Ceramic Bearing Systems and Marketing BS Systems
Benjamin Weiner
> [Zinn punted on the question and quoted Zipp: ]
> "For an average trained cyclist developing 250 watts, that's a savings of
> approximately 10 watts. At any level of competition, that is significant.
> The key is every part of the bearing system has seen marked improvements in
> precision resulting an a total benefit greater than the sum of its parts. ...
> But, as I said, I'm no engineer. So, for all you engineers out there, can
> there possibly be that big a difference in total system efficiency in
> changing from high-quality steel bearing systems to anything else?
No.
This question would be more appropriate for rec.bicycles.tech than
r.b.misc, so I'm crossposting it and setting followups there.
I calculated that the power dissipated in two standard steel
ball bearing hubs is about 0.33 watts for a 75 kg rider+bike
traveling at 10 m/s. It's hard to see how an improved bearing could
shave more than some percentage of that (10-20% ?), which verges on
negligible, I'm sure the aerodynamics of your shoe straps are more
important.
Since I already wrote up the calculation, here it is:
Typical coefficient of friction for a ball bearing is about mu=1.5e-3
(for example, http://www.ntnamerica.com/Engineering/PDFs/2200/frictemp.pdf)
Power dissipated in one bearing is
P1 = mu * (m/2) * g * V_bearing where m is mass of rider+bike
V_bearing is linear speed the
bearing rotates at.
P_bearingloss = 2*P1 total power lost in two wheels
V_bearing = V_bike * d/D d = bearing race diameter
D = wheel diameter
P_bearingloss = mu * mg * d/D * V_bike
P_bearingloss = C_bearingloss * mg * V_bike
where C_bearingloss = mu * d/D, can be compared to C_rolling resistance of tires
For a bearing race diameter of 20 mm and wheel diameter of 668 mm (700x23)
C_bearingloss = 4.5e-5
for comparison, C_rolling resistance is supposed to be 4e-3 for
smooth pavement (e.g. from www.analyticcycling.com), or 100x higher.
For a rider+bike of 75 kg traveling at 10 m/s, the power dissipated in
wheel bearings is 0.33 watts. A major technological breakthrough that
cut bearing friction in half would only gain 0.16 watts. It might be
significant in hour records or even the pursuit world record. At
normal levels of competition, I think the sleep gained in not worrying
about it has a greater performance benefit.
Dave Lehnen
I read that rather ludicrous answer by the Zipp representative about
ceramic bearings a couple of days ago. He claimed testing by Zipp-
sponsored CSC had shown a power reduction of 3 to 4 per cent with
ceramic balls compared to steel balls. This must have been a
percentage of the power lost in the steel-balled bearing, not the
total power supplied by the rider. But the Zipp guy applied the
efficiency gain to the whole system, not just to the wheel bearings.
There is no way you can save 10 watts in bearings that only
dissipate a fraction of a watt. Saving 4 per cent of 0.33 watts
sounds reasonable, gaining a whopping 0.0132 watts usable power.
Dave Lehnen
Owen Pope
news:3fdaef34$1@darkstar:
> Since I already wrote up the calculation, here it is:
>
> Typical coefficient of friction for a ball bearing is about
> mu=1.5e-3 (for example,
> http://www.ntnamerica.com/Engineering/PDFs/2200/frictemp.pdf
> ) Power dissipated in one bearing is
Do ceramic bearings even have ANY reduction in friction, in
practice? I guess that without grease, the ceramic may be
better, but when they're in a hub, there's no contact between
the bearing surfaces anyhow, so there's no difference.
Jobst tackled Zipp back in August, on the same subject:
http://groups.google.com/groups?q=zipp+%
2B+jobst+brandt+group:rec.bicycles.tech&hl=en&lr=&ie=UTF-8
&group=rec.bicycles.tech&selm=rRUZa.10833%24dk4.421696%
40typhoon.sonic.net&rnum=1
-Owen
frkrygow
> If shaving one's legs makes a real performance
> difference then ceramic bearings would most likely feel like riding with a
> tailwind.
:-) I love it!
But then there's the equally important issue:
"If" pigs could fly, what would ceramic bearings feel like?
Oh wait, I know: exactly the same!
--
Frank Krygowski [To reply, omit what's between "at" and "cc"]
Rocketman
news:3fdaef34$1@darkstar...
