The podium and 4th place for Thor and Roger at the 2009 Roubaix had been a real rush for the team, we were truly thrilled, but at the same time it felt like the learning had just begun.
First of all..
Not much about this type of engineering is very glamorous, looking back on it I realize now that all of these stories sound rather romantic, but being there at the time reminds me of that old joke about flying where the punchline is something like 'long periods of boredom interspersed with moments of shear terror.' Much of the time spent with the teams is spent cleaning things, fixing things, replacing things, and preparing for other things. It's non-stop work making sure that every single detail is covered, every bolt is tight, every tire is perfect and while enjoyable it generally isn't very exciting.
Riding in the team car at Roubaix is a truly bone shaking and mind numbing experience, the roads are narrow, the riders are far ahead of you and there is no way for the cars to get past each other in the Pave sections so it makes for hours spent listening intently to race radio (in French) to hear who had a flat, needs a bottle or when crashes occur. Every single puncture or crash announcement comes with a massive adrenaline rush, 'Oh $@#$ it is OUR guy? Is it OUR product?! Is it a wheel failure?..' Everybody experiences the same, the mechanics are thinking 'It is something I did (or didn't do)?!?!' the environment is unbelievably stressful, and unlike most races, you have these long periods where you can't get to them quickly..
Our biggest excitement from 2009 was tiny cracks in rims as well as some pretty serious tire cuts AFTER the race. These evoke that certain kind of fear about things that could have or might have happened, but like a scary movie you've seen before, the ultimate outcome is known already. So 2009 was mostly boredom during the race itself, with the fear happening before and after the event.
On the technical side, all of these emotions are really a mixed bag. The problem for the engineers is that you NEED to see failures in order to understand how and where to improve, but at the same time, you need to see them before the event an not during.. Yet, looking at the speed and power data, there is just nothing you can do in training or race prep that comes even close to the real thing, so technically, nothing you do in testing is going to come close to the real thing..
Some fun P-R race data I've seen:
1 minute power leading onto the Arenburg Pave: 658 watts
1 second wattage on the Arenburg Pave: 1584 watts
5 minute wattage covering run into Arenburg AND 2,400 meters of Pave: 561 watts
These are truly amazing numbers, and as you can imagine not something you get in training runs. We even tried motor pacing riders onto the Pave..but nothing matches the adrenaline and quite honestly the terror of running this stretch in th actual race!
The opposite of all that race adrenaline are the hours spend measuring the rate at which tires weep air, or impact testing tires and rims at a range of different pressures and sizes.
For 2010, test engineer John Fearncombe developed a highly automated impact test platform at Zipp which used the basic concept of the UCI test rig, but could automatically launch the impact sled at very precise speeds and energy. This allowed the team to very quickly test the sprint rates of tires at various sizes and pressures, find the ultimate failure energies and make extremely direct comparisons between options. For instance, a carbon wheel with 28mm tire might handle a 90 Joule impact before failure while an aluminum rim may dented at 70 Joule and rendered unrideable at 80 Joule when used with same tire. We could also create equivalency between systems, so if a rider liked one wheel and tire at one pressure, we could test that and them iterate the new wheel or tire using pressure to have identical spring rate. Over a few months we impacted literally thousands of combinations:
Solving for Latex Tubes
Thanks to Roger Hammond (who is not just one of the greatest english speakers and hard-men ever to ride the Pave, but is also a mechanical engineer) we would spend the run up to 2010 Roubaix evaluating the leak-down rate of the team tires while also convincing them to go ever wider! Turns out that a tubular tire with latex tube will lose 0.5-1.5psi per hour, which over a 7+ hour period (figure the race will be ~6:30 and the mechanics have to have the bikes ready at least 30 minutes before the start). This turned out to be a very critical aspect of pressure optimization and planning as the comes some 4 hours after initial inflation and the nearly equally bad (but the riders are strung out and going slower..) section at the Carrefour de l'Arbre comes nearly two hours after Arenburg.
We ran testing leading up to 2010 in the Carrefour de l'Arbre looking at the minimum allowable pressures for that sector and determined that those numbers would be used, plus the leak down rate to set the starting pressure. The wheels for the top riders were selected from the tires that had the lowest lead-down rates (near the 0.5psi per hour) and the numbers were written on the sidewalls to be sure and the race morning pressures were given to the mechanics with all of this factored in!
During the week long run up to Roubaix, some of the riders, Fabian chief among them still had some doubts. Remember that he had won the event previously on the old-world wheels and tires and wasn't sure that he wanted or needed the new technology. This is understandable as the large tires and carbon wheels seemed to offer small benefits yet come with large risks. One of the problems is that riders really don't 'feel' aerodynamics while riding..this is true for all of us, you go hard, you go fast, there is nothing really to compare to.
