Tire stats
Sandy
play host on a web site, so if someone would offer a place to put it, I can
e-mail the Excel file for posting on a web page.
A needed comment - there is data, and there is the source. I have
recommended it frequently, and the data is rather incomplete without the
commentary. If you seriously want the composite evaluations, recognizing
that there are personal and subjective elements, I suggest getting a copy of
the January issue of Le Cycle.
The address registering as mine in headers is legitimate, although it is
basically a sewer trap. Send an e-mail there with your address, and we can
get the debate on the road.
--
Sandy
Verneuil-sur-Seine FR
--
Il n'est aucune sorte de sensation qui soit plus vive
que celle de la douleur ; ses impressions sont sûres,
elles ne trompent point comme celles du plaisir.
- de Sade.
carl...@comcast.net
>I have tables for slip angle and rolling resistance for 85 tires. I don't
>play host on a web site, so if someone would offer a place to put it, I can
>e-mail the Excel file for posting on a web page.
>
>A needed comment - there is data, and there is the source. I have
>recommended it frequently, and the data is rather incomplete without the
>commentary. If you seriously want the composite evaluations, recognizing
>that there are personal and subjective elements, I suggest getting a copy of
>the January issue of Le Cycle.
>
>The address registering as mine in headers is legitimate, although it is
>basically a sewer trap. Send an e-mail there with your address, and we can
>get the debate on the road.
>
>
>--
>Sandy
>Verneuil-sur-Seine FR
Dear Sandy,
Email 'em to me as attachments!
Cheers,
Carl Fogel
Robert Chung
> I have tables for slip angle and rolling resistance for 85 tires. I
> don't play host on a web site, so if someone would offer a place to
> put it, I can e-mail the Excel file for posting on a web page.
>
> A needed comment - there is data, and there is the source. I have
> recommended it frequently, and the data is rather incomplete without
> the commentary. If you seriously want the composite evaluations,
> recognizing that there are personal and subjective elements, I
> suggest getting a copy of the January issue of Le Cycle.
I have posted Sandy's Excel file here:
http://anonymous.coward.free.fr/misc/TireStats09.xls
Thanks go to Sandy for entering and sharing these data. Please take note of
his caveat.
bjwe...@gmail.com
wrote:
Thanks to Sandy and to you.
One thing to note before everyone goes all crazy with this
data. If you look at the "rollout distance," which I assume
is some kind of roll-down test of rolling resistance, it
reports four digits, but the numbers are clearly quantized.
They're all multiples of approx. 2.0825 -
probably the circumference of the wheel+tire in meters.
So you should think of them as an integer number of
wheel revolutions, which also limits the precision
of the measurement.
I haven't read Le Cycle's accompanying article
for more information (will try, but don't expect to find a
copy in southern Arizona).
If anybody posts a calculation from these numbers
and shows four or more digits of precision, I'm going
to ride my fixed gear over to your house and break
your fingers in the drivetrain. Plus we'll have to
confiscate your calculator and give you a zero on
the homework assignment.
Ben
Nick L Plate
> One thing to note before everyone goes all crazy with this
> data. If you look at the "rollout distance," which I assume
> is some kind of roll-down test of rolling resistance, it
> reports four digits, but the numbers are clearly quantized.
> They're all multiples of approx. 2.0825 -
Mainland Europe use progression as a way to describe the gear used,
where UK and English speakers generally use equivalent diameter in
gear inches. 2.085 is the loaded rollout of a narrow 700c tyre in
metres. The distance the bike will travel with a rotation of this
wheel. It is multiplied by the gear ratio to obtain the
"progression", the distance obtained from a rotation of the pedal
crank. Gear ratio is chainwheel:sprocket
TJ
carl...@comcast.net
For those unable to read XLS files . .
A tubular column was added (where possible) to the second table.
