This used to be one of the first things you learned as a cyclist: If you want to go fast, make sure your tires are pumped up to the maximum pressure. The harder your tires are inflated, the faster they roll.
We now know that this isn’t true. The realization that tire pressure does not affect performance is the key to the revolution that has swept through the cycling world in recent years. Without this new-found knowledge, all-road bikes and their supple, wide tires would make no sense at all. Here is how it works.
Tire resistance is determined by two factors:
Hysteretic Losses: With each turn of the wheels, the tires flex. You can see that in the photo above: Both tires flatten where they touch the road. Flexing the tires takes energy – imagine squeezing a tennis ball. Energy on a bike can only come from one source: the rider’s power output. Reduce the energy lost to flex, and you’ll go faster.
The easiest way to reduce this energy loss: Inflate the tires more, and they’ll flex less. Less flex means less energy is converted to heat as the tire casing deforms under the weight of bike and rider.
You can also reduce the losses by making the tire casing easier to flex. A supple casing is easier to flex than a stiff one, so it absorbs less energy for the same amount of flex. Imagine squeezing a marshmallow instead of a tennis ball.
So we want a tire that is supple and runs high pressures. That means we have to make it narrow. Why can’t a wide tire be supple and run at high pressures?
Pressure is force per surface area. For example, PSI stands for Pounds per Square Inch. The larger the tire’s circumference (more inches), the more force will act on it (more pounds). Imagine the casing as a chain, with the pressure as an elephant standing on each link. If the chain is ten links long, it has to support the weight of ten elephants. Make the chain twice as long, and you’ll have twenty elephants standing on it. The chain has to be twice as strong. Or you have to reduce the weight of each elephant. Back to tires, this means that a wider tire either needs a stronger (read: stiffer) casing, or you need to reduce the pressure.
Summary: The best way to reduce hysteretic losses is to use supple tires, make them narrow, and inflate them hard. With wider tires, you either can have a supple casing or high pressure, but not both. Either way, wider tires will have more hysteretic losses. If you only look at hysteretic losses, ‘wide high-performance tires’ seems like an oxymoron.
That was the accepted wisdom when we started looking at tire performance way back in 2007. It’s not incorrect, but it overlooks a second factor that also affects how fast a bicycle rolls – and tire pressure works the opposite way there.
Suspension Losses: As the bike vibrates, energy is lost. Most of that energy is absorbed in the rider’s body, as soft tissues rub against each other. Decades ago, the U.S. Army studied tank seats and found that the discomfort we feel from vibrations is caused by friction between our body’s soft tissues. This friction consumes energy that is turned into heat. (Rub your hands against each other to see how friction creates heat.) The more uncomfortable the vibrations, the more energy is lost.
Mountain bikers have known for a long time that bouncing makes your bike slower. The fastest mtb is the one that absorbs shocks best. Road cyclists used to think that we had to ‘tough it out’ to go fast. We endured the discomfort of narrow high-pressure tires because we thought that they rolled faster. We thought that pavement was too smooth for suspension losses to matter.
At Bicycle Quarterly, we started to test the performance of tires on real roads in 2007. At first, we also assumed that higher pressures rolled faster. However, as long-distance cyclists, we suspected that there was a point of diminishing returns. Our thinking was this: In a short race, we may endure all kinds of discomfort if it makes us faster. But we can endure discomfort only for so long before it affects our power output. In a ride as long as the 1200 km (750 miles) of Paris-Brest-Paris, we might give up 5% in rolling resistance if we gain 20% in comfort. What we wanted to know: Where is this point of diminishing returns?
Back then, tire resistance was tested on steel drums that measure only the hysteretic losses. On steel drums, there is no doubt that higher pressures produce better results – as shown by all tires in the table above. Take the Vittoria Rubino Pro (second from bottom): At 60 psi, it requires 40% more energy than at 120 psi. I mention this tire, because later on, you’ll see how it performs under real-world conditions.
Drum tests also suggest that high pressure is more important than a supple casing: In the table above, the slowest tire at 120 psi has less resistance (13.4 W) than the second-fastest tire at 60 psi (13.9 W). That is why tire makers used to make their wide tires with stiff casings, so they could withstand high pressures. A wide, supple tire – limited to a low pressure rating of, say, 60 psi – would perform poorly on the steel drum. That is why they didn’t exist – who would want to make a slow ‘high-performance’ tire? It all made sense – if you test tires on steel drums.
Imagine our surprise when we found that in the real world – on real roads – tires perform very differently. We tested numerous tires, with two different methods (roll-down and power-meter), and always found the same: Higher pressures don’t make tires faster. And the advantages of supple casings are much larger than the steel drum tests suggest.
Why? Because the suspension losses are significant even on very smooth roads. And both lower pressures and supple casings reduce the vibrations of the bike and thus the suspension losses. But you cannot measure suspension losses unless you have a rider on the bike. That is why earlier studies (and many since) failed to give meaningful results…
Above are the real-road results for three Vittoria 700C x 25 mm tires, including the Rubino Pro. We tested on brand-new, ultra-smooth asphalt. For the Rubino, there is no difference in speed between 80 and 11o psi. It’s clear for all tires: Higher pressures don’t make them faster.
