Funny Chain Lubricant Story
Tom Kunich
then allowed it to swing back and forth and measured the friction on it.
After 10 minutes or so the friction would start picking up and then within a
half hour the chain would be essentially unlubricated.
I asked what happened if they tested motor oil. The comment was that it
require weeks for the chain to start getting increased friction so they
didn't want to run such tests.
What does that tell you?
raa...@gmail.com
that you can't compare apples to oranges, that an engine has a
enclosed drivetrain and a bike has an exposed drivetrain; the dry lube
is to help prevent accumulation of gunk; further, the chaindrive is
something like 99% efficient, so you can basically do whatever you
like- the difference is how messy do you really want to get ?
carl...@comcast.net
Dear Tom,
It may tell us that the online magazine mis-measured friction.
Spicer tested lubricated and unlubricated bicycle chains in 2000 and
found that lubrication had no significant effect on transmission
efficiency--even when the lubricant was removed by cleaning.
After testing a chain with Castrol Wrench Force Dry Lube, Pedro’s Syn
Lube, Generation 4 White Lightning, Spicer thoroughly cleaned the
chain and tested it dry.
No significant differences were noted by his testing equipment:
"However, these results also indicate that the actual lubricant used
has little effect on the overall performance of the drive under
laboratory conditions given the precision of the measurement.
In addition, the chain used for the lubrication study was fully
degreased and was re-tested for efficiency. This degreasing operation
consisted of a five-minute scrub with kerosene followed by a cleaning
with Castrol Degreaser. The measured efficiency of the de-lubricated
chain for the 52–15 combination at 60 RPM and 100 W was 90.3% and at
200 W was 96.5%. These efficiencies are essentially the same as those
measured for the chain in the re-lubricated condition."
It's the first article here:
http://www.ihpva.org/HParchive/PDF/hp50-2000.pdf
Cheers,
Carl Fogel
Tom Kunich
news:15635415vnsgs4d41...@4ax.com...
>
> Spicer tested lubricated and unlubricated bicycle chains in 2000 and
> found that lubrication had no significant effect on transmission
> efficiency--even when the lubricant was removed by cleaning.
>
> After testing a chain with Castrol Wrench Force Dry Lube, Pedro's Syn
> Lube, Generation 4 White Lightning, Spicer thoroughly cleaned the
> chain and tested it dry.
>
> No significant differences were noted by his testing equipment:
That's pretty interesting. You do understand that the idea of testing a
CLEAN chain isn't of much use? Or that lubrication is supposed to allow the
chain to run low friction despite dirt etc. in the mechanism?
carl...@comcast.net
Dear Tom,
As Spicer's testing showed, lubrication has next to no effect on
bicycle chain transmission efficiency.
The chief effect of lubrication on bicycle chains is to keep grit out.
Anyone can read Spicer's test:
http://www.ihpva.org/HParchive/PDF/hp50-2000.pdf
Maybe you can tell us more about whatever test you had in mind.
Cheers,
Carl Fogel
Frank Krygowski
>
>
> The chief effect of lubrication on bicycle chains is to keep grit out.
I'd have said the chief effect of lubrication is to trap the grit on
the chain, and distribute it on other parts of the bike. And the
rider's leg.
That's been true of every wet lube I've tried, anyway.
- Frank Krygowski
carl...@comcast.net
Dear Frank,
Wet lube initially keeps road dust out. (Nothing larger will fit
between pins and rollers.)
Any wet lube traps the road dust flung up by the tires, so the wet
lube turns black within a few miles.
After that, the outside of the chain is covered with an extremely fine
polishing sludge, which gradually mixes with the thin film of clean
lube inside the rollers as the minuscule pumping action draws it in
and out.
Eventually, the area between the pins and rollers builds up a thin
layer of polishing paste.
Chain wear rates suggest how long the process takes.
Each roller turns a tiny bit as it engages the top of the front
sprocket, a tiny bit as it exits, and ditto for the rear sprocket and
any idler pulleys.
But only two of the turns are under any significant load, when the
chain is pulled onto the front sprocket and pulls off the rear
sprocket.
At 100 rpm on a 53-tooth front sprocket, the chain moves at a mere 2.5
mph. (At the same 100 rpm on a 175 mm crank, the rider's foot whirls
at a blistering 4.1 mph.)
At 100 rpm on a 53-tooth sprocket, an individual roller on a
106-roller chain makes its small partial turn under load (entering the
front or leaving the rear sprocket) at only a rate of once every
3/5ths of a second.
How small is that turn?
As it pulls onto the front 53-tooth, the roller turns and locks in
1/53rd of a circle--a bit less than 7 degrees. (About 9 degrees for a
39-tooth.) Then it just sits there until it eases off the sprocket.
The extreme case is an 11-tooth rear, where the chain pulling off has
its pin and roller turn about 32 degrees.
So an hour of 100 rpm riding produces only 6,000 partial polishing
rotations (7 to 32 degrees) on any single pin and roller, amounting to
much less than 400 full turns per hour.
That's about 0.25 rpm.
(The 106-roller chain is a convenient figure that's easy to calculate
and produces a slightly inflated result compared to the typical
~114-roller chain.)
In other words, most bicycle chains last over a thousand miles, even
on a diet of polishing paste made from oil and road dust, because the
individual pins and rollers turn so little and because the polishing
paste can contain only the grains of extremely hard road dust small
enough to fit between the pins and rollers, whose clearance is below
what a typical micrometer can measure.
As a crude example, a 53x19 at 91.5 rpm on a 2124 mm tires is doing 20
mph. Each pin on a 106-roller chain makes its partial turns under load
at 91.5 rpm, about 5500 partial polishing turns every hour or every 20
miles. That's 27,500 slight "grinds" averaging only about 20 degrees
in a thousand miles and fifty hours of steady riding at about 90 rpm.
When 25 of these pins (the ends of a foot-long ruler) wear 0.0025",
the chain elongates a full 1/16th of an inch (0.0625") and should be
replaced. The 0.0025" wear on each pin-roller combination amounts to
about the thickness of a sheet of flimsy phone-book paper.
***
Such incredibly tiny wear explains why it's almost impossible to clean
real wet-lube chains that have gotten dirty.
Dunk a dirty, oily chain in solvent in an ultrasonic cleaner, shake it
in a bottle, do whatever you please--
Eventually, you'll probably run out of patience before fresh solvent
stops producing wisps of filth from what looks like an immaculate
chain.
The solvent acts only on the incredibly thin exposed edge of the film
of oil-and-grime trapped between each pin and roller, a surface that
isn't much wider than a human hair. Given such poor access, the
solvent takes forever, even with shaking or ultrasonic action, to eat
into the mess trapped between the pins and rollers.
***
Dry wax has the advantage that it draws no road dust into the
pin-roller interface. Instead of a oily film pumping in and out, the
dry lube flakes outward under pressure from between the pin and roller
and never returns.
The price for this is that the dry lube needs to be re-applied more
often than a wet lube (with exceptions of all kinds for different
lubes and conditions).
Frank is quite happy with wax that has some oil added, others swear by
various oils, and I've been reasonably pleased with Dupont Teflon
spray wax.
(I can't recommend it over oil or melted wax, but it's fairly cheap,
easy to apply, and has a pleasant new-toy effect that hasn't worn off
yet.)
***
As far as I know, lubrication makes no significant difference to chain
friction in terms of power transmission. Spicer's article explains the
theory behind this unexpected result, which could be grossly
simplified to a matter of how little actual polishing action takes
place in the lazily moving chain of a bicycle.
Even a dry chain that squeaks like a box full of bats still transmits
almost exactly the same power as a brand-new factory-lubed chain--the
noise is annoying, but it's apparently all out of proportion to the
increase in friction.
Cheers,
Carl Fogel
Ben C
[...]
> Each roller turns a tiny bit as it engages the top of the front
> sprocket, a tiny bit as it exits, and ditto for the rear sprocket and
> any idler pulleys.
>
> But only two of the turns are under any significant load, when the
> chain is pulled onto the front sprocket and pulls off the rear
> sprocket.
>
> At 100 rpm on a 53-tooth front sprocket, the chain moves at a mere 2.5
> mph. (At the same 100 rpm on a 175 mm crank, the rider's foot whirls
> at a blistering 4.1 mph.)
>
> At 100 rpm on a 53-tooth sprocket, an individual roller on a
> 106-roller chain makes its small partial turn under load (entering the
> front or leaving the rear sprocket) at only a rate of once every
> 3/5ths of a second.
>
> How small is that turn?
>
> As it pulls onto the front 53-tooth, the roller turns and locks in
> 1/53rd of a circle--a bit less than 7 degrees. (About 9 degrees for a
> 39-tooth.) Then it just sits there until it eases off the sprocket.
>
> The extreme case is an 11-tooth rear, where the chain pulling off has
> its pin and roller turn about 32 degrees.
>
> So an hour of 100 rpm riding produces only 6,000 partial polishing
> rotations (7 to 32 degrees) on any single pin and roller, amounting to
> much less than 400 full turns per hour.
>
> That's about 0.25 rpm.
It is said that cross-chaining increases rate of wear. Why is that? Are
the pins still only worn as they pull on and off the front and rear
sprockets, and it's made worse by having to pull them back into line, or
do they get worn continuously in some way in a cross-chaining situation?
Frank Krygowski
>
>
> It is said that cross-chaining increases rate of wear. Why is that? Are
> the pins still only worn as they pull on and off the front and rear
> sprockets, and it's made worse by having to pull them back into line, or
> do they get worn continuously in some way in a cross-chaining situation?
My guess would be this: With the chain running at an angle, the loads
transmitted between links and pins (at the same parts of the chain
travel Carl describes) are not supported by the entire width of the
pin. They're supported by a larger pressure load applied to the end
of the pin, while the rest of the pin gives little help. More
pressure leads to more wear. Similar problems exist when mechanical
shafts are supported by plain bearings that are not in proper
alignment; and for shafts that are insufficiently rigid, and deflect
under load as they spin. All this stuff works better when things are
straight.
In addition, IIRC, chain efficiency is lower when the chain doesn't
run in a straight line. I imagine this is partly due to the effect I
just described, and partly because a laterally-bent chain generates
more friction between the side plates.
- Frank Krygowski
jobst....@stanfordalumni.org
Effects of running chains diagonally, between sprockets not on the
same line, is apparent from worn chain pictures on the web that show a
worn link pins that have barrel shaped wear grooves at mid section.
The depth of the wear groove at its ends indicating canted loading.
Even with more closely spaced (10) sprockets, a greater angle occurs
than was common with five slightly wider spaced sprocket. In the days
of 5-sprocket clusters riders generally used the inner chainwheel to
drive the lower three and the large chainwheel to drive the upper
three, seldom riding in extreme cross-over mode. It was once a topic
of discussion on this newsgroup, but now that 30 combinations are
possible, little consideration is given to resulting chain life,
constant cadence having higher priority regardless of what chain line
this requires.
Jobst Brandt
carl...@comcast.net
Aaargh! Off by more than order of magnitude!
Absolutely no idea where I came up with 0.25 rpm--probably just
looking at the wrong figure or not noticing that a numeral was missing
from a number when I punched enter. That'll teach me not to show my
work.
Slightly short 106-roller chain on 53x11, 100 rpm, single pin-turn
with each pedal stroke, once as it pulls onto front 53, once as it
exits rear 11.
Average turn in degrees is:
( (1/53 * 360) + (1/11 * 360) ) / 2
or (6.8 + 32.7 ) / 2
or ~20 degrees
At 100 rpm, there are 6,000 20-degree turns per hour, the equivalent
of (20/360) * 6,000, or 333.3 full turns in 60 minutes, which is 5.5
rpm, not the mysterious and mistaken 0.25 rpm.
Of course, 5.5 rpm is still pretty slow for polishing chain rollers.
***
A equally embarrassing pair of mistakes in basic arithmetic involve
the 53x19 at 91.5 rpm and 20 mph on a 2124 mm rear tire.
At 91.5 rpm, any pin on the slightly short 106-roller chain makes
about 5500 partial turns (5,490), so in fifty hours (1,000 miles at 20
mph) any pin goes through about 275,000 partial turns (274,500).
That's 275,000 partial turns, ten times the 27,500 that I scribbled
due to my careless, dim-witted reading of a calculator.
