Login to VeloClub|Not a member?  Sign up now.
  • George Darroch

    Tom Anhalt’s data is interesting, and suggests (from my reading) that the sweet spot for 23c tyres can be found between 80-100psi, with reasonable adjustments for the rider and conditions.

    I’m not sold on the idea of oversize (25-28) tyres for road. The losses to aero outweigh the gains, especially as most wheels aren’t optimised for wider widths.

    • George Darroch

      And thanks to Matt, Josh, Damon, and others for such an interesting and comprehensive article.

    • Superpilot

      But it’s just not the width though, it is also the profile in its interface with the rim.

      I read somewhere some of the latest aero rims are less efficient with a 23mm as they are designed for 25mm in order to achieve a release of the air from the tyre across the rims profile. In that instance, a 23mm would be less aero efficient. Depending upon your riding level, a 25mm for a sportive rider is going to provide quite a bit more comfort that will matter more than a slight aero disadvantage. But you may be a superior rider than that. We also know a lot of pro teams are riding 25mm, with the latest aero wheels, so that might also be an indicator they trut in the aero of that package.

      28mm+ is more multi purpose riding, gravel or cobbles where absolute aero is less important than functionality over various terrain.

    • jycouput

      Might be true in some specific riding conditions. But something has to be said: What is your objective? Riding faster or riding further? Is it the fatigue accumulated with the distance that slows you down, or is it your physiological limitations?
      Ask yourself this question first, then the whole equation with regards to tire sizes and inflation becomes less of a dilemma

    • Steve

      Further to George’s concern stated just above, all of this discussion about rolling resistance, wider tires and lower pressures must be considered along with the effects those changes can have on aero performance and handling. This article gives little attention to them.

      Indeed, if you have a rim with any type of aero profile (say a U-shaped, 40mm or deeper all around carbon wheel), mounting and inflating a tire on the rim that measures wider than the rim will reduce your aero performance at aero speeds (20mph/32kph and up) by a far greater amount than the improvement in rolling resistance performance. (Matt, btw, per Josh’s research, for best aero performance the rim width should be at least 5% greater than the tire width, not the tire width no more 5% greater than the rim width, as you wrote above.)

      For example, if you’ve got a 17C wheelset with a 25mm or so brake track width (not max width for you toroid wheel owners), you really don’t want anything wider than a 23C tire, most which will run 24.5mm or so when inflated to 90-100psi. If you ‘ve got a 17C rim with a 27mm brake track width or a 19C rim with a 28 or 29mm brake track width, 25C tires can work within that 5% parameter. To complicate matters further, some tires of the same size (e.g. 23C Conti GP4K) measure much wider than others (e.g. 23C Zipp Tangente) inflated on the same rim to the same pressure.

      And rounding a turn at normal speeds on a tire inflated for max comfort, i.e. low, can get squishy or worse. So you have to prioritize or at least evaluate rolling resistance, aero performance and handling together and make your tire width and inflation level decisions with all of those in mind.

      • DT Swiss did a study a few months ago into the trade-off between rolling resistance and aero penalty of wider tyres using their new ERC1100 28mm wide aero wheelset http://www.roadrevolution18.dtswiss.com/endurance/

        Here’s the short version:
        Above 35km/h the aero benefit made a 25mm tyre faster than a 28mm tyre.
        Below 35km/h the rolling resistance make a 25mm tyre slower than a 28mm tyre.

        I would suggest that for an older spec 25mm wide aero wheelset the findings would be similar between 23mm and 25mm tyres. There is little if any speed to lose by going wider for those of us who spend more time on this side of 35km/h. For the classics, and those of us who ride rough bitumen, the equation is even more in favour of going wide.

        The future is wider wheels and tyres.

  • Simon Wile

    Rattling around on 120psi 23s must be awful. Aero gains from -2mm of rubber? What about the 2cm around your neck/arms/waist/extra fork spacer? Give me a set of gumwall 25/28s any day. Smooth as fast and less fatiguing.

    • Gavin Adkins

      When I was a teenager I would race on 18c tyres at 140psi. 20 years later, 25c tyres at 80psi still feels like half-inflated tractor tyres. Old habits die hard.

      • Interesting. My experience with 18c tyres at the same kind of pressure 20 years ago has had the opposite effect. To my mind, this setup represents the very worst-case scenario for a road cyclist. Never again!

