I’ve been wanting to write about rolling resistance for years, and I’ve had ongoing e-mail conversations with a number of you on the subject. Indeed, I’ve built up enough of it to compile a collection of some of the most interesting.
As I was riding though an unexpected rainstorm, I noticed that, even with the rain, it seemed to take less effort to ride on the wet road than when it is dry (all other things – especially wind and temperature – being constant). I know that there is less friction when it comes time to stop or to turn. Could it be possible that the rolling resistance is also reduced?
Yes, a film of water on the road reduces rolling resistance. This is something that has been understood by time trialists for a long time. For example, do you remember how Lance Armstrong, despite anticipated rain during the stage, predicted a new record for the final time trial of the 2003 Tour de France? He and Jan Ullrich, a minute apart on G.C. and with all of the marbles on the line, indeed went at record pace in pouring rain until Ullrich crashed with 10km to go. To ride that fast despite having to go slower on the wet corners and numerous roundabouts with slick painted lines, you can at least be sure the water on the road was not slowing them down.
David Millar ultimately went on to win that stage with Tyler Hamilton nine seconds in arrears, but Armstrong and Ullrich were ahead of both of their splits until Armstrong eased off when he heard that the German had crashed. Ullrich hit the second time check averaging 55.21kph, two seconds ahead of Armstrong, who was averaging 55.16kph. At that point, both were ahead of Greg LeMond’s 1989 time trial pace of 54.545kph, which was the Tour’s fastest until David Zabriskie broke it in the 2005 Tour prologue.
My theory is that the water effectively makes the road smoother, filling in spaces between pieces of gravel in the road.
Perhaps we are way past the point of diminishing returns, but what effect does a product like Slime, Stan’s, Specialized Airlock, or True Goo have on mountain bike UST performance? It seems like the more viscous the product, the more rolling resistance would increase.
Are you aware of any objective comparisons regarding the performance or lifespan of these products? Do the pros every use this stuff or is it just for the average Fred?
Though I have not seen test results on it, I imagine that you are right. And some pros definitely do use Stan’s NoTubes; I don’t know about the other products.
It makes sense to me that more liquid stuff, like Stan’s, would result in smaller energy losses, although the masses involved might be so small that the differences between sealants are irrelevant, as you suggest. On the other hand, saving the frictional energy losses between the inner tube and the tire casing is probably a net rolling resistance improvement with any sealant. But I am just guessing.
Tubes vs. tubeless
I’m curious what you think about the “25 percent decreased rolling resistance” of the new Dura-Ace/Hutchinson tubeless system. I would imagine a rough road surface might yield a significant difference but that a smooth road would be negligible.
I don’t know about the magnitude, but I think the improvement is real, due to the elimination of friction between the tire and tube. But I would guess that even a 25 percent improvement in rolling resistance is hard to measure in time savings without a very rigidly controlled experiment.
Rolling resistance is subtle at high speed relative to air resistance and at low speed on climbs relative to gravity. Though I ride the Hutchinson road tubeless tires faithfully on one of my bikes and think I can notice a rolling resistance improvement sometimes, I haven’t been able to measure it on my speedometer.
Choice of glue
I was recently reading through some literature on the rolling resistance of various tires and found one report of tests done by IRC that found tubulars glued with “hard track glue” had a lower rolling resistance than tires attached with “soft glue”. I have a couple questions about these findings: (1) Can you confirm this is true and (2) would using gluing tape (e.g., Tufo Tubular Tire Gluing Tape) affect the rolling resistance?
I can confirm that IRC made those tests a while back, but I can’t confirm its truth or not. It makessense, given that hysteresis would be reduced with harder glue. Certainly using gluing tape would affect rolling resistance.
One of my riding buddies has shown me some of the many e-chats about the “new” clincher versus tubular rolling resistance debate. The claim running around is that a top clincher has a much lower rolling resistance than a top tubular, owing mostly due to tire squirm of the tubular-glue contact. I have seen a claim at a “top” analytic cycling time trial computing site that states that on a three-kilometer track pursuit, the extra resistance of using a tubular with road glue (extra squirm as compared to the harder drying track glue) the clincher would be 2.5 seconds. Can this be true?
At a 40km distance, at that 3km pursuit speed anyway (no easy feat I know), that would be more than a 30 second advantage just by using clinchers, assuming identical aerodynamics. My friend is already selling off his tubular aero wheels in favor of clinchers, and he is pulling my leg to do the same for ITT.
See the tables at the end of this article.
Apparently there was an article in the German magazine Tour at the end of the 90′s that started this whole thing.
Clinchers are certainly easier, and if they are also faster, then it would hardly seem worth the extra effort to keep tubulars any more for a time trial bike. Is this true or is this bunk?