The other mistake is that with a 53x19, the average turn isn't ~20
degrees, it's ~13 degrees.
The average degrees for the two turns:
( (1/53) * 360 ) + ( (1/19) * 360 ) ) / 2
or (6.8 + 18.9) / 2
or only 12.9 degrees average turn for 53x19, not 20 degrees.
At 91.5 rpm, that's about 5500 13-degree turns per hour
or (13/360) * 5500 = ~200 full turns per hour, about 3 rpm.
Careless but detail-obsessed nitwits should always show their work.
Cheers,
Carl Fogel
carl...@comcast.net
Dear Frank & Ben,
As Frank points out, the twisting concentrates the force at one side
and puts things out of line, which is always bad for wear--you want
the load distributed as evenly as possible.
This twisting and sideways rocking also tends to squish more of the
oil-and-dust (or water-and-mud) in and out of the chain crevices,
which encourages chain wear. Sideways flex is just plain bad because
it helps work road dust into the chain guts.
As for the related matter of efficiency, there is some loss with
cross-chaining, but it's small compared to the penalty for using
smaller gears.
The testing already mentioned by Spicer showed that offsetting
(cross-chaining 52x11, for example) produced only about 0.5% power
losses, while smaller sprockets lost about 3% more power than larger
sprockets (52x11 was about 3% less efficient than 52x21).
Big sprockets good, small sprockets bad.
The details are near Table 1 and the conclusion, after the statistics
are pondered, is that "it appears that the offset has a negligible
effect on efficiency."
jim beam
> On Thu, 12 Jun 2008 15:16:05 -0700, "Tom Kunich" <cyclintom@yahoo.
> com> wrote:
>
>> <carl...@comcast.net> wrote in message
>> news:15635415vnsgs4d41...@4ax.com...
>>> Spicer tested lubricated and unlubricated bicycle chains in 2000 and
>>> found that lubrication had no significant effect on transmission
>>> efficiency--even when the lubricant was removed by cleaning.
>>>
>>> After testing a chain with Castrol Wrench Force Dry Lube, Pedro's Syn
>>> Lube, Generation 4 White Lightning, Spicer thoroughly cleaned the
>>> chain and tested it dry.
>>>
>>> No significant differences were noted by his testing equipment:
>>> http://www.ihpva.org/HParchive/PDF/hp50-2000.pdf
>> That's pretty interesting. You do understand that the idea of testing a
>> CLEAN chain isn't of much use? Or that lubrication is supposed to allow the
>> chain to run low friction despite dirt etc. in the mechanism?
>
> Dear Tom,
>
> As Spicer's testing showed, lubrication has next to no effect on
> bicycle chain transmission efficiency.
>
> The chief effect of lubrication on bicycle chains is to keep grit out.
it's there to lubricate. it can't and doesn't keep grit out. if
anything, lube retains it.
>
> Anyone can read Spicer's test:
> http://www.ihpva.org/HParchive/PDF/hp50-2000.pdf
while that article sincerely seeks to address the causes, it's
incredibly naive in terms of reality. "degreasing" new chains is highly
ineffective because:
1. surface adsorption still retains a layer of grease/lube.
2. even if a chain were assembled without lube at the factory, it would
still have the processing lubes on it used during each component's
forming process.
the chain therefore runs with this surface layer until such time as it
wears out or otherwise becomes contaminated. this is far removed from
real world service where abrasives and water can destroy surface
lubrication and thus allow friction to become much more significant.
"true" degreasing basically involves removing a surface layer of the
material that was previously greased. if that were to be done, surface
friction welding will follow in double-quick time and friction would
become very significant very quickly. real world service in fact sees
two "true" degreasing mechanisms in action - physical abrasion and
chemical action. road grit performs the former, water the latter once
it undercuts and corrodes any surface layers.
>
> Maybe you can tell us more about whatever test you had in mind.
http://en.wikipedia.org/wiki/Adsorption
surface chemistry has attracted a lot of attention in recent years. in
fact, i believe there was even a nobel prize for one of its research
pioneers recently.
travis...@gmail.com
> carlfo...@comcast.net wrote:
> > On Thu, 12 Jun 2008 15:16:05 -0700, "Tom Kunich" <cyclintom@yahoo.
> > com> wrote:
>
The discussion hasn't mentioned protection from corrosion, but it
should have. Wear of corroded surfaces, with their dramatically
greater surface area, is much greater.
Calculations of chain efficiency are interesting, and no doubt well
done. But, they are beside the point if little energy is needed to
remove material, with abrasives, from the wearing parts of the chain.
And consider the noise. If the wear on the chain were as bad as the
noise of a squeeking drive train is offensive, you'd certainly lube
your chain.
Great feature of this group is that Carl might find the numbers to
measure just how efficiently a dry, unlubricated chain, can drive the
next rider nuts. <smile> Hint: Requires truly minimal energy.
Harry Travis
jobst....@stanfordalumni.org
> The discussion hasn't mentioned protection from corrosion, but it
> should have. Wear of corroded surfaces, with their dramatically
> greater surface area, is much greater.
I'm not sure what sort of corrosion you have in mind, but my chain
sees nothing but oil and on occasion rain water, that in time washes
all oil out of the chain. Water works as a good lubricant but
unfortunately evaporates readily and leaves the chain unlubricated.
http://www.sheldonbrown.com/brandt/chain-care.html
> Calculations of chain efficiency are interesting, and no doubt well
> done. But, they are beside the point if little energy is needed to
> remove material, with abrasives, from the wearing parts of the
> chain.
I think you are looking in the wrong place. The smaller the sprocket,
the larger the articulation angle and proportionately the wear and
minuscule energy loss in the hinge-pin. For instance, two 60t
sprockets make the chain bend 24° in one revolution of the chain. a
53t-11t combination make the chain bend 79°.
> And consider the noise. If the wear on the chain were as bad as the
> noise of a squeeking drive train is offensive, you'd certainly lube
> your chain.
If it squeaks it must be clean. Grit does not allow metal-to-metal
stick-slip squeak.
> Great feature of this group is that Carl might find the numbers to
> measure just how efficiently a dry, unlubricated chain, can drive
> the next rider nuts. <smile> Hint: Requires truly minimal energy.
Do your own analysis if you have a grasp of the subject that you feel
allows you to write these lines.
Jobst Brandt
jim beam
<snip for cleanliness>
>
> If it squeaks it must be clean. Grit does not allow metal-to-metal
> stick-slip squeak.
that's not true, unless you're talking earth-mover quantities of dirt.
grit very effectively removes surface layers, and thus allows
metal-to-metal contact. asperity welding follows, with squeaking.
>
>> Great feature of this group is that Carl might find the numbers to
>> measure just how efficiently a dry, unlubricated chain, can drive
>> the next rider nuts. <smile> Hint: Requires truly minimal energy.
>
> Do your own analysis if you have a grasp of the subject that you feel
> allows you to write these lines.
bothering to do analysis never stopped you!
Ralph Barone
jobst....@stanfordalumni.org wrote:
Admittedly, there's little to no tension in the rear derailleur idlers,
but wouldn't that be where the greatest chain bend angle would be found?
carl...@comcast.net
>I think you are looking in the wrong place. The smaller the sprocket,
>the larger the articulation angle and proportionately the wear and
>minuscule energy loss in the hinge-pin. For instance, two 60t
>sprockets make the chain bend 24° in one revolution of the chain. a
>53t-11t combination make the chain bend 79°.
[snip]
>Do your own analysis if you have a grasp of the subject that you feel
>allows you to write these lines.
>
>Jobst Brandt
Dear Jobst,
Aha! A chance to test the newest part of my spreadsheet.
A chain bends only twice under load, not four times.
It bends under load twice on the top run, as it pulls off the rear and
pulls onto the front.
There's no significant load on the lower run as the chain exits the
front or engages the rear sprocket, unless it's a fixie and the rider
brakes through the pedals.
Nor is there any significant load as the chain wraps around the two
idler pulleys--which would have been 8 bends.
(That's one reason why derailleur chains don't wear out twice as fast
as single-speed chains.)
For a 60x60 (a silly but easily calculated example), that's (360/60) +
(360/60) degrees, a total of only 6 + 6 = 12 degrees of rotation under
load as a pin makes a trip around the sprockets.
For a 53x11 (verging on silly, but I pedal it and it exercises a
spreadsheet), that's (360/53) + (360/11) degrees, a total of only 32.7
+ 6.8 = 39.5 degrees of rotation under load.
For a 39x20 (just an example for those grinding up hills), that's
(360/39) + (360/20) degrees, a total of only 9.2 + 18.0 = 27.2 degrees
of rotation under load.
On the other six bends of a derailleur, there is the slight load of
the feeble derailleur spring and the weight of the chain.
Just how little load exists on the lower chain-run was shown by some
early riders who happily let their lower chain-runs dangle loose, with
only an unsprung roller to guide the chain onto the rear sprocket, a
strange approach known as a floating chain:
http://i16.tinypic.com/4gj8g2c.jpg
For true nit-pickers, the length of the chain and the size of the
sprockets affects the rotation-under-load RPM as compared to the pedal
cadence, meaning that longer chains wear at a lower rate.
For a 52x12 derailleur 114-link at 90 RPM, the pins turn under load at
4.21 RPM.
For a 52x12 fixie with a shorter 108-link, the bike goes the same
speed at the same 90 RPM, but the pins turn under load slightly
faster, 4.44 RPM.
Switch to a giant double-length 232-link chain for recumbents and
tandems, and the 52x12 at 90 RPM turns the pins under load at only
2.07 RPM.
***
Incidentally, the real loss of chain efficiency is not so much the
tiny bending of the chain at the pin, but what's called chordal
action. As any chain run (slack or under load) enters the sprocket, it
abruptly decelerates as the link moving in a straight line snaps down
and takes a short-cut to form a chord across the inside of the
"circle" of the sprocket.
The greater the angle, the greater each link decelerates as it enters
the sprocket, and the more the chain run vibrates, which is why small
sprockets are less efficient.
Here's a nice explanation of chordal action, with equations and the
vibratory point:
http://chain-guide.com/basics/2-2-1-chordal-action.html
The vibration wastes far more power than the amazingly tiny friction
that takes thousands and thousands of miles to wear the pin-roller
interface the 0.0025" that adds up to 1/16th of an inch elongation in
a foot of chain.
Using the chordal action equation, here are some sample chain speed
variations for X teeth.
teeth
chordal action
53 0.1756%
52 0.1824%
50 0.1973%
46 0.2331%
42 0.2776%
39 0.3243%
38 0.3146%
34 0.4266%
32 0.4815%
30 0.5478%
28 0.6288%
24 0.8555%
21 1.1169%
20 1.2312%
19 1.3639%
18 1.5192%
17 1.7027%
16 1.9215%
15 2.1852%
14 2.5072%
13 2.9058%
12 3.4074%
11 4.0507%
10 4.8943%
9 6.0307%
That's why big sprockets are more efficient at transmitting power--the
heavy chain run isn't being sped up and slowed down as much every time
a new link engages the sprocket. Wear is more a matter of how gritty
the debris in the polishing paste--
Er, how dirty the chain oil is.
As someone said, "Do your own analysis if you have a grasp of the
subject that you feel allows you to write these lines."
:-)
Cheers,
Carl Fogel
carl...@comcast.net
wrote:
Dear Ralph,
Yes, there's no load on the chain on the bottom run, except for its
own weight and the feeble derailleur spring.
Jobst mistakenly included the lower run in his explanation, using 4
turns, instead of just the upper run's 2 turns, where the chain is
under load.
The upper run is where the wear occurs because the pins turn under
load as the chain exits the rear and engages the front.
Power loss is different than wear.
The power lost polishing the pins under load is quite small. It takes
thousands of miles to polish each pin interface about 0.0025" and
elongate a foot of chain a whole 1/16th of an inch.
Most of the loss occurs because of what's called chordal action, which
is why lubrication makes little difference to power transmission.
When the chain engages the sprocket, the speed changes. In crude
terms, the link snaps down as it pulls onto the front sprocket and
takes a shortcut across the "circle" of the sprocket, a tiny chord
across the inside of the circle.
So the long, heavy chain run is constantly speeding up and slowing
down a little bit, which means that it vibrates. Accelerating that
mass in a twanging motion takes power.