        • Gavin Adkins

          Oh I ride around on 25c tyres at 80psi now, but mainly because I can’t be bothered pumping my tyres up!

      • Simon Wile

        Interesting. Still of the “rough = fast” mindset perhaps? “In contrast, most accept (and even revel in) suspension losses when it takes the form of the buzz and rumble of firm tyres, choosing to interpret this kind of feedback as a measure of speed.” That said the first time I rode my roubaix on 28s after a 6 weeks off it I was constantly looking at the rear tyre wondering if I had a flat it was so soft/smooth.

        • I think that has to do with the increased frequency of vibration transferred through the tires. I think one of the most important things that sensible tire pressures do is increase people’s confidence in cornering

        • Chuck6421

          It’s been a while since I listened to that interview, but if I correctly recall Jan also noted the added stress on the body from that vibration. It takes energy to absorb and react to it, subtle as it may be, and that was a non-trivial factor to consider.

    • Nathan

      I have always ridden 23c at around 120psi – probably because I am used to higher pressures on the track, and because when I first threw my leg over a roadie in the 90’s, 23 was almost considered a fat tyre – lot’s of riders were on much narrower rubber. It comes down to personal preference…. for most riders below the elite level, the physical gains of ‘correct’ tyre size and pressure is less then the psychological gain of ‘feeling’ comfortable/fast. I feel great riding 120psi over rough gravel roads with corregations….. but others will not.

  • Richard Watson

    Great article. Tyre pressure is clearly very important. But how accurate are the pumps we use? How can we accurately set pressure? I’d like to see an article on this topic!

    • Michael_Fink

      The whole point of the article is that there is no single ideal tyre pressure. The best pressure for your tyres is contingent upon a whole slew of factors only relevant to you.

      So you should determine your optimum pressure by experimentation, not by following some recommended pressure. Let’s say you determine the optimum pressure for your tyres is 80psi, it doesn’t matter that your mate’s pump’s gauge would record that as 60psi. So long as you use the same pump each time, and its gauge is consistent, it doesn’t matter whether it is accurately recording the true pressure.

      • Simon Wile

        Whoosh is the sound of a joke going overhead.

      • I think you should let some air out.

      • Richard Watson

        Agreed. But your caveat “So long as you use the same pump each time, and its gauge is consistent” captures the nature of my problem. I do use different pumps at times, and wonder about their consistency over time.

    • slowK

      Good point. For years, I thought my clinchers were losing about 8-10psi a week. But then I reconnected my pump immediately after pumping the tires up, and found that the tire pressure was constantly around 6psi lower than what I had inflated the tire to. I guess that the brief “pshhht” of air as I disconnected the pump head from the valve was largely responsible for this drop (for me anyway, there’s a much softer/briefer release of air when I engage the valve with the pump head). So I’d probably been running lower pressures than I thought.

      • Michael_Fink

        Running the risk that I’m missing another joke can I try to clear up this misconception…

        Your clinchers were not losing any appreciable amount of pressure when you disconnected the pump. The tyres only lost 6psi when you reconnected your pump. At that point some of the air that had previously filled your tyre flowed out of the tyre and into the chamber of the pump, until the pressure in the tyre and the pump itself was the same.

        • slowK

          OK – thank you. I did not know that. I would have thought there would be a one way valve type of configuration as close as possible to the valve end (to minimise “deadspace” of the pump head/hose/chamber). So the “pshhht” comes from the pump equalising with atmospheric pressure?

          • Michael_Fink

            That would be my assumption. If you use a track or floor pump then there’s a decent amount of air at high pressure when you disconnect it. I’d presume the ‘pshhht’ sound would be shorter/quieter with a mini pump, where there’s less air involved.

            • Alan Clark

              Even further to your point, there is more air in the closed system of a pump that has a gauge at the top vs. one with a gauge at the bottom. There’s a tube that that runs from the bottom of the pump to the gauge at the top that’s a part of that system.

          • Bmstar77

            Some pumps, like Lezyne, have a little pressure release button to let the pressure out of the pump line before you disconnect the pump from the valve. Only works on presta valves of course.

            • BobB59

              I have two Lezyne pumps, one with and one without the bleed valve. When I bleed the pump and then unscrew the pump, there is no Pfft sound. This tells me the sound is the release of air from the hose, not the tire.

        • Pete

          Unless when you remove the pump, there is a moment where there is no seal, but you are still depressing the valve. Allowing air out from the tube. (I’m not sure how frequently this occurs.)

