
>think I can figure those out, it would be nice to see the specs in
>writing form the manufacturer, and the Shimano website's exploded view
>does not give this. any advice on if this is worth the effort, and any
>info on bearing sizes would be greatly appreciated.
Have you heard of the Law of Diminishing Returns?
jeverett3<AT>earthlink<DOT>net http://home.earthlink.net/~jeverett3
John Dacey
>Congrats! Nice photo.
>
>You named your dog Keirin?
Thank you. Yes, that's the dog's name. The event's connotation of
controlled roughhouse seemed apt for a bullmastiff. He's old enough to
well predate the co-option of keirin-related stuff by certain segments
of the fix-gear road crowd.
-------------------------------
John Dacey
Business Cycles, Miami, Florida
Since 1983
Comprehensive catalogue of track equipment: online since 1996
http://www.businesscycles.com
-------------------------------
Kinky Cowboy
wrote:
>R Brickston wrote:
>
>I've never seen a full ceramic bearing with seals. If I could find a
>sealed version I might actually invest a couple of c notes in buying a
>pair just for the heck of it.
I'm slightly tempted by seal-less and lube-less zirconia full ceramics
in the hubs of my next project, which won't be seeing any rain. I
suspect the elimination of seal and grease drag will be more
significant than the difference in rolling resistance at 3-400rpm
between ordinary steel bearings and ceramics.
There was a fashion, before ceramic bearings were ever mentioned in
cycling circles, for running your hub bearings (all loose ball cup and
cone in those days) with just a little light oil instead of grease in
your 'best' wheels, i.e. the ones you only used for dry races on fast
courses.
Kinky Cowboy*
*Batteries not included
May contain traces of nuts
Your milage may vary
frkr...@gmail.com
John Dacey wrote:
> On 17 Sep 2006 20:58:57 -0700, frkr...@gmail.com wrote:
>
> >
> >John Dacey wrote:
> >>
> >>
> >> On 17 Sep 2006 08:52:55 -0700, "G8RRPH" <g8r...@gmail.com> wrote:
> >>
> >>
> >> > If I am willing and able to dump
> >> >cash into my bike, and if it gets me out more, whether from guilt or
> >> >whatever, then I am all the better.
> >>
> >> I concur.
> >
> >Then you'll both want to invest in the new self-adhesive racing stripes
> >I've just developed! My own tests show that their carefully calibrated
> >roughness can cause them, when properly applied, to energize boundary
> >layer flow and reduce air drag of vertical bicycle tubes by up to 2%.
> >And they're red, which everyone knows is important for psychological
> >boost, plus opponent intimidation!
> >
> >The beneficial effects are guaranteed to be present, even when
> >undetectable, just like ceramic bearings!
> >
> >Cost is only $99.99 per bike. Not available in stores. Just send your
> >credit card info, bank account numbers, social security number, all
> >passwords, and mother's maiden name to:
> >www.racers_buy_it_so_it_MUST_work.com
> >
> >- Frank Krygowski
> I find it curiously inconsistent
> that you lobby so vociferously for the right to choose whether to wear
> a helmet,yet are so ready to hold up to ridicule those who don't
> conform to your values in the instant matter.
>
> Maybe somebody should seek legislation to make ceramic bearings
> mandatory. . .
No, no, no, that would be a travesty. What I want is to make my newly
invented, self-adhesive racing stripes mandatory! ;-)
But seriously, your "curiously inconsistent" statement is odd indeed.
The consistency is this: for both ceramic bearings and styrofoam hats,
people are ignoring technical truths and measured evidence. They are
imagining (or pretending) that a microscopic improvement is likely to
have significant effect.
It really is possible to either calculate or measure the amount of
benefit you'll get from either of these talismans. Ditto from, say,
carbon fiber seat stays in place of steel ones; or an aerodynamic front
derailleur, or drilling holes in the pulley cage of your Campy
derailleur - to recall "magic" fashions of the past.
>
> The fractional performance enhancements offered by lighter weight
> bikes, aero wheels and components that lower rolling resistance are
> often of little import to riders outside venues of competition. That
> you may not find the cost of exotic componentry justified for the kind
> of riding you do doesn't mean that all riders will share the way you
> structure priorities for yourself.
Or, more succinctly, "you pays your money and you takes your choice."
And I agree, people should be free to waste ... um, make that "spend"
their money on whatever they like, within reason. (And they should not
be forced to buy products they don't want or need.)
But this is a technical bicycle discussion group. If it becomes
forbidden to point out that certain components are of negligible value,
the group will have much less value.
Now, do you want a set of my aero-enhancing racing stripes or not?
BTW, they are also resistant to impacts from SUVs, and they're high in
vitamin C. ;-)
- Frank Krygowski
Jay Beattie
30 years ago, I rode seal-less Campagnolo NR hubs (I don't count those
useless dust caps as seals) with very light grease. I was much faster
back then. QED. -- Jay Beattie.
jobst....@stanfordalumni.org
>>>>> If I am willing and able to dump cash into my bike, and if it
>>>>> gets me out more, whether from guilt or whatever, then I am all
>>>>> the better.
>>>> I concur.
>>> Then you'll both want to invest in the new self-adhesive racing
>>> stripes I've just developed! My own tests show that their
>>> carefully calibrated roughness can cause them, when properly
>>> applied, to energize boundary layer flow and reduce air drag of
>>> vertical bicycle tubes by up to 2%. And they're red, which
>>> everyone knows is important for psychological boost, plus opponent
>>> intimidation!
>>> The beneficial effects are guaranteed to be present, even when
>>> undetectable, just like ceramic bearings!
>>> Cost is only $99.99 per bike. Not available in stores. Just send
>>> your credit card info, bank account numbers, social security
>>> number, all passwords, and mother's maiden name to:
>>> www.racers_buy_it_so_it_MUST_work.com
>> I find it curiously inconsistent that you lobby so vociferously for
>> the right to choose whether to wear a helmet, yet are so ready to
>> hold up to ridicule those who don't conform to your values in the
>> instant matter.
>> Maybe somebody should seek legislation to make ceramic bearings
>> mandatory...
> No, no, no, that would be a travesty. What I want is to make my
> newly invented, self-adhesive racing stripes mandatory!
> But seriously, your "curiously inconsistent" statement is odd
> indeed. The consistency is this: for both ceramic bearings and
> Styrofoam hats, people are ignoring technical truths and measured
> evidence. They are imagining (or pretending) that a microscopic
> improvement is likely to have significant effect.
> It really is possible to either calculate or measure the amount of
> benefit you'll get from either of these talismans. Ditto from, say,
> carbon fiber seat stays in place of steel ones; or an aerodynamic
> front derailleur, or drilling holes in the pulley cage of your Campy
> derailleur - to recall "magic" fashions of the past.
Known as the wonder material "drillium" that greatly improved raceing
performance... and we are all racers in the first place, so that
applies to US. It reminds me of the poor folks who believed in the
"Trickle down theory" and voted for the man... or more recently the
tax cuts that gave us deferred road maintenance and education that's
fit for no one, among other benefits. Self deception!
>> The fractional performance enhancements offered by lighter weight
>> bikes, aero wheels and components that lower rolling resistance are
>> often of little import to riders outside venues of competition.
>> That you may not find the cost of exotic componentry justified for
>> the kind of riding you do doesn't mean that all riders will share
>> the way you structure priorities for yourself.
> Or, more succinctly, "you pays your money and you takes your
> choice." And I agree, people should be free to waste... um, make
> that "spend" their money on whatever they like, within reason. (And
> they should not be forced to buy products they don't want or need.)
So in the recent past we have had the aluminum wave, the titanium
wave, the carbon fiber wave, and now the ceramic bearing wave. Of
course there are minor ones like tubular tires that are better than
clinchers for racing, and besides, we are all racers as if that
mattered.
> But this is a technical bicycle discussion group. If it becomes
> forbidden to point out that certain components are of negligible
> value, the group will have much less value.
Are you knocking religion? If these folks believe in this and
proselytize about its merits, we need to imprison them and torture
them until they confess. If the sink and drown then they weren't
witches, and if the float they are, and need to be burned at the
stake.
> Now, do you want a set of my aero-enhancing racing stripes or not?
> BTW, they are also resistant to impacts from SUVs, and they're high
> in vitamin C. ;-)
I'm for it!
Jobst Brandt
jim beam
> Tom Sherman writes:
>
>>>> I was wondering if anyone has retrofitted their bike to
>>>> aftermarket ceramic bearings? How hard would it be if one were to
>>>> get the properly sized sealed bearing to do this one their own? I
>>>> used to change the bearing on my inline skates and skateboards
>>>> myself, so I think technically it should be pretty straight
>>>> forward. I have tried to find info on the web about what size the
>>>> bearings are in the bottom bracket, the guide and tension pulleys,
>>>> and of course the wheelsets, though I think I can figure those
>>>> out, it would be nice to see the specs in writing form the
>>>> manufacturer, and the Shimano website's exploded view does not
>>>> give this. any advice on if this is worth the effort, and any
>>>> info on bearing sizes would be greatly appreciated.
>
>>> bit of research for you.
>
>> Replacing all the steel bearings on a bicycle with ceramic bearings
>> will increase the average riding speed by 0.0001487645676%!
