Ah, the joys of being known as doubtful!
The email helpfully linked to this ceramic bearing review, which is
pretty exciting:
"What we received from Ceramic Speed was a bottom bracket, wheel
bearings, and pulleys. Once we were able to have them installed on the
Bicycle.net road bike the difference was immediate."
"To start, we spun the wheels and they just kept spinning and
spinning. It was impressive how smooth the bearings were. Once we took
the bike on the road and got in some rides we got the feeling like the
bike just wanted to keep accelerating. There was an obvious reduction
in overall friction giving the rider that added boost of speed."
"In the end we are sold at Bicycle.net on the improvement Ceramic
Speed bearings will make in the bike’s overall performance. We
contacted Ceramic Speed for more details on our perceived improvement
and they claim that the power savings is up to 10-12 watts . . ."
http://www.bicycle.net/2008/ceramic-speed-bearings
Well . . .
I'm _more_ skeptical of the ceramic bearing claims.
After all, a 4% weight reduction can actually shave 2% off someone's
time on a climb.
But the bearing claim looks implausible from the start.
If we assume a 95% efficient transmission, then putting 300 watts into
the pedals would produce only 285 watts at the tire, a loss of only 15
watts.
It's hard to imagine the bottom bracket, wheel bearings, and two
pulleys accounting for 10 to 12 watts of the total 15 watts lost,
since there are also the pedal bearings (somehow they were forgotten)
and the chain rollers engaging and disengaging from each sprocket
under load.
But let's go along with the fantasy that sprinkling some ceramic
bearings designed for 20,000 RPM applications into a low-RPM bicycle
transmission will save 10-12 watts for our test rider.
Will a rider experience "the feeling like the bike just wanted to keep
accelerating"?
Will he notice "an obvious reduction in overall friction giving the
rider that added boost of speed"?
Well, yes, for values of acceleration and top speed that increase by
about 1%.
Here's a familiar side-by-side bike speed calculator:
http://bikecalculator.com/veloUS.html
Use on-the-drops and tubulars for both bikes, then crank one up to 300
watts and give give the other that extra alleged 12 watts:
mph
300 watts 25.08
312 watts 25.44
That's a sizzling 0.36 mph faster, a 1.43% speed increase. You could
see it on a cyclocomputer as you pedalled, but the display would be
jumping around from 25.1 to 25.4 mph anyway on any normal road.
On a 20-mile ride, the alleged 12 watt improvement will produce a
41-second lead in 47 minutes, about 1.4% faster.
Let's try a sprint calculator:
http://www.analyticcycling.com/DiffEqMotionFunctions_Page.html
With the defaults, 45.2 seconds for a 500 meter standing start.
Raise the maximum power and average power 12 watts, and time drops to
44.7 seconds, just over 1% faster.
Set the average power to one watt less than the maximum power, 662
max/661 avg versus 650 max/649 avg, and the times drop to 38.6 versus
38.9 seconds, a 0.3 second difference, just under 1% faster.
So rounding probably exaggerates or minimizes the tiny ~1% difference
in sprint calculations. Not exactly something that a rider could
"feel".
Anyway, I doubt that the ceramic bearings put an extra 12 watts into
the rear tire in the first place.
And even if they did, I doubt that any rider could sense the
difference, much less perceive it as the dramatic improvement promised
by the marketing department.
Cadel Evans apparently runs ceramic pulley bearings:
But he doesn't bother with ceramic rear hub bearings:
That's an odd choice.
Do ceramic bearings in two deraileur pulleys (which handle only the
tension and power of the nearly slack lower chain run) provide a
bigger payoff than ceramic bearings in the rear axle (which handles
over half the weight and all the power of the rider)?
Maybe ceramic bearings don't make any practical difference, even to a
Tour de France rider?
Cheers,
Carl Fogel
shimano already use ceramic journal bearings in the top pulley for
xt/ultegra and up. and have done so for some years. advantage is wear
resistance, especially for the mtb application.