***
Tire slips at this angle from horizontal
Tubular
Angle ? Wide Brand Model
----- - ---- ----- -----
41.00 N 23.5 Continental Grand Prix
42.50 N 22.0 Deda tre HST
42.50 Yt 21.5 FMB Soie Naturelle
43.20 Yt 21.5 Continental Sprinter
43.20 N 22.0 Continental Supersonic
43.20 N 22.0 Vredenstein Fortezza Superlite
43.20 N 23.5 Vredenstein Fortezza Tricomp
43.75 N 24.0 Continental Force
44.00 N 23.5 Deda tre RS Dual
44.00 N 23.0 Hutchinson Equinox
44.00 xx 22.0 Tufo C Giro Twix
44.00 xx 21.0 Tufo S3 Pro
44.00 N 23.0 Vredenstein Fortezza Quatro Tricomp
45.00 N 21.5 Continental Attack
45.00 N 23.0 Continental Grand Prix 4 Season
45.00 N 24.0 Hutchinson Fusion 2
45.00 N 22.5 Michelin Pro 3 Grip
45.00 N 23.0 Vredenstein Pro 3 Light
45.00 N 22.0 Panaracer Extreme Evo 2
46.00 N 22.0 Areo Jet Racing
46.00 N 23.0 Challenge Elite
46.00 Yt 22.0 Continental Competition
46.00 N 23.0 Deda tre Grinta
46.00 N 22.0 Deda tre RS Corsa
46.00 N 22.5 Gommitalia Targa
46.00 N 23.0 Hutchinson Fusion Ultra
46.00 N 22.0 Michelin Lithion
46.00 N 23.5 Ritchey WCS Race slick Open T
46.00 N 23.0 Specialized Mondo S Works
46.00 N 22.5 Vittoria Diamante Pro 2
46.00 N 23.0 Vittoria Open Corsa Evo Cx
46.00 N 24.0 WTB Solano
46.00 Yt 21.0 Zipp Tangente
47.00 N 23.5 Deda tre RS
47.00 Yt 22.0 Gommitalia Platinum
47.00 N 23.0 Michelin Krylion
47.00 N 22.0 Panaracer Stradius Elite
47.00 N 26.0 Specialized Roubaix 23/25 Armadillo
47.00 xx 25.0 Tufo C Elite Ride
48.00 N 23.5 Areo Strain
48.00 N 22.0 Gommitalia Blitz
48.00 Yt 22.0 Schwalbe Milano
48.00 xx 21.0 Tufo Elite Jet
48.00 xx 21.0 Tufo S3 Lite
48.00 N 22.5 Vittoria Rubino Pro
48.00 N 22.0 Vittoria Rubino Pro Tech
48.00 N 21.5 Vredenstein Fortezza Tricomp Pro
48.00 N 23.0 WTB Camino Alto
49.00 Yt 21.0 Challenge Forte
49.00 Yt 21.5 Deda tre Olimpico
49.00 N 22.0 Schwalbe Lugano
49.00 N 23.0 Specialized All Cond Armadillo Elite
49.00 N 24.0 Specialized All Cond Flak Jack
50.00 Yt 24.0 Challenge Criterium
50.00 N 24.0 Challenge Gara
50.00 N 22.5 Gommitalia Calypso
50.00 N 23.0 Gommitalia Logo
50.00 N 22.5 Gommitalia Logo Sport
50.00 N 22.5 Panaracer Stradius sport
50.00 N 24.0 Ritchey Race Slick Comp
50.00 N 22.0 Schwalbe Ultremo R
50.00 xx 23.0 Tufo Hi composite
50.00 N 22.0 Vittoria Rubino Pro Slick
50.00 N 24.0 Vredenstein Ricorso
50.00 N 21.0 Zipp Tangente
51.00 N 22.0 Bontrager Race X Lite
51.00 N 24.0 Bontrager Race X Lite AC
51.00 N 21.5 Bontrager Race XXX Lite
51.00 N 23.0 Challenge Triathlon
51.00 N 22.0 Michelin Pro 3 Race
51.00 N 22.0 Schwalbe Blizzard
51.00 Yt 22.0 Schwalbe Montello
51.00 xx 21.0 Tufo CS 33 Special
51.00 Yt 21.5 Vredenstein Fortezza Pro
52.00 N 22.0 Areo Extreme Evo 2
52.00 Yt 21.5 Challenge Criterium
54.00 N 23.5 Challenge Forte
54.00 N 22.5 Hutchinson Carbon Comp
54.00 N 23.0 Hutchinson Intensive Longue Dist
54.00 Yt 22.5 Vittoria Rubino Pro
54.00 Yt 21.5 Vredenstein Volante
55.00 N 23.5 BBB RoadBeat App
55.00 Yt 21.5 Bontrager Race X Lite Pro
55.00 Yt 21.0 Gommitalia Espresso
58.00 N 24.0 Hutchinson Fusion 2 Tubeless
Average
48.13
***
Rollout distance
[Some models could not be identified
as tubulars or clinchers from the previous
table because they had the same names.]
Tubular
? Rollout Brand Model
- ------- ----- -----
49.98 Hutchinson Fusion 2 Tubeless
Yt 49.98 Vredenstein Fortezza Pro
Yt 49.98 FMB Soie Naturelle
Yt 45.81 Challenge Criterium [21.5 or 24.0?]
45.81 Challenge Triathlon
Yt 41.65 Schwalbe Milano
41.65 Gommitalia Calypso
41.65 Gommitalia Blitz
41.65 Challenge Elite
41.65 Hutchinson Fusion Ultra
41.65 WTB Camino Alto
41.65 Vredenstein Fortezza Tricomp
41.65 Vittoria Open Corsa Evo Cx
41.65 Michelin Pro 3 Grip
41.65 Michelin Pro 3 Light
? 41.65 Challenge Forte [tub 21? clinch 23.5?]