To try and demonstrate the benefits in the real world (the wind tunnel doesn't always 'feel' real). We did a test both on the cobbles and on a few pavement sections between some of the key sectors where it seemed an attack might be likely or where a rider might find himself isolated. The results were eye opening.
Fabian's mechanic Roger Changing Wheels on the Recon Day (AFP)
With the larger tires, we found that the riders went faster over the Pave as pressure was reduced..until the point that the rim was bottoming out, and then the speeds reduced again.
This is similar to the effect of why increasing air pressure can make you slower on rough roads, rather than the tires absorbing the imperfections in the road, the bike is being lifted or rather bounced over the bumps. A tire bottoming on the rim dramatically increases the spring rate of the system causing the bicycle to bounce off of the cobble which can cause loss of traction, discomfort and loss of speed.
Coming out of the Pave portion of the testing the riders seemed to be truly sold on the larger tires once and for all. The 28mm tire in the low-mid 60psi range proved to be nearly 1km/hr faster than the 24mm tire with pressure in the mid 70 psi range (which is what is required to prevent bottoming the tire) when the data was normalized for rider power output.
Moving on to the pavement testing, we had Fabian do some interval efforts on the new wheels and the old wheels to try and demonstrate the aero differences. The results even shocked our engineers! The traditional wheels with 24mm tires required 24-26 additional watts to go the same speed as the aero wheels with the 28mm tires!
Finally, we had officially debunked the conventional wisdom that 24mm were 'faster'.
Change in Tactics
This new data was worked into all of the computer simulation models and in many ways the overall picture was even more promising than just looking at small sectors of the race.
During a Paris-Roubiax, the top riders are expected to burn 6,500-7,500 calories. The field data pointed to a 500+ calorie savings due to improved aerodynamics (remember, generally you aren't using aero to go any faster, it is buying you the same speed at a lower power).
The analysis of the final 50km of the race pointed that the aero wheels would allow for an extended breakaway compared to the traditional wheels. One estimate was that using Fabian's previous data a 20km solo effort could work and with the new tires and wheels a 30 or possibly 40km final effort might be possible!
Ultimately, races are not won on technology or computer screens. It's real people suffering at the edge of what's possible, making thousands of decisions per hour with fatigued minds and bodies. However, all of this data, technology, and testing started to swing the belief of these riders in the favor of technology. It felt as if the conventional wisdom was turning in our favor.
Fabian's Bike Race Morning, with hand written notes to mechanics on tire sidewalls (James Huang)
I've seen over and over that riders who believe in something are more willing to commit to it, and it becomes a self-fulfilling prophecy. When you believe the thing is more comfortable, you fatigue slower. When you believe a tire has more grip, you will push it closer to the limit. When you feel that something gives you an advantage that nobody else has, you feel empowered to use it.. and so on. In many ways, all of this technological improvement led to the biggest improvement of all, which was that these athletes now had access to better equipment, but they were willing to optimize around that equipment, and were then able to come into the race knowing AND believing in what they had done.
The beauty (not so much..) of being embedded with the team all week sweating all of the little details in the run up to Roubaix is that you get to fly home to the US on race day to be ready for work on Monday. I headed for the airport while the team headed to the race depart, knowing that every possible thing that could be done was done.
Some nine hours later I landed in the US and turned on my phone as the plane taxied to the jetway. Nothing..and then suddenly, text after text and message after message of congratulations.. Fabian had ridden an unbelievable race as it turns out. Funny enough, he attacked the lead group on the exact section of smooth road where we had done the power/aero testing during the week prior.
You might remember that this was the attack that started the ridiculous rumor about him having a motor in the bike...the attack is simply phenomenal to watch!
Speaking with him after the fact he had some really awesome things to say about that day, but most impressive was his mindset. He told us, 'I remembered what you said about the advantages of technology, I felt like I was on a time trial bike while everybody else was chasing me on equipment from the Eddy Merckx era..' which was something we had told the riders over and over. In the end, it wasn't an equipment advantage, but rather a technology and knowledge advantage that had translated into an incredible confidence and belief.
Cancellara later commented to the media that 'Roubaix will never again be won on the old wheels,' as there was 'too much advantage' to the new technology. Michael Hall, Director of R&D now at Zipp said at the time, 'I think we just created the future.'
Fabian has turned out to be thus far correct about he new technology. 2011 Roubaix was won on a Mavic Roubaix wheel nearly identical in all measurements to that original 2010 wheel. In 2012, 2013 and 2014 was won on again on the 2010 design wheel. Best of all, the frame makers during this time joined the trend and pushed tire clearance further. For 2014 the race was won on a 30mm rear / 28mm front tire and looking to 2015 we are working with more than half a dozen teams on gauges, pumps and other inflation related items and it is brilliant to see many of these teams running 30mm tires front and rear. The conventional wisdom has changed, and the riders will be faster, happier and less likely to suffer equipment issues on their fatter, lower pressure tires.