New tire rolling resistance data.

Back by popular demand, the general all-things Road forum!

Moderator: robbosmans

mattydubs
Posts: 82
Joined: Mon Jan 14, 2013 10:57 pm
Location: Denver

by mattydubs

Drop the PSI and... uh, 25s? I thought 25s were the new black in clincher widths. At least that's what all Cat 3+s I know train on and that's what Mr Zinn says.

by Weenie


Visit starbike.com Online Retailer for HighEnd cycling components
Great Prices ✓    Broad Selection ✓    Worldwide Delivery ✓

www.starbike.com



jano
Posts: 151
Joined: Sun Oct 21, 2012 8:15 pm

by jano

fdegrove wrote:Hi,

I firmly believe that unless you actually test on an imperfect surface, that atleast somewhat replicates the surface of a real road, you've got nothing.


There is no reason for that. It would only change the numbers but not the rankings per se.

IOW a topformer in the lab should perform equally well on a road.

Ciao, ;)


Exactly.
A comparison can only be done in a controlled environment, otherwise it's useless.

User avatar
WMW
in the industry
Posts: 893
Joined: Sun Jul 29, 2012 2:59 pm
Location: Ruidoso, NM

by WMW

I avoided posting anywhere else, because it was already taking too much time to maintain the slowtwitch thread... but what the hell.

Testing for Crr on a smooth roller can only give you the energy loss due to tire casing hysteresis and deflection. On the road, surface roughness and hardness also play a role. I'm not measuring that... only the tire deflection component. A 1.5x factor is applied to the measurements to bring them more in line with field test results for a smooth road. On a rough road the losses would be much higher.

Even though the contact patch on an 80mm roller is very different (and the resistance much higher) than it is on the road, the results can be scaled to smooth road conditions. I'm frankly surprised that it works this well, but the people who have done meticulous and difficult field testing to isolate Crr have seen the same relative Crr difference as on roller tests. And roller tests are much easier to perform accurately.
formerly rruff...

bm0p700f
in the industry
Posts: 5777
Joined: Sat May 12, 2012 7:25 pm
Location: Glermsford, Suffolk U.K
Contact:

by bm0p700f

I agree the data will provide a ranking that will be the same in the real world. Tests like this have to be done under controlled conditions otherwise the data is worthless. WNW what are the errors in the data or do you not know.

petromyzon
Posts: 781
Joined: Mon Apr 05, 2010 4:14 pm

by petromyzon

WMW/rruff, firstly, thank you very much for providing this data, lots of work!

Was the Ultremo ZX the 2012 version with their HD speed guard breaker or the 2013 version with the V guard Vectran breaker?

I'd like to see fdegrove's hypothesis tested.... Would a Vittoria carcass with Black Chilli rubber and a Vectran breaker be the best of all worlds?

User avatar
WMW
in the industry
Posts: 893
Joined: Sun Jul 29, 2012 2:59 pm
Location: Ruidoso, NM

by WMW

bm0p700f wrote:WNW what are the errors in the data or do you not know.


I did 4 repeats of Conti SS 20mm dispersed over 2 days and the max scatter was less than 2%. I'm happy with that. I'll know more about it after I've retested more tires. If you are wondering how much of an error there is in translating the numbers to on-road Crr, it appears to be low enough that the roller results are valid for making comparisons.

There are other factors I need to get refined like the temperature effect on Crr. I'm using .6%/F currently, which is what I read in Bicycling Science. I normalize to a fixed ambient temperature because I figured the temperature difference would be nil on the road. Tom Anhalt thinks the effect is larger based on his testing... and also believes no correction should be applied at all since the tire would heat up on the road also. I still doubt that there will be much of a difference between the tire temperatures on the road, but that is something I can test.
formerly rruff...

User avatar
WMW
in the industry
Posts: 893
Joined: Sun Jul 29, 2012 2:59 pm
Location: Ruidoso, NM

by WMW

The ZX I tested has the HD Speed Guard. Maybe the new one is better, but it really has a long way to go before it makes it to the "good" range.

I agree that a high TPI casing and the Conti BC tread compound would likely be the best. Maybe they should collaborate with the Veloflex people to make a new line...
formerly rruff...

User avatar
Kraaf
Posts: 557
Joined: Mon May 22, 2006 8:18 pm
Location: Netherlands

by Kraaf

Thanks for chiming in, Andy.
And thanks, Ron.
Cool stuff.
.
I love you guys. Seriously.
_________________

fdegrove
Tubbie Guru
Posts: 5894
Joined: Tue May 03, 2005 2:20 am
Location: Belgium

by fdegrove

Hi,

It's no more than a wild guess but I suppose Conti historically opted for a lower TPI count in order to prevent the sidewalls from being cut too easily.

Even now I still own some tubular tyres which do not have any anti-puncture belt whatsoever. They ride amazingly well and provided you avoid urban trash areas they do not puncture any more or less than a more "modern" tyre. They're not even anything special as they come with a butyl tube and are dirt cheap too.
Incredibly comfortable, even more so than a Veloflex Carbon at the same pressure. All that with a butyl inner tube.
As you would expect, that particular model has been phased out just recently only to be replaced by one that does come with a PRB which, would you believe, adds 50g per tyre. I want a gun. A big one..... :roll:

All this makes you wonder what could be made if someone really put some effort in it and build a real racing tyre. Tubular and clincher, no frills just speed. :lol:

Sure, it would be great if the impact of a PRB, TPI count and the nature of a compound could be tested thouroughly and individually but how?
AFAIK all PRB have a negative impact on weight, Crr and comfort. Most aren't even that effective to start with.
Casing's TPI, higher is better and how many times in a lifetime do we slash a sidewall anyway?
Compound? Other than Black Chili (excellent in wet conditions) I'd say keep it KISS. Veloflex's natural rubber is a fine example.
And what's the impact of vulcanizing over glueing, thread profiles versus slick?

