Part 4A in this Series covers the history of Bicycle Rolling resistance and the how and why Pneumatic tires are so awesome. If you just want to see the data, you can jump to Part B HERE
Back in 2007-2008 during the Paris-Roubaix wheel development, I had an interesting moment in the Arenberg Forest. I was working with one of the most famous Roubaix winners of the last 10 years, one whom I had also worked with to win Tour Stages, Tour time trials and even a World Championship. We had been running tires at ever lower pressures trying to find the point at which a rim/wheel failure was inevitable and right there plotted out on the screen was a trend which has been frozen in my brain for these last years: every time we lowered pressure, he went faster.
It has long been known in CX and Mountain Bike racing that lower pressures are faster, but in road racing and triathlon we have long held onto the belief that most road and even cobble surfaces are smooth enough that higher pressures will be faster, at the expense of comfort. Even at the beginning of my history with Paris Roubaix testing (~2005), the belief was that we needed to find pressures high enough to be fast, yet low enough that the riders could handle the bikes over cobbled sections. And yet, right there, every which way we looked at it on the computer, repeated across multiple riders: Lower Pressure was Faster.
Fast forward to today and we have numerous good sources for Crr (Coefficient of Rolling Resistance) testing, and we have a real movement to identify and improve aspects of high performance tires. We are, in many ways, in a golden age of tire Rolling Resistance advancement, much in the way the 2000's were the age of massive aerodynamic advancement. However, none of Crr studies in the lab are yet to truly explain or predict the phenomenon we saw in the Arenberg Forest.
A Theory in the Making
In the last 10 years, two sources that I know of have identified similar effects in their data, Jan Heine of Bicycle Quarterly has written about the effect he calls 'Suspension Losses' which you can read HERE. Some of Jan's most interesting work is in looking at the power required to ride on different surfaces, including very aggressive ones such as highway rumble strips.
The theory behind 'Suspension Losses' is rooted in pre-pneumatic tire experience and is also a topic of discussion amongst in-line skating athletes. Solid tires make surface roughness incredibly apparent to the athlete both in terms of comfort AND speed.
Imagine a rigid tire and wheel rolling over a 5mm bump in the road. In this case the tire is rigid, so the entire wheel/tire and therefore bicycle will be raised and lowered by 5mm
Model of a Rigid Tire on 5mm Bump. This scenario is quite literally how the early 'Boneshaker' bicycles earned their names, it was neither fast nor comfortable.
The rider of the bicycle becomes the suspension system to absorb the bump as the tire is incapable of handling it at the point of impact. The forward momentum of the bike is converted into a vertical force which is partially absorbed within the rider's body as well as absorbed in friction at the contact points between the bike and rider.
Another way of describing it is that the bump is essentially lifting the entire system by 5mm and dropping it in a sort of pavement bench-press of the bike and rider. Think of 1000 5mm bumps in the road as the road doing 1000 mini-bench presses of a 180lb object and it becomes clear that energy is not being used wisely in this scenario.
Pneumatic tires were such a revolution as they were not only more comfortable, but proved significantly faster than the solid tires they replaced.
Looking at the similar bump with a tire modeled at 100psi and we see that rather than lifting the system by 5mm, the system is only lifted 1mm off of the ground, with the other 4mm of displacement being absorbed by the tire. As the pneumatic tire is very efficient, much of the energy absorbed is returned with the primary losses being small amounts of heat produced in the tire casing.
Model of 23mm Tire at 100psi Absorbing 5mm Bump. The entire system is lifted 1mm with the rest absorbed by the tire.
Our second data point came from Tom Anhalt who has been studying Rolling Resistance and other bicycle physics on his website HERE
Tom has picked up the baton from Al Morrison and had been measuring and posting bicycle tire rolling resistance data taken on rollers. Tom posted a very interesting piece in 2009 related to the differences between roller testing and real world testing, where he was able to roughly match roller data at lower pressures, but saw a divergence in the data at higher pressures. Tom's article mentioning this was published at Slowtwitch.com and can be found HERE
Similar to Jan Heine's data, Tom found that rolling resistance decreased as pressure increased up to a point, and then began increasing again as shown below:
Tom Anhalt's Real World Tire Test on 'Good' Asphalt Surface, Compared to Identical Tire Tested on Rollers by Al Morrison
Tom coined the phrase 'Break-Point Pressure' to describe the point at which the Crr changed from decreasing with pressure to increasing with pressure. Tom was also the first to theorize that we could estimate what he called 'Transmitted Losses' which were the losses due to vibration and roughness and that we could (and should) model them into our theories about optimal tire pressure.
A new term: Rolling Impedance or just Impedance
For the rest of this series we will be using the term Impedance to define this resistance to forward motion caused by surface roughness. I have stolen the term impedance from electrical engineering where it is defined as the resistance of a circuit to an alternating current. The phrase feels more natural to me than any used previously and was also approved by Tom Anhalt, so we hope it sticks.
Part 4B will take the concept of Impedance to the next level and help us begin to understand how to compensate with our tire pressures. Click HERE to Read Part 4B