While tubulars are generally lighter for similar tire quality, and tubular wheels can save a lot of weight over clincher wheels, rolling resistance may not fall to the advantage of the tubulars. I still cannot say which is the appropriate choice for you, given the superiority of a clincher in price, installation, and probably in rolling resistance, and the superiority of a tubular in weight, round cross-section for cornering performance, capacity for high tire pressure, and staying on the rim in the case of sudden air loss.
I am attaching correspondence below from an interested engineer and bike racer named Tom who has studied the subject quite a bit and is now convinced of the superiority of clinchers, and from Alberto De Gioannini, an engineer who used to work for Vittoria and who now makes super-high-quality torque wrenches for bikes.
My conclusion is based on a careful study of the relevant data available to the general public. Check out: Check outthe results of this study. Germany’s Tour magazine has published the results of rolling resistance tests for clinchers to go along with their previously published testing of tubulars. All the tests were performed at the Continental Tire facility, and I’ve attached a summary of Coefficient of Rolling Resistance (Crr) of both types of tires tested in a single list. [Editor's note: The following table represents a combination of these twolists.]
|Deda Tre Giro d’Italia||0.0038|
|Vittoria Open Corsa Evo CX||0.0039|
|Michelin Pro 2 Race||0.0042|
|Vittoria Diamante Pro Rain||0.0044|
|Michelin Megamium 2||0.0047|
|*Veloflex Carbon (Tubular)||0.0049|
|*Gommitalia Route du Nord (Tubular)||0.0050|
|Panaracer Stradius Pro||0.0051|
|Schwalbe Stelvio Plus||0.0052|
|*Gommitalia Platinum (Tubular)||0.0053|
|*Vittoria Corsa Evo CX (Tubular)||0.0054|
|Schwalbe Stelvio Evolution Front||0.0056|
|Continental GP Force (rear specific)||0.0057|
|Schwalbe Stelvio Evolution Rear||0.0057|
|*Vittoria Corsa Evo KS (Tubular)||0.0057|
|Continental Ultra GatorSkin||0.0058|
|Ritchey Pro Race Slick WCS||0.0058|
|*Continental Competition (Tubular)||0.0059|
|*Veloflex Roubaix (Tubular)||0.0059|
|*Continental Podium (Tubular)||0.0060|
|Specialized S-Works Mondo||0.0061|
|Continental GP 3000||0.0067|
|Hutchinson Top Speed||0.0069|
|*Schwalbe Stelvio (Tubular)||0.0069|
|Continental GP Attack (front specific)||0.0073|
|*Tufo Elite Jet (Tubular)||0.0073|
|*Schwalbe Montello 300 (Tubular)||0.0075|
|*Tufo Hi-Composite Carbon (Tubular)||0.0077|
*denotes tubular tire test
As can be seen by that list, from a rolling resistance standpoint, the best clinchers have a significant rolling resistance advantage over the best tubulars tested. Particularly interesting to me were the results of the Vittoria Corsa CX and Open Corsa CX tires. These tires are literally identical (same casing, same tread) except for the means of attachment to the rim.
In this testing, the tubular version had an almost 40 percent higher rolling resistance, 0.0039 vs. 0.0054.
Using those results, I calculated that using identical wheelsets (I used Zipp 404 clincher and tubular models shod with Vittoria Corsas) it would take a grade of more than 8 to 9 percent before the weight advantage of the ~1 lb. lighter tubular wheelset/tire/tube combo would finally offset the lower rolling resistance advantage of the clincher versions. At all grades below that, the clinchers would save power.
You are correct that thread count is a large determiner of rolling resistance. Obviously, the higher the (true) thread count, the thinner the threads of the casing will be which lessens the thickness of the casing wall and results in less hysteresis losses. But, there are other factors in play as well, such as tread thickness and material type, anti-puncture strips, and even thickness and material of the tube.
Tubulars have an additional potential hysteresis loss due to the glue used to attach the tire to the rim. Typical “road glue” has measurable losses (as shown in the data above) whereas hard “track glue” would minimize these losses.
Tubulars attached to rims using gluing tape will inherently have even higher losses than a standard gluing since the thickness of glue is greater and the tape in the middle also contributes a hysteretic loss.
If you care to see what the predicted differences indicated by the data linked to above would be for typical TT performances, check out the online calculator Tom Compton coded up at his website.
Be sure to put in realistic entries and not just accept the default entry values.
Here are the results of some recent on-the-road tests (not rigorously performed, but eye opening).
I did four trials both up and down the hill, with Ritchey clincher set (Runs 1 and 2) with 23c Michelin Pro Race tires and 2 with Zipp 404s with Tufo Elite Road tires applied with Tufo tape (runs 3 and 4). The conditions were ideal, with very consistent temperature and no perceptible wind due to the sheltering.
The course I used omitted the very top and very bottom sections of the road and ended up being 410 feet of climbing over 0.74 miles for an average grade of 10.6 percent.
I targeted 225 watts and did pretty well with my runs showing average powers of 227, 223, 226, and 226W respectively.