The smaller the sprocket, the greater the shortcut. The bigger the
shortcut the link takes across the inside of the sprocket "circle",
the greater the change in chain speed, vibration, and power loss.
Here's a page that gives the equation for such chordal action:
http://chain-guide.com/basics/2-2-1-chordal-action.html
As the graph at the bottom shows, the chain speed variation due to
chordal action is almost nothing at 53 teeth (0.1756%), but is about
twenty times as much at 11 teeth (4.0507%).
Keep in mind that the % of chain speed variation is not a direct
measure of power loss--that's a different percentage. But the two are
fairly well related, which is one reason why small sprockets lose more
power than large sprockets.
The speed-change rises very steeply as the tooth count approaches 11
teeth:
teeth
chordal-speed-variation
increase-from-16-teeth
16 1.9215%
15 2.1852% +13.7%
14 2.5072% +30.5%
13 2.9058% +51.2%
12 3.4074% +77.3%
11 4.0507% +110.8%
That's why land-speed-record bikes with two chains connected by a
jackshaft use much bigger sprockets than necessary to obtain their
high gearing.
A pair of ordinary 52x12's coupled by a jackshaft would produce
18.7-to-1 gearing, roughly what's used.
But the two chains would be vibrating badly because of the 12-tooth
sprockets, whose average chain-speed variation is 3.4%.
So Rompelberg used a 70x15 and 60x16 (only 17.5-to-1 gearing) at first
and then 70x15 and 60x14 (20-to-1 gearing) for his land-speed records.
That reduced the average chain-speed-variation down to around 2% and
2.3%, with the mismatched tooth-counts avoiding the two chains
vibrating in synchronization.
Cheers,
Carl Fogel
jim beam
good informative post.
Tom Kunich
<carl...@comcast.net> wrote in message
news:m9jb54dga83i16639...@4ax.com...
So, Carl, do you lube a chain or leave it dry?
carl...@comcast.net
Dear Tom,
As I said earlier in this thread, I'm now spraying my chain with
Dupont Teflon wax.
But I doubt that it's much better overall (or worse) than all the
other schemes . . .
Such as the exotic oils at incredible prices used by many happy RBT
posters and local bike shops, melted wax blended with just a hint of
an unassuming table oil (Frank's preference, with years of good
results), the liquid drip-on waxes at perfume-bottle prices that some
people swear by, or the drippings from discarded motor-oil cans
extracted from dumpsters (Jobst has mentioned doing this to stop
squeaks after rain in the Alps).
It's about $6 a can at Lowes hardware.
Like just about everyone, I end up replacing exposed chains.
My impression is that people who ride where it's wet and muddy buy
chains more often--gosh, what a surprise!
But every bike's tires stir up that invisible cloud of fine road dust,
which gets into the chain guts no matter what magic lube is used.
Small as the dust particles are, they're harder than the chain.
And the road dust particles have to be small--they can't do their
dirty work unless they fit between a new pin and roller, a space
smaller than most micrometers can measure.
Bicycle chains that run in cases last longer, partly because they're
protected a little from road dust, but mostly because most such bikes
aren't ridden very hard--the more gently you pedal, the longer the
chain lasts.
If the oil is black, it's full of road dust. Every bike chain in a
case that I've seen (not that many) was black as sin.
Some motorcycles ran primary chains in sealed oil baths (truly sealed,
not just a bike chain case). They essentially lasted forever, partly
because there was virtually no polishing action with the clean oil and
partly because they were double-row chains, which spread the load out.
It's worth noting that no one has produced a credible test showing a
chain in action transmitting significantly different power when
different lubes are used--most of the power loss is due to chordal
action, the variation in chain speed caused by straight links of chain
suddenly wrapping around the "circle" of a sprocket. That vibration is
why even the motorcycle chains running in sealed oil baths lose around
3% power.
The wear, annoying though it is, is a separate matter and doesn't cost
much power. It takes forever to polish about 0.0025" of metal off all
those rollers, and that wear depends on the road dust between the
surfaces.
Cheers,
Carl Fogel
jobst....@stanfordalumni.org
http://www.sheldonbrown.com/brandt/chain-care.html
>> Admittedly, there's little to no tension in the rear derailleur
>> idlers, but wouldn't that be where the greatest chain bend angle
>> would be found?
> Yes, there's no load on the chain on the bottom run, except for its
> own weight and the feeble derailleur spring.
> Jobst mistakenly included the lower run in his explanation, using 4
> turns, instead of just the upper run's 2 turns, where the chain is
> under load.
The example I cited only compares what the ratio of bends in chains is
for the combinations chosen. It does not say where the load is or
what wear there is. Derailleur idlers cause bends that when clean and
unlubricated, cause the squeak s that people mention. It is not the
loaded entry or exit from sprockets.
Jobst Brandt
Frank Krygowski
It's certainly true that low tooth counts give more chordal action and
less smooth power transmission. But I'm not convinced that this, in
itself, causes a loss in efficiency (or power, or energy).
If you look at the chain drive as a thermodynamic system, any energy
wasted has to leave as heat. Friction generates heat and (usually)
wastes energy. Do you have evidence that chordal action does the
same, beyond what comes from the friction of the usual flexing the
links at the bends?
You said "So the long, heavy chain run is constantly speeding up and
slowing down a little bit, which means that it vibrates. Accelerating
that mass in a twanging motion takes power." I'm not convinced that
it does - or rather, I think that if it does, it must be buried in
some secondary or tertiary effect. IOW, I think the energy invested
in the vibration is negligible, and/or almost all recovered.
As an analogy, riding a bike on a smooth road with a series of gentle
hills or undulations (say, on a 100 foot wavelength) has the
bike&rider accelerating slightly down hills, then decelerating on the
way back up. But if it weren't for the secondary effect of increased
aero losses on the higher speed portions, there wouldn't be any
significant energy loss caused by the accelerations. And in fact,
those sorts of roads tend to give higher TT speeds, IIRC.
So: more roughness in the drive motion, I agree. Larger bend angle,
therefore a tad more friction loss, I agree. But aside from that, do
you have an explanation of where chordal action causes the energy
actually to leave the system?
- Frank Krygowski
carl...@comcast.net
Dear Frank,
Changing the speed of the chain run every time a link engages the
sprocket means accelerating and decelerating the chain run.
Acceleration and deceleration take power.
At 90 RPM on a 53x11, pins are engaging and disengaging on the top run
4770 times per minute. The chain speed change on the front 53 is
0.1976%, but 4.0507% on the 11-tooth rear.
It takes power to speed up and slow down the chain run ~ 4% almost
5,000 times per minute.
Cheers,
Carl Fogel
Tom Kunich
news:db4d54961e9ju2qp3...@4ax.com...
> Changing the speed of the chain run every time a link engages the
> sprocket means accelerating and decelerating the chain run.
>
> Acceleration and deceleration take power.
>
> At 90 RPM on a 53x11, pins are engaging and disengaging on the top run
> 4770 times per minute. The chain speed change on the front 53 is
> 0.1976%, but 4.0507% on the 11-tooth rear.
>
> It takes power to speed up and slow down the chain run ~ 4% almost
> 5,000 times per minute.
Still - what's the size of the power loss? Are you accelerating and
decelerating ONLY the single link?
jobst....@stanfordalumni.org
> Changing the speed of the chain run every time a link engages the
> sprocket means accelerating and decelerating the chain run.
> Acceleration and deceleration take power.
Don't confuse accelerating and slowing down on a bicycle with
mechanical action and losses. A swinging pendulum accelerates from
standstill to maximum speed every cycle and does so in a vacuum for a
long time, demonstrating that there is no power required.
> At 90 RPM on a 53x11, pins are engaging and disengaging on the top
> run 4770 times per minute. The chain speed change on the front 53
> is 0.1976%, but 4.0507% on the 11-tooth rear.
> It takes power to speed up and slow down the chain run ~ 4% almost
> 5,000 times per minute.
Where does the power go? What power does it take to slow down a
moving chain and where is it extracted from the mechanism?
Jobst Brandt
Frank Krygowski
>
>
> Dear Frank,
>
> Changing the speed of the chain run every time a link engages the
> sprocket means accelerating and decelerating the chain run.
>
> Acceleration and deceleration take power.
Acceleration takes power or energy. Deceleration (in this instance)
would give back power or energy.
I'm sure we've previously discussed the idea of a bike with large-mass
wheels, like flywheels, but with the same total mass as a normal
bike. Yes, it takes more energy to accelerate the flywheel bike up to
speed, and it wouldn't be good for sprints. But if you were to point
that bike up a hill, you'd recover your acceleration energy, as it
helped prevent the bike from decelerating. It wouldn't decelerate as
quickly as a normal bike of equal total mass.
A similar industrial application is a stamping press with a large
flywheel driven by a small motor. Energy is stored in the flywheel as
the motor accelerates it up to speed. That energy is given back to
the system when the press is activated and the flywheel decelerates.
I think your chain is doing the same when it's in its deceleration
phase. I don't see that energy being wasted into heat, except for the
previously discussed pin friction during bending.
- Frank Krygowski
Ben C
[...]
>
> It's worth noting that no one has produced a credible test showing a
> chain in action transmitting significantly different power when
> different lubes are used--most of the power loss is due to chordal
> action, the variation in chain speed caused by straight links of chain
> suddenly wrapping around the "circle" of a sprocket. That vibration is
> why even the motorcycle chains running in sealed oil baths lose around
> 3% power.
So why does efficiency increase with tension? I suppose a tighter chain
vibrates less.
Thinking of your picture of the "floating" chain, you might think well
you've got to lift that dangling chain up each time it goes onto the
sprocket, but then of course it falls back down giving you the energy
back.
jobst....@stanfordalumni.org
You seem to be using wreck.bike as a thought scratch pad. I notice
there is no question mark or indication of a proposal included. By
airing such dabblings, some readers may take up the thread and build
on it. That seems to be occurring in technology lately.
Jobst Brandt
John Henderson
> You seem to be using wreck.bike as a thought scratch pad. I
> notice there is no question mark or indication of a proposal
> included. By airing such dabblings, some readers may take up
> the thread and build on it. That seems to be occurring in
> technology lately.
But balancing that negative assessment is the fact that the
growth of knowledge would be slowed greatly without dialectics,
including the re-asking of seemingly naive questions.
John
carl...@comcast.net
Dear Frank,
I'm baffled by arguments that power is somehow not required to vibrate
a long chain run by increasing and decreasing the speed up to 4% as
links engage gear teeth.
Cheers,
Carl Fogel
Tom Kunich
news:ghnd54h4cbv50825p...@4ax.com...
> On Mon, 16 Jun 2008 13:33:39 -0700 (PDT), Frank Krygowski
>
> a long chain run by increasing and decreasing the speed up to 4% as
> links engage gear teeth.
Carl, I don't think that's the question. I think that we're all wondering
about the actual losses. Obviously there are losses but are they
significant?
jobst....@stanfordalumni.org
As tom Sherman would say "But, but, but,..." That is what we have
here today. Problems introduced here are neither new nor not
understood but by your "dialectic"
http://www.merriam-webster.com/dictionary/dialectic
these threads go far afield, contributors foreign to the concepts
adding their imagination of what strikes their fancy. By this method,
wreck,bike has become an unreliable source, filled with of bicycle
lore.
Use of the construct "the fact that" already degrades scientific
discussion. This "begging the question", because establishing facts
is the goal of the exchange.
Jobst Brandt
jobst....@stanfordalumni.org
>>> Changing the speed of the chain run every time a link engages the
>>> sprocket means accelerating and decelerating the chain run.
>>> Acceleration and deceleration take power.
>> Acceleration takes power or energy. Deceleration (in this
>> instance) would give back power or energy.
>> I'm sure we've previously discussed the idea of a bike with
>> large-mass wheels, like flywheels, but with the same total mass as
>> a normal bike. Yes, it takes more energy to accelerate the
>> flywheel bike up to speed, and it wouldn't be good for sprints.
>> But if you were to point that bike up a hill, you'd recover your
>> acceleration energy, as it helped prevent the bike from
>> decelerating. It wouldn't decelerate as quickly as a normal bike
>> of equal total mass.
