One of the 5 Fastest Tires in the World

Recently, the German magazine TOUR published a table showing the ‘five fastest tires in the world.’ We are excited to see our Compass Bon Jon Pass 700C x 35 mm tires on this list, in the company of the fastest racing tires. A 35 mm-wide tire on a list that otherwise includes only tires between 23 and 26 mm wide! That by itself is already cause for celebration. It means that our casings really are among the very fastest in the world.

And since all our tires use the same casings and construction, TOUR’s results apply not just to the Bon Jon Pass, but to all Compass tires. I was surprised that they tested the Standard casing. I would love for them to test the Extralight, which we know from our own experience to be even faster.

What is interesting is that the Compass tire scored superbly on smooth asphalt (light gray bars), but a little less well on rough asphalt (dark bars). This doesn’t match our experience, where wider tires provide advantages especially on rough roads. The reason is simple: TOUR tested without a rider on the bike. This measures the hysteretic losses in the tire, but it neglects the (much more important) suspension losses that occur as the rider’s body and bike vibrate. (Click here to learn more about suspension losses.)

This means that TOUR’s testing overlooks one of the main advantages of wide tires: their superior comfort, which also makes them faster. In other words, with a rider on the bike, especially on rough asphalt, the Compass tire probably is even faster than it appears in TOUR’s testing.

We are proud that the Compass Bon Jon Pass scored so well, especially since it is intended as an all-round tire, not an all-out racing tire. The Bon Jon Pass is suitable for gravel racing and has 3 mm-thick tread for many miles on the road. Compare that to the Vittoria with its 0.8 mm-thick tread, which is intended only for time trials, and even then, it’ll wear out quickly.

The excellent performance of the Compass tire shows once again why wide tires have revolutionized cycling: You wouldn’t want to ride the other tires on TOUR’s list on anything but the smoothest, cleanest roads for fear of flats and premature wear. And yet with wider tires, we can ride some of the world’s fastest tires on the backroads where cycling is at its most beautiful.

Further reading:

Correction (1/31/2018): The testing method of TOUR’s test had been reported as a steel drum. Also, TOUR tested the Standard casing, not the Extralight, as we previously assumed.

About Jan Heine, Editor, Bicycle Quarterly

Spirited rides that zig-zag across mountain ranges. Bicycle Quarterly magazine and its sister company, Compass Cycles, that turns our research into the high-performance components we need for our adventures.
This entry was posted in Tires. Bookmark the permalink.

51 Responses to One of the 5 Fastest Tires in the World

  1. Mike Stead says:

    Hi,

    I’m testing the Barlow Pass for Road.cc – what’s th me difference between that and the Bon Jon?

    Thanks

    Mike

  2. Harald says:

    It’s good to see your products included in these tests now.

    If they’d measured suspension losses, you may not have done that well after all — 6 bar/87 psi on a 35mm tire must make for an awful ride. The 25mm tire at 7 bar/100 psi seems much more sensible.

    Also, do you know if they tested the Extralight or the regular casing version of the Bon Jon Pass?

    • Even at 6 bar (87 psi), a 35 mm tire will absorb vibrations much better than a 23 mm tires at 7 bar (100 psi). On real roads, you can run it at much lower pressures without giving up any performance – and not risk pinch flats, as you would on the narrower tire.

      TOUR tested the Standard casing. The Extralight rolls considerably faster, so it should score even better.

  3. Jens says:

    Hi Jan,
    no doubt that the Compass performed well. It’s not an Extralight, but Standard Bon Jon.
    Please correct the explanations to our test protocol. We’re not using steel drums.
    http://www.tour-magazin.de/service/so_testet_tour/rennradreifen-rollwiderstand-test/a37337.html

    • Thank you, Jens. I corrected the errors in the post.

    • DaveS says:

      Is it possible to get a transcription of the video in the link above? I don’t speak/understand German. The text in the link above can be translated by google, but I don’t know what was said in the video.

