“There’s fine and then there’s Zipp,” according to one of Josh Poertner’s sale agents. Josh is the Technical Director of legendary wheel company Zipp and he thinks this a good summary of how Zipp position themselves, and also how they are regarded in the market place. Zipp is synonymous with quality, but it comes at a price; you get what you pay for.
Josh works with 27 engineers in America’s auto racing heartland, Indianapolis. It is a revelation speaking with someone who knows carbon fibre wheels so well, someone who can cut through the myths and the marketing to lay the facts on the table. If you are prepared to let down your defences and indulge in a very technical insight, it is well worth tapping Josh Poertner’s fountain of knowledge and learning a lot more about carbon fibre wheel technology.
A Single Strand of Carbon Fibre
Our journey with Josh Poertner begins with carbon fibre fabrication. I started by asking about the selection of carbon fibre for Zipp wheels.
JP: People are often surprised at how hands-on carbon fibre production is. People like to use the term hand-made and pretty much all of the carbon is hand-made.
The actual carbon production is the graphitisation of the polyacrylonitrile fibre, it’s turned into carbon strands. It’s bundled into a K count, how many thousands per bundle, typically 3K or 6K or 12K and sometimes 24K.
Most of what we use is unidirectional carbon and that is a bunch of parallel running 12K’s or 24K’s that has been pre-impregnated with an epoxy resin and then it’s frozen. The other one that you see a lot in the brake track and the tyre weld of the carbon clinchers is a woven fabric. Typically they have different stiffness properties. When you weave an over-under, the fibre bends. The more bends you have, the poorer it’s going to perform in those directions. If you fracture it, the fracture will run one weave length and terminate at the cross points. Unidirectional carbon aren’t nearly as damage tolerant – it can run the entire length of the strand.
We tend to put woven [carbon fibre] where we drill holes. If you think of drilling in uni-directional [carbon] you can get a run [fracture] like in nylon. We use woven where we need toughness and also use woven where we have impacts. Woven has higher elongation failure rates, because of the crimping. It can take compression before it actually begins to interlaminar shear.
Three Dimensional Kevlar Stitching
I am standing with Josh in front of the complete series of ZIPP wheelsets, from the 202 tubular and firecrest carbon clincher through to the 303?s, 404?s, 808?s and disc wheels. Picking up and looking at the construction in more detail Josh continues.
JP: The core of the rim is uni-directional and here, where the powder holes are drilled and the impact zones are, you see woven. Even the tubulars have woven under the ceramic brake surface, just for impact toughness. One of the things that is new on the firecrest wheels is that we have replaced the woven material on the inner diameter with the three dimension stitching technology.
Tens years ago all of the rims were woven; we were really the first to go to an all uni-directional, but we still had woven on the inner and outer diameter. You will notice every single rim on the market today has a [carbon fibre] layup that looks just like my rim. We really set the standard on that. That was a lot of [us] saying “let’s take the woven away from the areas you don’t need it”, which is in the sidewalls.
In the inner diameter we have replaced woven with sewing. The fibre is continuous, tip to tail, so there is no actual seam in the entire rim. Where you have the holes you have what is known as open hole compression, and open hole compression is a hard thing for carbon to resist. You can either drill the hole where the fibers end, and that’s not so strong, or you can push a pin through and mould the hole, and the fibers bend around the hole and they are still not so strong. Nobody has really come up with a great open hole compression solution. If you look at the Boeing 787 they are doing the window surrounds with a three dimensional weave, almost like a grading. It is three dimensionally woven and from that we actually had the idea; what if we three dimensionally stitch or sew?
Conventional wisdom was that you can’t sew prepreg, it’s not going to work. It took about two years to develop the sewing technique, the machines, the coding, and the process.