41.65 Deda Tre RS Corsa
41.65 Continental Force
41.65 Schwalbe Ultremo R
41.65 Hutchinson Carbon Comp
x 41.65 Tufo C Elite Ride
Yt 41.65 Bontrager Race X Lite Pro
41.65 Vredenstein Fortezza Tricomp Pr
Yt 41.65 Deda Tre Olimpico
? 39.56 Vittoria Rubino Pro [tub 22.5? clinch 22.5?]
39.56 Vredenstein Fortezza Quattro Tricomp
? 39.56 Zipp Tangente [tub 21? clinch 21?]
? 39.56 Zipp Tangente [tub 21? clinch 21?]
37.48 Challenge Gara
37.48 Continental Grand Prix
37.48 Vittoria Diamante Pro 2
37.48 Hutchinson Fusion 2
37.48 Michelin Pro 3 Race
37.48 Continental Supersonic
37.48 Ritchey WCS Race S Open T
37.48 FMB Targa
? 37.48 Challenge Forte [tub 21? clinch 23.5?]
Yt 37.48 Continental Competition
35.40 Areo Jet Racing
x 35.40 Tufo CS 33 Special
x 35.40 Tufo C Grio Twix
Yt 35.40 Challenge Criterium [21.5 or 24.0?]
Yt 33.32 Schwalbe Montello
Yt 33.32 Gommitalia Platinum
33.32 Schwalbe Lugano
? 33.32 Vittoria Rubino Pro [tub 22.5? clinch 22.5?]
33.32 Vittoria Rubino Pro Tech
33.32 Hutchinson Intensive Longue Dist
33.32 Panaracer Extreme Evo 2
33.32 Area Extreme
33.32 Michelin Krylion
33.32 Bontrager Race X Lite
33.32 Bontrager Race X Lite AC
x 33.32 Tufo Hi composite
33.32 Continental Grand Prix 4 Season
33.32 Continental Attack
Yt 33.32 Continental Sprinter
33.32 Specialized Mondo S Works
33.32 WTB Solano
x 33.32 Tufo S3 Lite
x 33.32 Tufo Elite Jet
31.23 Panaracer Stradius Sport
31.23 Ritchey Race Slick Comp
Yt 31.23 Vredenstein Volante
31.23 Deda Tre HST
31.23 Vredenstein Fortezza Superlite
31.23 Bontrager Race XXX Lite
29.15 Gommitalia Logo
29.15 Deda Tre RS Dual
29.15 Specialized All Cond Flak Jack
29.15 Vittoria Rubino Pro Slick
29.15 Michelin Lithion
29.15 BBB RoadBeat App
x 29.15 Tufo S3 Pro
29.15 Specialized All Cond Armadillo Elite
29.15 Specialized Roubaix 23/25 Armadillo
Yt 29.15 Gommitalia Espresso
27.07 Deda Tre Grinta
27.07 Schwalbe Blizzard
27.07 Hutchinson Equinox
27.07 Deda Tre RS
27.07 Gommitalia Logo Sport
***
Cheers,
Carl Fogel
carl...@comcast.net
Dear Ben,
Another point of interest is slip-out angles that are given to 0.1
degrees for seven of the top eleven tires.
Admittedly, four are at 43.2 degrees, two are at 42.5 degrees, and one
is at 43.75 (yes, 43.75) degrees.
I wonder what the precision was on the speed, weight, water film, and
tire pressure that produced 43.75 degrees?
Maybe they had an oddly degree-scale?
Cheers,
Carl Fogel
Chris
Less numbers, more riding. After all, it's the rider, not the bike.
Chris
carl...@comcast.net
Dear Chris,
It looks as if the rider will roll fast in a straight line on the
Hutchinson Tubeless, but he'd better brake well before wet corners.
It's dead last in wet slip-angle, but tied for best in rolling
resistance.
I imagine that the lack of a tube helps reduce rolling resistance.
I wonder if the kind of rubber needed to seal a tubeless tire is just
rotten for wet traction, or if the poor traction is some mechanical
property of tubeless construction?
Cheers,
Carl Fogel
John Forrest Tomlinson
>[Some models could not be identified
> as tubulars or clinchers from the previous
> table because they had the same names.]
I thought it was because they were some strange hybrid of tubular and
clincher.
carl...@comcast.net
Dear John,
Sorry, I was too brief to be clear.
Yes, the Tufos were all originally marked with an X in the first table
because, as you say, they're neither tubular nor clincher. They're no
problem.
But four other tires came in two versions with the same name.
They were marked in the first table (wet-slip).
But the second table (rollout) had no column showing which tires were
tubulars.
So I added that column by hand, looking up what the tires were in the
first table.
All the Tufos are marked xx.
Tubulars are marked Yt.
Clinchers were left blank (easier to read).
And the pairs of these three tires in the second table are marked ?:
Challenge Forte [tub 21? clinch 23.5?]