    • Joe Wein says:

      The following is an English translation of the German test protocol video:

      “(subtitle:) Rolling Resistance Test Setup

      (voice-over:) The improved rolling resistance measurement by “Tour” uses an oscillator on a rough, level road surface in the lab.

      For this new test two wheels with the tyres to be tested are attached to an axle and loaded with 110 kg. 15 kg of that mass are fitted excentrically below the axle.

      The apparatus uses the principle of a pendulum: The lower the rolling resistance, the longer the machine will swing.

      We measure the rolling distance until the stop and determine from that the rolling resistance in W for a system weight of 85 kg including bicycle at 30 km/h.

      Except for the low speed the measurement is very close to reality. Reproducibility is very good.”

  4. Johan Brox says:

    So many questions. I do not have the magazine in front of me and can only go by what they’ve published online, but I’m wondering about the following:

    – How did they determine the given pressures for each tire? Did they do several runs with different pressures for each tire? 87 PSI sounds extremely high for Bon Jon @ 85 kg, which would diminish performance especially on the “rough” drum.

    – Did they test the Bon Jon as tubeless or with inner tubes?

    – Have they tested Vittoria’s claims that the graphene layer mitigates the short service life one would expect from the razor-thin tread?

    – How is the “rough” drum profiled – like a compact gravel road or rougher? How do they take into account that the steel drum cannot simulate a loose, ie. shifting surface, which I would expect to also affect rolling resistance?

    – How is power transfer handled – is there an actual person riding on a bike or is the tire weighted against the drum and powered by a motor?

    Anyway, congratulations on a remarkable result!

    • Johan Brox says:

      Sorry, just saw the link to their test rig. That answers some questions, but not all. I think they lose quite a lot of insight by testing the wheels alone and not as a system together with rider and bike.

    • Some of the answers to your questions are in the link – TOUR didn’t use a rider, just a weighted pendulum. I doubt they ran the tires at multiple pressures, probably just used the max. pressure for each tire. As to the Vittoria, modern tread compounds do miracles combining good traction with longevity, when in the past (say 20 years ago), you had to choose one or the other. But any rubber that grips also rubs off, and with 0.8 mm tread on the Vittorias, there isn’t much to wear until they get so thin that they will flat too easily. In fact, I usually replace my tires before they get to 0.8 mm tread thickness. I did set my fastest PBP time on a set of well-worn tires that had only a relatively thin tread left… Maybe we should send a set of worn-out tires to TOUR for the next test! 😉

  5. Doug L. says:

    Would like to see test results if all these tires were ridden on real roads in East TN. Did notice the compass tires were inflated to 6 bar! It would appear that testing on a metal drum has its advantages for competing manufacturers.
    All else aside, I have been riding 32 and 38 mm compass tires for over a year. With my own A-B testing found them to be faster than any of the others I have used, that includes Continental, Pana racer, Specialized and Vittoria. As stated previously the drum test does not measure enjoyable rides off the beaten path.

  6. Peter says:

    I routinely run the JBP EL tyres tubeless at 3 (rear) and 2.5 (front) bar without problems; I’m 60kg though, heavier riders may want a little more air. The one problem with these tyres is that I can’t get the sidewalls completely airtight; esp. on those vertical lines (near the “E” on the rim at the top-left side of the first photo).

    Otherwise, wonderful tyres. For smooth asphalt, gnarly British roads and German gravel paths. Fairly durable as well, after about 6000km the rear became more prone to punctures and I replaced it just prior to riding LEL.

    And speaking of tubeless, is there a roadmap when the other tyres (Loup Loup!) will be available as TL? And/or a statement when a size is not likely to be remade as TL at all?

    • Sealing: Make sure you shake the sealant well before injecting it. In theory, a sealant that can seal significant cuts in the tire should be able to seal a hairline that appears where the rubber of the sidewall has been pushed into the mold line. However, we are also working on eliminating as many of those mold lines as possible to get a more uniform rubber covering of the sidewalls. (The alternative is to make the rubber coating thicker, which would make the tires slower.)