The stitch on the outside is called the the cover stitch. There are three needles, three puncturing threads and two covering threads. The covering threads give you essentially the same keel resistance as the woven fibre would give you and the added benefit is if the carbon fails, the first thing that fails is the glue. The carbon fibre is ten times stronger than the resin. The failure, whether it is impact, or whether it’s open hole, it is always going to start with an interlaminar shear. You have two carbon parts that are really strong, they are only held together by glue. When there is an impact, or a spike in the load, they shear. What we have done is sown a re-enforcement. Carbon is on the XY [plane] and on the Z is typically glue. We sew through in the Z axis with kevlar, so now to have interlaminar shear you have to physically break those [kevlar] fibers.
It is the nature of wheels, [where] you have to have holes for the nipples, you have to strengthen them. We have found that this [technique] is stronger than any preexisting technology.
The three dimensional stitching is patented, as is the kevlar stitching that runs along the top of the tubular rims for reinforcement, called Carbon Bridge Technology.
The Carbon Bridge Technology allowed this to be the first carbon wheel to finish the Paris-Roubaix and, ultimately, win the Paris-Roubaix. We spent two years developing the technology and trying to convince the riders to try it and the first year we tried it [in 2008] Martijn Maaskant finished 4th, and the next year we won. Tom Boonen later said “I don’t think the race can be won on an aluminium wheel ever again”. We worked with Boonen and Specialized this year where took them into the Arenberg forest, the Carrefour de l’Arbre and to some of the flat paved sections at the end. You could see in the power meter 24 – 26 watts lower at the same speed on the carbon wheel and over the 265km distance, that’s 700 plus calories of energy not burned by riding the carbon wheel.
The Comfort of Carbon Fibre
Just like steel, alumnium and titanium, carbon fibre has material character traits, though it is not necessarily what it seems to be.
JP: People tend to want to always believe that carbon is comfortable and that is not necessarily true. We laughed at the early days when they had carbon seat posts and riders would say it has so much damping. We also make speaker cones out of carbon fibre because it has so little damping. If you want damping, you have to design it in. Take an old school Zipp 440 wheel that we created in 1991, that has a V-shape. It is essentially a triangle and there is nothing more rigid than a triangle. You can add all of the damping you want, if you don’t have the spring rate it really doesn’t matter.
The History of Zip 404 Wheels beginning with the original ‘straight edge’ 404
If you look at our wheels [now] there are no straight surfaces anywhere in the rim. It is all curved and the rim bulges out. You actually have a bit of compliance within the rim section. Now that I have a spring rate, I can then actually add damping. It is really about systems design. Something doesn’t damp because it is carbon.
American Verses Chinese Carbon Production
The majority of the bicycle frames and parts are made in Asia, though ZIPP is one of the few exceptions to the rule with design and production in the United States. Josh Poertner share his position on keeping it local.
JP: The western world has done a really good job of giving its knowledge to Asia in the search for higher profits. We have stayed away from that; everything we sell, we make. Everything we do we consider to be our secret and I consider that to be our advantage. My engineers sit 50 feet away from the guys who actually make the product and that gives us a really different outlook on things. We can move quickly, we can take big design risks in advance development of products.
The original Zipp 303 was a 28mm wide tubular rim with sidewalls that intentionally give about 1.5mm radial compression within the rim section and we started sewing the outer diameter. We started in 2006 saying “we’re going to win Roubaix on this wheel”, while the conventional wisdom was “we’re not even going to ride carbon wheels at Roubaix”. That is the kind of process that I don’t think anybody is willing to undertake in China. You spend the time and money travelling there and back, and now you have a mould – if you don’t use it, they’re going to sell it to some other guys. You need to get to market and make money.
For us, everyone in my company comes from an auto racing background. We are in the home of auto racing in America – Indianapolis. We have 30 composite shops in the greater Indianapolis area, guys making racing car chassis, wings, custom driver seats. For years we made driver seats, steering wheels and wings. We have done engineering work for multiple Indy 500 race winning cars. That’s the background of our crew and our team, so we really try to leverage that to change the cycling experience.