>
> ... while shortening the life of the bearing races as their stress
> increases from a smaller contact patch with the harder bearing balls.
since when does hardness have anything to do with modulus? we're not
talking plastic deformation, we're talking elastic so it's stiffness
[you know, the straight line part of the stress/strain part of the
graph, the bit before yield] of the ball that determines contact patch
size, not hardness.
sheesh jobst, you're the one that likes to b.s. about hertzian bearing
deformation - check the math. shame you don't bother to read the books
you brag about owning.
>
> And, what is the purpose of this exercise?
>
> Jobst Brandt
jobst....@stanfordalumni.org
This comes from the manufacturer's web site (smaller contact area):
> sheesh jobst, you're the one that likes to b.s. about hertzian
> bearing deformation - check the math. shame you don't bother to
> read the books you brag about owning.
>> And, what is the purpose of this exercise?
I was expecting to hear that it made the rider faster, but maybe not.
Jobst Brandt
jim beam
for elastic deformation, smaller contact area is a function of modulus,
not hardness. ceramic rolling elements can indeed have a higher modulus
so they can have a smaller contact area, but it still doesn't follow
that bearing races have a shorter life due to their use. in fact hybrid
bearings usually enjoy longer life due to better surface wear
characteristics.
Peter Cole
> for elastic deformation, smaller contact area is a function of modulus,
> not hardness. ceramic rolling elements can indeed have a higher modulus
> so they can have a smaller contact area, but it still doesn't follow
> that bearing races have a shorter life due to their use. in fact hybrid
> bearings usually enjoy longer life due to better surface wear
> characteristics.
As Jobst often points out, ball bearings in bicycles operate in a low
speed, high load mode. Ceramic ball bearings are optimal for high speed,
low load operation -- a poor match even if there was some other marginal
benefit.
jim beam
bike bearings are not necessarily higher load than any other bearing, so
that excuse doesn't work. and there's no reason a ceramic bearing has
to be operated at high speed. they work well at high speed because the
rolling elements are typically lighter, but they work just as well at
low speed, if not better because there's less contact welding.
all this bleating is typical r.b.t irrational baseless fear of anything
new and hysterical distaste about expense. if no one wants them, they
don't have to buy them - ordinary bearings work fine. it's not like
anyone has a gun to anyone's head.
frkr...@gmail.com
jim beam wrote:
>
>
> all this bleating is typical r.b.t irrational baseless fear of anything
> new and hysterical distaste about expense. if no one wants them, they
> don't have to buy them - ordinary bearings work fine. it's not like
> anyone has a gun to anyone's head.
Wrong. This discussion is not about fear. It's about irrational
belief that expensive, negligible differences will somehow magically
improve one's performance on a bicycle. We get this regularly, whether
the subject is rubber-filled holes in frames, carbon fiber that is both
rigid and flexible, any spoke count greater than one, etc.
Nobody is pretending either ceramic or steel bearings should be either
mandatory or forbidden. We are simply examining the science, and
realisitically evaluating claims.
And, as is often the case, all the "bleating" seems to be from you,
directed at Jobst. Give it a rest.
- Frank Krygowski
carl...@comcast.net
Dear Frank,
What are your thoughts on the effect of mixing ceramic balls with
metal races?
Will the races wear out noticeably more quickly?
There seems to be some disagreement here.
Cheers,
Carl Fogel
Alex
G8RRPH wrote:
> I was wondering if anyone has retrofitted their bike to aftermarket
> ceramic bearings? How hard would it be if one were to get the properly
> sized sealed bearing to do this one their own? I used to change the
> bearing on my inline skates and skateboards myself, so I think
> technically it should be pretty straight forward. I have tried to find
> info on the web about what size the bearings are in the bottom bracket,
> the guide and tension pulleys, and of course the wheelsets, though I
> think I can figure those out, it would be nice to see the specs in
> writing form the manufacturer, and the Shimano website's exploded view
> does not give this. any advice on if this is worth the effort, and any
> info on bearing sizes would be greatly appreciated.
I'm sure you could find the proper bearings. How easy it will be to do
the work depends on the parts. If you have loose bearings, swapping
them out is easy. If you have cartridge bearings, then it depends on
the design. Some cartridge bearings are easy to change, others require
specials tools. Is it worth the effort? Depends on how much
disposable income you have. If you can afford the ceramic bearings,
you can afford to pay someone else to do all the work for you. It;'s a
lot of money and work for really trivial gains. The difference in
friction between a decent steel ball and a ceramic ball is minimal.
------------------
Alex
Ron Ruff
jim beam wrote:
> all this bleating is typical r.b.t irrational baseless fear of anything
> new and hysterical distaste about expense. if no one wants them, they
> don't have to buy them - ordinary bearings work fine. it's not like
> anyone has a gun to anyone's head.
I think it is more the distaste of being relentlessly pummeled with
fashion driven "improvements" in cycling... and the marketplace in
general... and how following these trends has become the main reason to
get up in the morning for many people. If everyone road a bike (or
lived) for the sheer enjoyment of the experience, things would be very
different from what they are...
Pat Lamb
Er, I thought we were discussing a cost/benefit analysis of ceramic
bearings. The benefit is grams per bike, and the cost is tens of
dollars. As you just wrote, "ordinary bearings work fine," implying
there's no technical benefit to ceramic bearings. Ergo, the discussion
is narrowed to whether the weight benefit is worth the cost -- the
discussion reasonably devolved to this, so it's not "hysterical distaste
about cost."
Pat
Peter Cole
> Peter Cole wrote:
>> As Jobst often points out, ball bearings in bicycles operate in a low
>> speed, high load mode. Ceramic ball bearings are optimal for high
>> speed, low load operation -- a poor match even if there was some other
>> marginal benefit.
>
> bike bearings are not necessarily higher load than any other bearing, so
> that excuse doesn't work.
<http://en.wikipedia.org/wiki/Spalling>
"Spalling describes surface failure in which chips are shed from a
contact point, for example, in a ball bearing. This is due to the
maximal shear stress being not at the surface but just below it leading
an overloaded bearing to loose flakes rather than brinelling"
Wheel cones and spindle faces wear out on my bikes via spalling, so I
conclude they are overloaded, I don't know about yours.
> and there's no reason a ceramic bearing has
> to be operated at high speed.
I never said so.
> they work well at high speed because the
> rolling elements are typically lighter, but they work just as well at
> low speed, if not better because there's less contact welding.
Better than steel (or better than ceramic at high speed)?
The question is whether they would induce extra stress in the race
because of their higher modulus (smaller contact patch).
<http://www.tribology-abc.com/abc/hybrid.htm>
"Ceramic balls
The centrifugal load is described by F= -mT2R where m is the weight of
the ball. Because the density of silicium nitride is 60% of steel, the
centrifugal load is reduced by using these balls with 60%. As a
consequence the running speed can be increased up to 30%.
Because the E-modulus of silicium nitride is about 1.5 times that from
steel, a smaller contact surface exists with the consequence of a high
contact pressure. So, the benefit of a small centrifugal load is limited
by the large E-modulus. From Hertz theory it follows that the maximum
load for the combination steel- silicium nitride reduced with 30%."
> all this bleating is typical r.b.t irrational baseless fear of anything
> new and hysterical distaste about expense. if no one wants them, they
> don't have to buy them - ordinary bearings work fine. it's not like
> anyone has a gun to anyone's head.
Engineers like to use the right tool for the job, that's all. Bikes are
a bad application for ceramic balls even if the cost is ignored.
G8RRPH
I can afford the around $200 to retrofit my bike, but I like working on
my bike, cleaning it, tuning it etc., so if I am doing this anyway it's
not a big deal to make the switch. I am reentering cycling after a 8
year hiatus for a variety of reasons, so all is new again, and I am
seeking this groups advice, which has been very helpful (thnaks to
all). I think I will pick up a few for my wheels, change as needed,
then the bottom bracket if I see a noticeable difference from the
wheels.
Dave
jim beam
> jim beam wrote:
>> Peter Cole wrote:
>
>>> As Jobst often points out, ball bearings in bicycles operate in a low
>>> speed, high load mode. Ceramic ball bearings are optimal for high
>>> speed, low load operation -- a poor match even if there was some
>>> other marginal benefit.
>>
>> bike bearings are not necessarily higher load than any other bearing,
>> so that excuse doesn't work.
>
>
> <http://en.wikipedia.org/wiki/Spalling>
> "Spalling describes surface failure in which chips are shed from a
> contact point, for example, in a ball bearing. This is due to the
> maximal shear stress being not at the surface but just below it leading
> an overloaded bearing to loose flakes rather than brinelling"
that needs editing. yes, there is shear stress below the surface, but
spalling is a form of fatigue which can also be initiated by the tiniest
surface imperfections and scratches. that's why "extra polish" is an
available spec for bearings. it's also why vacuum degassed steels are
used extensively in these applications.
>
> Wheel cones and spindle faces wear out on my bikes via spalling, so I
> conclude they are overloaded, I don't know about yours.
probably, yes, but overloaded by what?
>
>
>
>> and there's no reason a ceramic bearing has to be operated at high speed.