>> A similar industrial application is a stamping press with a large
>> flywheel driven by a small motor. Energy is stored in the flywheel
>> as the motor accelerates it up to speed. That energy is given back
>> to the system when the press is activated and the flywheel
>> decelerates.
>> I think your chain is doing the same when it's in its deceleration
>> phase. I don't see that energy being wasted into heat, except for
>> the previously discussed pin friction during bending.
> I'm baffled by arguments that power is somehow not required to
> vibrate a long chain run by increasing and decreasing the speed up
> to 4% as links engage gear teeth.
That depends on what you call significant. Even a guitar string
vibrating in air gradually damps asymptotically to stillness. That
isn't what we consider significant losses in bicycle chains whose
vibrations are substantially lower frequency than those of an acoustic
element.
Jobst Brandt
Bill Sornson
>
>> You seem to be using wreck.bike as a thought scratch pad.
This from a man who posts off-topic political scat (seldom his own thoughts,
btw, just ideological boilerplate) on a fairly regular basis.
Bill "ironic hypocrisy or hypocritical irony?" S.
John Henderson
> As tom Sherman would say "But, but, but,..." That is what we
> have here today. Problems introduced here are neither new nor
> not understood but by your "dialectic"
>
> http://www.merriam-webster.com/dictionary/dialectic
>
> these threads go far afield, contributors foreign to the
> concepts
> adding their imagination of what strikes their fancy. By this
> method, wreck,bike has become an unreliable source, filled
> with of bicycle lore.
>
> Use of the construct "the fact that" already degrades
> scientific discussion. This "begging the question", because
> establishing facts is the goal of the exchange.
Finding a balance isn't easy. You might enjoy this
(non-cycling) article:
http://www.caravanandmotorhomebooks.com/articles/fact_opinion.html
I did when I first read it.
John
Andre Jute
> jobst.bra...@stanfordalumni.org wrote:
> > As tom Sherman would say "But, but, but,..." That is what we
> > have here today. Problems introduced here are neither new nor
> > not understood but by your "dialectic"
>
> >http://www.merriam-webster.com/dictionary/dialectic
>
> > these threads go far afield, contributors foreign to the
> > concepts
> > adding their imagination of what strikes their fancy. By this
> > method, wreck,bike has become an unreliable source, filled
> > with of bicycle lore.
>
> > Use of the construct "the fact that" already degrades
> > scientific discussion. This "begging the question", because
> > establishing facts is the goal of the exchange.
>
> Finding a balance isn't easy.
I was a mistake to let the public in on Einstein's Relativity. It was
sure to spread into public and private morality (alas no longer
private) and from there to return in a hostile mutated form to haunt
science, as it now has. But those who will agree with me in principle
will not like the worst example I can find to hold up: scientist who
pay lip service to "global warming" when the very data in the article
in which they natter on about global warming proves the opposite;
scientists who say that global warming is a crusade, that the facts
must be distorted to force political action. Moral relativity isn't
knocking on the door of science, as Jobst fears, it has taken over the
priesthood.
Andre Jute
Thumbs well clear of the bricks
PS I had no problem with Tom Sherman's dialectic; he rather reminded
me of the Marxists I knew in my youth -- and in fact rescued from the
police on more than one occasion. But Tom didn't mix his politics with
scientific facts; he left that sort of lysenkoism to city boys.
jim beam
>> jobst....@stanfordalumni.org apparently wrote:
>>
>>> You seem to be using wreck.bike as a thought scratch pad.
>
> This from a man who posts off-topic political scat (seldom his own thoughts,
> btw, just ideological boilerplate) on a fairly regular basis.
>
come on bill, you should know by now; anyone not a fawning sycophant of
the great jobst brandt is an object for his diminution and/or derision,
rhyme, reason or none. gross hypocrisy is nothing compared with
inability to learn, inability to check facts and shameless repeated
underinformed opinion presented as fact.
Frank Krygowski
Well, then ask some questions. We should be able to work this out!
I'm thinking along several separate lines. One is, I see no obvious
way for energy to leave the system as heat. A second is, the mass of
the chain is tiny anyway, so this effect should be negligible, even if
it is a real effect (which I doubt).
But most important, acceleration and deceleration themselves simply
don't cause loss in energy.
- Frank Krygowski
- Frank Krygowski
jim beam
> Carl Fogel wrote:
>
>> Changing the speed of the chain run every time a link engages the
>> sprocket means accelerating and decelerating the chain run.
>
>> Acceleration and deceleration take power.
>
> Don't confuse accelerating and slowing down on a bicycle with
> mechanical action and losses. A swinging pendulum accelerates from
> standstill to maximum speed every cycle and does so in a vacuum for a
> long time, demonstrating that there is no power required.
ah, the jobstian conceptual black hole of misunderstanding... what
happens if gravity disappears jobst? how much does the acceleration
keep changing then?
>
>> At 90 RPM on a 53x11, pins are engaging and disengaging on the top
>> run 4770 times per minute. The chain speed change on the front 53
>> is 0.1976%, but 4.0507% on the 11-tooth rear.
>
>> It takes power to speed up and slow down the chain run ~ 4% almost
>> 5,000 times per minute.
>
> Where does the power go?
you need to brush up on your basics.
> What power does it take to slow down a
> moving chain and where is it extracted from the mechanism?
>
wrong tree. wrong bark.
Frank Krygowski
> You might enjoy this
> (non-cycling) article:http://www.caravanandmotorhomebooks.com/articles/fact_opinion.html
>
>
Not totally non-cycling! It did mention a bicycle wheel! (Sure, it
was as an antenna, but still...)
- Frank Krygowski
Ed Pirrero
> Bill Sornson wrote:
Since when has that ever stopped Sorni or Kunich from doing it?
Spouting ill-informed or non-informed emotional rantings at the world
is the unique province of the far right, fueled by the echo-chamber of
talk radio.
Sorni is one of those folks who routinely killfiles folk with whom he
disagrees. Thus ensuring, eventually, that all of usenet will agree
with his take on every issue.
Sorta like the Vandeman of r.b.t. politics.
E.P.
Ben C
It's true my post was a bit rambling.
The question was just: why does higher chain tension improve efficiency?
Carl posted a link to a pdf earlier in this thread. It says, in a
section called "Theory":
If it is assumed that friction between contacting components performs
work during drive operation, then power losses from the drive
necessarily reduce efficiency. The normal force producing friction is
related to the chain tension in the link during articulation and
engage- ment. An analysis for this tension can be found in the work
by Tordion (1996) and in the work by Kidd et al. (1998). Since the
friction depends on chain tension, there are perhaps two major loca-
tions for loss in the drive that should be identified beforehand
since the chain tension is large and is transferred from the chain to
the sprocket at these loca-tions. These include the surfaces between
the inner link bushing and chain pin and between the sprocket tooth,
link roller and inner link bushing Rather than derive in detail the
func- tional form for the losses, the results of models will be
presented to give the reader a feeling for the anticipated results of
experiment. The interested reader can find the full derivation else-
where (Spicer et al. 1999).
I haven't read Kidd, Tordion or Spicer, but this seems to be saying
higher tension is expected to be less efficient because the normal force
is higher, hence so is the frictional force you are using to dissipate
energy.
But that's not what they found, a bit further on, my *s:
Experimental results indicated that he efficiency of the chain drive
varied as a function of chain tension. It was found that the
efficiency varied linearly with the *reciprocal* of the average chain
tension with the highest efficien- cies occurring at high chain
tensions and lowest at low chain tensions. For example, the highest
efficiency mea- sured in the study, 98.6%, was mea- sured at a chain
tension of 305 N and he lowest, 80.9%, at 76.2 N.
Conclusion:
Owing to the high efficiencies measured under high chain tensions,
friction can account for only a few percent of the overall losses.
Most probably, mechanical losses that are not converted to thermal
energy in the drive account for the remainder of the measured loss.
So I guess that's this vibration. So my question is, is the proposed
mechanism here that higher tension means less vibration?
Ben C
> jobst....@stanfordalumni.org wrote:
>> Carl Fogel wrote:
>>
>>> Changing the speed of the chain run every time a link engages the
>>> sprocket means accelerating and decelerating the chain run.
>>
>>> Acceleration and deceleration take power.
>>
>> Don't confuse accelerating and slowing down on a bicycle with
>> mechanical action and losses. A swinging pendulum accelerates from
>> standstill to maximum speed every cycle and does so in a vacuum for a
>> long time, demonstrating that there is no power required.
>
> ah, the jobstian conceptual black hole of misunderstanding... what
> happens if gravity disappears jobst? how much does the acceleration
> keep changing then?
A universe without gravity would be different in all kinds of ways and
so it's difficult to imagine what would happen, especially if inertial
and gravitational mass are considered the same thing.
Here's another example: if you have two masses either end of a spring
bouncing towards and away from each other, that little system can go on
bouncing practically forever in a vacuum without a power source.
Kenny
motorcycle that I changed the crankcase motor oil every 1500k. Each
time I changed the oil I could feel the grit in the oil. This was a
four stroke engine and I assume the grit was coming from either crappy
gas or gas tank and somehow blowing by the piston rings. Either way
there was grit getting in there and this would shorten the life of the
engine. Not that I minded because it was a beater anyway. What I
didn't like to do was changing the oil all the time and having to
dispose of it. In my quest to find a more efficient way of maintaining
it I ventured into a motorcycle parts shop and stumbled upon a product
called liquid teflon. The directions for its use was to add 1 .oz of
liquid teflon to one quart of motor oil. This stuff worked like a
champ. I never changed the motor oil on this motorcycle for the next
12000k. The kickstand fell off due to corrosion but the engine still
went like a bat out of hell. I eventuallly had to put the motorcycle
to rest due to wear and tear on the exterior. I wish I could put
liquid teflon on a bike chain, but you need a certain amount of heat
for the teflon to bond to the metal and I don't have the tools to do
this. I'd imagine teflon bonded to the chain rollers would make the
chain last at least 15000k.
Kenny
carl...@comcast.net
Dear Ben,
Yes, there is that pesky detail that higher tension leads to higher
chain efficiency.
If most of the power loss occurred when the links rotate at the
sprockets, then more tension would produce the opposite effect.
On the other hand, more tension should reduce the power loss of the
~30 free links vibrating on each long chain run.
The tauter the free chain runs are, the closer they are to an
efficiently vibrating guitar string. The slacker they are, the closer
they are to a guitar string that needs to be tightened.
In the case of the upper and lower chain runs, on a 53-tooth at 90
rpm, each run receives up and down mismatched impulses from both ends
at about 4800 rpm, with the vibration constantly stopping as links
feed onto the sprocket and constantly starting as they emerge on the
other side.
(Mismatched unless someone is running a 53x53.)
Another point of interest is that the addition of two dreadfully small
idler pulleys on a derailleur fail to affect the chain's efficiency.
One possible explanation is that they're not under load, so there's no
friction.
Another possible explanation is that their chain runs are so short
that there's scarcely any inefficient jangling of the two new chain
runs.
Cheers,
Carl Fogel
carl...@comcast.net
Dear Ben,
To head back to the chain question, if you have ~30 free links on the
upper chain run receiving close to 5,000 impulses per minute from both
ends on a 53-tooth at 90 rpm, then the those links will be closer to
chaotic jangling at lower tension and closer to a "spring" or guitar
string at higher tension.
As each vibrating link feeds onto a sprocket, its vibration is lost to
the sprocket as the link hits the tooth. That's an impact rather than
a rotation, a tiny hammering rather than the tiny sliding of rotation.
Cheers,
Carl Fogel
Ben C
So that might mean more vibration means more wear, not just less
efficiency.
I think I've caught up with what Jim's point was.
If my two masses on a spring aren't on a spring, but you're holding one
of them in each hand, then it requires continuous power to oscillate
them backwards and forwards.
This is because your arms don't store energy like a spring, and it takes
just as much energy to pull the weights in in as to push them out.
Similarly, a jigsaw requires continuous power to jig the blade up and
down (even when you aren't cutting anything).
So does a bicycle chain jig like a jigsaw-- requiring to be continuously
driven-- or does it bounce like a guitar string, oscillating by itself?
Well there's nothing much springy about a bicycle chain, so I think it's
quite reasonable to believe that when it vibrates it _is_ drawing energy
continuously from the sprockets.