Computers verses Wind Tunnels verses Marketing
Advanced modelling and prototyping on computers aides the design process, though this hasn’t replaced wind tunnel testing. If anything, it makes it more complex, particularly with market pressure to constantly be innovative, lighter, stiffer and faster. So what do you get when you put a marketing in the same room as engineering?
JP: You can’t delude yourself. Everything we do in Computation Fluid Dynamics (CFD) that you have to make a decision on, you also have to take to the wind tunnel, because there is always that risk. It’s not easier on the computer, but you can pick out a lot of data that you can’t get in the wind tunnel. The wind tunnel essentially gives you three data points and a consumer could say “but you didn’t have a bike”. Then you put a bike in there and they say “you didn’t have a rider”.
The problem with any sort of measurement device is that you have an uncertainty that is a percentage of your number. If I put a wheel in there, my uncertainty is maybe 8 grams. If I put a bike in there, it is maybe 20 grams. If I put a bike in there with a rider, it could be 100 grams. The problem is that if you are trying to find a change in your wheel, that may only be 15 grams, but there is too much noise. In the wind tunnel, you can see a lot of numbers that the marketing guy calls “better”, but the engineer has to call the same. “That’s 15 grams better, but our uncertainty is 20?. As a responsible engineer, I can’t call that better.
Ninety nine out of one hundred companies in this industry are marketing companies and are buying stuff that someone else is making. Of course, they can go in and say “this is the best”, “this is better”, and we can’t do this.
What we can do in the computer is steering torque. None of the wind tunnels worldwide is measuring this and that is something that we can model. We can measure the steering torque and the centre of pressure and iterate that design in the computer and watch that move.
We sent a team to work in the wind tunnel to develop Yaw moment measurement. Because we have made changes and helped the wind tunnel make changes, other companies are benefiting from that. We’re willing to accept that. We were there first and basically created the science behind wheel stability, and our whole industry is going to benefit from that. Not a single wheel company here isn’t talking about stability, whether they have any idea what that is or not.
When Carbon Fails
With the current trends, your next road bike is most likely to be carbon fibre, but not necessarily your next wheelset. Carbon wheelsets are often still reserved for performance rather than everyday riding.
JP: The problem with the wheel is that it sees a lot of heat because there is a braking surface, and of course the area you damage is also the brake surface. We spent three years developing resins that can handle the high temperatures for carbon clinchers; you will see over 400°F (200°C). Our rims can handle transience on the surface of 600°F (315°C). That is another thing with the wheels out of China that people don’t think about. They can soften up, warp and come apart. Tour (magazine) Germany did a nine wheel carbon clincher shootout and Zipp and Xenti were the only two that didn’t melt on the descent and fail through heat. It’s a very hard problem to solve.
We were late to the carbon clincher business because we were busy trying to solve that problem. I was beat about the head and neck by distributors and sales people saying “everybody’s doing it, it works, it works”… but guys, it doesn’t pass our testing! By the time we launched, it was a really well known problem “Oh, it’s hilly were I live – we can’t sell carbon clinchers, they melt”. Melt is not technically right, though is the word that people understand.
There is a thing called the Glass Transition Temperature, that engineers call TG, which is essentially how hot it can be heat-soaked before it starts to move, the epoxy will soften and it will be able to deform. Most of the product out of Asia is between 260 and 280°F (125 – 140°C), some of the good ones are 300°F (150°C). Some of the good US and European made wheels are in the 320 to 350°F (160 – 180°C) range. We are 450°F (232°C) and upwards. The Zipp 202 is a little bit above that still because it is a lighter weight product, so there is a risk that it gets hotter. With the 202 we have launched a new brake pad geometry that is 1.5mm taller radially and that is a 15% increase in surface area. Because it is radial, you are putting in more “swept” area on the rim, so the temperatures tend to run lower.