>
> I never said so.
forgive me - i misunderstood your meaning when you said "Ceramic ball
bearings are optimal for high speed...".
>
>> they work well at high speed because the rolling elements are
>> typically lighter, but they work just as well at low speed, if not
>> better because there's less contact welding.
>
> Better than steel (or better than ceramic at high speed)?
ceramic can be better than steel at low speed because of less contact
welding.
>
> The question is whether they would induce extra stress in the race
> because of their higher modulus (smaller contact patch).
>
>
> <http://www.tribology-abc.com/abc/hybrid.htm>
>
> "Ceramic balls
> The centrifugal load is described by F= -mT2R where m is the weight of
> the ball. Because the density of silicium nitride is 60% of steel, the
> centrifugal load is reduced by using these balls with 60%. As a
> consequence the running speed can be increased up to 30%.
>
> Because the E-modulus of silicium nitride is about 1.5 times that from
> steel, a smaller contact surface exists with the consequence of a high
> contact pressure. So, the benefit of a small centrifugal load is limited
> by the large E-modulus. From Hertz theory it follows that the maximum
> load for the combination steel- silicium nitride reduced with 30%."
i'm glad you looked up that excellent site.
regarding loading, most bike bearings are operated /well/ inside safe
parameters, even for ceramic elements, so again, i really don't see the
problem.
>
>
>> all this bleating is typical r.b.t irrational baseless fear of
>> anything new and hysterical distaste about expense. if no one wants
>> them, they don't have to buy them - ordinary bearings work fine. it's
>> not like anyone has a gun to anyone's head.
>
> Engineers like to use the right tool for the job, that's all. Bikes are
> a bad application for ceramic balls even if the cost is ignored.
for what reason? it's not load, it's not speed, it's not friction, it's
not lifetime. so what is it other than cost or distaste?
and with respect, invocation of the term "engineers" is mighty thin ice.
some "engineers" feel free to presume proficiency on a number of
subjects on which they have no expertise whatsoever, so i don't think
there's mileage in trying to imply "engineering" application inadequacy.
frkr...@gmail.com
I suggest you find a way to do careful measurements of your speed
before and after the magic bearings. That is, don't just try to judge
a noticeable difference by feel - because believing buyers frequently
"feel" imaginary improvements.
Try, say, a multi-sample coast down test. Use very consistent riding
position, identical clothing, identical water bottle placement, zero
wind speed, and perfectly consistent temperatures - because any changes
in those parameters will overwhelm any difference from bearings.
And whatever you do, don't make any other changes to the bike. Like,
for example, adding aerodynamic racing stripes! ;-)
But do report back.
- Frank Krygowski
frkr...@gmail.com
carl...@comcast.net wrote:
>
>
> Dear Frank,
>
> What are your thoughts on the effect of mixing ceramic balls with
> metal races?
>
> Will the races wear out noticeably more quickly?
>
> There seems to be some disagreement here.
I'd hesitate to guess without seeing data. And I think it will take
quite a few data points to tell. Here's why:
Failure of bearings is a statistical phenomenon. Bearings are rated by
their L10 life - that's actually L(subscript 10) - which is the number
of revs that 90% of the bearings will last. For a given load, 10% will
fail earlier. Some will last much much longer. And the life of any
bearing is extremely load sensitive, inversely proportional to the cube
of the load.
Put this all together, and here's what you get: two "identical" riders
on "identical" bikes doing "identical" riding can produce much
different bearing lives. This can be caused by one rider being less
skillful at bunny-hopping potholes, or inflating his tires less
frequently, or by an anonymous steel mill worker having had a bad day,
etc.
Personally, I think these variations may be much more important than
the competing material differences - that is, the smaller contact patch
on the one hand, and the supposed reduction in adhesion on the other
hand. (I don't know much about the adhesion thing anyway.)
- Frank Krygowski
carl...@comcast.net
Dear Frank,
Thanks.
A stray thought occurs to me.
If the ceramic balls reduce adhesion enough, could they eliminate most
fretting? Or does micro-welding occur no matter what?
If the ceramic balls can't micro-weld to the metal races to cause
fretting, then any headset dimpling in a mixed bearing would be due to
impact.
Conversely, if no dimpling develops in mixed headsets, then that would
show that only fretting can cause dimples.
I doubt this theoretical test will survive practical criticism, but
I'm going to enjoy the illusion for a few moments that ceramic balls
in steel races could provide a clear demonstration of whatever causes
dimpled headsets--fretting, impact, or a combination of the two.
Cheers,
Carl Fogel
Peter Cole
It *is* load/lifetime, ceramic balls would require a 30% load de-rate.
For someone like myself who experiences spalling failures on wheel cones
and crank spindles this would reduce service life further.
As far as cost goes, ceramic balls cost about $5/g in weight savings,
not a bargain.
> and with respect, invocation of the term "engineers" is mighty thin ice.
> some "engineers" feel free to presume proficiency on a number of
> subjects on which they have no expertise whatsoever, so i don't think
> there's mileage in trying to imply "engineering" application inadequacy.
I'm talking philosophy, not proficiency. Most engineers are trained to
weigh trade-offs to find the best, of many possible, solutions. Ceramic
balls may be the best solution for some, but I'd have a hard time
determining who that would be.
jim beam
> jim beam wrote:
>> Peter Cole wrote:
>
>>> Engineers like to use the right tool for the job, that's all. Bikes
>>> are a bad application for ceramic balls even if the cost is ignored.
>>
>> for what reason? it's not load, it's not speed, it's not friction,
>> it's not lifetime. so what is it other than cost or distaste?
>
> It *is* load/lifetime, ceramic balls would require a 30% load de-rate.
> For someone like myself who experiences spalling failures on wheel cones
> and crank spindles this would reduce service life further.
ok, let's get away from the ceramic question for a moment: what are the
details of your failures? which cones? brand? model? front/rear?
and what about crank spindles? even with steel, if you're using decent
componentry and it's well adjusted, you should be getting reasonable
life. and what is the time frame for these failures?
fyi, spalling and abrasion are the commonest failure modes on a bike.
if you're keeping things clean and sealed, abrasion is not the problem,
so that leaves spalling. cheap components will spall very quickly - the
classic is cheapo case-hardened stuff which can fail in only a few
hundred miles.
dvt
> jim beam wrote:
>> Peter Cole wrote:
>
>>> Engineers like to use the right tool for the job, that's all. Bikes
>>> are a bad application for ceramic balls even if the cost is ignored.
>>
>> for what reason? it's not load, it's not speed, it's not friction,
>> it's not lifetime. so what is it other than cost or distaste?
>
> It *is* load/lifetime, ceramic balls would require a 30% load de-rate.
> For someone like myself who experiences spalling failures on wheel cones
> and crank spindles this would reduce service life further.
I'm smaller, and I rarely see spalling failures on hubs and bottom
brackets. But they do occur; I have a freehub on my 'cross bike with a
rough race, although it's not bad enough to warrant immediate replacement.
So if mid-weight riders like me see a few failures, and heavier riders
experience a higher failure rate, the load on the steel bearings is
approaching the upper limit. A bearing allowing 30% less load is not a
good choice for direct replacement if durability is desired.
On the other hand, all these bearings could be redesigned to reduce the
load. A larger bottom bracket shell would help. And it would be a simple
matter to enlarge the front hub and the left side of the rear hub. I
don't think it's a simple matter to fit larger bearings in a freehub,
though.
Summary: maybe it will make sense once the kinks get worked out, but I
suspect early adopters will pay the price of reduced durability.
--
Dave
dvt at psu dot edu
Everyone confesses that exertion which brings out all the powers of body
and mind is the best thing for us; but most people do all they can to
get rid of it, and as a general rule nobody does much more than
circumstances drive them to do. -Harriet Beecher Stowe, abolitionist and
novelist (1811-1896)
jobst....@stanfordalumni.org
>>> What are your thoughts on the effect of mixing ceramic balls with
>>> metal races?
>>> Will the races wear out noticeably more quickly?
>>> There seems to be some disagreement here.
>> I'd hesitate to guess without seeing data. And I think it will take
>> quite a few data points to tell. Here's why:
>> Failure of bearings is a statistical phenomenon. Bearings are
>> rated by their L10 life - that's actually L(subscript 10) - which
>> is the number of revs that 90% of the bearings will last. For a
>> given load, 10% will fail earlier. Some will last much much
>> longer. And the life of any bearing is extremely load sensitive,
>> inversely proportional to the cube of the load.
>> Put this all together, and here's what you get: two "identical"
>> riders on "identical" bikes doing "identical" riding can produce
>> much different bearing lives. This can be caused by one rider
>> being less skillful at bunny-hopping potholes, or inflating his
>> tires less frequently, or by an anonymous steel mill worker having
>> had a bad day, etc.
>> Personally, I think these variations may be much more important
>> than the competing material differences - that is, the smaller
>> contact patch on the one hand, and the supposed reduction in
>> adhesion on the other hand. (I don't know much about the adhesion
>> thing anyway.)
> A stray thought occurs to me.
> If the ceramic balls reduce adhesion enough, could they eliminate
> most fretting? Or does micro-welding occur no matter what?
> If the ceramic balls can't micro-weld to the metal races to cause
> fretting, then any headset dimpling in a mixed bearing would be due
> to impact.