      Tubeless: Most of our wider tires are already tubeless-compatible. The Loup Loup Pass 650B x 38 is the only one that isn’t yet. We’ll see whether we can make a case for making a new mold that makes this tubeless-compatible.

  7. Tom A. says:

    A couple of comments:
    – I think it’s great that they included your tires.
    – You should point out that on the smooth surface, the testing being done at 6 bar actually helped the Bon Jon roll faster than it would at lower (more typical/appropriate) pressures. For example, I tested the Bon Jon at 65 psi (~4.5bar) and it rolled signficantly slower than a Conti GP4K at 7bar (.0058 vs .0035 Crr). In other words, “you can’t have your cake and eat it too” on the pressure 😉
    – My testing has also implied that your EL casing is only “worth” ~1-1.5W per tire between 20-30kph, as compared to the standard casing. So, apparently it’s not quite the dramatic improvement you’ve been assuming. For that reason, coupled with my experience with the EL casing being more susceptible to sidewall cuts and sealant “bleed-through”, when people ask me I steer them towards your standard casing models.

    • Tom, thank you for your comments. I haven’t seen your testing, but assuming that you tested without a rider, it well known that in that scenario, lower pressure makes a tire slower. So I am not surprised that our tires at almost half the pressure of a Conti GP4K will show worse results. Even TOUR tested our tires at a lower pressure than the others, but the difference was smaller, hence our tire performed better.

      However, on real roads, the lower pressure reduces the suspension losses, and thus supple tires roll as fast at lower pressures as they do at high pressure even on very smooth roads – as we have shown conclusively in a variety of tests. On rough roads, lower pressures are faster.

      The same applies to the Extralight casing: The main benefit is reduced vibration and hence lower suspension losses. Your tests neglect that factor, which is why you don’t see a big benefit. Ride them on a real road, and you’ll notice the difference immediately.

      • Tom Anhalt says:

        Actually, my testing IS with a rider on board (I’m the “power source”, after all ;-)…but, I think you are unfairly dismissing out of hand any testing that doesn’t include the rider. After all, I know you’ve seen the work that Josh Poertner has done at Silca, which confirmed my own previous observations: https://silca.cc/blogs/journal/part-4b-rolling-resistance-and-impedance
        It’s pretty clear that if one makes certain to not run overly-high pressures (i.e. below the “breakpoint pressure” for a given condition), then the Crr performance of a given tire is dominated by the tire internal loss properties. And, since those losses in that regime (below “breakpoint”) tend to be proportional to input energy (a function of roughness AND speed) a ranking of tire hysteresis losses on smooth surfaces will stay the same, just at proportionally higher levels.

        In other words, it’s perfectly legitimate to judge the “speed” potential of various tires using smooth surface testing and applying a roughness factor. After all, what you call “suspension losses” only start dominating under cases where the pressures are too high for the conditions. Up until then, the tire itself is acting as the majority of the “suspension”, and those losses are the same hysteretic losses occuring at the contact patch.

        I like to think of it this way…if you’re running pressures to high to avoid “suspension losses”, then you’re running too high of pressure for the conditions. If you can’t lower pressure any further to avoid bottoming the tire, then you need to run larger tires.

        Course “roughness”, and speed (i.e. energy input) guide pressure selection: pressure selection guides tire size 🙂

        BTW, I HAVE ridden both Extra LIght and your regular casing tires on many roads. There really is NOT a palpable difference. The roller testing shows that they don’t have dramatically different hysteretic losses (and remember, if the pressure is below “breakpoint” for the conditions, body “suspension” losses don’t come into play)

      • Hi Tom, thank you for clarifying. I now remember – you test by riding a hilly course on the road. What I worry about with that testing is the noise in the data. Even on a track, with no wind, and with a rider who keeps the position constant, there is some noise in the data – about 1-2%. So if our testing shows a tire to be 2% faster, we don’t know whether it really is faster, or whether it was just noise. Fortunately, the tires we tested showed much bigger difference. But with less-controlled conditions, the noise will increase exponentially.