If you lift the top chain run up a bit and drop it again on a stationary
bike, it doesn't exactly bounce.
On the other hand the derailleur cage (where fitted) is on a spring so
there is potential for springy-style oscillation on the bottom run.
I'm also wondering is which is more efficient, a fixed-gear bike with
high chain tension, or a normal derailleur bike?
Frank Krygowski
> The tauter the free chain runs are, the closer they are to an
> efficiently vibrating guitar string. The slacker they are, the closer
> they are to a guitar string that needs to be tightened.
...
> To head back to the chain question, if you have ~30 free links on the
> upper chain run receiving close to 5,000 impulses per minute from both
> ends on a 53-tooth at 90 rpm, then the those links will be closer to
> chaotic jangling at lower tension and closer to a "spring" or guitar
> string at higher tension.
I'm not getting much out of these guitar string analogies - and I'm a
guitarist.
I do a lot of thinking using analogies, but not all analogies are
valid. That includes my favorite pair: "Time flies like an arrow.
Fruit flies like bananas."
(OK, not really analogies, but I had to throw them in.)
Anyway, I'd like more detail on where you think the energy is going.
- Frank Krygowski
Dan O
> On 2008-06-16, carlfo...@comcast.net <carlfo...@comcast.net> wrote:
> [...]
>
>
>
> > It's worth noting that no one has produced a credible test showing a
> > chain in action transmitting significantly different power when
> > different lubes are used--most of the power loss is due to chordal
> > action, the variation in chain speed caused by straight links of chain
> > suddenly wrapping around the "circle" of a sprocket. That vibration is
> > why even the motorcycle chains running in sealed oil baths lose around
> > 3% power.
>
> So why does efficiency increase with tension? I suppose a tighter chain
> vibrates less.
>
I'm going to take a totally uneducated WAG that (assuming a tighter
chain does in fact transfer power more efficiently (?)) maybe it's
because more of the sprocket teeth/cogs that are engaged by the
wrapped chain are contributing some little driving force to help the
top ones that are under load (??)
Michael Press
John Henderson <jhenRem...@talk21.com> wrote:
Dialectics among those who can formulate their thoughts
in an exact and verifiable way: physical theory and mathematics.
Hand waving is best left to those skilled in the craft;
they can back it up. And if not will own up to being wrong.
Most hand waving on rbt is done by those who cannot even
recognize falsification of their pet theories.
--
Michael Press
Ben C
> On Jun 16, 1:46 pm, Ben C <spams...@spam.eggs> wrote:
>> On 2008-06-16, carlfo...@comcast.net <carlfo...@comcast.net> wrote:
>> [...]
>>
>>
>>
>> > It's worth noting that no one has produced a credible test showing a
>> > chain in action transmitting significantly different power when
>> > different lubes are used--most of the power loss is due to chordal
>> > action, the variation in chain speed caused by straight links of chain
>> > suddenly wrapping around the "circle" of a sprocket. That vibration is
>> > why even the motorcycle chains running in sealed oil baths lose around
>> > 3% power.
>>
>> So why does efficiency increase with tension? I suppose a tighter chain
>> vibrates less.
>>
>
> I'm going to take a totally uneducated WAG that (assuming a tighter
> chain does in fact transfer power more efficiently (?))
That was what they found in the document Carl posted a link to. Here is
the link again: http://www.ihpva.org/HParchive/PDF/hp50-2000.pdf
ibc....@gmail.com
> >On 2008-06-17, jim beam <spamvor...@bad.example.net> wrote:
> >> jobst.bra...@stanfordalumni.org wrote:
> >>> Carl Fogel wrote:
>
> >>>> Changing the speed of the chain run every time a link engages the
> >>>> sprocket means accelerating and decelerating the chain run.
>
> >>>> Acceleration and deceleration take power.
>
> >>> Don't confuse accelerating and slowing down on a bicycle with
> >>> mechanical action and losses. A swinging pendulum accelerates from
> >>> standstill to maximum speed every cycle and does so in a vacuum for a
> >>> long time, demonstrating that there is no power required.
>
> >> ah, the jobstian conceptual black hole of misunderstanding... what
> >> happens if gravity disappears jobst? how much does the acceleration
> >> keep changing then?
>
> >A universe without gravity would be different in all kinds of ways and
> >so it's difficult to imagine what would happen, especially if inertial
> >and gravitational mass are considered the same thing.
>
> >Here's another example: if you have two masses either end of a spring
> >bouncing towards and away from each other, that little system can go on
> >bouncing practically forever in a vacuum without a power source.
>
> Dear Ben,
>
My practice is to replace my chain about every four months (possibly
earlier if there's more than 1/8 inch "stretch" over a measured foot),
and to slather on some oil if the chain gets squeaky. Occasionally I
leave it too long, and I experience poor shifting until the new chain
cuts the chainrings and cassette back into shape. Am I doing this
wrong?
Disclaimer: I am relatively lazy.
--ibc
Michael Press
<15263fbe-99b2-42c7...@j33g2000pri.googlegroups.com>,
Kenny <Posto...@gmail.com> wrote:
You left out the part where you took an oil sample
ever 1500k, and rubbed it between your fingers.
--
Michael Press
jobst....@stanfordalumni.org
> Yes, there is that pesky detail that higher tension leads to higher
> chain efficiency.
> If most of the power loss occurred when the links rotate at the
> sprockets, then more tension would produce the opposite effect.
> On the other hand, more tension should reduce the power loss of the
> ~30 free links vibrating on each long chain run.
I'm unclear to which tension is being referred to in this respect. Is
it derailleur chain tensioner or pull in the power transmitting run of
the chain? A bit of explanation would go a long way in possibly
making this credible.
> The tauter the free chain runs are, the closer they are to an
> efficiently vibrating guitar string. The slacker they are, the
> closer they are to a guitar string that needs to be tightened.
Guitar strings have a huge area impinging on surrounding air in
comparison to a steel and bulky bicycle chain that vibrates at rates
far below acoustic frequencies causing velocities that experience
practically no aerodynamic losses as musical strings do. Just think
of a piano tone with the key held down after striking. Even there,
air losses are minimal.
> In the case of the upper and lower chain runs, on a 53-tooth at 90
> rpm, each run receives up and down mismatched impulses from both ends
> at about 4800 rpm, with the vibration constantly stopping as links
> feed onto the sprocket and constantly starting as they emerge on the
> other side.
This keeps dipping into the concept that positive and negative
acceleration causes mechanical losses. That is not true.
> (Mismatched unless someone is running a 53x53.)
> Another point of interest is that the addition of two dreadfully small
> idler pulleys on a derailleur fail to affect the chain's efficiency.
They take energy to rotate but it is insignificant to the propulsion
of the bicycle. The magnitude of these losses are evident in spinning
the pedals backward with a well oiled freewheel. Most of those
losses are the derailleur idler wheels. That is easier to prove by
taking the same chain drive with a shorter chain that does not pass
through the idlers.
> One possible explanation is that they're not under load, so there's no
> friction.
That is readily disproven by the above experiment.
> Another possible explanation is that their chain runs are so short
> that there's scarcely any inefficient jangling of the two new chain
> runs.
I have never heard that "jangling" in many years of bicycling. What
causes that sound? The only sound with which I am familiar is the
sound of the links impinging on the derailleur idlers that are light
enough to resonate, slight as it is, but that is only when spinning
the pedals backward at a high rate.
Jobst Brandt
jobst....@stanfordalumni.org
>>> It's worth noting that no one has produced a credible test showing
>>> a chain in action transmitting significantly different power when
>>> different lubes are used--most of the power loss is due to chordal
>>> action, the variation in chain speed caused by straight links of
>>> chain suddenly wrapping around the "circle" of a sprocket. That
>>> vibration is why even the motorcycle chains running in sealed oil
>>> baths lose around 3% power.
>> So why does efficiency increase with tension? I suppose a tighter
>> chain vibrates less.
> I'm going to take a totally uneducated WAG that (assuming a tighter
> chain does in fact transfer power more efficiently (?)) maybe it's
> because more of the sprocket teeth/cogs that are engaged by the
> wrapped chain are contributing some little driving force to help the
> top ones that are under load (??)
On chains, only the first two engaging cogs of the driving sprocket
bear on the chain, and only the last or last two (depending on number
of cogs) bear on the driven sprocket. This arises from two effects,
elastic deformation and chain pitch elongation cause by wear.
Therefore this effect is not tension related. Depth of engagement of
a wear elongated chain is obvious and apparent from chain skipping
increase of a new chain on worn sprockets. The greater the driving
tension the more the chain skips out of engagement.
Jobst Brandt
Ben C
[...]
> My practice is to replace my chain about every four months (possibly
> earlier if there's more than 1/8 inch "stretch" over a measured foot),
Between 1/16" and 1/8" over a foot: new chain. More than 1/8", the
cassette is probably also toast.
> and to slather on some oil if the chain gets squeaky. Occasionally I
> leave it too long, and I experience poor shifting until the new chain
> cuts the chainrings and cassette back into shape.
Usually if the chain or sprockets are worn badly you get skipping, which
isn't just ordinarily annoying like poor shifting, but really annoying.
> Am I doing this wrong?
Not really. You could clean the chain occasionally. It wouldn't be any
more efficient but might last longer.
> Disclaimer: I am relatively lazy.
So not much chance of that chain getting cleaned then :)
jobst....@stanfordalumni.org
>>>> It's worth noting that no one has produced a credible test
>>>> showing a chain in action transmitting significantly different
>>>> power when different lubes are used--most of the power loss is
>>>> due to chordal action, the variation in chain speed caused by
>>>> straight links of chain suddenly wrapping around the "circle" of
>>>> a sprocket. That vibration is why even the motorcycle chains
>>>> running in sealed oil baths lose around 3% power.
>>> So why does efficiency increase with tension? I suppose a tighter
>>> chain vibrates less.
>> I'm going to take a totally uneducated WAG that (assuming a tighter
>> chain does in fact transfer power more efficiently (?))
> That was what they found in the document Carl posted a link to.
> Here is the link again:
http://www.ihpva.org/HParchive/PDF/hp50-2000.pdf
This is relative efficiency based on what appears to be relatively
constant frictional losses that become less significant as transmitted
power increases. This is misplaced scrutiny and presented in a way
that one might imagine that having greater derailleur chain tensioning
improves efficiency.
Writers on this thread have not helped by focusing on this effect that
is insignificant in bicycling.
Jobst Brandt
jobst....@stanfordalumni.org
> My practice is to replace my chain about every four months (possibly
> earlier if there's more than 1/8 inch "stretch" over a measured
> foot), and to slather on some oil if the chain gets squeaky.
> Occasionally I leave it too long, and I experience poor shifting
> until the new chain cuts the chainrings and cassette back into
> shape. Am I doing this wrong?
Probably, especially if you don't ride much in a month. Besides, a
new chain does nothing for driven sprockets, the ones that skip under
load with a new chain.
Jobst Brandt
A Muzi
What was the mechanism by which foreign material was removed so as to
not erode the crank bearing sleeves and piston rings in the cylinders?
--
Andrew Muzi
<www.yellowjersey.org/>
Open every day since 1 April, 1971
** Posted from http://www.teranews.com **
clareatsnyderdotontariodotcanada
>On 2008-06-16, carl...@comcast.net <carl...@comcast.net> wrote:
>[...]
>>
>> It's worth noting that no one has produced a credible test showing a
>> chain in action transmitting significantly different power when
>> different lubes are used--most of the power loss is due to chordal
>> action, the variation in chain speed caused by straight links of chain
>> suddenly wrapping around the "circle" of a sprocket. That vibration is
>> why even the motorcycle chains running in sealed oil baths lose around
>> 3% power.
>
>So why does efficiency increase with tension? I suppose a tighter chain
>vibrates less.
>
>Thinking of your picture of the "floating" chain, you might think well
>you've got to lift that dangling chain up each time it goes onto the
>sprocket, but then of course it falls back down giving you the energy
>back.
It doesn't give the energy back very effectively? Is the "falling"
chainlink falling onto the sproket in such a way that the energy is
fully recaptured?
clareatsnyderdotontariodotcanada
Less (or no) "over-wrap"?