> Conversely, if no dimpling develops in mixed headsets, then that
> would show that only fretting can cause dimples.
Fretting is not a operating feature of rolling element bearings and
not a cause of ball and race damage in a rotating bearing. Fretting
is a stationary effect caused by microscopic oscillating motion and
has nothing to do with bearing life as is being discussed here, We
are, after all, talking abut rotating bearings as in wheels and
pedals, not static head bearing damage while costing down hills, a
problem that has been solved by better bearing design.
> I doubt this theoretical test will survive practical criticism, but
> I'm going to enjoy the illusion for a few moments that ceramic balls
> in steel races could provide a clear demonstration of whatever
> causes dimpled headsets--fretting, impact, or a combination of the
> two.
Strange that you should bring that aspect into the mix, considering
that just the concept of fretting is foreign to the faithful in this
forum. You'd think that people like BMW and Daimler Benz would
immediately jump to ceramic ball bearings, as they apparently have the
most failures of differential bearings, judging from the number of
rushing rear axles I hear rolling by. These are outdone only by the
number of SUV's and pickup trucks with delaminating tires slapping
down the pike. Maybe, they also, should have ceramic tire treads.
Jobst Brandt
dvt
> jobst....@stanfordalumni.org wrote:
>> jim beam wrote:
>>> since when does hardness have anything to do with modulus? we're
>>> not talking plastic deformation, we're talking elastic so it's
>>> stiffness [you know, the straight line part of the stress/strain
>>> part of the graph, the bit before yield] of the ball that determines
>>> contact patch size, not hardness.
>> This comes from the manufacturer's web site (smaller contact area):
>>
>> http://tinyurl.com/khpwp
Interesting web site. They claim the ceramic bearings work to much
higher temperatures. They also point out that the coefficient of thermal
expansion (CTE) is much different in the balls and races. Wouldn't that
lead to changing preload with temperature? I'd think that matching CTE
between balls and races would be a good idea for use at a wide range of
temperatures.
> for elastic deformation, smaller contact area is a function of modulus,
> not hardness.
The web site in question implies that smaller contact area comes from a
combination of modulus, hardness, and Poisson's ratio.
> ceramic rolling elements can indeed have a higher modulus
> so they can have a smaller contact area, but it still doesn't follow
> that bearing races have a shorter life due to their use. in fact hybrid
> bearings usually enjoy longer life due to better surface wear
> characteristics.
Do you have a cite for this? I'd like to know the true cost/benefit
ratio. For example, let's say that ceramic bearings in bicycle-type
applications typically last 20x longer but cost 10x more, maybe they
have some justification for high mileage riders.
Peter Cole
> 3/16" SiN4 5 for $4, for the Zr and Al, 10 for $3.50, same grades.
>
>
> G8RRPH wrote:
>> FWIW, I was at Boca bearings dot com, they have:
>> 1/4" silicon nitride (Si3N4) bearings, 5 for $5 US (grade 5)
>> 1/4" alumina or zircon, 10 for $4 US (grade 10)
>>
>> 5/16" Si3N4, $9 for 5, Al and Zr, $7 for 10
>>
>> The 6805 hybrid ceramic balls in steel races, with seals, $119 for a
>> pair.
>>
>> I am sure you could find better prices, but the cost does not seem to
>> be that much relative to what a new set of mid to high level wheels
>> will set you back.
>>
>> Dave
>
I think you're misreading the price list -- confusing grade # with quantity.
jim beam
> jim beam wrote:
>> jobst....@stanfordalumni.org wrote:
>>> jim beam wrote:
>
>>>> since when does hardness have anything to do with modulus? we're
>>>> not talking plastic deformation, we're talking elastic so it's
>>>> stiffness [you know, the straight line part of the stress/strain
>>>> part of the graph, the bit before yield] of the ball that determines
>>>> contact patch size, not hardness.
>
>>> This comes from the manufacturer's web site (smaller contact area):
>>>
>>> http://tinyurl.com/khpwp
>
> Interesting web site. They claim the ceramic bearings work to much
> higher temperatures. They also point out that the coefficient of thermal
> expansion (CTE) is much different in the balls and races. Wouldn't that
> lead to changing preload with temperature?
not to sufficient extent as to make a significant difference to life at
normal bearing temps, and certainly not at bike temps.
> I'd think that matching CTE
> between balls and races would be a good idea for use at a wide range of
> temperatures.
>
>> for elastic deformation, smaller contact area is a function of
>> modulus, not hardness.
>
> The web site in question implies that smaller contact area comes from a
> combination of modulus, hardness, and Poisson's ratio.
the modulus [and poisson] determine contact size for a given load.
harness determines the extent of that given load.
>
>> ceramic rolling elements can indeed have a higher modulus so they can
>> have a smaller contact area, but it still doesn't follow that bearing
>> races have a shorter life due to their use. in fact hybrid bearings
>> usually enjoy longer life due to better surface wear characteristics.
>
> Do you have a cite for this?
http://www.bearingworks.com/fnb.htm
[too simplistic.]
> I'd like to know the true cost/benefit
> ratio. For example, let's say that ceramic bearings in bicycle-type
> applications typically last 20x longer but cost 10x more, maybe they
> have some justification for high mileage riders.
contact a manufacturer. 20x is way too long. 5x is pushing it.
jim beam
?!!!
>
>> I doubt this theoretical test will survive practical criticism, but
>> I'm going to enjoy the illusion for a few moments that ceramic balls
>> in steel races could provide a clear demonstration of whatever
>> causes dimpled headsets--fretting, impact, or a combination of the
>> two.
>
> Strange that you should bring that aspect into the mix, considering
> that just the concept of fretting is foreign to the faithful in this
> forum.
is or is not false brinelling "a problem that has been solved by better
bearing design"??? you're sitting both sides of the fence.
> You'd think that people like BMW and Daimler Benz would
> immediately jump to ceramic ball bearings, as they apparently have the
> most failures of differential bearings, judging from the number of
> rushing rear axles I hear rolling by.
jobst, with respect, you're not listening to failed bearings, you're
listening to tire scrub. bearing failure noises in cars are /most/
distinctive, and "whoosh", "rush" or any of the other adjectives you've
used to describe this imagined problem are most definitely /not/ it.
> These are outdone only by the
> number of SUV's and pickup trucks with delaminating tires slapping
> down the pike. Maybe, they also, should have ceramic tire treads.
they effectively already do - it's called "silica" and it's added to
tire compounds in place of carbon black.
jim beam
> Peter Cole wrote:
>> jim beam wrote:
>>> Peter Cole wrote:
>>
>>>> Engineers like to use the right tool for the job, that's all. Bikes
>>>> are a bad application for ceramic balls even if the cost is ignored.
>>>
>>> for what reason? it's not load, it's not speed, it's not friction,
>>> it's not lifetime. so what is it other than cost or distaste?
>>
>> It *is* load/lifetime, ceramic balls would require a 30% load de-rate.
>> For someone like myself who experiences spalling failures on wheel
>> cones and crank spindles this would reduce service life further.
>
> I'm smaller, and I rarely see spalling failures on hubs and bottom
> brackets. But they do occur; I have a freehub on my 'cross bike with a
> rough race, although it's not bad enough to warrant immediate replacement.
in my experience, this is a function of component quality first,
adjustment precision second. quality componentry can take a lot of
abuse and lasts a really long time. cheap stuff, even when correctly
adjusted, hardly lasts at all. i'm #205 and my hubs/bb's last forever,
but i don't ride anything less than xt/105 hubs, xt/ultegra bb's
[shimano] or record [campy]. every time i've used stuff like veloce
hubs or lower end shimano, it's needed replacing in much shorter order.
Robin Hubert
Not quite. I have seen the Shimano headsets you currently use fretted
and dimpled. I'm sure you're aware, however, to take your cartridges
apart and inspect/lube them.
jim beam
they can dimple from true brinelling - i.e. overload - not /all/
dimpling is false brinelling ["fretting"] as jobst seems to think.
dvt
>> Interesting web site. They claim the ceramic bearings work to much
>> higher temperatures. They also point out that the coefficient of thermal
>> expansion (CTE) is much different in the balls and races. Wouldn't that
>> lead to changing preload with temperature?
> not to sufficient extent as to make a significant difference to life at
> normal bearing temps, and certainly not at bike temps.
What you say is true; thermal expansion is small at temperatures
experienced by most bicycles. But since they claim improved performance
at elevated temperatures, do the different CTEs change preload? I'm
thinking of temps like 500° C, which I hope my bicycle never has to
experience if I'm in the neighborhood.
>>> for elastic deformation, smaller contact area is a function of
>>> modulus, not hardness.
>>
>> The web site in question implies that smaller contact area comes from a
>> combination of modulus, hardness, and Poisson's ratio.
>
> the modulus [and poisson] determine contact size for a given load.
> harness determines the extent of that given load.
What do you mean by "extent?"
>>> ceramic rolling elements can indeed have a higher modulus so they can
>>> have a smaller contact area, but it still doesn't follow that bearing
>>> races have a shorter life due to their use. in fact hybrid bearings
>>> usually enjoy longer life due to better surface wear characteristics.