Zipp Tangente [tub 21? clinch 21?]
Vittoria Rubino Pro [tub 22.5? clinch 22.5?]
The Zipp Tangente doesn't really matter, since both models had the
same rollout.
But the rollouts were different for the Challenge Forte (41.65 or
37.68) and Vittoria Rubino Pro (39.56 or 33.32).
The Challenge Criterium came as a tubular in two widths, but there was
no way to tell which had the good rollout of 45.81 or the medium
rollout of 33.32 (probably how many 2.085 circumference spins on a
drum):
Challenge Criterium [21.5 or 24.0?]
If both the 21.5 and the 24.0 Challenge Criterium were made with the
same casing and tread and both glued the same way, they'd still have
to be inflated to different pressures for a realistic RR test.
At 21.5, the cross-section of Challenge Criterium is 363 mm^2, versus
452 mm^2 for the 24.0 version
In the wet-slip test, the wider 24.0 Challenge Criterium did a little
better, 50 degrees versus 52 for the narrow 21.5 Challenge Criterium.
Cheers,
Carl Fogel
dusto...@mac.com
> Rollout distance
(Snipped for purpose of comparison)
> 49.98 Hutchinson Fusion 2 Tubeless
> Yt 49.98 Vredenstein Fortezza Pro
> 41.65 Vredenstein Fortezza Tricomp
> 41.65 Vittoria Open Corsa Evo Cx
> 41.65 Michelin Pro 3 Grip
> 41.65 Michelin Pro 3 Light
> 41.65 Hutchinson Carbon Comp
> 41.65 Vredenstein Fortezza Tricomp Pr
> Yt 41.65 Deda Tre Olimpico
> 37.48 Continental Grand Prix
> 37.48 Continental Supersonic
> ? 33.32 Vittoria Rubino Pro [tub 22.5? clinch 22.5?]
> 33.32 Michelin Krylion
> Yt 33.32 Continental Sprinter
> 29.15 Vittoria Rubino Pro Slick
> 29.15 Michelin Lithion
Hey, what's with that Vreddie tubular equaling the best clincher, and
rolling up to 20 revs further than some of the "tires I see people
using in Texas"?
At 2 meters and change advantage per rev, the best tubular and
clincher (which had the same rollout figure) are better by more than
40 meters than the last poor Michelin I left on the list? And 8 x 2,
for 16 (!!!) meters better than a peloton of commonly seen "high
performance" clinchers? 53 feet approx?
I can't help but wonder what difference this would make in a hundred-
mile road race! Or a 40k TT! --D-y
jobst....@stanfordalumni.org
>> Rollout distance
You'll notice that there is no mention of how the tubulars were
attached to the rim, that being their main flaw, accepted in the old
days because riders had to change their own flats or have someone do
it for them. Hard glue, that has essentially no rolling losses was
not used except on the track for two reasons. First, it was not known
how much RR road glue cost and second, it is hell to change tires with
hard glue and it doesn't work unless the tire has a bare (cloth) base
tape which most don't.
Jobst Brandt
Sandy
news:4994dc63$0$1618$742e...@news.sonic.net...
It was easy to guess you weren't one to buy the mag, as you never even
bothered to read the free snippings I sent you a couple of years ago (2007
results, as I recall). I even doubt you will divert your eyes to the free
data that has been provided.
Well, your glue knowledge remains constant while the world has moved on.
Indeed, the article for 2009 addresses glue, in particular, and the glue-on
intermediates strips as well. Should you have a shred of intellectual
curiosity, such as might even be used to reinforce your statement by
serendipity, go ahead - spend your lunch money one day and get a copy. Then
you'll know all about Lepici Paska.
--
Bonne route !
Sandy
Verneuil-sur-Seine FR
carl...@comcast.net
Dear D,
Depending on how heavy the drum is and how hard it's spun, the rollout
test can exaggerate the differences--which is the whole idea, since
very small differences are being measured.
As an analogy, flip a bike upside-down, crank the rear wheel up to 30
mph, and count the seconds for it to slow down to 10 mph. That's a
quick way to measure spoke wind drag plus bearing and seal drag for
comparison. Good or bad, wheels spin for an amazing time because
there's very little drag slowing them down. You're just measuring very
tiny differences, so you have to find a way to exaggerate or magnify
them.
In the drum rollout test, the tires that you show are ~50 spins versus
~30 spins, meaning that the good tire rolls about 66% further than the
other. In other words, each spin of the good tire eats up only 1/50th
of the same original force, while the bad tire wastes 1/30th of the
same force with each spin.
But it's a very small force.
Al Morrison's data has a nice column showing the watts per wheel:
http://www.biketechreview.com/tires/images/AFM_tire_testing_rev8.pdf
His tests show a range from 11.5 to 24.0 watts per wheel, meaning that
for a pair of tires the total difference between the best and the
worst in his basic test table of road tires is 25.0 watts--23.0 to
48.0 watts will be lost to the two tires.
