Wheel brand Alto Cycling recently completed a destructive braking heat test with Spark Wheel Works that shows carbon clinchers by Zipp, ENVE, Mavic and others failing, often in spectacular fashion. While Alto Cycling CEO Bobby Sweeting said the point of the test wasn't to call his competitors' products unsafe, the visual of rims blowing up wasn't reassuring for riders. BikeRadar talked to representatives for all the brands involved for their thoughts on the test, and on the safety of carbon clinchers in general.
Alto's carbon clincher heat test
In conjunction with Spark Wheel Works, Alto tested carbon clinchers from Mavic, ENVE, Zipp, Knight, Boyd, FSE, Bontrager and Roval as well as the brand's own product. The test consisted of running the one wheel from each brand on a drum at 1,200w, with a 7lb hanging load applied at the brake lever and a set of new SwissStop Black Prince pads used for each wheel. The wheels were run until failure, with the rim temperature measured constantly.
Sweeting said the idea for the test came about as Alto researched new resins with high heat resistance.
Like all carbon cycling products, the carbon strands in carbon clinchers are held together with resin. To make carbon wheels, pieces of carbon fiber impregnated with resin are placed in molds and heated up, which effectively melts the resin. When it cools, you are left with a solid wheel. The issue with carbon clinchers is that rim braking generates heat, and, when pushed to extremes, can cause the resin to soften, which can lead to the rim bubbling or folding open. In worst-case scenarios, the rim can fail enough that the tire can come off or the inner tube can burst from the heat. Years ago, a few events like Levi's Gran Fondo in California banned carbon clinchers after multiple incidents on steep, winding descents.
Prior to the test, Sweeting said that he was often asked by customers how Alto compared to other brands in terms of heat resistance, and that he honestly didn't know.
"So we wanted to show off our new product while simultaneously creating a metric that would display how rims truly perform under the exact same braking scenario, and how they would delaminate (if at all)," Sweeting wrote in a report on the test.
Running at 1,200 watts, Alto and Spark Wheel Works were able to get the rims from ENVE and Zipp to fail in a little over 300 seconds, with other brands' rims delaminating more quickly. Alto's wheel ran for 1,200 seconds before failing.
The brands involved cried foul, citing a number of issues with the test. Sweeting stood by his results.
"We want to make super clear that we aren't claiming that a rim is unsafe with their brand-specific brake pads and real world conditions," Sweeting said. "We simply wanted to test resin heat transfer capabilities, and allow people to translate the results however they wish."
Big brands' objections and responses
Engineers and spokespeople at the brands involved expressed various levels of frustration with the test and the publication of the results. Some were cautious in their responses, but nearly all wondered why existing test protocols weren't used, such as those specified by the International Organization for Standardization (ISO).
At Trek, Bontager brand manager Alex Applegate had perhaps the most measured response. "We have questions on the test. We don’t know all the parameters," Applegate said. "There are plenty of standards out there for wheel testing, as specified by ISO. It is interesting that they [tested differently]."
"We test to very stringent standards that go above and beyond accepted standards. We base those on real-world conditions of normal riders and professionals at the highest team levels," he said. "In practice, we’ve had the latest Aelous TLR 3 for six years, and are closing in on 100,000 wheels. There are effectively no failures. We have seen two. And those are slight [delaminations]: the wheel deforms, it slows down, you stop."
Knight Composites director of engineering Kevin Quan, said he couldn't gauge the fairness of the test because braking force wasn't measured at the rim.
"Was it consistent? Was it as high as they claim?" Quan said. "There is an industry standard for this sort of thing and it's in ISO 4210. The standard heat resistance is 75 watt hours in 15 minutes. That's the equivalent of a 220lb rider stopping 30 times from 30mph."
"I do not think it replicates a real world condition as it does not allow any stoppages in brake force at any time during its duration," Quan said. "I think the momentary stoppages (10 maximum in ISO) better approximate a real world condition like mountain switchbacks. It's rare or impossible to conceive of a situation whereby a rider squeezes the brakes for 20 continuous minutes under high speed."
At ENVE, spokesman Jake Pantone said the Alto test "has many flaws that lend to them gaming a test that isn’t meaningful in the real world," citing a lack of correct pad use, a test that penalized rims with better 'bite', and the input power not being dynamically monitored.
"Safety has always been priority number one for ENVE," Pantone said. "We solved the issue of melting brake tracks years ago and this isn’t even part of the conversation for us now. We literally haven’t had a single rim come back for a heat-related issue since we launched our latest technologies a few years ago. Even if Alto’s rim is exponentially better in terms of heat management than ours, it’s irrelevant to real-world needs and likely compromised in terms of ride quality or weight."
At Specialized, Roval spokesman Sean Estes said that "while we appreciate any efforts made in an attempt to quantify and therefore improve rider safety, we have some concerns specific to the testing protocol used in this specific scenario."
"The goal of any test such as this one is to replicate real-world scenarios and we are not confident the testing protocols that yielded these outcomes reflected accurate representations in that regard," Estes said. "Riders can rest assured that all Roval wheels are rigorously tested – both in-house and third-party – to ensure they’re able to handle a wide variety of real-world scenarios one might encounter."