>>
>> Do you have a cite for this?
>
> http://www.bearingworks.com/fnb.htm
>
> [too simplistic.]
>
>> I'd like to know the true cost/benefit
>> ratio. For example, let's say that ceramic bearings in bicycle-type
>> applications typically last 20x longer but cost 10x more, maybe they
>> have some justification for high mileage riders.
>
> contact a manufacturer. 20x is way too long. 5x is pushing it.
Thanks for the link. I'll look at it and comment if I find anything of
interest.
jim beam
> jim beam wrote:
>> dvt wrote:
>>> jim beam wrote:
>>>> jobst....@stanfordalumni.org wrote:
>
>>>>> http://tinyurl.com/khpwp
>
>>> Interesting web site. They claim the ceramic bearings work to much
>>> higher temperatures. They also point out that the coefficient of thermal
>>> expansion (CTE) is much different in the balls and races. Wouldn't that
>>> lead to changing preload with temperature?
>
>> not to sufficient extent as to make a significant difference to life
>> at normal bearing temps, and certainly not at bike temps.
>
> What you say is true; thermal expansion is small at temperatures
> experienced by most bicycles. But since they claim improved performance
> at elevated temperatures, do the different CTEs change preload?
i don't imagine so. being able to run at elevated temperatures is a
function of being able to retain sufficient hardness to do so -
something not possible with steel bearings as they can start to soften
[ever-so-slightly] above about 120°C - it's not anything to do with
thermal expansion or its effect on preload.
> I'm
> thinking of temps like 500° C, which I hope my bicycle never has to
> experience if I'm in the neighborhood.
if you need rolling element bearings, ceramics are your only hope afaik.
>
>>>> for elastic deformation, smaller contact area is a function of
>>>> modulus, not hardness.
>>>
>>> The web site in question implies that smaller contact area comes from a
>>> combination of modulus, hardness, and Poisson's ratio.
>>
>> the modulus [and poisson] determine contact size for a given load.
>> harness determines the extent of that given load.
>
> What do you mean by "extent?"
all other things being equal, the harder material can take a higher load.
dvt
> dvt wrote:
>> jim beam wrote:
>>> dvt wrote:
>>>> jim beam wrote:
>>>>> for elastic deformation, smaller contact area is a function of
>>>>> modulus, not hardness.
>>>> The web site in question implies that smaller contact area comes from a
>>>> combination of modulus, hardness, and Poisson's ratio.
>>> the modulus [and poisson] determine contact size for a given load.
>>> harness determines the extent of that given load.
>> What do you mean by "extent?"
> all other things being equal, the harder material can take a higher load.
Sorry, I didn't get that. In the context of contact area, I don't
understand the meaning of "take a higher load." If all other things are
equal, do you mean that the contact area depends on hardness?
jim beam
> jim beam wrote:
>> dvt wrote:
>>> jim beam wrote:
>>>> dvt wrote:
>>>>> jim beam wrote:
>>>>>> for elastic deformation, smaller contact area is a function of
>>>>>> modulus, not hardness.
>
>>>>> The web site in question implies that smaller contact area comes
>>>>> from a
>>>>> combination of modulus, hardness, and Poisson's ratio.
>
>>>> the modulus [and poisson] determine contact size for a given load.
>>>> harness determines the extent of that given load.
>
>>> What do you mean by "extent?"
>
>> all other things being equal, the harder material can take a higher load.
>
> Sorry, I didn't get that. In the context of contact area, I don't
> understand the meaning of "take a higher load."
a ceramic with Rc78 will take more load before deformation than a steel
with Rc60.
"in the context of contact area" is confusing. a hybrid bearing with
ceramic balls will not take the same load as a steel bearing because the
contact area is smaller and affects the steel. /but/ real-life bearing
lives of hybrids are often much longer because there is less
micro-welding. and an all ceramic bearing can often take a higher load
than a steel one and give a longer life again. how much do you want to pay?
> If all other things are
> equal, do you mean that the contact area depends on hardness?
no, stiffness [modulus].
dvt
> dvt wrote:
>> jim beam wrote:
>>> dvt wrote:
>>>> jim beam wrote:
>>>>> dvt wrote:
>>>>>> jim beam wrote:
>>>>>>> for elastic deformation, smaller contact area is a function of
>>>>>>> modulus, not hardness.
>>
>>>>>> The web site in question implies that smaller contact area comes
>>>>>> from a
>>>>>> combination of modulus, hardness, and Poisson's ratio.
>>
>>>>> the modulus [and poisson] determine contact size for a given load.
>>>>> harness determines the extent of that given load.
>>
>>>> What do you mean by "extent?"
>>
>>> all other things being equal, the harder material can take a higher
>>> load.
>>
>> Sorry, I didn't get that. In the context of contact area, I don't
>> understand the meaning of "take a higher load."
>
> a ceramic with Rc78 will take more load before deformation than a steel
> with Rc60.
So you *were* talking about deformation. So when you said "all other
things being equal, the harder material can take a higher load," you
were saying that harder materials deform less, resulting in a smaller
contact area. But that contradicts your other statements.
I'm thoroughly confused. Maybe you meant that harder = higher modulus,
rather than harder = greater hardness?
> "in the context of contact area" is confusing.
Heh. I don't think I'm the one doing the confusing in this discussion. :^)
> a hybrid bearing with
> ceramic balls will not take the same load as a steel bearing because the
> contact area is smaller and affects the steel.
No argument.
> /but/ real-life bearing
> lives of hybrids are often much longer because there is less
> micro-welding.
Is micro-welding a problem at bicycle speeds and temperatures?
> and an all ceramic bearing can often take a higher load
> than a steel one and give a longer life again.
I'll take your word on that one.
> how much do you want to pay?
I believe that's the crux of the question, and it's the question that
started this subthread. What is the cost/benefit of ceramic bearings?
>> If all other things are equal, do you mean that the contact area
>> depends on hardness?
>
> no, stiffness [modulus].
Back up a few steps, let's try again. If all other things are equal
(including modulus), does contact area depend on hardness? When I say
"all other things are equal," I'm saying that modulus, Poisson's ratio,
and any other relevant parameter is the same. Perhaps you meant
something else when you said "all other things are equal."
carl...@comcast.net
[snip]
>> /but/ real-life bearing
>> lives of hybrids are often much longer because there is less
>> micro-welding.
>
>Is micro-welding a problem at bicycle speeds and temperatures?
[snip]
Dear Dave,
Last year, Jobst listed a variety of fretful bicycle problems, the
last of which involves headset bearings:
"High forces and low speeds on bicycles are fraught with fretting
problems. . . .
"Aluminum crank/square taper steel spindle interfaces . . ."
"Pedal shaft/crank attachments . . ."
"Quill stems fret in steertubes . . ."
"Knurled jam nuts on QR axles eat into dropouts . . ."
"Clicking spoke nipples . . ."
"Head bearings develop dimples . . ."
"I'm sure there are other examples."
http://groups.google.com/group/rec.bicycles.tech/msg/b74e85c36d059867?hl=en&
Cheers,
Carl Fogel
jim beam
<snip>
> Maybe you meant that harder = higher modulus,
> rather than harder = greater hardness?
no. stiffer = smaller contact area. harder = higher load. hardness !=
stiffness.
<snip>
>
> Is micro-welding a problem at bicycle speeds and temperatures?
yes. there is no elasto-hydrodynamic separation at most bike bearing
speeds, so the bearing elements make contact, hence there /is/
micro-welding. it is worse with all-steel bearings and minimized by
ceramic hybrids.
dvt
> dvt wrote: <snip>
>> Maybe you meant that harder = higher modulus, rather than harder =
>> greater hardness?
> no. stiffer = smaller contact area. harder = higher load. hardness
> != stiffness.
jim, I don't know if you're coming or going. Let me put the confusing
quotes all in one place...
> smaller contact area is a function of modulus, not hardness.
That one is pretty clear on its own.
> the modulus [and poisson] determine contact size for a given load.
> harness determines the extent of that given load.
I presume you mean hardness instead of "harness."
> all other things being equal, the harder material can take a higher load.
> harder = higher load. hardness != stiffness.
I think it's a pretty simple thing. You're probably being a bit sloppy
with the term "hard." Sometimes you mean hardness as measured on a
Rockwell scale, and sometimes I think you mean something else that I
just can't fathom.
jim beam
> jim beam wrote:
>> dvt wrote: <snip>
>>> Maybe you meant that harder = higher modulus, rather than harder =
>>> greater hardness?
>
>> no. stiffer = smaller contact area. harder = higher load. hardness
>> != stiffness.
>
> jim, I don't know if you're coming or going. Let me put the confusing
> quotes all in one place...
>
>> smaller contact area is a function of modulus, not hardness.
>
> That one is pretty clear on its own.
>
>> the modulus [and poisson] determine contact size for a given load.
>> harness determines the extent of that given load.
>
> I presume you mean hardness instead of "harness."
>
>> all other things being equal, the harder material can take a higher load.
>
>> harder = higher load. hardness != stiffness.
>
> I think it's a pretty simple thing. You're probably being a bit sloppy
> with the term "hard." Sometimes you mean hardness as measured on a
> Rockwell scale, and sometimes I think you mean something else that I
> just can't fathom.