        My questions about your testing:
        – How repeatable are the runs? If you test the same setup 3 times, how much variation do you get?
        – How do you adjust for temperature changes? It’s well-known that temperature has a very significant effect on rolling resistance.
        – How to you make sure there is no wind? In our tests, we only need about 3 minutes of totally calm conditions, and we throw out any test runs that have even the slightest breeze. It seems that with your much-longer test runs, you’ll always have a little wind. And that will affect your test runs more than anything else. (And no, wind doesn’t cancel out on a circular course, because a cross-wind actually is the worst, so if the wind blows steady from one direction, you go much slower than if it’s calm.)

        If I understand your testing correctly, it needs a huge number of repeats to get statistically significant results. If you test one tire 100 times, and another tire also 100 times, selecting each tire at random during the test period, then you might get decent results, because all the noise averages out. Even then, I’d want to see a statistical analysis to make sure. There is a reason why tests of new medication use thousands of participants.

        The alternative is to control your conditions as closely as possible. The best setup there is the laboratory, where temperature and wind don’t ever change. The problem is replicating real-world conditions. That is why we test in the real world, but under carefully controlled conditions. This is a huge amount of work – more than once did we show up at the test track at 5 a.m. (the calmest period of the day is at sunrise in Seattle), only to have a slight wind come up. It meant we had to go home and try again another day.

        And finally, suspension losses always occur, and they are important even on very smooth roads. Witness that we found both a RockShox fork and a fork made from flexible Kaisei ‘Toei Special’ steel blades to be faster (statistically significant!) than a super-stiff fork from a hybrid bike, because it reduced vibrations on brand-new, smooth asphalt.

      • Tom Anhalt says:

        Actually, your memory isn’t correct. The majority of my testing is on small diameter aluminum rollers (Kreitler) using a bicycle with the front in a stationary fork mount. The description is here: http://bikeblather.blogspot.com/2013/02/tire-crr-testing-on-rollers-chartand.html

        I HAVE confirmed those results at times (see the Silca charts referenced earlier) using “Virtual Elevation” (AKA “Chung”) methods. That’s where the observation of the “breakpoint” pressure was obtained. Up until that pressure, the on-road results were directly proportional to the smooth roller results (i.e. “roughness factor” x smooth results). It’s not until the “breakpoint pressure” is reached that the energy losses at the contact points of the rider come into play.

        “– How repeatable are the runs? If you test the same setup 3 times, how much variation do you get?”:
        Test runs for the same tire invariably fall within .0001 Crr of each other. In fact, I have a particular tire that I use ONLY as a control during the runs just to make sure the setup is correct.

        “– How do you adjust for temperature changes? It’s well-known that temperature has a very significant effect on rolling resistance.”
        That’s covered in the link above, but I made measurements that showed a typical tire varies by ~1.4% of Crr per deg C. My results are all “normalized” to 20C.

        “– How to you make sure there is no wind?”
        For the roller testing, that’s a non-issue. For any VE field testing, that requires careful course and time selection. Although I mostly use VE for measuring CdA differences, I think I’ve shown that I’ve got a good handle on what it takes to get reliable (and high resolution) results. An example of how sensitive VE testing can be is here: http://bikeblather.blogspot.com/2013/08/aero-field-testing-using-chung-method.html

        You know, that “Tom Compton Challenge” mentioned in the link just above would be a good idea for you to try out with your field testing method. In other words, make a KNOWN change (e.g. empty bottle on board vs. lead shot filled bottles) that should affect the Crr value and see if your methodology can detect it. That would be the corollary to the different sized spheres used in the drag coefficient sensitivity study.