The Devil in the Detail
Josh is realistic about the performance gains that are possible when it comes to tuning your bike and wheels. It makes sense to split the elite performance athlete from the sports enthusiast. When it comes down to nitty gritty details of tuning a racing bicycle and wheels, when you are paying for your own equipment, there are limits, but also affordable options.
JP: The Mission Statement of my department is “Ultimate Customer Experience” and that goes for every aspect. People get so hung-up on one thing “We build the lightest”, but is that really the ultimate experience. [One of the problems of] an 800gram wheelset, is that part of what you are feeling in instability is that there is not much inertia. There are so many factors that influence it. For bearings, we are the only company left in the world spec’ing DIN P5 bearings from Switzerland. We use a steel bearing that is more precise than almost any ceramic bearing on the market and that’s a real part of the value proposition. Ceramic bearings tend to have low friction if they have ultra-high quality, but a lot of times to get the price point right, you will put a ceramic ball in a Chinese [ball] race. You can say you have ceramic, but essentially you are an ABEC 1, a very low grade race with a high grade ball. That’s not doing anything.
This industry can be really guilty of going for the marketing angle, but guys, if the ceramic bearings upgrade is less that $1000 for the wheelset, it’s not worth it. You get what you pay for. To give you a benchmark, our Zipp hub continually achieves top one, top two in independent mags. Tour magazine did a test a few years ago and we were number two in rolling efficiency behind the DT190 hub, with full ceramic bearings with ABEC 7 races. I look at that and go ‘we’ve got 30% larger flange diameter, we’re a couple of grams lighter, we’ve got a 2mm larger axle and we’re a good bit less expensive’. I think that’s pretty good.
Nine out of ten ceramic bearing upgrades will make my [Zipp] hub less efficient. Ceramic doesn’t necessarily mean efficient. There are a million Chinese made ceramic bearings that are being used for furnace carts and anywhere where you have something that is hot. That is suited to the application and ceramic doesn’t mean high quality and high grade. It is heat resistance against thermal growth, it needs to get hot and not change size like a steel ball. There are a lot of people out there trying to leverage that, but any one of those upgrades in my hub is going to make it roll way less efficiently, so you, as the consumer, may as well put the money in a bin and set it alight.
The best case scenario of ceramic bearings in a wheelset is 1 watt at 30 miles and hour. We sell ceramic bearings and they’re $1000, but we’re honest: “guys, it’s 1000 bucks for a watt”. Tony Martin, at the worlds, rides thousand dollar ceramic bearings. For the consumer, you can buy a better tyre and you’ll save twice that. You put latex tubes in there you’ll save 6 times that amount. A new chain on your bike can be as much as 6 watts compared to a worn chain. Just cleaning your drive chain is a couple of watts. 1 watt of ceramic bearings is about 3 seconds per 40 kilometres. I can find you a minute for the 40k for very little money.
With Chris Langdon, our distributor [echelon], we spent a year working with Kristin Armstrong on every little detail with multiple wind-tunnel tests, looking at different tyres and tyre pressures and clincher verses tubular. Really sweating the details, and she won, and in the end said she felt she really had THE advantage. That’s when she won by 18 seconds and had ceramic bearings and latex tubes, had every little thing dialled and you go “that’s what that’s for”. For consumers who say “I put ceramic bearings in my wheel and now none of my friends can catch me on the hill”, that’s not ceramic bearings. If you loose the national time trials by three seconds, ceramic will buy you three.
There’s fine and then there’s ZIPP
Zipp wheelsets and products are not within the reach of all riders, though Josh knows this: “We try and be at the pinnacle, but are not all things to all people.” For cyclists and teams who are in the line of sight of ZIPP, they can rely on an enviable reputation of quality and reliability that the company has achieved.
Josh feels a personal responsibility to his customers, and while marketing is part of any successful business, the innovation is not dictated by marketeers. “We try to make decisions based on research and based on science and the hardest part is at the end when we try to be honest.”