>
hardness is related to strength. look it up. hardness is not
stiffness. look that up too.
you now have enough to do your own homework. bye.
Peter Cole
By my calculations, typical bicycle bearings have a lubricant film
thickness of about 10 micro-inches at typical operating speeds (100-200
rpm). The question is the ratio of film thickness to composite surface
roughness. Grade 25 balls have a max surface roughness of 2
micro-inches. If bearing races had similar finish, the film/composite
roughness ratio would be >3, within the elastohydrodynamic operating zone.
Ben C
> jim beam wrote:
>> dvt wrote:
>> <snip>
>>> Maybe you meant that harder = higher modulus, rather than harder =
>>> greater hardness?
>>
>> no. stiffer = smaller contact area. harder = higher load. hardness !=
>> stiffness.
>>
>> <snip>
>>>
>>> Is micro-welding a problem at bicycle speeds and temperatures?
>>
>> yes. there is no elasto-hydrodynamic separation at most bike bearing
>> speeds, so the bearing elements make contact, hence there /is/
>> micro-welding. it is worse with all-steel bearings and minimized by
>> ceramic hybrids.
>>
>
> By my calculations, typical bicycle bearings have a lubricant film
> thickness of about 10 micro-inches at typical operating speeds (100-200
> rpm).
That's surely not the operating speed of a headset though? They're
perhaps the bearings that are most likely to be affected by
micro-welding on a bike.
jim beam
dvt
> dvt wrote:
>> jim beam wrote:
>>> dvt wrote: <snip>
>>>> Maybe you meant that harder = higher modulus, rather than harder =
>>>> greater hardness?
>>
>>> no. stiffer = smaller contact area. harder = higher load. hardness
>>> != stiffness.
>>
>> jim, I don't know if you're coming or going. Let me put the confusing
>> quotes all in one place...
>>
>>> smaller contact area is a function of modulus, not hardness.
>>
>> That one is pretty clear on its own.
>>
>>> the modulus [and poisson] determine contact size for a given load.
>>> harness determines the extent of that given load.
>>
>> I presume you mean hardness instead of "harness."
>>
>>> all other things being equal, the harder material can take a higher
>>> load.
>>
>>> harder = higher load. hardness != stiffness.
>>
>> I think it's a pretty simple thing. You're probably being a bit sloppy
>> with the term "hard." Sometimes you mean hardness as measured on a
>> Rockwell scale, and sometimes I think you mean something else that I
>> just can't fathom.
>>
> are you bored? - there seems to be an unnatural degree of obtusity here.
Thanks for the insult. "harder = higher load?" What the heck does that mean?
> hardness is related to strength. look it up. hardness is not
> stiffness. look that up too.
Hardness is related to strength in metals. I don't believe that's true
for ceramics. We're talking about ceramic bearings.
> you now have enough to do your own homework. bye.
I see you're not willing to clarify. You've gone a long way towards
proving your lack of knowledge.
By the way, I'm basing most of my arguments on conversations with a PhD
in materials science and the book Callister, "Materials Science and
Engineering." The book is in my lap as I type this. Does that qualify as
doing my own homework?
Michael Press
dvt <dvt+u...@psu.edu> wrote:
> By the way, I'm basing most of my arguments on conversations with a PhD
> in materials science and the book Callister, "Materials Science and
> Engineering." The book is in my lap as I type this. Does that qualify as
> doing my own homework?
Yes, and it is masking a lurid volume called
`The Return of Dr. Damocles' where he tampers with the
very essence of forces sustaining of the universe. Hah!
You thought we couldn't see it.
Oh, and the universe hangs for the top. Hope I'm not
giving anything away.
--
Michael Press
Jonathan D. Proulx
> Oh, and the universe hangs for the top. Hope I'm not
> giving anything away.
No it doesn't it hangs from the bottom, stop spreading FUD.
jim beam
<snip>
> Hardness is related to strength in metals. I don't believe that's true
> for ceramics. We're talking about ceramic bearings.
<snip>
> I see you're not willing to clarify. You've gone a long way towards
> proving your lack of knowledge.
>
> By the way, I'm basing most of my arguments on conversations with a PhD
> in materials science and the book Callister, "Materials Science and
> Engineering." The book is in my lap as I type this. Does that qualify as
> doing my own homework?
bottom line dave, there are two possible scenarios for this little game:
1. you have too little knowledge and too much time, but you still want
to try catching me out.
2. you genuinely have a problem with some pretty basic concepts.
if it's the latter, there are a multitude of sources on the web
explaining properties of materials. you therefore need to look into
"strength" and "modulus". then you can cite what you've read and get
back with questions on what's confusing you. [but given that you're a
phd student, i assign a low probability to this being the case.]
if it's the former, you're deliberately trying to twist two real basic
concepts as if you'll confuse me into making a mistake which you'd try
to use as discreditation. well, let's just say this: i know next to
nothing about acoustics, so it would be pretty damned stupid for me to
try catching out an acoustics phd student on that subject. now, guess
how your efforts look when you're trying to be a smart-ass on materials
theory?
btw, faith-based assumptions about engineering ceramics won't get you
very far - they're not pottery so their properties are somewhat
different from what seems to be your experience in the kitchen. and if
your "materials phd" friend is more than imaginary, let them respond
direct - otherwise it sounds too much like "my brother is bigger than
your brother" childishness.
Peter Cole
Not really. Using the calculator above for the example SKF 16004 (11,
5mm balls), at 200 rpm, it gives a film thickness of 0.152 micron for a
100N load, and 0.134 for 600N.
Those numbers would put that bearing in the "mixed mode" area for
bicycle loads and speeds if you accept their numbers for "asperity".
They use .001 micron for the balls (equivalent to 25 micro-inches for
grade 25 balls), but the ball bearing spec for grade 25 surface finish
is less than .0001 micron.
ABMA definitions:
"2.12) Grade: A specific combination of dimensional form and surface
roughness tolerance. A ball grade is designated by a grade number."
grade 25 -- deviation from spherical form = 0.000025, max. surface
roughness (in micro-inches "Ra") = 2.0
"Surface Roughness: Surface roughness consists of all those
irregularities which form surface relief and which are conventionally
defined within the area where deviations of form and waviness are
eliminated.
Waviness: The more widely spaced circumferential component of surface
texture."
The tribology site suggests using the much larger numbers for Ra
(deviation from spherical form) rather than surface roughness for the
balls, and I assume they do the same for the race. Assuming race surface
roughness is much better than "out of round" (they use 0.1 micron) spec,
they are over-estimating "asperity" by an order of magnitude or more.
Revising those numbers to reflect surface roughness would put bearings
in this size range well within elastohydrodynamic mode for typical
bicycle rotational rates (and loads).
dvt
> dvt wrote:
> <snip>
>> Hardness is related to strength in metals. I don't believe that's true
>> for ceramics. We're talking about ceramic bearings.
>
> <snip>
>> I see you're not willing to clarify. You've gone a long way towards
>> proving your lack of knowledge.
>>
>> By the way, I'm basing most of my arguments on conversations with a
>> PhD in materials science and the book Callister, "Materials Science
>> and Engineering." The book is in my lap as I type this. Does that
>> qualify as doing my own homework?
>
> bottom line dave, there are two possible scenarios for this little game:
>
> 1. you have too little knowledge and too much time, but you still want
> to try catching me out.
>
> 2. you genuinely have a problem with some pretty basic concepts.
I'll freely admit it's #2.
> if it's the latter, there are a multitude of sources on the web
> explaining properties of materials. you therefore need to look into
> "strength" and "modulus". then you can cite what you've read and get
> back with questions on what's confusing you. [but given that you're a
> phd student, i assign a low probability to this being the case.]
I did give citations. I understand strength and modulus very well, and I
get the concepts of hardness and fracture toughness. You apparently use
the words "hard," "harder," and "hardness" to mean different things, and
I'm trying to understand.
> if it's the former, you're deliberately trying to twist two real basic
> concepts as if you'll confuse me into making a mistake which you'd try
> to use as discreditation.
Definitely not the case. I think you'll see that if you reread the
exchange. The sentence that triggered my questions was "har[d]ness
determines the extent of that given load." What does that mean?
> now, guess
> how your efforts look when you're trying to be a smart-ass on materials
> theory?
I think it's pretty obvious that I'm not trying to be a smart aleck. I
truly don't understand, and I suspect that you don't know what you're
talking about either. My questions have been pretty simple, and you've
been unwilling to clarify.
> btw, faith-based assumptions about engineering ceramics won't get you
> very far - they're not pottery so their properties are somewhat
> different from what seems to be your experience in the kitchen.
Red herring, and incorrect besides. What faith-based assumptions have I
made?
What makes you think that my experience with ceramics is kitchen? Would
I have this copy of "An introduction to the mechanical properties of
ceramics" by David J. Green on my desk if that were true?
I'm not claiming to know very much about Green's book. If I did, I
probably wouldn't be asking questions, I'd be answering them. I'm using
it as a reference to learn about fracture toughness for reasons
unrelated to bicycles.