      • A small-diameter roller has so many issues that don’t occur in the real world that I’d be very careful before making any assumptions of real-world performance.

        Regarding the known changes, you get into circuitous reasoning when you say “should affect…” In fact, I doubt that a little extra weight will affect tire resistance significantly, otherwise, we all would speed up as we empty our water bottles during a ride.

        The proper scientific way is to
        a) validate your model, by showing that the changes you find replicate what happens on the road. In our case, we ride around a track made of asphalt, so that step isn’t necessary. In your case, it must be shown that differences you observe on the roller are differences that occur on real roads.
        b) repeat your measurements for the same setup, to get a handle on your measurement accuracy. We both have done that and found that both our methods have little noise.
        c) replicate your results with a different methodology. We did that by testing the same tires in a rolldown test and on the track with a Power Meter. The results were the same – the fastest tires always were fastest, the slowest were slowest, and the differences between tires remained the same.

        Based on all these, I have a lot of confidence that our tests represent real-world riding conditions accurately and yet have reduced extraneous variables (“noise”) enough to provide meaningful results.

    • HaloTupolev says:

      >”In other words, “you can’t have your cake and eat it too” on the pressure”

      That test was a dramatic outlier, though. For instance, the Snoqualmies scored *much* better despite being pumped even squishier. While it’s certainly true that erring low can cost a few watts in hysteresis, that doesn’t seem to be the interesting story behind whatever happened with your Bon Jon result.

      • When testing anything, it’s absolutely crucial to repeat each test run several times – we do at least three repeats – and then do a statistical analysis of the results to make sure you are seeing actual differences between tires/pressures/etc., rather than just noise in the data.

      • Tom Anhalt says:

        Yes, the Snoqualmies in my tests scored better…but, they are also wider, remember 😉

        The Bon Jon result for me is quite curious…and has me wondering about lot to lot variation of the tires.

      • There isn’t much variation between Compass tires. I think these erroneous results point to problems with your testing (see comment above). We’ve tested all widths of Compass tires under carefully controlled conditions, and they all performed the same (within the 2-3% variability that we get with all our testing). So we can say with confidence that there isn’t more than 2-3% variability among the tires, and that the different sizes also have the same resistance (rolling resistance, suspension losses, air resistance) at our test speed of 29.5 km/h (18.3 mph).

      • Tom Anhalt says:

        Right…but have you tested them over time and over multiple lots? Do the current lot of BJPs roll the same as earlier lots? Based on the newer style packaging, I’m pretty sure the one I tested is newer.

      • We did test tires from multiple lots and in multiple sizes. I also know how these tires are made. There is some variation in how much rubber covers the casing, but it’s not so huge that it’ll affect the performance to a great degree.

  8. codadelgruppo says:

    Very interesting that Tour uses a rolling pendulum to test tires on a simulated pavement surface. That would appear to reduce some of the potential inaccuracies as compared to using a rotating drum. However, doesn’t simulate suspension losses because there is no “body tissue” to dissipate kinetic energy in the form of vibration. That might be possible by replacing the iron weights with ballistic gel, or some other surrogate for body tissue that absorbs and dampens vibration. An even more realistic test may require require a bumpier surface, something that causes more “bounce” in the simulated rider, as happens on real roads.

    For years I have been participating in a spirited mixed-surface group ride. Many of us have ridden together for a long time, so everyone’s pace and ability is well-sorted. After switching to Compass EL tires, several of my riding pals asked why I was speeding up every time we hit the dirt. On further reflection, it wasn’t that I was speeding up, I just wasn’t slowing down as much as the riders on stiffer tires.

    Eric Nichols

    • HaloTupolev says:

      >”However, doesn’t simulate suspension losses because there is no “body tissue” to dissipate kinetic energy in the form of vibration.”