A yeoman-like workup, but you're ignoring a salient fact. The placebo
effect does not stop at the doors of science. It rings the bell,
announces itself, walks in and helps itself to the contents of your
fridge.
If someone believes there will be a difference, there will be a
difference. The fact that there is actually a difference, albeit
small, is gravy.
R
Ceramic plain bearings are a very different creature from the ceramic
ball bearings currently being touted as magic charms. Such bearings
were never claimed to have lower friction losses than properly
maintained ball bearings, and in fact were positioned to compete with
the metal or plastic plain bearings included with practically every
other manufacturer's derailleurs during the same span of time.
Chalo
How dare you sully up this discussion with facts!!
Ceramic bearings are one of the biggest hypes put onto the bike biz
since biopace.
I just saw an add for a complete set of 5/32 ceramic balls for Campag
wheels. 60 balls for only $330......$5.50 per ball..YGBSM....
ceramic bearing balls do in fact cost a good deal more to produce than
steel ones. and their precision, fwtw, it typically about 10 times
better. how much is a single steel bearing ball?
grade 25, we sell for $.10 per ball..for bicycles, more than adequate.
So, ceramic is 55 times more expensive for 10 times rounder.
grade 25, we sell for $.10 per ball..for bicycles, more than adequate.
So, ceramic is 55 times more expensive for 10 times rounder.
---------------
Not to mention a point of diminishing returns on roundness.
> jim beam wrote:
>> shimano already use ceramic journal bearings in the top pulley for
>> xt/ultegra and up. and have done so for some years. advantage is wear
>> resistance, especially for the mtb application.
Chalo wrote:
> Ceramic plain bearings are a very different creature from the ceramic
> ball bearings currently being touted as magic charms. Such bearings
> were never claimed to have lower friction losses than properly
> maintained ball bearings, and in fact were positioned to compete with
> the metal or plastic plain bearings included with practically every
> other manufacturer's derailleurs during the same span of time.
Yeah but ceramic sleeves don't shatter. Where's the fun in that?
--
Andrew Muzi
<www.yellowjersey.org/>
Open every day since 1 April, 1971
** Posted from http://www.teranews.com **
Every product has a high-end price parabola. One is tempted to ask
about cost/benefit or where adequate features for the intended purpose
are optimized. In a jet engine, ceramic balls may represent a cost
savings over the unit's life. Hard to justify in our hubs, cranks etc.
genuine campy are more like $0.80 each. so ceramic is only 6.9 times
more expensive for 10 times rounder...
<SNIP>
> > I just saw an add for a complete set of 5/32 ceramic balls for
> > Campag wheels. 60 balls for only $330......$5.50 per ball..YGBSM....
>
> ceramic bearing balls do in fact cost a good deal more to produce
> than steel ones. and their precision, fwtw, it typically about 10
> times better. how much is a single steel bearing ball?
>
I pay 2.5 AU cents a 1/4" bearing. The original review reminded me of a
couple of articles a co-worker of mine wrote for the local bicycle rag
on chain lubes and a black(!!) hi-visibility back-pack cover for
cyclists.
Every negative thing she wrote about the products was edited out and
replace with praise.
Cheers
Joel
The ultimate is the customer who wants ceramic balls in his headset. Talk about the opposite of high-rpm...
--Mike-- Chain Reaction Bicycles
www.ChainReactionBicycles.com
Dear Mike & Andrew,
Does anyone sell ceramic bearings for pedals?
Come to think of it, what about cartridge bearings for pedals?
Cheers,
Carl Fogel
>> "A Muzi" <a...@yellowjersey.org> wrote
>> | Every product has a high-end price parabola. One is tempted to ask
>> | about cost/benefit or where adequate features for the intended purpose
>> | are optimized. In a jet engine, ceramic balls may represent a cost
>> | savings over the unit's life. Hard to justify in our hubs, cranks etc.