A tight chain breaks cleanly from the sprocket. A loose chain can
follow the sprocket and requires power to "pull" it off.
clareatsnyderdotontariodotcanada
Energy can leave as heat or sound. Low frequency sound waves could be
produced by the oscillating chain.
Ben C
I figure the sprocket is lifting the sagging bottom run up, but the run
of chain falling off the front chainwheel is counterbalancing it. So
energy changes hands continuously but isn't leaving the system, at least
not by that mechanism.
John Henderson
> What was the mechanism by which foreign material was removed
> so as to not erode the crank bearing sleeves and piston rings
> in the cylinders?
The power of positive thinking?
John
jobst....@stanfordalumni.org
>> So I guess that's this vibration. So my question is, is the
>> proposed mechanism here that higher tension means less vibration?
> Less (or no) "over-wrap"?
> A tight chain breaks cleanly from the sprocket. A loose chain can
> follow the sprocket and requires power to "pull" it off.
What you describe is "chain suck", a phenomenon that occurs with badly
worn drive sprockets with a new chain. Worn driven sprockets with a
new chain cause skipping. Let's not get all the marbles mixed up on
microscopic efficiency assessments. Just ride bile and forget about
the hour record... for now.
Jobst Brandt
jobst....@stanfordalumni.org
>> So I guess that's this vibration. So my question is, is the
>> proposed mechanism here that higher tension means less vibration?
> Less (or no) "over-wrap"?
> A tight chain breaks cleanly from the sprocket. A loose chain can
> follow the sprocket and requires power to "pull" it off.
What you describe is "chain suck", a phenomenon that occurs with badly
worn drive sprockets with a new chain. Worn driven sprockets with a
new chain cause skipping. Let's not get all the marbles mixed up on
microscopic efficiency assessments. Just ride bike and forget about
jim beam
"ride bile"? indeed, it's what you seem to do best!
what a fabulously jobstian freudian slip.
carl...@comcast.net
[snip]
>Just ride bile
[snip]
Dear Jobst,
Some typos are priceless.
Cheers,
Carl Fogel
Kenny
Aside from the air filter that filters air for the carburetor there is
also a gas filter that leads from the fuel tank and the carburetor.
I'm kind of sure that the grit came from corrosion of the fuel tank.
There is also, as I remember, a sieve like object that sat at the
bottom of the crank case which I surmised was used to filter out some
of the junk out of the oil.
Michael Press
carl...@comcast.net wrote:
> On 19 Jun 2008 03:06:20 GMT, jobst....@stanfordalumni.org wrote:
>
> [snip]
>
> >Just ride bile
>
> [snip]
>
> Dear Jobst,
>
> Some typos are priceless.
I would not give 5 cents for any of them.
--
Michael Press
Frank Krygowski
>
> Energy can leave as heat or sound. Low frequency sound waves could be
> produced by the oscillating chain.
Consider how little energy it takes to produce a loud sound. As an
example, examine a cricket's muscle structure. Not very impressive,
I'm sure. Yet it can produce sounds far louder than any made by a
bike chain, for hours on end.
- Frank Krygowski
jobst....@stanfordalumni.org
>> Energy can leave as heat or sound. Low frequency sound waves could
>> be produced by the oscillating chain.
> Consider how little energy it takes to produce a loud sound. As an
> example, examine a cricket's muscle structure. Not very impressive,
> I'm sure. Yet it can produce sounds far louder than any made by a
> bike chain, for hours on end.
Or, take the example I mentioned earlier. Play a middle-C note on a
piano and hold the key down to see how long and loud acoustic
vibrations take to dampen in air to silence. Where do writers get
technical information to post this absurd energy loss stuff? At least
Buck Rogers had a thread of credibility in his future technology.
Alley Oop!
http://www.comics.com/comics/alleyoop/index.html
Jobst Brandt
carl...@comcast.net
Dear Jobst,
Stand on your chain to put it under tension, have someone tap it, and
tell us how long the chain vibrates.
Where does the energy go?
Do the same thing again, only this time tap the lower chain run, which
is under scarcely any tension, and tell us how long the chain
vibrates.
Where does the energy go?
Cheers,
Carl Fogel
jobst....@stanfordalumni.org
>>>> Energy can leave as heat or sound. Low frequency sound waves
>>>> could be produced by the oscillating chain.
>>> Consider how little energy it takes to produce a loud sound. As
>>> an example, examine a cricket's muscle structure. Not very
>>> impressive, I'm sure. Yet it can produce sounds far louder than
>>> any made by a bike chain, for hours on end.
>> Or, take the example I mentioned earlier. Play a middle-C note on
>> a piano and hold the key down to see how long and loud acoustic
>> vibrations take to dampen in air to silence. Where do writers get
>> technical information to post this absurd energy loss stuff? At
>> least Buck Rogers had a thread of credibility in his future
>> technology.
>> Alley Oop!
http://www.comics.com/comics/alleyoop/index.html
> Stand on your chain to put it under tension, have someone tap it,
> and tell us how long the chain vibrates.
> Where does the energy go?
Unless the chain is struck with a hammer or other heavy object, it
will not deflect or it will lift the rider standing on the pedal, a
highly unlikely event.
> Do the same thing again, only this time tap the lower chain run, which
> is under scarcely any tension, and tell us how long the chain
> vibrates.
You8 can do this yourself. Just the same, tapped with a small screw
driver it will vibrate a few milliseconds. So what?
> Where does the energy go?
It goes into friction in oily chain links that are natural losses of
chain articulation. This is not a reasonable method for measuring
that loss. Typically, on a one speed bicycle with freewheel, without
derailleur, pedals will stop spinning in about a second if spun in
reverse at average cycling speed. That gets closer to demonstrating
losses.
Jobst Brandt
travis...@gmail.com
Since this thread is about losses, I'll take it to lost metal, on
chainwheels, and to especially to the omniscient Madisonian, Muzi:
50 old-school (no pins, no ramps, all teeth cut the same) chain wheels
are put in front of you, all from bikes that have been pedaled 100km.
10 have had the chainrings (cr) advanced 72 degrees on the spider
every 20K, 10 have never been removed from the spider, and the other
30 have advanced intermediate amounts.
And, lets assume randomization of pedaling styles and hygiene, so the
fastidious cr rotator is no less likely to be a brutal masher than she
is a smooth-as-silk spinner. etc.
Can you sort the chainrings into piles according to their experience,
by eye, with little classification error? Can you at least distinguish
the most evenly rotated chains from the least? Would the task be
easier if the crs had seen twice the number of rotations? Is this is a
way to get customers back into the shop to get their chainrings
rotated, at risk of needing new $120 rings every 30km?
Harry Travis
carl...@comcast.net
Dear Jobst,
On a 53-tooth at 90 rpm, the two long chain runs bump up and down in
chordal action in four places as it enters and exits the two main
sprockets.
The chain vibrates--or deflects, to use your term--whether you press
gently on the pedals or heave on it with your whole weigh--at close to
5,000 impacts per minute at each point where it meets the sprockets.
Cheers,
Carl Fogel
Adam Kadlubek
<snip>
> Carl Fogel
Good post - on the other hand tho, chain efficiency goes up when chain
tension goes up. The effect is that 44-22 gear has less chordal action
then 22-11, but the efficency is similar due to 22-11 having twice the
chain tension of the 44-22. There was an ihpva article on this, where,
when you interpolate the results, it shows that smaller gears are
actually more efficient.
Regards
--
Adam Kadlubek
Ben C
> Frank Krygowski wrote:
>
>>> Energy can leave as heat or sound. Low frequency sound waves could
>>> be produced by the oscillating chain.
>
>> Consider how little energy it takes to produce a loud sound. As an
>> example, examine a cricket's muscle structure. Not very impressive,
>> I'm sure. Yet it can produce sounds far louder than any made by a
>> bike chain, for hours on end.
>
> Or, take the example I mentioned earlier. Play a middle-C note on a
> piano and hold the key down to see how long and loud acoustic
> vibrations take to dampen in air to silence.
Yes but the string in the piano is elastic, so it oscillates by itself,
gradually losing energy as sound and other things.
I don't believe a bicycle chain is like that. It isn't like a guitar or
piano string, or a pendulum: it requires continuous power to keep it
vibrating just like the blade of a jigsaw or like a dumb-bell you lift
up and down with your arm. There's no spring, no bounce, and no cycle of
energy storage and unstorage to keep it going (except perhaps a small
amount due to gravity since the vibration is up/down rather than
left/right).
Where does the lost energy go? If you pick up a heavy object and shake
it around, it doesn't make a deafening noise, emit gamma rays, or heat
up (much), but your muscles will heat up because they are working
constantly to accelerate and decelerate the object.
If you turn a crank that makes a chain vibrate the effect is just the
same: the energy is dissipated in your legs not in the chain.
Tom Kunich
news:bkkm541bvkesbo9ah...@4ax.com...
> On 20 Jun 2008 04:59:26 GMT, jobst....@stanfordalumni.org wrote:
>
>>carl...@comcast.net wrote:
>>
>>>>>> Energy can leave as heat or sound. Low frequency sound waves
>>>>>> could be produced by the oscillating chain.
>
> chordal action in four places as it enters and exits the two main
> sprockets.
>
> The chain vibrates--or deflects, to use your term--whether you press
> gently on the pedals or heave on it with your whole weigh--at close to
> 5,000 impacts per minute at each point where it meets the sprockets.
Carl, that means that the chain is vibrating so slightly that you or I
cannot see it at about 83 hertz. At that frequency and an amplitude of
extremely tiny vibrations it can't lose measurable energy to the surrounding
air.
jobst....@stanfordalumni.org
> On a 53-tooth at 90 rpm, the two long chain runs bump up and down in
> chordal action in four places as it enters and exits the two main
> sprockets.
> The chain vibrates--or deflects, to use your term--whether you press
> gently on the pedals or heave on it with your whole weigh--at close
> to 5,000 impacts per minute at each point where it meets the
> sprockets.
Whether this is functional or merely mathematically determinable is
under discussion here. Texture of a smooth, well paved road is a
significantly greater cause of vibration, one that can be heard and
felt to such a degree that the chain motions mentioned are entirely
masked and insignificant. As I said, the chain motions cannot be
heard even on a test stand turning the pedals backward. Even with a
derailleur, the sound is hardly audible.
Jobst Brandt
jobst....@stanfordalumni.org
>>>> Energy can leave as heat or sound. Low frequency sound waves could
>>>> be produced by the oscillating chain.
>>> Consider how little energy it takes to produce a loud sound. As an
>>> example, examine a cricket's muscle structure. Not very impressive,
>>> I'm sure. Yet it can produce sounds far louder than any made by a
>>> bike chain, for hours on end.
>> Or, take the example I mentioned earlier. Play a middle-C note on a
>> piano and hold the key down to see how long and loud acoustic
>> vibrations take to dampen in air to silence.
> Yes but the string in the piano is elastic, so it oscillates by
> itself, gradually losing energy as sound and other things.
The string is a steel wire and has insignificant elastic elongation,
similar to the bicycle chain in the examples. Losing energy as sound
needs to be defined. Besides, the tone lasts for ever compared to the
excitation of chain chordal motion.
> I don't believe a bicycle chain is like that. It isn't like a
> guitar or piano string, or a pendulum: it requires continuous power
> to keep it vibrating just like the blade of a jigsaw or like a
> dumb-bell you lift up and down with your arm. There's no spring, no
> bounce, and no cycle of energy storage and unstorage to keep it
> going (except perhaps a small amount due to gravity since the
> vibration is up/down rather than left/right).
I think your model of the event is glossing over elasticities and
energy sinks that are essential to the concept... and have been
ignored in the course of this thread.
> Where does the lost energy go? If you pick up a heavy object and
> shake it around, it doesn't make a deafening noise, emit gamma rays,
> or heat up (much), but your muscles will heat up because they are
> working constantly to accelerate and decelerate the object.
The losses are in the human body, not in the inanimate object. The
losses in the bicycle chain are not there, however and they don't
involve the human body, never getting that far.
> If you turn a crank that makes a chain vibrate the effect is just
> the same: the energy is dissipated in your legs not in the chain.
If you are worried about that vibration, I think photons of sunlight
when riding toward the light compared to riding away from it should
not be overlooked.