      I’m not sure this is true. While a squishy material with lots of damping provides a good visualization for where the energy is going, I don’t see how a rigid undamped metal object is going to be good at returning road buzz energy into forward motion. I wouldn’t be surprised if the exact behavior differ considerably compared with a human rider, but I’d expect suspension effects to still *exist.*

      It would be interesting to see wide wattage-versus-PSI distributions, to try and draw a picture of just what is going on.

      • I don’t see how a rigid undamped metal object is going to be good at returning road buzz energy into forward motion.

        It’s easier to visualize when you ask where the energy of the suspension losses is dissipated. The weights don’t dissipate much energy. My hypothesis is that it gets returned when the tire pushes off the bump.
        Energy return as tire pushes off bup

  9. Scott Bontz says:

    I (a Bon Jon rider) tried but could not tell from the German text how many tires were tested to come up with the 5 fastest in the world. Nor can I find explanation in your report. (Which also doesn’t explain the metric of the bar chart. Google translation: rolling resistance, in watts.) I do see in Tour’s story “85 kg.” Is this the weight on the wheel? (Seems a lot.) How would such a test weight affect hysteretic losses than would a rider’s weight? Can you tell how they applied the weight?

    • You raise a good point – if they tested five tires total, and we were fourth, that wouldn’t be all that impressive. However, TOUR probably test more tires than anybody in the world – it’s not a small number. Also, the other tires on the list are generally accepted to be the fastest tires in any tests I have seen. So this really means that in this type of testing, Compass tires are among the very best-performing tires you can find anywhere.

      • Gert says:

        As I recall the test results of the Bon Jon tires is from a test of gravel bike tires from last year, where the tires were tested at a lower pressure. In the text it was noted that the Bon Jons were faster at 6 bar. That result has then put the Bon Jons into the top five.
        So without rereading a lot of back issues of Tour to confirm it. I am almost certain that it is top five all time. (With that test setup)

        I thought You read Tour regurlarly Jan, so I did not point out the test at the time

  10. thebvo says:

    Are the Extralight tires “considerably faster?”
    Has Bicycle Quarterly published a roll-down test of the Compass tires? Standard vs XLight?
    And Compass vs other wide tires… I can’t recall, but I don’t remember seeing the results.
    Those roll down tests, although time intensive, are a truly great way of rating performance because they use time as its quantifier. It makes it easier, as a consumer, to translate the results.

    • Bicycle Quarterly isn’t the marketing arm of Compass, so our goal isn’t to show the superiority of Compass tires. What we’ve done recently is basic research by testing the same tires in all widths, to see whether wide tires roll slower (they don’t).
      We did test the Extralight and Standard casings internally, and the Extralights were faster. The standard casings (for which the results have been published) also roll very well, but riding the two back-to-back, you’ll notice a significant difference in speed and comfort.

      We’ve switched to the track tests with power meters, because a) the conditions are more closely controlled as the speed is truly constant and b) our test track for the rolldown tests no longer exists, and we haven’t been able to find a new one that meets the requirements of a steep start to get up to speed quickly, followed by a long, slightly downhill rollout for constant speed during the actual timed distance.

  11. Nik says:

    I wonder why they tested the 35mm tire at a pressue of 6 bar = 87 psi. I’m 90kg/190lbs and I use these tires at 60 or 70 psi, and even lower when I briefly had them working without tubes.

    Is Tour magazine mistaking the maximum possible pressure for the recommended pressure ?

    • Their test doesn’t include a rider, so a lower pressure will make the tire slower (more tire deformation). So they test at the highest pressure the tire can support. On real roads, we all know that lower pressures don’t slow down the bike, because vibrations are reduced as well, but TOUR‘s test cannot capture that.

  12. Noel Hoffmann says:

    Congratulations, Jan. Despite the flawed testing, it must feel good to receive a measure of justification from the “mainstream” cycling press.

  13. calcagnolibero says:

    Steel drums tests do not match reality and should be avoided once for all. I trust more the BQ test.