The best RR in that table is 0.00233, a Vittoria EVO Corsa Crono 20 mm
tubluar, glued with the 5 coats of glue that Al found produces
significantly better RR than 2 coats of glue--see his note at the top
of the page.
The worst RR is 0.00488, a used Conti 3000 19 mm clincher with a butyl
tube.
So the ~30 to ~50 spins of a single tire in the rollout tables
correspond roughly to ~11.5 to ~24.0 watts on a single wheel in Al
Morrison's tables.
Notice at the top of that page that Al shows six tires testing between
11.5 and 11.9 watts, a range of 0.4 watts.
To get an idea of 0.4 watts and how sensitive the testing is, think of
a big keychain LED flashlight, which is rated at about 0.5 watts, and
1+ hours for a AAA battery:
http://store.advancedmart.com/twonhawaledp.html
Al's testing would detect a keychain LED flashlight at four levels of
brightness.
Cheers,
Carl Fogel
Peter Cole
>
> Al Morrison's data has a nice column showing the watts per wheel:
> http://www.biketechreview.com/tires/images/AFM_tire_testing_rev8.pdf
>
> His tests show a range from 11.5 to 24.0 watts per wheel, meaning that
> for a pair of tires the total difference between the best and the
> worst in his basic test table of road tires is 25.0 watts--23.0 to
> 48.0 watts will be lost to the two tires.
Jobst's test from 20+ years ago (& maybe others, I don't know) showed
large rr differences between various tires. The thing I always found
frustrating was the lack of info on the specific tires that were
available to me.
Al Morrison's data (& the 2006 Continental data) confirm that there's
still a wide rr variation. 25W is huge, the absolute value may even be
larger (1.5-2x) on pavement.
> Notice at the top of that page that Al shows six tires testing between
> 11.5 and 11.9 watts, a range of 0.4 watts.
What was news to me was that powertaps have gotten good enough to allow
this amount of precision and that people (at least one, anyway) are
measuring a large variety of tires. I'm never going to buy tubulars or
$60 a pop tires, but I'm hopeful that someday tires in my price range
will be tested. The Bontrager Selects look pretty good & they're even
made in 35mm.
Despite all the bickering and posing, a very useful thread.
dusto...@mac.com
(I asked):
> >Hey, what's with that Vreddie tubular equaling the best clincher, and
> >rolling up to 20 revs further than some of the "tires I see people
> >using in Texas"?
>
> >At 2 meters and change advantage per rev, the best tubular and
> >clincher (which had the same rollout figure) are better by more than
> >40 meters than the last poor Michelin I left on the list? And 8 x 2,
> >for 16 (!!!) meters better than a peloton of commonly seen "high
> >performance" clinchers? 53 feet approx?
>
> >I can't help but wonder what difference this would make in a hundred-
> >mile road race! Or a 40k TT!
(CF replied):
> In the drum rollout test, the tires that you show are ~50 spins versus
> ~30 spins, meaning that the good tire rolls about 66% further than the
> other. In other words, each spin of the good tire eats up only 1/50th
> of the same original force, while the bad tire wastes 1/30th of the
> same force with each spin.
> But it's a very small force.
If I coast 66% further down the road per packet of power input than an
identical me except for tires, I'm rolling 66% better, is another way
to look at it.
> Al Morrison's data has a nice column showing the watts per wheel:
> http://www.biketechreview.com/tires/images/AFM_tire_testing_rev8.pdf
>
> His tests show a range from 11.5 to 24.0 watts per wheel, meaning that
> for a pair of tires the total difference between the best and the
> worst in his basic test table of road tires is 25.0 watts--23.0 to
> 48.0 watts will be lost to the two tires.
Club riders make maybe 230 watts, the upper echelon of racers might
make 480 (for purposes of discussion).
Ten percent isn't anywhere near "vanishingly small".
> The best RR in that table is 0.00233, a Vittoria EVO Corsa Crono 20 mm
> tubluar, glued with the 5 coats of glue that Al found produces
> significantly better RR than 2 coats of glue--see his note at the top
> of the page.
Mr. Morrison's gluing technique is far superior to mine because I
couldn't seem to get layers anywhere near that thin. Even accounting
for overnight drying, whatever, two tubes of glue per wheel done "my
way" (and I wouldn't use that much glue in the first place), when
everything went molten in the Texas heat would have glue flying
everywhere including rim sidewalls, which is unimportant only on bikes
without rim brakes <g>.
That said, back in the day we in my cohort sought the strongest glue
joints between tire and rim so the tire wouldn't roll off in corners--
I did see a good racer "roll" one going in a straight line once when
he hit a depression in an asphalt parking lot roadway-- or flop off
after an impact flat while riding dirt/gravel roads, where a loose
tire might interfere with the wheel's rotation and cause a crash.