        • Alan Doughty

          Heisenberg’s uncertainty principle, as applied to tyre pressure. The act of measuring changes the measurement.

    • Simon Wile

      I got you Richard ;)

    • George Darroch

      Silca make some very good pumps. If you have $650 you can buy air-compression perfection.

    • Very much with you on this one. I used a Schwalbe digital gauge for a while (cheap to buy) but the thing stopped working again and again. Anyway, would love to read a test of floor pumps on how accurate their readings are. I dont believe though that any of us would feel a difference of 5psi. I constantly lower the pressure on my wife’s bike (she is one of the 100psi crowd) and she doesnt feel a thing ;-)

      • Alan Clark

        One thing you can do is to make sure your pump gauge has the correct pressure range. Most average Made In Taiwan pumps that you can buy at your LBS have the best accuracy in the middle third of their total range. If your gauge is 0-160 psi, it’s most accurate at between ~60-120 psi. 3% is the nominal accuracy for most LBS-quality pumps within this range, but your results may vary.

  • RayG

    Sometimes you can take something fun and over-think it.

  • Michael_Fink

    Old habits die hard apparently:

    Conditions: Any kind of water on the road decreases the grip of the tyres, so it’s prudent to lower tyre pressure ~10psi. A passing shower or shaded roads that haven’t dried after overnight rain can complicate tyre pressure choice.

    Everything else in this article, and what I remember of last year’s podcast on this topic, suggests that if you feel your bike handles better in the wet with ~10psi less in the tyres than you usually run then you’re simply over-inflating your tyres in the dry.

    Doesn’t seem to be anything to be gained by pumping up your tyres by ~10psi than their optimal inflation pressure, just because it’s a dry day.

    • Cyco

      They sure do. I stopped lowering my wet pressures 15ish years ago after talking to a Michelin engineer.

      His data showed that you couldn’t aquaplane a road tyre under 140km/h – and told me if I was having issues he’d build me some tyres!

      He said that any extra slip in the wet was caused by surface contamination, and pressure changes would not help. It took a few more years before I started to drop my dry pressure, though.

  • This might be my inability, but I can’t see that Figure 1 says anything at all about rim width. In fact, that looks like the only variable not part of the graphs. All tyres are on a 17c rim, aren’t they?

    • Sorry, my fault. That was my figure for the influence of tyre pressure on tyre height and width. I’ve replaced it with the correct figure showing the influence of rim width, which by the way, is more profound than tyre pressure for any given tyre size.

  • Mathias Holtedahl Thorp

    Someone should make a calculator/app where you can insert data for all variables, then get the suggested psi for your ride. Anyone? :)

  • velocite

    I’m not impressed by the idea that it’s up to every individual to establish what’s best for him/her. To minimize rolling resistance you need a controlled test, and there aren’t that many variables. I can imagine a helpful table showing two tyres, supple and not, in two sizes, say 23 and 28, 2 road surfaces, say rough chip and smooth, and a range of pressures – and maybe a couple of rider weights. We could then interpolate to get reasonable pressures for ourselves.

    Also, how to reconcile the ‘lower is always best’ with the chart indicating that for coarse asphalt rolling resistance is minimized at 100 psi and for new asphalt, at 110 psi?

    I used the Berto 15% drop approach for years because it seems reasonable that the right pressure will be a function of air volume and rider weight. But how that might relate to rolling resistance I do not know.

    • Michael_Fink

      Given current thinking is basically ‘lower is always better, up until the point you pinch flat / damage your rim’ the most influential bit on any such table might simply be ‘How often are you prepared to suffer a mechanical breakdown?’

    • HamishM

      Try Jan Heine’s take home: https://janheine.wordpress.com/2016/03/09/tire-pressure-take-home/ . The two sentence summary is:
      1. Ride the tire pressure that feels good to you.
      2. When in doubt, let out some air.

      • velocite

        I’ve read all the words, thanks. What’s missing is quantities – that was my point.

        • HamishM


          • velocite

            Ha ha!

  • Larry @CycleItalia

    I’ve been ranting about this for decades – use a larger tire and run less air pressure in it. Thanks to Josh and Jan for singing this loud and long enough recently that some (though not all) are finally paying attention!!
    Enough air to prevent a pinch-flat seems to be about right, or a bit more if you’re the kind who can’t/won’t avoid running into or over any and every thing on the road.
    Anyone else notice how inflation devices now all look like SILCA products? There’s the best…and the imitators.
    (Disclaimer: Silca is an official supplier to CycleItalia though we happily bought them before this!)