Here were the times for the four runs:
1. – 7:49.5
2. – 7:56.2
3. – 8:05.8
4. – 8:02.8As you can see, runs No. 3 and 4 took a range of seven to 15 seconds longer. It was not looking good for the Tufos, especially since the wheelset was about one pound lighter than the Ritcheys.
I then took all the data (weather conditions, altitude, weight, etc.) and plugged them into www.analyticcycling.com to see how the power compared to what would be predicted for this climbing speed. The model predicted that the clinchers would take on average 5.1W more power than I actually recorded, while it predicted that the Tufos would take on average 4.3 W less than I recorded. This means that at this speed, the difference in power between the two sets was an average of 9.4W…in other words, the Tufo’s rolling resistance at this speed would require 9.4W more power in order to climb at the same speed (~5.6mph).
An important thing to remember is that the power required to overcome rolling resistance is proportional to speed and weight. So, for a heavier rider going at a faster speed, the power differential would be even greater. Back-calculating for the Coefficient of Rolling Resistance (Crr) from the power difference and the speed, I found that the difference in the Crr was .0047.
This number can then be used to calculate the power loss under other conditions. For example, using my weight traveling at 25 mph if I had Tufos mounted with tape on my wheels, it would require over 40W extra to go the same speed.
As a check, I used coast-down tests…the surprising thing is, that even though the 404s are much more aero than the Ritcheys, my “terminal velocity” was actually greater by 1mph with the clinchers (41.8mph versus 40.8mph). Plugging these numbers into www.analyticcycling.com. I get the same Crr difference to account for that speed difference.
So, it looks like I repeated the results from other tests (I recall one showing a Crr difference of 0.002 for changing just one tire…so that matches) for Tufos with tape vs. good clinchers.
Even though I took great care in performing this test (to the point of even using the same cassette on both wheelsets) the number of runs are small and an argument can be made that the differences seen are not statistically significant. I’m confident that more runs would change that…but I just don’t have the time. However, this test was just an I’ll-believe-it-when-I-see-it type of confirmation for me after seeing results of similar testing by other riders.
Of course, many people question the applicability of smooth drum testing vs. on-road performance, but I can explain why drum tests ARE applicable. Basically, the same mechanisms that “rank” the tires on a smooth drum (i.e. hysteresis losses) will give them the same “rank” on a rough surface.
I’ve read with great interest the debate on rotating weigh and wheel inertia. All cyclists know by experience that light wheels make you climb faster, so all that have been said related to ‘climbing’ makes good sense (and it’s well documented too).Results seem to confirm that, given exactly the same materials (poly-cotton) and construction, a clincher tire has a lower rolling resistance than a tubular.
Results so far:
Vittoria Crono EVO tubular is the best rolling product tested: its thin casing, high tpi count (it’s higher than 290 TPI), casing material (special poly-cotton), thin tread and overall lightweight (170g), make it an awesome performerVittoria Open Corsa Evo absorbs 6.5 percent less power than Corsa EVO tubular (but tubulars are safer in case of puncture and can be glued to oh-so-light wheels that will make them faster on any uphill)The TPI theory is confirmed: looking at nylon tires, Vittoria Diamante Pro (220 TPI) rolls better than Rubino Pro (120 TPI)A lighter tire will roll better than a heavier one, given the same casing material: Vittoria Diamante Pro Light and Diamante Pro are both 220 TPI.Lightweight isn’t everything, though: the lightest in our test, (135g.) is by far the worst roller, probably because of its specific construction and of the nylon casing, as opposed to poly-cotton or poly-cotton+Kevlar.
Here’s alink where you’ll find some interesting data on the whole picture: aerodynamics, rolling resistance, weight.
I have some power-absorption data I just got from an independent Finnish laboratory that might raise some interest: an Open Corsa EVO CX 23-622 (inflated at 8 bar, with a mass of 50kg on, simulating a rear wheel) rolling at 40kph requires 46 watts of a cyclist’s power… Rubino Pro (120 TPI nylon tire, same conditions) absorbs 59 W… and Hutchinson Fusion 64 W.
Alberto De Gioannini
Bicycle tire engineer
VeloNews technical writer Lennard Zinn is a frame builder (www.zinncycles.com),a former U.S. national team rider and author of numerous books on bikesand bike maintenance including the pair of successful maintenance guides”Zinnand the Art of Mountain Bike Maintenance” and “Zinnand the Art of Road Bike Maintenance” as well as “Zinn’sCycling Primer: Maintenance Tips and Skill Building for Cyclists.”
Zinn’s VeloNews.com column is devoted to addressing readers’ technicalquestions about bikes, their care and feeding and how we as riders canuse them as comfortably and efficiently as possible. Readers can send brieftechnical questions directly to Zinn (firstname.lastname@example.org)Zinn’s column appears each Tuesday here on VeloNews.com.