Jobst Brandt
clareatsnyderdotontariodotcanada
Grab hold of a 50 foot hunk of 1 inch Sisal or Hemp rope and tie it to
a post about 5 feet from the ground.
Grab the other end of the roap and start whippining it up and down to
get a "standing wave" going. After about 5 minutes you tell me if it
doesn't require the input of significant energy. It's only going up
and down - so where does the energy go - other than as sweat.
carl...@comcast.net
Dear Tom,
I keep waiting for you guys to figure out where energy can be lost on
a chain forced to vibrate.
Maybe a hint will help.
There is no test involving just a pair of sprockets joined by a chain
that magically rotates under power in mid-air.
Put the whole bike inside a bell jar and remove the air and the sound
question.
What else must vibrate on a bicycle with a real rider or laboratory
testing equipment, if the upper run of the chain, which is under
tension, is vibrating?
That is, if the chain under tension jiggles, what else jiggles?
If you want to explore an analogy, consider a mute for a violin. It's
just a fancy wooden paperclip that you stick on the bridge, which
holds up the vibrating strings.
The frequency of the sound waves remains the same, but the amplitude
the sound waves drops because the energy from the bow that vibrates
the string is being diverted and wasted before it reaches the body of
the instrument.
Same power from the bow, same instrument, but something is reducing
the efficiency of the transmission of the power to the body, damping
the sound output.
Is the sprocket/vibrating-chain/sprocket combination coupled to any
things that don't jiggle efficiently?
Cheers,
Carl Fogel
Ben C
>
>>>>> Energy can leave as heat or sound. Low frequency sound waves could
>>>>> be produced by the oscillating chain.
>
>>>> Consider how little energy it takes to produce a loud sound. As an
>>>> example, examine a cricket's muscle structure. Not very impressive,
>>>> I'm sure. Yet it can produce sounds far louder than any made by a
>>>> bike chain, for hours on end.
>
>>> Or, take the example I mentioned earlier. Play a middle-C note on a
>>> piano and hold the key down to see how long and loud acoustic
>>> vibrations take to dampen in air to silence.
>
>> Yes but the string in the piano is elastic, so it oscillates by
>> itself, gradually losing energy as sound and other things.
>
> The string is a steel wire and has insignificant elastic elongation,
> similar to the bicycle chain in the examples.
How can that be true? Yes they are both steel but the piano wire is much
thinner so subject to much greater stress for a given force. I don't see
how it vibrates if it isn't oscillating basically like a mass on a
spring.
If you strung a piano with bicycle chains you would need much heaver
hammers to hit them with to get the same effect.
> needs to be defined. Besides, the tone lasts for ever compared to the
> excitation of chain chordal motion.
Exactly-- because the chain doesn't vibrate elastically like the piano
wire. You need continuous power to drive the chain excitation, which is
why it works as a way of losing power.
You can probably bounce a basketball up and down quite a few times
without getting tired as you only have to top up the energy stored in
the bouncing ball with small inputs.
But to "bounce" a beanbag of the same mass the same number of times (by
manually raising and lowering it) would consume much more energy.
>> I don't believe a bicycle chain is like that. It isn't like a
>> guitar or piano string, or a pendulum: it requires continuous power
>> to keep it vibrating just like the blade of a jigsaw or like a
>> dumb-bell you lift up and down with your arm. There's no spring, no
>> bounce, and no cycle of energy storage and unstorage to keep it
>> going (except perhaps a small amount due to gravity since the
>> vibration is up/down rather than left/right).
>
> I think your model of the event is glossing over elasticities and
> energy sinks that are essential to the concept...
Well I have been talking about elasticities and energy sinks, but I
don't really understand what your point is.
[...]
>> Where does the lost energy go? If you pick up a heavy object and
>> shake it around, it doesn't make a deafening noise, emit gamma rays,
>> or heat up (much), but your muscles will heat up because they are
>> working constantly to accelerate and decelerate the object.
>
> The losses are in the human body, not in the inanimate object.
Exactly.
> The losses in the bicycle chain are not there, however and they don't
> involve the human body, never getting that far.
You may be right: I don't know exactly how large the losses are or
whether they are all dissipated before they get as far as being "damped"
by the rider's legs.
>> If you turn a crank that makes a chain vibrate the effect is just
>> the same: the energy is dissipated in your legs not in the chain.
>
> If you are worried about that vibration, I think photons of sunlight
> when riding toward the light compared to riding away from it should
> not be overlooked.
I always take care to wear a jersey that is black on the front and white
on the back when riding to work in the morning (when I am travelling
east).
carl...@comcast.net
Dear Adam,
That may be the Spicer article that I keep mentioning, the first one
in this issue:
http://www.ihpva.org/HParchive/PDF/hp50-2000.pdf
I'm not sure about the details of what you're saying, but you may well
be right.
In testing, Spicer found that:
A) More teeth are more efficient.
Both theories explain that result.
If power is lost primarily in the friction as the links rotate upon
entering or exiting the sprockets, there should be less friction and
power loss with a big sprocket because the links rotate through a
smaller angle on 53-tooth (6.8 degrees) than on an 11-tooth (32.7
degrees).
If power is lost primarily due to chain vibration, there should be
less vibration and power loss with a 53-tooth because the chordal
action produces about twenty times less chain-speed variation on a
53-tooth (0.18%) than on an 11-tooth (4.05%), with vibration
corresponding roughly to how much the chain-speed varies.
B) At the same power, more tension (which is also lower chain speed
and thus chain vibration at a lower rate) is more efficient.
More tension at the same power means lower chain speed and a lower
rate of vibration, so the chain-vibration theory squares with this
observed effect.
More tension doesn't seem to be likely to reduce the friction when
links rotate. Usually, friction rises when surfaces are pressed
together harder. But the lower chain-speed might mean higher friction
on each rotation at a lower rate of rotations per minute.
C) Different kinds of lubrication doesn't seem to have any effect--and
even removing the lubricant doesn't seem to affect efficiency!
Lubricant doesn't matter much to the vibration theory, so the
vibration theory works fine here. Chordal action has little to do with
lubrication.
But it's hard to see why chain efficiency should be apparently
unaffected by lack of lubrication if link-rotation friction is the
primary cause of power loss. Lubricant should reduce friction
significantly.
D) Cross-chaining doesn't seem to have much effect, either.
Cross-chaining doesn't matter much to the vibration theory--the
chordal action proceeds at the same rate, no matter what.
But it's hard to see why chain efficiency should be apparently
unaffected by cross-chaining if link-rotation friction is the primary
cause of power loss. Twisting the chain sideways by cross-chaining
should increase link-roller friction, but the effect doesn't seem to
show up in the testing.
So it seems as if increased vibration explains observed power losses
better than link-rotation friction.
If you think about the whole system for a few seconds, it's not hard
to see two places where a taut, vibrating chain on a bicycle, either
in real life or reduced to a laboratory test-stand, would lead to
power losses that would increase with vibration.
I keep wondering why posters are having so much trouble seeing
anything except two sprockets and a chain, apparently floating in
mid-air, attached to nothing else, and somehow powered to rotate
against resistance.
Cheers,
Carl Fogel
Tom Kunich
news:605o549nck79g38fr...@4ax.com...
> On Fri, 20 Jun 2008 07:05:47 -0700, "Tom Kunich" <cyclintom@yahoo.
> com> wrote:
>
> I keep waiting for you guys to figure out where energy can be lost on
> a chain forced to vibrate.
And of course you don't seem to be able to understand that the amplitude of
this "vibration" is too small to have any significant effect. But you
apparently believe it is without offering any opinion on the value of this
loss. Here's a hint - the power in a violin string is very small yet you can
plainly hear it. It will fill an entire concert hall. Whereas the chain is
essentially quiet in the smallest quietest room.
Ben C
So the ideal solution is 34-tooth chainrings front and rear and huge
wheels to get sufficient gear inches.
Or, even more efficiently, leave out the chain altogether and connect
the cranks directly to one of the huge wheels. Then you could make the
other wheel tiny, which would also offset the extra weight of the large
wheel.
I think I'd better get down to the patent office with this idea asap.
Tom Kunich
news:l46o54l4pjlofjmvn...@4ax.com...
> On Fri, 20 Jun 2008 03:01:48 -0700 (PDT), Adam Kadlubek
>
> anything except two sprockets and a chain, apparently floating in
> mid-air, attached to nothing else, and somehow powered to rotate
> against resistance.
I keep wondering why you have so much trouble understanding that pedaling
causes the cranks to pull into the bottom bracket bearings which is an area
of major power losses.
jobst....@stanfordalumni.org
>>>>>> Energy can leave as heat or sound. Low frequency sound waves
>>>>>> could be produced by the oscillating chain.
>>>>> Consider how little energy it takes to produce a loud sound. As
>>>>> an example, examine a cricket's muscle structure. Not very
>>>>> impressive, I'm sure. Yet it can produce sounds far louder than
>>>>> any made by a bike chain, for hours on end.
>>>> Or, take the example I mentioned earlier. Play a middle-C note
>>>> on a piano and hold the key down to see how long and loud
>>>> acoustic vibrations take to dampen in air to silence.
>>> Yes but the string in the piano is elastic, so it oscillates by
>>> itself, gradually losing energy as sound and other things.
>> The string is a steel wire and has insignificant elastic
>> elongation, similar to the bicycle chain in the examples.
> How can that be true? Yes they are both steel but the piano wire is
> much thinner so subject to much greater stress for a given force. I
> don't see how it vibrates if it isn't oscillating basically like a
> mass on a spring.
If you consider elongation of the piano wire when struck by a padded
piano hammer, you should consider the angular excursion that is less
than 0.5 degree. Lengthening that takes place for that is given by
the cosine (0.5)=0.99996 or practically none. The vibration is causes
by tension, not elongation (consider the length involved).
> If you strung a piano with bicycle chains you would need much heaver
> hammers to hit them with to get the same effect.
HOLD IT!!! It is the mass of the wire that makes it vibrate at a
specific frequency when tensioned to its specific note. The mass of a
chain is greater and would resonate at a lower frequency, the
restoring to center rate depending on tension, not elasticity.
>> needs to be defined. Besides, the tone lasts forever compared to the
>> excitation of chain chordal motion.
> Exactly-- because the chain doesn't vibrate elastically like the piano
> wire. You need continuous power to drive the chain excitation, which is
> why it works as a way of losing power.
Neither vibrates elastically! Consider guitar strings that are all at
the same tension but make different tones dependent on their mass.
> You can probably bounce a basketball up and down quite a few times
> without getting tired as you only have to top up the energy stored in
> the bouncing ball with small inputs.
Oooooh. Your analogies are so far afield and not analogous to the
bicycle chain.
> But to "bounce" a beanbag of the same mass the same number of times
> (by manually raising and lowering it) would consume much more
> energy.
You might just as well say swimming is like a bicycle chain and loses
power with each stroke.
>>> I don't believe a bicycle chain is like that. It isn't like a
>>> guitar or piano string, or a pendulum: it requires continuous
>>> power to keep it vibrating just like the blade of a jigsaw or like
>>> a dumb-bell you lift up and down with your arm. There's no
>>> spring, no bounce, and no cycle of energy storage and unstorage to
>>> keep it going (except perhaps a small amount due to gravity since
>>> the vibration is up/down rather than left/right).
>> I think your model of the event is glossing over elasticities and
>> energy sinks that are essential to the concept...
> Well I have been talking about elasticities and energy sinks, but I
> don't really understand what your point is.
The point is that you don't understand the magnitude of any of these
quantities and are trying to support the concept that they are
significant in bicycling.
> [...]
>>> Where does the lost energy go? If you pick up a heavy object and
>>> shake it around, it doesn't make a deafening noise, emit gamma
>>> rays, or heat up (much), but your muscles will heat up because
>>> they are working constantly to accelerate and decelerate the
>>> object.
>> The losses are in the human body, not in the inanimate object.
> Exactly.
>> The losses in the bicycle chain are not there, however and they
>> don't involve the human body, never getting that far.
> You may be right: I don't know exactly how large the losses are or
> whether they are all dissipated before they get as far as being
> "damped" by the rider's legs.
Well, let's say damped by the soft tissue in the rider's foot bearing
on the pedal. It's that chain buzz that I notice over pavement
texture, having musically sensitive feet.