I can't say we got consistent results-- maybe we weren't so consistent
with our protocols <g>. But the bonds resulting from one full layer on
both rim and tire (full = "full width of rim, matching width of glue
stripe on tire"), wait until dry to touch and install were often
strong enough that a tool was needed to find a place to start the tire
off the rim, and the rim strip might well stay on while the tire
carcass peeled off if you got in a hurry. "In harm's way"; the point
was avoiding mishap.
In other words, I can't understand why someone would want less than a
secure tire-rim bond, and I especially can't understand bitching about
high rolling resistance of intentionally poorly glued-on tubular
tires. But that's California for you! --D-y
carl...@comcast.net
[snip]
Dear D,
A) A bicycle is not likely to roll 66% farther with the best versus
the worst tires if the two bikes start from any relevant speed. Think
about tire drag compared to air drag.
Consider that the power per wheel (watts) is calculated at 25 mph with
a 100 lb load (200 lbs on each).
B) Take the top twenty tires in that test and compare them, not the
very best to the very worst from several years that Al has spent
testing every road tire that he can get his hands on:
http://www.biketechreview.com/tires/images/AFM_tire_testing_rev8.pdf
The top 20 tires range from 11.5 to 13.2 watts per wheel.
That's a range of 1.7 watts per wheel, or 3.4 watts per pair for a
200-lb load at 25 mph.
The top 20 tires were measured at 19 to 27 mm wide.
All twenty top tires were tested at 120 psi, including the 27 mm rated
for only 100 psi max. On real roads with real pressures, the
suspension factor may produce considerable differences RR for tires
whose widths vary 8 mm.
As Al's header points out, "Crr values are typical for very smooth
surfaces--Crr on typical road surfaces may be 50 to 100% higher."
Where the same tires would end up on real roads is hard to say, partly
because rollers aren't roads and partly because the watts per wheel
isn't adjusted for air drag.
Of the top 20 tires, eight were tubulars. Al's data suggests that four
of the tubulars would have dropped out of the top twenty with the
original gluing method.
The top 20 tires were a mixture of new and used tires.
The top 20 tires included no butyl tubes.
Near the end of the PDF, Al compares a few tires with latex versus
butyl tubes. In the same tire, a latex tube could change the results
from 0.7 to 2.8 watts per wheel.
That's a variation in a small sample of 2.1 watts per wheel for tubes,
versus a variation of only 1.7 watts for the top 20 tires.
Look just below those Latex vs. butyl tube tests and there's an
interesting section on Michelin (thin) vs. Vredstein (thick) latex
tubes. No difference was detected between the tubes in the same used
23 mm Michelin Pro 2 Lite tire, but 0.3 watts per tire was detected
between the two tubes in a narrower 20 mm used Conti Supersonic. (And
the watts for the Michelin dropped from 13.0 to 12.1 on a "warm day"
test.)
That 0.3 watt difference detected between two latex tubes in one tire
(but not in another) would be enough to rearrange the order of the top
20 tires considerably.
Al's work is impressive, but it may not be saying quite what you'd
think at first glance.
Cheers,
Carl Fogel
carl...@comcast.net
wrote:
Dear Peter,
It isn't clear from Al's comment at the top of his PDF what the
50-100% RR increase for real roads means.
It could be that everything just scales up proportionally, so that the
absolute watt differences increase 50% to 100%. In that case, the
power differences on real roads between the very best and the very
worst would rise from 25 watts per pair of wheels to 50 watts per pair
of wheels.
But it could be that the RR rises and the absolute differences stay
about the same, or even drop. When the top 20 tires are all inflated
to 120 psi, range from 19 mm to 27 mm, and vary from 11.5 watts to
13.2 watts per wheel for a 200-lb load at 25 mph, it's hard to figure
out what will happen on real roads.
Look at the tests at the bottom of the PDF and see what you think of
the curious data concerning latex and butyl tubes and thin versus
thick latex tubes.
Cheers,
Carl Fogel
dusto...@mac.com
> A) A bicycle is not likely to roll 66% farther with the best versus
> the worst tires if the two bikes start from any relevant speed. Think
> about tire drag compared to air drag.
Air drag could be equal between two bikes with different tires. Then
one bike would roll 66% further.
> Consider that the power per wheel (watts) is calculated at 25 mph with
> a 100 lb load (200 lbs on each).
That sounds like a fairly heavy rider cranking along at a good clip.
Did I miss something?
> B) Take the top twenty tires in that test and compare them, not the
> very best to the very worst from several years that Al has spent
> testing every road tire that he can get his hands on:
> http://www.biketechreview.com/tires/images/AFM_tire_testing_rev8.pdf
>
> The top 20 tires range from 11.5 to 13.2 watts per wheel.
>
> That's a range of 1.7 watts per wheel, or 3.4 watts per pair for a
> 200-lb load at 25 mph.