> "Mike Jacoubowsky" <mik...@ix.netcom.com> wrote:
>> The ultimate is the customer who wants ceramic balls in his headset. Talk about the opposite of high-rpm...
carl...@comcast.net wrote:
> Does anyone sell ceramic bearings for pedals?
> Come to think of it, what about cartridge bearings for pedals?
Ceramic are obviously available for loose-ball pedals, ceramic
cartridges maybe (size?), Bremen cups no. The trend is to sintered
bronze sleeves, akin to an electric motor journal.
The same ceramic balls for your 2008 Record hubs fit most classic pedals
along the lines of an MKS Sylvan; more modern designs are seldom loose-ball
A very common arrangement with clipless pedals these days is to have a
plain Oilite beaing on the inboard end of the spindle, and a very
small cartridge bearing (6mm bore) on the outboard end.
My last experiment with clipless pedals was with plastic-bodied Time
ATACs. I chose these because they used a cartridge bearing at each
end, and the spindle diameter at the inboard bearing was 12mm. In
order to get the comparatively generous spindle and bearing sizes, I
was willing to accept the plastic bodies.
Atomlab (a BMX parts manufacturer) for years has made pedals featuring
"DU" bearings. I thought at first that they meant bearings made from
depleted uranium, but fortunately that isn't the case. DU is a type
of plain bearing with a steel shell bonded to a layered amalgam of
Babbitt-type material sintered together with PTFE and other solid dry
lubricants. The advantage of plain bearings in a freestyle pedal is
that they allow a pedal with a thick spindle to have a relatively thin
body.
http://atomlab.com/pedals.html
Chalo
Dear Andrew,
Yikes! I googled and found ceramic cartridge bearings for pedals:
http://www.glorycycles.com/fsacecabefor.html
They cost only $139.90, a small price to pay for going . . .
Er, the PDF seems very confused about how much faster we'll be going.
http://lib.store.yahoo.net/lib/glorycycles/fsacer.pdf
First the PDF claims that using ceramic bearings can save "22m in just
55 seconds at 32kph. In short, astounding speed improvement of 4%."
In other words, you might go 33.28 kmh instead of 32.0 kmh, 4% faster.
But then the PDF claims that a "Record BB @ 100rpm and 400W consumes
0.6W, the same BB with ceramic bearings consumes 0.02W."
Let's round the ceramic drag down to zero and use this calculator:
http://bikecalculator.com/veloMetric.html
Tubulars, on-the-drops watts km/h
Record BB bearings 400.0 44.94
magic ceramic bearings 400.6 44.97 0.067% faster
Hmmm . . . a 0.067% speed increase is only about one-sixtieth of the
astounding 4% speed increase that we were promised.
To go 4% faster than 44.94 kmh speed with 400 watts, we need to go
46.74 kmh with . . .
This calculator tells us how many watts we need for 46.74 kmh:
http://bikecalculator.com/wattsMetric.html
We need 447 watts to go 4% faster than we went at 400 watts.
It's darned hard to get 47 extra watts by sprinkling ceramic bearings
in the transmission if the bottom bracket uses only 0.6 watts.
So I may wait a bit before I order those $140 ceramic cartridge pedal
bearings.
Cheers,
Carl Fogel
Ceramic bearings sometimes be found much less expensively than the
prices quoted in these postings. It takes a different talent than CF's
calculations of cost/watt to explain why there isn't price competition
for these. Could be in their failure rate?
I don't know why 3 fine shop owners here disparage them. While there
may be no way to honestly tout them (except for wear in some
applications?) why not let the fool and his money soon be parted if
the client wants them (unless warrantee replacement is too
frequent)?
But, I'd really rather know why I can't find replacement pulley wheels
for less than the price of entire cheap rear ders. Wait, that
dastardly Chris at Velo-Orange has 'em for $4/pair.