  • Type100

    I swapped from 28 mm Continental UltraSport IIs to 32 mm Clement Strada LGGs on my Giant TCX and its 19 mm-wide wheels.

    The UltraSport IIs had a pressure range of 80-115 psi; I kept to the low end of that scale. Mounted to my bike and its wheels, it rode and gripped well even as I slacked off and let pressure drop spontaneously to 65 psi.

    The Strada LGGs were a whole other can of worms as far as pressures were concerned. Rated from 40-80 psi, I initially pumped 70 psi into them and proceeded to lose front wheel grip on wet asphalt with zero warning. Even when the roads dried up later that day, I could feel them actively disliking any attempt at turn-in, feeling nervous with any sort of lean angle.

    After much experimentation I found their sweet spot: 45 psi front, 50 psi rear. The transformation was huge. They still rolled well, but were much happier customers and could accept as much lean-in as I dared. No longer temperamental in the wet, either.

    Color me amazed: just 4 mm of tire width brought such huge variability in tire pressure, ride, and grip characteristics. I know that it would have been better if I had 32 mm UltraSport IIs instead for a like-for-like comparison, but they’re not available in my country and this is the best comparison I could share from my experience.

    • Michael_Fink

      Interesting. But there’s potentially much more at play there than just 4mm. The real test would be to compare riding characteristics of two different widths of the Clement (or the Continental) tyre at different pressures.

  • Luis Pombo

    please remenber for the Continental Grand Prix 4000 S2 700 X 25 C, the min. pressure is 95 psi.

    • Rodrigo Diaz

      No way I run them at that pressure. I run the 23s at 85/90 psi on the first generation of wider HED Ardennes rims (they measure close to 24.5 mm on them). No problems – no popping off the beads, no noticeable increase in pinch flats.

      Go figure, I also run the Schwalbe G-One (35s) at 40 psi and below (tubeless) for rough road riding when the recommended minimum 45-70 psi. It took a while to over the the “fast” feeling of bouncing around every little bump, but even on my TT bike on a very smooth course I am running 90 PSI now for 23s.

      Give it a test and see if it works for you. I’m not saying I would go against manufacturer’s advice for everything, but I trust my judgement and ability, and I do test things before riding them hard (I am an engineer, worked as a bike mechanic and I accept my responsibilities when running things “off spec”. I am not litigious).

      • Geoff

        My suspicion is growing that the minimum pressures specified by tyre manufacturers assume you are running a rim that is 18mm to 19mm wide. These days lots of us are running 23mm or wider rims (HED C2, Pacenti SL23 etc.) It would seem to make sense that a lower pressure is still safe on a wider rim.

        • Superpilot

          I run as low as 70psi on a 25mm GP4000s2 with narrow Open Pro rims at 75-80kg for rough chip seal. Fast and smooth surface usually 80-90psi. It is the suggested pressure only.

        • Rodrigo Diaz

          maybe. I still one own a Mavic Open rim from 2009 or so. I think that’s a 15 mm width, very narrow even then.

      • Bmstar77

        I’m interested in your running 23mm on wide rims. The ETRTO standards say that for a 17mm internal rim width, 25mm is the narrowest recommended tyre, and for a 19mm internal rum width, 28mm is the smallest recommended tyre. My Velofex 23’s have a warning on the box about using a rim no wider than 15mm. I think the HEDs are 21mm internal width, which is huge. You don’t experience any problems?

        • Rodrigo Diaz

          Nothing negative in 4+ years, both with the early Ardennes at 17 mm and the newer Ardennes Plus. I think those are 2 mm wider externally, no idea internally but definitely wider. I’ve run the 25 as low as 70 in the front. I’m about 78 kg but ride “gently” on the road, I normally don’t smash stuff too much. But even from HED there are some discrepant instructions regarding tire pressure – I’ve found “weight in kg x 1.3 = pressure in PSI” to “maximum pressure 100 psi” to “70 to 80 PSI”, in a HED video from Tetmeyer. These are all from HED itself!

          For what is worth, HED reports tests with 23 mm tires and recommends 22+ tires for its wheels – that’s from them so your mileage may vary.