The bottom 20 or so on the list contain 10 or more tires that are held
in high (Vred)-esteem, Common-tinently used, and at least a few are
slow even with latex tubes, which for most people do not give
sufficient service life.
I guess I could add "rough figure" addition to see how much worse
they'd be with butyl tubes, but I don't even want to know <g>.
Which just adds a little weight to my line of questioning: What's that
Vreddie tubular doing up there as co-valedictorian when the "new
clinchers" are just as good or *better* than tubulars, particularly
IRT rolling resistance?
(noting, my favorite tire, if it is indeed the same tire as in days of
old, the Vittoria CG tubular, was/is a "slow" tire compared to some
modern clinchers with latex tubes)
> All twenty top tires were tested at 120 psi, including the 27 mm rated
> for only 100 psi max. On real roads with real pressures, the
> suspension factor may produce considerable differences RR for tires
> whose widths vary 8 mm.
>
> As Al's header points out, "Crr values are typical for very smooth
> surfaces--Crr on typical road surfaces may be 50 to 100% higher."
>
> Where the same tires would end up on real roads is hard to say, partly
> because rollers aren't roads and partly because the watts per wheel
> isn't adjusted for air drag.
(snippage of "confusion factors" for brevity)
It's not hard to imagine a tire that would be faster on a given
surface than a different tire that might be faster on a different
surface.
Or that the difference might be enough watts to matter.
"Al is doing the best he can here".
--D-y
carl...@comcast.net
>On Feb 13, 1:01 pm, carlfo...@comcast.net wrote:
>
>> A) A bicycle is not likely to roll 66% farther with the best versus
>> the worst tires if the two bikes start from any relevant speed. Think
>> about tire drag compared to air drag.
>
>Air drag could be equal between two bikes with different tires. Then
>one bike would roll 66% further.
[snip]
Dear D,
Yes, it's assumed that the two would have equal wind drag.
Think some more about the overall effect of equal wind drag and
different tire drag on total distance.
Cheers,
Carl Fogel
dusto...@mac.com
> On Fri, 13 Feb 2009 11:38:09 -0800 (PST), "dustoyev...@mac.com"
>
> <dustoyev...@mac.com> wrote:
> >On Feb 13, 1:01 pm, carlfo...@comcast.net wrote:
>
> >> A) A bicycle is not likely to roll 66% farther with the best versus
> >> the worst tires if the two bikes start from any relevant speed. Think
> >> about tire drag compared to air drag.
>
> >Air drag could be equal between two bikes with different tires. Then
> >one bike would roll 66% further.
>
> [snip]
>
> Dear D,
>
> Yes, it's assumed that the two would have equal wind drag.
>
> Think some more about the overall effect of equal wind drag and
> different tire drag on total distance.
Equal wind drag means that bike A will roll 66% further before it
stops than bike B if bike B has tires that roll 66% less far than bike
A's tires due to rolling resistance, and the overall wind resistance
for both bikes is the same. --D-y
Carl Sundquist
The different rr would mean that the bikes are slowing at different
rates. Therefore the bike with the superior rr would be subject to
greater wind drag at a given distance, right?
carl...@comcast.net
wrote:
Dear D & Carl,
Maybe it's not as obvious as I think, so . . .
http://www.biketechreview.com/tires/images/AFM_tire_testing_rev8.pdf
That's Al Morrison's Crr table with explanation at the top, pointing
out that the watts are per wheel when he's doing ~25 mph on rollers.
Imagine two bicycles doing 25 mph, side by side, with equal wind drag,
but two different tires, one the best in Al Morrison's table at 23
watts drag for a pair of 11.5 watt drag tires, the other the worst
with 48 watts drag for a pair of 24 watt drag tires.
Side by side, the two riders are putting out different power to
maintain 25 mph with 200 lbs load (100 on each tire).
But the different power doesn't matter at all because both riders stop
pedaling at the same instant and start coasting.
All the power now is just their perfectly equal momentum. There are no
transmission losses, no tricky power differences. They're equivalent
to a pair of bowling balls, one of which has 24 watts of rolling drag
when the coasting starts, while the other has 48 watts of rolling
drag.
The tire drag is _much_ smaller than the wind drag.
How far they roll is mainly determined by wind drag. The _difference_
separating the winner from the loser will be due to tire drag, but the
total distance is mostly a matter of wind drag.
As they slow down, _both_ drag effects drop. The wind drag drops
faster than the tire drag, but wind drag starts out so much higher
than the tire drag that it's the dominant effect.
Here's a crude graph to illustrate the idea of how far the two bikes
roll and what's eating up their momentum:
wind drag
tire drag
A 123456789012345
12345
---total-distance---
B 123456789012345
123
--total-distance--
Bicycle A rolls 66% farther for tire drag alone (5 units versus 3).
Overall, one bike rolls only a little bit further because most of the
overall slowdown is due to wind drag (the 15 units). Instead of 66%
farther, one bike rolls about 10% farther.