But, we've been here before. Suntour put ball bearing cartridges in
their excellent rear ders long ago, then smartly sold the pulley
wheels separately as an upgrade. Not that they claimed the wheels with
those bearing is why their ders shifted better. But, if the consumer
wanted to respect ball bearings more and believe that Campy bronze
bushings were primitive, well Maeda industries was happy to make many
extra bucks on the upgrades. As were shops.
Harry Travis
Well, this shop owner doesn't like to make promises he can't keep.
Even if they say they make no real difference, in fact, the customer
thinks they do. So, after lightening his wallet by hundreds of
dollars, he gets mad at me for STILL not being able to get up that
hill. Same for wheelsouttaboxes and many other things that most other
bike shops shill about.
>
> But, I'd really rather know why I can't find replacement pulley wheels
> for less than the price of entire cheap rear ders. Wait, that
> dastardly Chris at Velo-Orange has 'em for $4/pair.
We have pulleys in the $10 per range, less than any rder.
both campy and shimano use bushings for a reason - they accommodate
lateral play, something essential to quiet operation on modern
close-spaced sprockets.
Subtle shift if time and tense, Jim.
"Use" instead of "used". Historians: Please correct me. Bushings were
used in pulley wheels through years and years and millions of rear
ders BEFORE the additional advantage of providing lateral play for
self-centering in index shifting. Why? cheap, reliable, that's a
filthy area of the bike, wear causes no great harm, and the frictional
losses are so small--here I can make it much shorter than CF: The only
thing to blow up is your calculator from rounding either substantive
energy consumption or savings in the denominator, and effectively
trying to divide by zero.
To understand use of ceramic bearings on bikes, at current prices, you
have to understand peacock feathers. Or not. I just suggested that
Maeda/Suntour, enshrined in boomer biking lore for technical quality,
knew something about selling techno, too, with its after-market
cartridge-bearing pulley wheel upgrade packs.
Memories of youthful techno lust were so long lasting that I bought a
boxed NOS pair a year ago in a garage sale, realized my real ardor had
cooled, and found someone who wanted a 4th pair to complete
restorations.
Harry Travis
you seem to have missed the fact that i have never claimed anything
about friction for journal bearings in this application. don't
incorrectly presume i have a position which you then proceed to argue
against.
>
> To understand use of ceramic bearings on bikes, at current prices, you
> have to understand peacock feathers. Or not. I just suggested that
> Maeda/Suntour, enshrined in boomer biking lore for technical quality,
> knew something about selling techno, too, with its after-market
> cartridge-bearing pulley wheel upgrade packs.
no. the reason to use a ceramic for a derailleur pulley journal bearing
is wear resistance. nothing else can compete. end of story.
> Memories of youthful techno lust were so long lasting that I bought a
> boxed NOS pair a year ago in a garage sale, realized my real ardor had
> cooled, and found someone who wanted a 4th pair to complete
> restorations.
boring.
I didn't know wear was a problem with steel derailleur pulley bearings/
bushings. They see such a small load,
The ads claim (among other things) that ceramic bearings have less
friction than ordinary steel balls and save watts -- but I wonder what
effect the wipers have and whether any savings is lost in wiper drag.
I had some of the original Bullseye pullleys, and they had quite a
bite of drag -- more than my old Campy NR sloppy bushing pulleys.--
Jay Beattie;
campy use a sintered bronze outer on a steel inner.
> They see such a small load,
indeed. but they still wear out, and it's because of crud ingress,
especially if you ride in the wet.
>
> The ads claim (among other things) that ceramic bearings have less
> friction than ordinary steel balls and save watts
for rolling element bearings, they do. by a small degree.
> -- but I wonder what
> effect the wipers have and whether any savings is lost in wiper drag.
more than is lost in bearing drag!
You would miss the parallel between the sale today of ceramic
bearings in pulley wheels and the replacement of bushings with ball
bearing cartridges BEFORE index shifting.