Have I proven myself worthy of an explanation yet? Or do I need a PhD in
materials science before you clarify?
dvt
> In article <efe2vl$ute$1...@f04n12.cac.psu.edu>,
> dvt <dvt+u...@psu.edu> wrote:
>> By the way, I'm basing most of my arguments on conversations with a PhD
>> in materials science and the book Callister, "Materials Science and
>> Engineering." The book is in my lap as I type this. Does that qualify as
>> doing my own homework?
>
> Yes, and it is masking a lurid volume called
> `The Return of Dr. Damocles' where he tampers with the
> very essence of forces sustaining of the universe. Hah!
> You thought we couldn't see it.
I know this was a joke, but even after rereading the story about the
sword of Damocles, the joke went over my head.
> Oh, and the universe hangs for the top. Hope I'm not
> giving anything away.
Now that's a good one. What is the "load path" through a preloaded
ceramic bearing? :^)
Michael Press
dvt <dvt+u...@psu.edu> wrote:
> Michael Press wrote:
> > In article <efe2vl$ute$1...@f04n12.cac.psu.edu>,
> > dvt <dvt+u...@psu.edu> wrote:
> >> By the way, I'm basing most of my arguments on conversations with a PhD
> >> in materials science and the book Callister, "Materials Science and
> >> Engineering." The book is in my lap as I type this. Does that qualify as
> >> doing my own homework?
> >
> > Yes, and it is masking a lurid volume called
> > `The Return of Dr. Damocles' where he tampers with the
> > very essence of forces sustaining of the universe. Hah!
> > You thought we couldn't see it.
>
> I know this was a joke, but even after rereading the story about the
> sword of Damocles, the joke went over my head.
The joke is that you would keep a comic book in your heavy
textbook, reading it while pretending to study materials'
science. I imagined a name for a comic book villain that I
do not think has been used.
> > Oh, and the universe hangs for the top. Hope I'm not
> > giving anything away.
>
> Now that's a good one. What is the "load path" through a preloaded
> ceramic bearing? :^)
--
Michael Press
jim beam
dave, with respect, if you truly understood strength, modulus and
hardness, we wouldn't be having this discussion. however, since you ask
one more time, let's go back to basics:
on a stress-strain graph with say an engineering steel, before onset of
plastic deformation, we have a straight line region that relates the two
quantities. the slope of that line is the stiffness or modulus, the
"height" of the line, the strength.
any questions on that? [a ceramic effectively omits the plastic portion
and goes straight to fracture]
next, we have the relation [or distinction] between hardness and
strength. generally, stronger materials are harder, so there's a
roughly linear correlation between the two. that's a gross
over-simplification, but hopefully the principle can be understood.
so, getting back to your first question, the hardness [strength] of the
material determines the extent of the load for the bearing - as in,
higher strength [hardness] has a longer line on its stress-strain graph
[load] before the onset of deformation.
wrt to ceramics, they still have a stress strain graph, with slope and
length, but it's usually much shorter in the tensile portion, [often
effectively zero] compared to the compressive portion. the graph axes
may have changed sign, but slope still indicates stiffness, and the
magnitude still indicates strength [which in turn is related to hardness].
does that make sense?
>
>> btw, faith-based assumptions about engineering ceramics won't get you
>> very far - they're not pottery so their properties are somewhat
>> different from what seems to be your experience in the kitchen.
>
> Red herring, and incorrect besides. What faith-based assumptions have I
> made?
"Hardness is related to strength in metals. I don't believe that's true
for ceramics."
there's a belief [faith] problem - strength is still related to hardness
[among other things]. see above for the difference between ceramics in
tension and compression. high density engineering ceramics [unlike
familiar pottery] otoh can be very strong in tension - or at least,
strong enough for applications like bearings.
bottom line on bearings: you want them both strong [hard] and stiff. a
titanium bearing, while theoretically some ti alloys are as strong as
some steels, is roughly half the stiffness. that would lead to
distortion and reliability problems. steels are a great mix between
stiffness and hardness. ceramics are stiffer and harder still.
>
> What makes you think that my experience with ceramics is kitchen? Would
> I have this copy of "An introduction to the mechanical properties of
> ceramics" by David J. Green on my desk if that were true?
>
> I'm not claiming to know very much about Green's book. If I did, I
> probably wouldn't be asking questions, I'd be answering them. I'm using
> it as a reference to learn about fracture toughness for reasons
> unrelated to bicycles.
>
> Have I proven myself worthy of an explanation yet? Or do I need a PhD in
> materials science before you clarify?
>
since you have access to a tech library, one of the great books on basic
principles is dieter - mechanical metallurgy. don't let the title
deceive: it'll talk you through all the materials concepts and theory on
stress, strain, and their relationship. once you have that under your
belt, you can move to ceramics and be confident you know what you're
looking at.
dvt
I have to laugh when I read your "with respect." It always means you're
about to show disrespect. Like the rest of that sentence:
> ...if you truly understood strength, modulus and hardness, we wouldn't
> be having this discussion. however, since you ask one more time,
> let's go back to basics:
>
> on a stress-strain graph with say an engineering steel, before onset
> of plastic deformation, we have a straight line region that relates
> the two quantities. the slope of that line is the stiffness or
> modulus, the "height" of the line, the strength.
>
> any questions on that? [a ceramic effectively omits the plastic
> portion and goes straight to fracture]
You know that I'm well aware of all of those concepts.
> next, we have the relation [or distinction] between hardness and
> strength. generally, stronger materials are harder, so there's a
> roughly linear correlation between the two. that's a gross
> over-simplification, but hopefully the principle can be understood.
From Green's book (see cite upthread):
"The resistance of a material to the formation of a permanent
surface impression by an indenter is termed *hardness*. The
deformation process must be inelastic and, hence, it is inherently
related to the resistance of a material to such a deformation
(indentation). Hardness impressions can be formed even in brittle
materials, though at higher loads this is usually accompainied by
localized cracking. For more ductile materials, however, one would
expect hardness to be related to the yield stress of a material."
Green implies that brittle materials don't show a relationship between
hardness and yield/ultimate strength. Callister (see other cite
upthread) also says that yield and hardness are related in metals, but
says nothing of the sort for ceramics.
> so, getting back to your first question, the hardness [strength] of
> the material determines the extent of the load for the bearing - as
> in, higher strength [hardness] has a longer line on its stress-strain
> graph [load] before the onset of deformation.
That is much clearer. Thank you. You defined hardness and extent, and
now I understand what you were trying to say.
> wrt to ceramics, they still have a stress strain graph, with slope
> and length, but it's usually much shorter in the tensile portion,
> [often effectively zero] compared to the compressive portion. the
> graph axes may have changed sign, but slope still indicates
> stiffness, and the magnitude still indicates strength [which in turn
> is related to hardness].
>
> does that make sense?
Yep, I finally understand what you said. But I'm still not in agreement.
>>> btw, faith-based assumptions about engineering ceramics won't get
>>> you very far - they're not pottery so their properties are
>>> somewhat different from what seems to be your experience in the
>>> kitchen.
>>
>> Red herring, and incorrect besides. What faith-based assumptions
>> have I made?
>
> "Hardness is related to strength in metals. I don't believe that's
> true for ceramics."
>
> there's a belief [faith] problem - strength is still related to
> hardness [among other things].
Cite? I have found cites that imply the contrary (see above).
And the faith thing is still a red herring.
> since you have access to a tech library, one of the great books on
> basic principles is dieter - mechanical metallurgy. don't let the
> title deceive: it'll talk you through all the materials concepts and
> theory on stress, strain, and their relationship. once you have that
> under your belt, you can move to ceramics and be confident you know
> what you're looking at.
Thanks for the tip.
jim beam
ok, like i said before, i made gross generalizations to keep it simple.
how deep you want to go depends on how much time you have and whether
you want to sign up for class!
ultra-simplified, the difference between ductile and brittle materials
is the lack of deformation mechanism in the latter. plastic deformation
is due to dislocations and their movement. ceramics don't have this
property - all they can do is fracture. you said you had some
familiarity with fracture mechanics - essentially, ceramics have
inherent flaws which when analyzed from a fracture mechanics viewpoint,
show how these flaws suddenly propagate. the greater the number and
extent of flaws [low density pottery] the easier it is for failure to
occur. this is tackled in toughened glass by keeping the surface, the
region from which almost all cracks propagate, under compression thus
limiting the ability of surface flaws to go beyond critical.
engineering ceramics can't usually be treated in this way, so the way to
address strength is with high densification. as green says, the
connection between hardness and strength is much more vague in ceramics
- and this is because of the flaw factor. the lower the flaw density,
the better the relationship between hardness and strength. ceramic
bearings have a /much/ lower flaw density than pottery.
extrapolating from there, that's why we have fiber reinforced materials
using glass and carbon. the smaller the material cross section, by
definition, the smaller the flaws. neither carbon or glass fibers have
ductility, but they're very strong because their failure mechanism has
been curtailed. [there are other ceramic fiber materials, but glass is
the easiest and cheapest to produce in bulk.]
dvt
> ok, like i said before, i made gross generalizations to keep it simple.
> how deep you want to go depends on how much time you have and whether
> you want to sign up for class!