    • The drum tests (and TOUR‘s conceptually similar test) do test the hysteretic losses, so they give us information about the quality of the casing. When you take them for what they are, they can be quite useful.

      However, you cannot deduct from TOUR‘s tests that the Bon Jon Pass tires roll faster at 6 bar than at lower pressures, because they don’t measure the reduction in suspension losses at lower pressures. That means also that generally, wide tires perform (even) better on the road than in these rider-less tests.

  14. Jacob Musha says:

    I’m sure this has been covered before but I can’t find the information… Have you tested tubulars against clinchers? Is a comparable tubular (similar casing and width) faster or slower than a clincher?

    These results make a convincing argument for me to get rid of my last remaining “narrow” tire road racing bike (27mm tubulars) and use a bike with wider Compass clinchers instead!

  15. Conrad says:

    Has anyone else noticed that when riding rollers (drums, right?) there is a huge difference in resistance between say 75 and 100 psi. Whereas on real roads, I can’t really say that a Rat Trap at 30 psi is slower or faster than a Bon Jon at 45 psi or a Vittoria Corsa at 90 psi. Unless its a rough road, in which case the wider tire is definitely faster. So I think the roll down tests have much more real world applicability.

    • Part of the reason why the resistance on roller trainers is so large is that the drums are so small. For rolling resistance testing of tires, much larger drums are used that don’t dig as deep into the tires. Even so, your general observation is correct: On a drum, higher pressure is faster. On the road, it isn’t.

    • Tom Anhalt says:

      Yes, I’ve definitely noticed it. In fact, that “amplification” is one of the main reasons small diameter drum testing can be so useful. The differences are magnified and it’s easier to “tease them out” using a standard power meter. Luckily, there are well established equations from the tire industry that then allow those differences to be translated into a flat surface equivalent difference.

      This really is one case where what some might think there’s a “bug”, it’s actually a feature! 🙂

      • Luckily, there are well established equations from the tire industry that then allow those differences to be translated into a flat surface equivalent difference.

        I am not aware of any testing that shows how this relates to real-road riding. The tire industry, for the most part, still believes that higher pressures make tires roll faster. When you test on small-diameter rollers, you are far removed from the real world. All kinds of factors will affect your results that don’t play a role on real roads. Obviously, tire pressure will determine your results more than anything else, as it determines how far the small roller digs into your tires. Another issue are the suspension losses that you cannot simulate on a smooth steel roller. If you could do your testing on very large rollers (as large as possible while still fitting underneath the bike) with a textured surface, you might get better results. The steel rollers used for testing bike tires in the best labs are about 6 feet (1.8 m) in diameter to eliminate the effects of the roller digging into the tire.

        At least, TOUR‘s test uses real tires on a real, flat, paved surface. There are still some questions about whether something funny happens when the tire reverses direction, but otherwise, it seems like a good way to test the hysteretic losses (but not the suspension losses).

  16. Philip says:

    I was thinking about the effects of lower diameter bicycle wheels and it seems to me that for most traditional bicycles, this would lower the bicycle+rider`s clearance and thus reduce air drag. In the article `Aerodynamics of Real-World Bicycles` you mention “simply raising your stem by 20 mm (3/4 in) increases the drag by 5%“.

    One of the bikes I am interested in buying features 20 inch wheels while many other wheels are 26 inch, this comes to a 6 inch difference or 152.4 mm. Isn`t this an advantage for shorter wheels to taller wheels? Are there any overriding factors that would make a difference here?

    • Wheel size doesn’t affect where the rider sits on the bike – the pedals must still clear the ground! What matters is the frontal area, and since the wheels align with the tubes of the bike, any benefit from smaller wheels will be small.

      Where there would be a greater benefit is that smaller wheels allow closer drafting…

  17. viskoworld says:

    About the frame flex myth… going mainstream – https://www.youtube.com/watch?v=BH_AL4rxrp8

Comments are closed.