Tell is if there's a hamster's murmur of performance difference
between the long-traditional oiled bronze bushing and any kind of ball
bearing in this application and what that area of performance is.
Friction, durability, life, noise, risk of catastrophic failure,
maintenance....whatever. Tell us.
HPT
Exactly.
I have a bike with a rear changer equipped with adjustable loose-ball
rollers. They are heavier, no better in function and of course that's
one more thing to maintain. Bronze or ceramic sleeves are perfectly
adequate; expensive and complex ball pulleys are no better in any respect.
Trivia buffs/ Bar Bet dep't: Campagnolo's #97 wrench for adjusting
loose-ball pulleys is one of small set of items not stamped
"Campagnolo", like the #G chromed straightedge.
no, that is what you're arguing against, but it doesn't exist for
journal bearings and i've never argued that it does - try and follow the
thread.
>
> Tell is if there's a hamster's murmur of performance difference
> between the long-traditional oiled bronze bushing and any kind of ball
> bearing in this application and what that area of performance is.
> Friction, durability, life, noise, risk of catastrophic failure,
> maintenance....whatever. Tell us.
>
see above. "ceramic" does not automatically translate to "bullshit".
nor is an invitation to jump on the wrong bandwagon - journal bearings
are not rolling element bearings.
no one said they were - travis is trying to use the argument against
rolling element bearings on journal bearings. and looking foolish in
doing so because they're a totally different beast.
>
> Trivia buffs/ Bar Bet dep't: Campagnolo's #97 wrench for adjusting
> loose-ball pulleys is one of small set of items not stamped
> "Campagnolo", like the #G chromed straightedge.
now that /is/ trivia!
Have at it Jim: Let's stipulate that the differences in frictional
loss between bearings of different types in the pulley wheels are
trivial. Now, tell us what your preferred bearing type and material is
for each pulley wheel, and how much more the consumer should be
willing to pay for them vs the simple oiled bronze bushing A Muzi
seems to be satisfied with for his own rides. Assume, as in
restaurants, a $1 increase in mfg cost translates to a $4 increase in
the price at the retail level, and that we're talking about $40 -$70
rear ders. Hell, show your erudition and expound on what we should
want and why for the road, and then again for the dirt.
Harry Travis.
you're missing the point. again. the reason for the ceramic journal in
a derailleur pulley is longevity and grit resistance.
metal journals just don't last as long when contaminated.
> Assume, as in
> restaurants, a $1 increase in mfg cost translates to a $4 increase in
> the price at the retail level, and that we're talking about $40 -$70
> rear ders.
a restaurant analogy for bearing technology??? dude, that's bullshit.
> Hell, show your erudition and expound on what we should
> want and why for the road, and then again for the dirt.
shimano xt and up, ultegra and up have had ceramic journals for nearly a
decade. no price premium, all inclusive, no upgrade required. you
might never ride shimano so you wouldn't know, but if you do, the first
way to notice what these bearings are made of is to ride bike in all
kinds of filthy wet weather for a few years, then wonder why your campy
derailleur pulleys are shagged out, and your shimano ones aren't.
>> > I just saw an add for a complete set of 5/32 ceramic balls for Campag
>> > wheels. 60 balls for only $330......$5.50 per ball..YGBSM....
>>
>> ceramic bearing balls do in fact cost a good deal more to produce than
>> steel ones. and their precision, fwtw, it typically about 10 times
>> better. how much is a single steel bearing ball?
>
>grade 25, we sell for $.10 per ball..for bicycles, more than adequate.
>So, ceramic is 55 times more expensive for 10 times rounder.
As a comparison, I just bought a "Mini-Pak" of 175 3/16" Grade 25
balls from Motion Industries for $5.35(US); which works out to $.03
per ball.
--
jeverett3<AT>sbcglobal<DOT>net (John V. Everett)