          The only problem I have found is that in my bicycle (an older Argon Gallium) 25s rub on the chainstay with the Ardennes plus, so I’m limited to 23s. I guess 24s could work but the brands normally carried by my LBS and team sponsor aren’t usually on stock. 25 Continental GP 4000 and Michelins rub on standing climbs under heavy torque. So I’ve used 23s exclusively for a long time.

        • ETRTO are very conservative. HED, Enve, Zipp, Roval and now DT Swiss make road wheels at 28mm+ widths, because a wheel 5% wider than the tyre is the most aerodynamic.

    • velocite

      With me on my bike the load on its wheels is 36kgs/51kgs front/rear. The Frank Berto 15% drop chart says that on 25mm tyres pressures should be 68 psi/100 psi. No problems with that – or lower, for that matter.

  • Samuel_in_France

    “It was Tom Anhalt that first raised the possibility that there was more to rolling resistance than friction alone.”

    Tom Anhalt’s contributions were great, but the above* was described on Usenet by Jim Papadopoulos in the mid 1990s or possibly earlier.

    The theories were roughly outlined by Archibald Sharp in 1896.

    * Or what is meant by it. Friction plays an insignificant role in rolling resistance and I don’t like including “suspension losses” in rolling resistance.

    • tanhalt

      Agreed…I am by far not the first person to think of this. I was just following up using field testing techniques to confirm or deny. In any case, I think Matt did a great job on this overview/summarization. Nice work.

  • winkybiker

    Isn’t feeling fast more important than being fast? I’m serious.

    • Alan Clark

      It depends on if the person next to you in the sprint is feeling fast or is actually fast.

      • winkybiker

        There’s always someone faster, regardless of tyre pressure. Only the identity of that person changes in response to your own speed.

  • Nikolas Gloy

    “c” is not a unit of measurement. “700C” is an ancient term for a specific wheel + tire size combination. You are using “c” as a synonym for “mm” but it’s not. Please stop.

    • I made the decision about 2 years ago that unless I was reporting a size that I had measured for a tyre, I was never going to refer to it in mm. As you can see from the data above, the width of any designated tyre size can vary a lot, so to use a formal unit of measurement is inaccurate. What is far more accurate is to refer to any tyre as it is designated, because a 25C tyre will always be a 25C tyre, regardless of the rim it is mounted on and the tyre pressure that is used.

      I take your point with the capitalisation and I promise to adhere to it in the future.

      • Nikolas Gloy

        Capitalization is irrelevant. Neither “c” nor “C” is a unit of measurement.

        To write about a “17c” rim (as in those charts) is gibberish.

        For tires, if the manufacturer’s claimed width is inaccurate, it still doesn’t help to write about it as “25C” or whatever. For the purposes of choosing tire pressure, it’s the actual width in mm, not some arbitrary number in “C” or “D” or “Z”, that is relevant.

        • Oh, I see, I didn’t realise you were referring to rim bed width. In this instance I was staying true to the data as supplied as well as what appears to be an industry trend for describing rim width. There is no consensus though, and the whole issue of measuring and reporting rim width is plagued by variables (eg hookless versus hooked rims, bed width versus hook-hook). But then that’s just a symptom of an inventive industry that is only very loosely standardised, so I expect the gibberish is going to persist.

      • George Darroch

        I C what you’ve done and I agree.

  • Pro Cat 4

    Good article.

  • Geoff

    One question that I have also is how seriously we should take the minimum pressure specified by the tyre manufacturer. In one instance, I use 25mm tyres on a 23mm wide rim and have for a while been inflating to 6 bar. I changed to a different tyre brand, where the manufacturer specifies a higher minimum pressure – 6.2 bar. If I inflate to 6.2bar, I do feel more vibration, which is irritating. My question is whether I am really at risk inflating below the recommended pressure.

    • I have ridden GP4000 28mm down to 40psi, but typically 60psi, which are both below recommended pressure, with no ill effect after 5000km. They cornered great, and rolled smoothly. Now on Compass 28mm tyres running the same pressures and again, no issues. Try dropping your pressure 5psi per ride, and see the effect for yourself.

  • I’ve been a deciple of Jan Heine for some years now, and this is expressed in the bikes I ride. Road bike: 700×32 Compass tires 70/60psi. Randonneuring bike: 650bx42 Compass tires 50/40psi. Both are fast. I’m the one who’s not fast! Supple tires not handicapped with thick sidewalls or “anti-puncture” layers will always be fast.


Pin It on Pinterest

November 20, 2017
November 18, 2017
November 17, 2017