There's no wind drag in the RR calculations from Al Morrison or Le
Cycle. Al's Crr and the Le Cycle spin-down tests isolate the
comparatively minor tire drag from the huge wind drag.
That's why Al has to work like crazy with power meters and correction
spreadsheets to give us such wonderful data--he's looking at very
small differences. His extremely sophisticated method lets him
distinguish between 0.00233 and 0.0234, with details about
repeatability and standard deviation and temperature correction and so
on.
And that's why the Le Cycle tests don't differentiate as well as Al
does between the tires. They're counting ~2.085 meter spins on a drum,
so their figures really mean ~24 spins for the top three tires, ~22
for the next, and so on.
Here's a speed calculator:
http://bikecalculator.com/veloUS.html
Plug in 25 mph for both, 178 lbs to raise the weight to 200, change to
tubulars and on the drops, and 100% transmission efficiency.
(Tubulars for this calculator mean only that the RR is 0.0040, as
opposed to 0.0050 for clinchers, which was reasonable back when the
calculator was written. There's an all-numeric km/h version that lets
you enter any RR, or you can save the page to your hard drive, view
the source, and change the tubular-clincher-MTB values to suit
yourself.)
Anyway, the two bikes are rolling along at the same speed at 25 mph,
which means an instantaneous "power" proportional to ~288 watts on a
0.0040 RR bike when they cut the power and start coasting.
The point is not the precise details of power and distance. The point
is that when we cut the power, something like 24 to 48 watts was going
to tire drag, while somewhere around 240 to 260 watts (5 to 10 times
the tire drag) was going into wind drag.
The real brake on a flat rollout from 25 mph is the wind drag, not the
tire drag. Don't expect 66% distance differences when the tire drag is
only 10% to 20% of the braking force, not 100%.
(And yes, accurate calculations would be hideous. The slower you go,
the more the tire's portion of the total drag rises over time. I'd
love to find a flat coasting distance calculator, but so far I haven't
seen hide nor hair of such a beast. Maybe Tom Anhalt, whose
spreadsheet Al uses, could do it, or Tom Compton of Analytic Cycling,
but they tend to be interested in more practical calculations.)
Again, both wind and tire drag start dropping as soon as the riders
stop pedaling and start coasting. (Transmission efficiency vansishes.)
The wind drag drops much faster than tire drag, but it has to--the
wind drag started five to ten times as high as tire drag, and both
will reach 0 at the same instant when the rolling stops.
As an analogy, think of rolling down a steep hill to test equal brake
drag on two bikes. Now give the riders a stick to trail behind them.
One stick may be 66% more effective than the other, but sticks are far
less effective than real brakes, so the total stopping distance for
brakes plus sticks won't be 66% shorter for the rider trailing the 66%
more effective stick.
Cheers,
Carl Fogel
Robert Chung
> What was news to me was that powertaps have gotten good enough to
> allow this amount of precision
These particular data were collected with an SRM, not a Power Tap. However,
whichever power meter was used, the key is hardly ever the measurement tool.
Don't ignore the contributions of careful experimental technique and
analytical methods.
> Despite all the bickering and posing, a very useful thread.
"Despite?" I would have said, "In addition to."
dusto...@mac.com
> Maybe it's not as obvious as I think, so . . .
More obvious to one who has delved deeper.
> The point is not the precise details of power and distance. The point
> is that when we cut the power, something like 24 to 48 watts was going
> to tire drag, while somewhere around 240 to 260 watts (5 to 10 times
> the tire drag) was going into wind drag.
>
> The real brake on a flat rollout from 25 mph is the wind drag, not the
> tire drag. Don't expect 66% distance differences when the tire drag is
> only 10% to 20% of the braking force, not 100%.
Thanks, I understand.
>
> (And yes, accurate calculations would be hideous. The slower you go,
> the more the tire's portion of the total drag rises over time. I'd
> love to find a flat coasting distance calculator, but so far I haven't
> seen hide nor hair of such a beast. Maybe Tom Anhalt, whose
> spreadsheet Al uses, could do it, or Tom Compton of Analytic Cycling,
> but they tend to be interested in more practical calculations.)
If I were selling tires I knew would win the coast-down, I'd think
those calculations would be very practical.
It would be great fun to be able to set up a controlled experiment. I
can envision the easier-rolling tire to have enough advantage once
speed goes to near-wobble that the faster tire would coast further
than might be expected? Can you tell I like the idea of easy-rolling
tires?
The only discouraging thing here is, scrounging for cheap tires might
be a little less fun-- truth setting me more expensive instead of
free? --D-y
Marcus Coles
tire wins instead. ;-)
Marcus
Robert Chung
> Since on coast down the acceleration is already paid for maybe the
> heavy tire wins instead.
This is exactly why we good estimates of Crr either control or measure
accelerations.