Food for thought..........

\\RANT OFF. Apologies to rruf for the sidejacking.....

Ciao, ;)
Being a snob is an expensive hobby.

User avatar
WMW
in the industry
Posts: 893
Joined: Sun Jul 29, 2012 2:59 pm
Location: Ruidoso, NM

by WMW

fdegrove wrote:Sure, it would be great if the impact of a PRB, TPI count and the nature of a compound could be tested thouroughly and individually but how?


Continental's top tires all have the same casing material. The GP4000 has thicker tread and a Vectran belt... the Attack, Force and the TT tire all have thinner tread and a Vectran belt... although I'm not certain if the tread compound or Vectran belts are slightly different (probably not)... and the Supersonics have thin tread and no Vectran belt.

When Continental came out with the Black Chili compound, their rolling resistance went from bad to good. I think the tread compound was the only change. Plus the BC tread has good traction and great longevity. The only negative is that it seems to not ride as smoothly.
formerly rruff...

sedluk
Posts: 412
Joined: Mon Sep 07, 2009 3:10 am

by sedluk

The test is fundamentally flawed. There is no correlation between a tire on a smooth surface versus a rough surface. Roads are never a smooth surface so the results are meaningless.

As a thought experiment, take a hard solid rubber tire. It will have incredibly low resistance on a smooth surface and incredibly high energy loss on a rough surface. With this test, the solid tire would still be predicted to be the most efficient on a rough surface. Sorry but completely flawed.

tetonrider
Posts: 91
Joined: Wed Oct 06, 2010 3:38 am

by tetonrider

i really commend WMW and others who are putting their time and hearts into these comparisons and making the data available. i said the same over in the slowtwitch thread, but i'm hopeful that i can carve out some time to post data from a number of tires i own and use frequently -- including some vittoria clinchers as well as a number of tubulars (continental, vittoria, fmb, dugast). could be interesting.

i think there is quite a bit to be learned from this testing.

that said, while i want to believe that the order and magnitude of difference as exhibited on a roller scale linearly to rougher roads, i'm just not sure i'm there. i see references to "careful tests" that prove this, but i've never seen any of these tests (and i've been looking for a while, though perhaps in the wrong places). i'm open to the possibility that i am wrong.

i tend to agree with sedluk's point of view. if you take an extreme example (a tire that will not deform at all; hypothetical) it might rank #1 there but would fair poorly on the world's worst road. i notice that reference to "the real world" generally suggest that the roller results scale to smooth roads. hm.

while this example is extreme, this leads me to believe that there is somewhere along the spectrum between "smooth road" and "world's worst road" where the correlation to the rollers breaks down. would it be on chip seal? that's real world where i ride and race. also, there are different grades of chip seal.

i think the further trouble i have is that tires have different properties, some of which allow the tire to be more flexible at higher pressures -- or some allow the rider to run lower pressures without risking pinch flats. i know there must be an attempt to standardize, but comparing all tires at the same pressure may mean that some (all?) tires are running at sub-optimal pressures. this example becomes clear if you compare Crr on the rollers for a 19mm tire vs a 25. on the road one would never choose the same pressure for both, given a constant rider weight.

doing this kind of testing would be even more tedious. an example: do many runs with tires at 5psi increments to find the lowest Crr for each. then, compare those "optimal" conditions to one another.

still doesn't solve the chip-seal road issue -- but that could be done by creating/modifying a roller.

if all this DOES scale linearly, i'd be delighted.

petromyzon
Posts: 781
Joined: Mon Apr 05, 2010 4:14 pm

by petromyzon

Did you read the thread Sedluk? Separate data have suggested that as you increase pressure rolling resistance decreases, up to a break point where it increases sharply because compliance is important on rough roads. A solid tyre may well be a completely difficult case. However, previous data has also suggested that the rank order of rolling resistance determined by this method is preserved when the tyres are tested outside on real roads. Commercial testing labs use this method.

cyclenutnz
Posts: 854
Joined: Tue Jul 18, 2006 5:18 am
Location: Cambridge, New Zealand
Contact:

by cyclenutnz

I use a surface roughness factor in course modelling that scales Als CRR figures - I've not done formal testing of scenarios but once I've evaluated the surface the models work very well in predicting the effect of changing tyres for different riders.

And in formal testing I've seen predictable (per the AFM figures) changes in CRR when I've changed tyre set ups since the previous session (testing focussed on Aero so any CRR info is incidental).

by Weenie


Visit starbike.com Online Retailer for HighEnd cycling components
Great Prices ✓    Broad Selection ✓    Worldwide Delivery ✓

www.starbike.com



acoggan
Posts: 26
Joined: Thu Dec 22, 2011 8:28 pm

by acoggan

sedluk wrote:The test is fundamentally flawed. There is no correlation between a tire on a smooth surface versus a rough surface. Roads are never a smooth surface so the results are meaningless.


http://www.trainingandracingwithapowerm ... -part.html

http://xkcd.com/54/
Last edited by acoggan on Fri Feb 08, 2013 3:07 pm, edited 1 time in total.

Post Reply