>>> If you turn a crank that makes a chain vibrate the effect is just
>>> the same: the energy is dissipated in your legs not in the chain.
>> If you are worried about that vibration, I think photons of
>> sunlight when riding toward the light compared to riding away from
>> it should not be overlooked.
> I always take care to wear a jersey that is black on the front and
> white on the back when riding to work in the morning (when I am
> travelling east).
Good work!
Jobst Brandt
Tim McNamara
carl...@comcast.net wrote:
> I keep waiting for you guys to figure out where energy can be lost on
> a chain forced to vibrate.
>
> Maybe a hint will help.
>
> There is no test involving just a pair of sprockets joined by a chain
> that magically rotates under power in mid-air.
>
> Put the whole bike inside a bell jar and remove the air and the sound
> question.
>
> What else must vibrate on a bicycle with a real rider or laboratory
> testing equipment, if the upper run of the chain, which is under
> tension, is vibrating?
>
> That is, if the chain under tension jiggles, what else jiggles?
>
> If you want to explore an analogy, consider a mute for a violin. It's
> just a fancy wooden paperclip that you stick on the bridge, which
> holds up the vibrating strings.
>
> The frequency of the sound waves remains the same, but the amplitude
> the sound waves drops because the energy from the bow that vibrates
> the string is being diverted and wasted before it reaches the body of
> the instrument.
>
> Same power from the bow, same instrument, but something is reducing
> the efficiency of the transmission of the power to the body, damping
> the sound output.
>
> Is the sprocket/vibrating-chain/sprocket combination coupled to any
> things that don't jiggle efficiently?
Carl, I am not convinced that your conception of the question is, umm,
sound. If I put my bike's chain under tension by having my wife stand
on the pedals and tap the top run with a wrench, I see no sustained
vibration. If there is a sustained note sounded, it is beyond the range
of my hearing. There's a metallic tap and that's it. The chain doesn't
even appear to move. I'd have to really whack the chain to push it out
of the alignment that results from tension. If I tap it on the bottom
run, it bounces around more but settles almost immediately because of
the damping effect of the derailleur springs.
A chain is not a piano string or a guitar string. These are solid steel
wires, possibly wrapped with a brass, bronze or steel wire to add mass
and lower the resonant frequency. When you ask if the chain "is coupled
to things that don't jiggle efficiently," it seems to me that one of
those is the chain itself. With the assemblage of links, the inherent
slop in the sleeve bearings of the chain and the viscous damping of the
lubricant would prevent sustained vibrations all by itself.
Of course, this is not particularly germaine to actually riding a bike.
Use your nonpareil Google skills to find photos of chains in chordal
wave patterns. Paris-Roubaix or one of the other cobblestone classics
will provide photos of chains bouncing around while riding. Compare
those to photos of chains of bikes being raced on smooth roads, such as
in time trials.
Tim McNamara
<9bb7f045-8cbc-4bcc...@k30g2000hse.googlegroups.com>,
Adam Kadlubek <uzurp...@gmail.com> wrote:
The measurements, according to Frank Berto at least, are that smaller
cogs are markedly less efficient than larger cogs because of the greater
bending of the chain as it wraps around the smaller cog. There are both
friction and hysteresis losses.
r15...@aol.com
> On Thu, 12 Jun 2008 13:43:15 -0700, "Tom Kunich" <cyclintom@yahoo.
>
> com> wrote:
> >An online magazine ran an article on dry lubricants. They lubed a chain and
> >then allowed it to swing back and forth and measured the friction on it.
> >After 10 minutes or so the friction would start picking up and then within a
> >half hour the chain would be essentially unlubricated.
>
> >I asked what happened if they tested motor oil. The comment was that it
> >require weeks for the chain to start getting increased friction so they
> >didn't want to run such tests.
>
> >What does that tell you?
>
> Dear Tom,
>
> It may tell us that the online magazine mis-measured friction.
>
> Spicer tested lubricated and unlubricated bicycle chains in 2000 and
> found that lubrication had no significant effect on transmission
> efficiency--even when the lubricant was removed by cleaning.
>
> After testing a chain with Castrol Wrench Force Dry Lube, Pedro’s Syn
> Lube, Generation 4 White Lightning, Spicer thoroughly cleaned the
> chain and tested it dry.
>
> No significant differences were noted by his testing equipment:
>
> "However, these results also indicate that the actual lubricant used
> has little effect on the overall performance of the drive under
> laboratory conditions given the precision of the measurement.
> In addition, the chain used for the lubrication study was fully
> degreased and was re-tested for efficiency. This degreasing operation
> consisted of a five-minute scrub with kerosene followed by a cleaning
> with Castrol Degreaser. The measured efficiency of the de-lubricated
> chain for the 52–15 combination at 60 RPM and 100 W was 90.3% and at
> 200 W was 96.5%. These efficiencies are essentially the same as those
> measured for the chain in the re-lubricated condition."
>
> It's the first article here:
> http://www.ihpva.org/HParchive/PDF/hp50-2000.pdf
>
> Cheers,
>
> Carl Fogel
A thoroughly 'de-lubricated' chain will still run flawlessly smooth as
long as it has solvent on it. Try WD 40 for instance, it takes the oil
off the chain but also lubricates the chain, just not for very long.
Robert
Tom Kunich
news:timmcn-F9CF33....@news.iphouse.com...
>
> The measurements, according to Frank Berto at least, are that smaller
> cogs are markedly less efficient than larger cogs because of the greater
> bending of the chain as it wraps around the smaller cog. There are both
> friction and hysteresis losses.
The friction losses are from turning the chain under pressure sideplate to
sideplate.
Ben C
>>>> Yes but the string in the piano is elastic, so it oscillates by
>>>> itself, gradually losing energy as sound and other things.
>
>>> The string is a steel wire and has insignificant elastic
>>> elongation, similar to the bicycle chain in the examples.
>
>> How can that be true? Yes they are both steel but the piano wire is
>> much thinner so subject to much greater stress for a given force. I
>> don't see how it vibrates if it isn't oscillating basically like a
>> mass on a spring.
>
> If you consider elongation of the piano wire when struck by a padded
> piano hammer, you should consider the angular excursion that is less
> than 0.5 degree. Lengthening that takes place for that is given by
> the cosine (0.5)=0.99996 or practically none. The vibration is causes
> by tension, not elongation (consider the length involved).
How does tension cause vibration?
http://en.wikipedia.org/wiki/Simple_harmonic_motion
Force on the wire is proportional to its displacement. You strike it
with the hammer and displace it, that results in a force pulling it
back to the centre. But it overshoots and displaces the other side of
centre, which results in a force pulling it back... and so on, and that
is the vibration. Of course it loses a bit of energy, but it goes on for
quite a while.
The force proportional to displacement is a spring force (Hooke's law)
because the wire is a spring.
If the wire had an infinite modulus of elasticity, it would not vibrate
like this, no matter how much tension it was under.
>> If you strung a piano with bicycle chains you would need much heaver
>> hammers to hit them with to get the same effect.
>
> HOLD IT!!! It is the mass of the wire that makes it vibrate at a
> specific frequency when tensioned to its specific note. The mass of a
> chain is greater and would resonate at a lower frequency, the
> restoring to center rate depending on tension, not elasticity.
Yes resonance is important too-- and it's a function both of mass and
stiffness.
[...]
>> Exactly-- because the chain doesn't vibrate elastically like the piano
>> wire. You need continuous power to drive the chain excitation, which is
>> why it works as a way of losing power.
>
> Neither vibrates elastically! Consider guitar strings that are all at
> the same tension but make different tones dependent on their mass.
Yes-- different resonant frequencies. To fine-tune the guitar you adjust
the tension of the strings. You could equally adjust the mass, but
welding little beads onto the strings and filing them off again is less
convenient.
>> You can probably bounce a basketball up and down quite a few times
>> without getting tired as you only have to top up the energy stored in
>> the bouncing ball with small inputs.
>
> Oooooh. Your analogies are so far afield and not analogous to the
> bicycle chain.
Well you were the one who started talking about pendulums and pianos.
[...]
>> Well I have been talking about elasticities and energy sinks, but I
>> don't really understand what your point is.
>
> The point is that you don't understand the magnitude of any of these
> quantities and are trying to support the concept that they are
> significant in bicycling.
Not really, I'm just discussing how energy might be lost to vibration
and saying why I think a pendulum or piano wire might be a misleading
analogy for a vibrating chain.
jobst....@stanfordalumni.org
> A thoroughly 'de-lubricated' chain will still run flawlessly smooth
> as long as it has solvent on it. Try WD40 for instance, it takes
> the oil off the chain but also lubricates the chain, just not for
> very long.
As I have often mentioned, water (in heavy rain) will remove all oil
and dirt from a running chain and lubricate it well. Water is a poor
lubricant because it evaporates quickly when the weather becomes dry
and the road dries off. There are plenty of industrial applications
that use water as a lubricant, typically canneries and fish processing
plants.
Jobst Brandt
jobst....@stanfordalumni.org
>> The measurements, according to Frank Berto at least, are that
>> smaller cogs are markedly less efficient than larger cogs because
>> of the greater bending of the chain as it wraps around the smaller
>> cog. There are both friction and hysteresis losses.
> The friction losses are from turning the chain under pressure
> side plate to side plate.
I believe Berto did straight line tests to no confuse skewed with
in-line power transmission. Losses come from pin rotation and roller
rotation on smaller sprockets, the ratios of which are linear with
sprocket size number of teeth).
Jobst Brandt
jobst....@stanfordalumni.org
> [...]
http://en.wikipedia.org/wiki/Simple_harmonic_motion
You seem to avoid the magnitude of the restoring force, which is the
sine of the displacement angle from a straight line, times tension.
Because the cosine of that angle is minuscule, tension in the piano
string is not affected. The whole evaluation is normally done with
constant tension with insignificant change.
> If the wire had an infinite modulus of elasticity, it would not vibrate
> like this, no matter how much tension it was under.
I does and you are surmising what isn't.
>>> If you strung a piano with bicycle chains you would need much
>>> heaver hammers to hit them with to get the same effect.
>> HOLD IT!!! It is the mass of the wire that makes it vibrate at a
>> specific frequency when tensioned to its specific note. The mass
>> of a chain is greater and would resonate at a lower frequency, the
>> restoring to center rate depending on tension, not elasticity.
> Yes resonance is important too-- and it's a function both of mass and
> stiffness.
Stiffness has nothing to do with it. That is why increasing mass is
done with wire wrap that has essentially no stiffness. It is
computed that way and assumes constant tension when the wire is
struck.
> [...]
>>> Exactly-- because the chain doesn't vibrate elastically like the
>>> piano wire. You need continuous power to drive the chain
>>> excitation, which is why it works as a way of losing power.
>> Neither vibrates elastically! Consider guitar strings that are all at
>> the same tension but make different tones dependent on their mass.
> Yes-- different resonant frequencies. To fine-tune the guitar you
> adjust the tension of the strings. You could equally adjust the
> mass, but welding little beads onto the strings and filing them off
> again is less convenient.
Only to a standard, which is practically identical for all strings.
If this were not so, the neck of the guitar would slowly take a bend to
one side. The tuning changes are less than a percent of string
tension.
>>> You can probably bounce a basketball up and down quite a few times
>>> without getting tired as you only have to top up the energy stored
>>> in the bouncing ball with small inputs.
>> Oooooh. Your analogies are so far afield and not analogous to the
>> bicycle chain.
> Well you were the one who started talking about pendulums and pianos.
Those are analogous to a vibrating chain that was proposed but
erroneous as a vibrating element. That is why those are appropriate.
> [...]
>>> Well I have been talking about elasticities and energy sinks, but
>>> I don't really understand what your point is.
>> The point is that you don't understand the magnitude of any of
>> these quantities and are trying to support the concept that they
>> are significant in bicycling.
> Not really, I'm just discussing how energy might be lost to
> vibration and saying why I think a pendulum or piano wire might be a
> misleading analogy for a vibrating chain.
You are making waves in a pond that you don't understand and mislead
others by making claims that support mechanical voodoo.
Jobst Brandt
r15...@aol.com
Indeed one will notice that any creaks, groans and squeeks the bike
has due to exposure to past rainstorms will all disappear during the
next one -- only to return worse than ever in the following days.