(Note that the Watts are for the entire bike and rider, not just for one tire. That is why the power measurements is so much higher than in the drum tests.)
What happens is this: As tire pressure increases, the tire flexes less, and the hysteretic losses go down. However, the tire also vibrates more, and the suspension losses increase. The two roughly cancel each other, and that is why there is no clear trend in the table above.
Before we continue, it’s important to mention that we made sure these results are statistically significant. This means that we are seeing real differences in performance, not just ‘noise’ in the data.
The graph shows a few more things:
- Really low pressures make a tire slow, because it flexes way too much: the hysteretic losses are huge. The extreme is a totally flat tire – very slow. At some point, the tire has enough air to avoid excessive flex. Above this ‘break point,’ hysteretic and suspension losses start to balance each other.
- Hysteretic and suspension losses are non-linear, so they balance differently for different tires and different pressures.
- CX Tubular: A tubular tire sits on top of the rim, so it can flex around its entire circumference. It can run at very low pressures without excessive flex. The break point is low (80 psi).
- CX Clincher: A clincher rim constrains the tire around about 1/3 of its circumference, so higher pressure is needed to avoid excessive flex. The break point is a bit higher (87 psi).
- Rubino Clincher: The Rubino’s stiffer casing is harder to flex: The hysteretic losses for the same amount of flex are higher. The stiffer casing also transmits more vibrations, so low pressure doesn’t reduce the suspension losses to the same amount. This means that the break point is higher than for the more supple CX (95 psi).
- All three tires roll slowest at moderately high pressure: The tire is already too hard to absorb vibrations, so suspension losses are high. However, the casing still flexes, so hysteretic losses are also high. It’s better to run low or very high pressures, at least on the very smooth asphalt of our test track.
Summary: On real roads, even smooth ones, higher pressures don’t roll faster.
We tested many tires – above from our first tests in 2007 that measured the time for a roll-down on a carefully chosen test hill – and we always found the same: Above the break point, increasing the tire pressure doesn’t make you faster. The break point is higher for stiffer tires. For the tubulars, the break point is lower than the pressures we tested.
I wish we had done this testing when I was still racing. Back then, I ran 21.5 mm Clement Criteriums at a bone-rattling 130 psi. I would have been faster – much faster – on 28 mm Campione Del Mondos at much lower pressures.
The above results were on ultra-smooth pavement (Vittoria) and relatively rough, but not bumpy, pavement (others). Does the road surface affect a tire’s break point?
To test the extreme, we ran various tires on rumble strips, which are a good stand-in for cobblestones. (Unlike real cobblestones, rumble strips are very regular, so we could obtain repeatable results.) We also tested each tire on the smooth pavement right next to the rumble strips.
On the smooth pavement (left), the Compass 26 mm tires roll as fast at 75 psi as they do at 95 psi. This confirms what we’d found before: Even on smooth roads, tire pressure makes no difference.
On the rumble strips (right), higher pressure was slower: The 26 mm Compass tires used 20% more energy at 95 psi than they did at 75 psi.
Summary: On really rough surfaces, higher pressures roll slower. Wider tires roll faster on rough surfaces because they can handle lower pressures.
So now we know that higher pressures don’t make your bike faster – whether on ultra-smooth asphalt, on rough surfaces like cobblestones, or anywhere in between. For supple tires, the break point – even on smooth surfaces – is close to the point where the tire becomes unrideable, because the sidewalls collapse under hard cornering. On rough surfaces, it’s hard to reach the break point – the tire pinch-flats before its performance deteriorates.
Conclusion: If you want to go fast, you need supple tires. That is all. Pumping them up harder won’t make you faster. On rough surfaces, it actually makes you slower.
When we saw these results, we realized that this could revolutionize bicycles: If we didn’t need to high pressure to go fast, we could make wide tires with supple casings, run them at low pressure, and still roll as fast as we did on narrow ‘racing’ tires. With these wide, supple tires, we could go on roads and trails that wouldn’t be much fun on a traditional, narrow-tire racing bike.
We envisioned a new breed of bike – racing bikes with ultra-wide, supple tires – and we called them ‘Allroad Bikes’ (still without a hyphen). That was way back, in 2007.
The first step was to make the tires needed for these all-road bikes. First we worked with other manufacturers, but in order to get exactly the tires we wanted, we introduced our own Compass tires in truly wide widths (above). It’s taken the bike industry a little while to get on board, but now our ideas are generally accepted: all-road bikes are the most important segment in the performance bicycle market.
All-road bikes are more than a passing fad, because they finally correct one of the shortcomings of traditional performance bikes: the need to trade comfort for speed. Now we know that discomfort not only is unpleasant, it actually slows you down. Put more simply:
Comfort = Speed
Cycling has become much more fun on all-road bikes that combine the fun of riding a performance bike with the comfort and go-anywhere ability of wide tires. They are the bikes of the future, and they are here to stay.
As Compass becomes Rene Herse Cycles this year, we will continue to push the envelope beyond what most people think possible. The latest development are our dual-purpose knobbies that roll as well on pavement as many racing tires and offer great traction in mud and snow. It’s going to be a fun ride!