>
> ultra-simplified, the difference between ductile and brittle materials
> is the lack of deformation mechanism in the latter. plastic deformation
> is due to dislocations and their movement. ceramics don't have this
> property - all they can do is fracture. you said you had some
> familiarity with fracture mechanics - essentially, ceramics have
> inherent flaws which when analyzed from a fracture mechanics viewpoint,
> show how these flaws suddenly propagate. the greater the number and
> extent of flaws [low density pottery] the easier it is for failure to
> occur. this is tackled in toughened glass by keeping the surface, the
> region from which almost all cracks propagate, under compression thus
> limiting the ability of surface flaws to go beyond critical. engineering
> ceramics can't usually be treated in this way, so the way to address
> strength is with high densification. as green says, the connection
> between hardness and strength is much more vague in ceramics - and this
> is because of the flaw factor. the lower the flaw density, the better
> the relationship between hardness and strength. ceramic bearings have a
> /much/ lower flaw density than pottery.
If "the connection between hardness and strength is much more vague in
ceramics," then your earlier statement of "hardness determines the
extent of the load" doesn't apply to ceramic bearings.
jim beam
<snip>
>
> If "the connection between hardness and strength is much more vague in
> ceramics," then your earlier statement of "hardness determines the
> extent of the load" doesn't apply to ceramic bearings.
>
with an indenter as offering more resistance to loading, it /can/ take a
higher load!
think about it like this: there's two ways to use an hardness indenter-
1. the usual way - press with a fixed mass then gauge the indenter
impression.
2. indent for a given impression, then gauge the mass required to do so.
for the latter, isn't it obvious that the harder the material, the
more load it required to indent it?
dvt
> dvt wrote:
>> If "the connection between hardness and strength is much more vague
>> in ceramics," then your earlier statement of "hardness determines
>> the extent of the load" doesn't apply to ceramic bearings.
>>
> no dude! if it's harder [stronger], as in something you're measuring
> with an indenter as offering more resistance to loading, it /can/
> take a higher load!
You're confusing me again. The indentation test measures *hardness*, not
strength. We've already established that hardness != strength, and the
two are not related, in ceramics. Now you tie them together.
> think about it like this: there's two ways to use an hardness
> indenter-
> 1. the usual way - press with a fixed mass then gauge the
> indenter impression.
> 2. indent for a given impression, then gauge the
> mass required to do so.
> for the latter, isn't it obvious that the
> harder the material, the more load it required to indent it?
Yes, that's a measure of hardness. Your point?
jim beam
> jim beam wrote:
>> dvt wrote:
>
>>> If "the connection between hardness and strength is much more vague
>>> in ceramics," then your earlier statement of "hardness determines
>>> the extent of the load" doesn't apply to ceramic bearings.
>>>
>> no dude! if it's harder [stronger], as in something you're measuring
>> with an indenter as offering more resistance to loading, it /can/
>> take a higher load!
>
> You're confusing me again.
you cannot be serious.
> The indentation test measures *hardness*, not
> strength. We've already established that hardness != strength,
wow, there's a massive cognitive disconnect here. the correlation
between hardness and strength for ceramics depends on flaw density, not
the fact that they're not metal - a point you just don't seem to get.
> and the
> two are not related, in ceramics. Now you tie them together.
don't put [false] words in my mouth. how many more times do you need to
be told? - correlation between them for ceramics depends on flaw
density. since we're talking high grade engineering materials with low
flaw density, there is strong correlation, hence BEARINGS CAN AND ARE
MADE OF CERAMICS and they /can/ take the load. alternatively, i could
be in cahoots with manufacturers and the [hitherto] secret architect of
a vast global conspiracy to deceive - with a giant web of fabricated
data spun over the last 30 years.
>
>> think about it like this: there's two ways to use an hardness
>> indenter-
>
>> 1. the usual way - press with a fixed mass then gauge the
>> indenter impression.
>
>> 2. indent for a given impression, then gauge the
>> mass required to do so.
>
>> for the latter, isn't it obvious that the
>> harder the material, the more load it required to indent it?
>
> Yes, that's a measure of hardness. Your point?
>
you're being deliberately obtuse. get reality. read and learn. even
if you don't want to listen to me - pay attention to the fact that
ceramic bearings exist, are highly successful and are used where
cost/performance warrants them throughout the globe. that simply
couldn't happen if ceramics failed. if you don't want to acknowledge
that, you're just what i said right at the start - bored and fucking
about for the sake of it. if you rode your bike a bit more often, you'd
not be in that situation.
dvt
> dvt wrote:
>> jim beam wrote:
>>> dvt wrote:
>> The indentation test measures *hardness*, not
>> strength. We've already established that hardness != strength,
> wow, there's a massive cognitive disconnect here. the correlation
> between hardness and strength for ceramics depends on flaw density, not
> the fact that they're not metal - a point you just don't seem to get.
That's a point you've never made (at least cogently). You've brought in
a third variable: flaw density. So let's see if I have this straight...
Two ceramic samples, same strength, same flaw density. You predict the
same hardness for those two samples?
>> and the
>> two are not related, in ceramics. Now you tie them together.
> don't put [false] words in my mouth. how many more times do you need to
> be told? - correlation between them for ceramics depends on flaw
> density.
Well, since you just told me the first time, I guess that makes twice.
But I got it the first time, so I think that means once.
> since we're talking high grade engineering materials with low
> flaw density, there is strong correlation, hence BEARINGS CAN AND ARE
> MADE OF CERAMICS and they /can/ take the load. alternatively, i could
> be in cahoots with manufacturers and the [hitherto] secret architect of
> a vast global conspiracy to deceive - with a giant web of fabricated
> data spun over the last 30 years.
Did I say that ceramic bearings were a bad idea? Cite, please.
>>> think about it like this: there's two ways to use an hardness
>>> indenter-
>>> 1. the usual way - press with a fixed mass then gauge the
>>> indenter impression.
>>> 2. indent for a given impression, then gauge the
>>> mass required to do so.
>>> for the latter, isn't it obvious that the
>>> harder the material, the more load it required to indent it?
>> Yes, that's a measure of hardness. Your point?
even
> if you don't want to listen to me - pay attention to the fact that
> ceramic bearings exist, are highly successful and are used where
> cost/performance warrants them throughout the globe. that simply
> couldn't happen if ceramics failed.
Again, point out where I said ceramic bearings were a bad idea. Even if
I thought they were worth the money in a bicycle, but I *never* said
they didn't have their place.
> you cannot be serious.
> you're being deliberately obtuse.
> get reality. read and learn.
> if you don't want to acknowledge
> that, you're just what i said right at the start - bored and f@#$ing
> about for the sake of it. if you rode your bike a bit more often, you'd
> not be in that situation.
Your post was intentionally and blatantly offensive. I recommend an
apology.
jim beam
> jim beam wrote:
>> dvt wrote:
>>> jim beam wrote:
>>>> dvt wrote:
>
>>> The indentation test measures *hardness*, not
>>> strength. We've already established that hardness != strength,
>
>> wow, there's a massive cognitive disconnect here. the correlation
>> between hardness and strength for ceramics depends on flaw density,
>> not the fact that they're not metal - a point you just don't seem to get.
>
> That's a point you've never made (at least cogently).
eh?
"the lower the flaw density, the better the relationship between
hardness and strength."
jim beam, 09/29/2006 06:39 PM.
you think that's offensive??? the only reason you'd post such a
ridiculous statement is an attempt to try the aggressor/victim switcheroo.
if you want to know what's truly offensive, it's someone who doesn't
read, and who willfully won't understand ultra-basic concepts, handing
out criticism and trying to make out there's not a fundamental problem
with their cognitive abilities. you're an idiot dave. no, make that a
fucking idiot. now you can go ahead and throw your teddy in a corner.
jobst....@stanfordalumni.org
of which would be wrong because the differences argues are in the
inconsequential, imperceptible, expensive, name drop category.
> the lower the flaw density, the better the relationship between
> hardness and strength.
In that pursuit, one might imagine that many bicyclists are
experiencing bearing ball failure or similar problem. I don't see any
more merit in the ceramic bearing thread than in most of InterBike,
that is built on these gossamer threads. Grimacing grime faced racers
crossing some finish line (or posing as such) gracing the walls of
larger displays with the implication that you would be like them if
only you bought their product.
For engineers, I thought Campagnolo had the best video with the 3D CAD
assembly of their new BB. Without a sound track or text explanation,
I use the qualification of "for mechanical design engineers" because
unless you understand the problems this design is trying to solve, you
cannot appreciate why the various components are the way they are; for
instance, what the little wire clip or wave washer is doing.
Jobst Brandt
dvt
> you think that's offensive??? the only reason you'd post such a
> ridiculous statement is an attempt to try the aggressor/victim switcheroo.
>
> if you want to know what's truly offensive, it's someone who doesn't
> read, and who willfully won't understand ultra-basic concepts, handing
> out criticism and trying to make out there's not a fundamental problem
> with their cognitive abilities. you're an idiot dave. no, make that a
> fucking idiot. now you can go ahead and throw your teddy in a corner.
I guess that means you're unwilling to defend your argument any further.
jim beam
bother to get your head around it that's your problem. shit dave,
people spend whole lifetimes researching this stuff, and you want me to
spoon feed you a distillate, but then you complain when i do so? sign
up for class!!! idiot.