> Mike Kruger <Mik...@mouse-potato.com> wrote:
>
> > [Zinn punted on the question and quoted Zipp: ]
> > "For an average trained cyclist developing 250 watts, that's a savings
of
> > approximately 10 watts. At any level of competition, that is
significant.
> > The key is every part of the bearing system has seen marked improvements
in
> > precision resulting an a total benefit greater than the sum of its
parts. ...
>
> > But, as I said, I'm no engineer. So, for all you engineers out there,
can
> > there possibly be that big a difference in total system efficiency in
> > changing from high-quality steel bearing systems to anything else?
>
> No.
>
> This question would be more appropriate for rec.bicycles.tech than
> r.b.misc, so I'm crossposting it and setting followups there.
>
> I calculated that the power dissipated in two standard steel
> ball bearing hubs is about 0.33 watts for a 75 kg rider+bike
> traveling at 10 m/s.
That's higher than my seat-of-the-pants estimate; but in the same order of
magnitude.
> It's hard to see how an improved bearing could
> shave more than some percentage of that (10-20% ?), which verges on
> negligible, I'm sure the aerodynamics of your shoe straps are more
> important.
Yup. You got it. At racing speeds, bearing friction could be dropped out
of the equation and you wouldn't even notice.
Nice work! You should email your equations to the numerically-challenged
marketing department at ZIPP ;-)
-Rocketman
B.C. Cletta
>
> Typical coefficient of friction for a ball bearing is about mu=1.5e-3
> (for example, http://www.ntnamerica.com/Engineering/PDFs/2200/frictemp.pdf)
> Power dissipated in one bearing is
>
> P1 = mu * (m/2) * g * V_bearing where m is mass of rider+bike
> V_bearing is linear speed the
> bearing rotates at.
> P_bearingloss = 2*P1 total power lost in two wheels
is this per bearing or per wheel, i.e., two bearings in front, upwards
of six in rear (2 hub + 2 FH)?
jim beam
> Do ceramic bearings even have ANY reduction in friction, in
> practice? I guess that without grease, the ceramic may be
> better, but when they're in a hub, there's no contact between
> the bearing surfaces anyhow, so there's no difference.
the "no contact" [elasto-hydrodynamic separation, or ehs] argument is
over stated. check out the following page:
http://www.tribology-abc.com/calculators/e12_3.htm
scroll down to the electric conduction data and see why it's used to
measure ehs. than look at the rpm's necessary to achieve it.
at speeds of up to ~100 rpm, the sample bearing's surfaces are in pretty
much full contact for all loads. 100 rpm for a bike wheel = ~8mph.
that's hill climbing speed.
for a 50N [11lb] load, the bearing needs to be running >600rpm to be
achieving ehs - that's ~ 40mph. not even lance can cruise and do
significant milage at that speed. more realistically, lances cruising
speed would have to be even higher to achieve ehs with a more "real
world" front wheel load.
jb
Rick Onanian
<"frkrygow"@omitcc.ysu.edu> wrote:
>"If" pigs could fly, what would ceramic bearings feel like?
If pigs could fly, your main concern would be what's falling out of
the sky, not what your bearings feel like.
--
Rick "...I'm just glad that cows don't fly" Onanian
Benjamin Weiner
> > Power dissipated in one bearing is
> >
> > P1 = mu * (m/2) * g * V_bearing where m is mass of rider+bike
> > V_bearing is linear speed the
> > bearing rotates at.
> > P_bearingloss = 2*P1 total power lost in two wheels
> ^^^^^^
> is this per bearing or per wheel, i.e., two bearings in front, upwards
> of six in rear (2 hub + 2 FH)?
It doesn't matter, because if you have more bearings each carries
less weight. I divided it into two bearings each carrying m/2,
1/2 of the rider+bike, considering the bearings on both sides
of an axle as a unit. One could call it four bearings (2 per axle)
each carrying m/4. The answer is the same. I ignored the extra
force from bearing preload, by the way, but I don't think that's
important compared to the weight.
Freehub bearings don't enter into it anyway because they don't support
the rider's weight. Also, freehub bearings are fixed while the
rider is pedaling, and only move when coasting. To quote Sheldon,
"The freewheel is the least important bearing on a bicycle, since it
only turns when it is not carrying any load."