Mavic global brand manager Chad Moore took issue with the Alto test.
"I am not at all criticising their motives," Moore said. "I have to believe that their intentions were noble, but this was simply a failure in terms of offering consumers with some reliable information in terms of making an educated and confident purchase. I am more than surprised that when they saw the results that they didn’t question their testing method. It is unfathomable that a wheel company, that has only existed for less than five years, could suddenly have a wheel that performs so dramatically better than the others in the test."
Moore's primary concerns centered around the various levels of friction, and thus braking force and heat, that were seemingly unaccounted for in the Alto test, and how the speed of the wheels driven at 1,200w should, in Mavic's opinion, have differed more.
"The Alto rim has very little, if any, texture on the braking surface. So, you can use any pad you like - heck, you could use soap - and there will be less heat in those rims. We are 100% sure that if we sent our Cosmic Pro Carbon SL UST rim to them, without the laser treatment we apply to our braking surface, that we would match or surpass their benchmark. So, as the interface between their rim and pad has less friction there is absolutely less heat and thus, the wheel can run longer before there is damage."
"They mention the motor is set at 1,200w and the force on the brakes stays consistent," Moore said. "However, when you look at the speed of all of the wheels, including Alto, it doesn’t add up. In our testing, we have seen up to a 50% change in speed depending on the braking surface being tested. We have even seen these big differences between some of the exact wheels they used in their test. So, with a wheel that has an aggressive / high-performing braking surface - like the Enve, Mavic or Zipp – there should have been a much bigger difference in the speed if the motor was staying consistent at 1,200w. The issue is that in the real world that wheel would have adequately braked, or even stopped, well before that point was reached."
"It is not effective to apply equal braking load when testing different wheels. In our test, and those you’ll find from other brands I’m sure, we maintain a motor at 600w-700w and adjust the braking force based on the resistance on the motor. This completely levels the playing field for the brake surface treatment. Slicker surface means more braking force because there is less resistance on the motor. More aggressive braking surface means less braking force because of more resistance on the motor."
In a similar vein, Boyd company founder Boyd Johnson said the braking forces were not the same on all the wheels. "The braking load was the same, but when you look at differences in brake tracks, the forces were vastly different," Johnson said. "If we wanted to pass Bobby’s test, we would have put on a unidirectional carbon brake track with no texture at all, and a coating to make it slicker."
Regardless of their varying levels of response, all the brands involved vigorously defended their carbon clinchers as safe for riders.
"Are carbon clinchers as a category safe now? Yes," Johnson said. "You can’t make something where it is impossible to damage it. You can break a bowling ball in half. As far as real world conditions, carbon clinchers are safe to ride, and that includes on steep mountains, in hot temps and under heavier riders."
How big brands test for safety
Zipp declined to specify exactly how it tests for safety, but took issue with the implication that its wheels were unsafe.
“Safety has always been Zipp’s number one priority and it's the primary reason why we waited until 2010 to release our first Firecrest carbon clincher to the public. We’ve always tested our competitors’ wheels in addition to our own. That testing revealed how challenging it is to engineer a safe carbon clincher," said SRAM road marketing technical rep Nate Newton.
Newton said that since the introduction of the Zipp Firecrest design, the company has not had a single heat-related rim failure with more than 100,000 carbon clinchers sold.
"Our testing protocol is based on data gathered from the most extreme possible real world riding and racing conditions, including extensive racing for seven years at the highest level," Newton said in an email to BikeRadar. "The specifics regarding the testing and engineering benchmarks behind our record of carbon clincher safety represents a proprietary competitive advantage we hold, but we proudly stand behind these aspects of our test protocol: All testing is founded in a deep understanding of bicycle dynamics and loading, acquired through precise measurements in the field. Zipp brake testing equipment is capable of isolating static, dynamic, and thermal loads during duty cycle tests and riding simulations. We control for key variables such as wheel energy dissipation and braking power to replicate, and then exceed, the harshest conceivable real-world conditions. Every product undergoes rigorous validation testing before coming to market to ensure durability, strength, and safety at both the laminate and full wheel system levels. We maintain the utmost confidence in our compliance with international safety standards, including ISO 4210, EN 14781, and UCI certifications. Complete records and traceability of the production process at our Indianapolis factory ensure every wheel sold meets our proven safety standards."
Among other things, Mavic uses Mont Ventoux as a real-world test, with a 100kg/220lb bike + rider descending the steepest 10km at an average gradient of 10% with the rear brake engaged the entire time.
Alto responds to critcism
Alto CEO Sweeting said that this test was not intended to be a real-world test, but instead an accelerated failure test.
"As a 'real world' test we would use an interval protocol and a wattage more appropriate for an average 80kg rider going down a hill," he said. "However, our test protocol was accurate in being able to compare resin and composite properties, which is all that we wanted to do. Accelerated failure tests are very useful; you simply have to extrapolate the results. One good example of a similar test is something that I heard recently from a marine industry engineer. He said that, in order to test corrosion resistance of their bolts, they put them in a cup of salt water. That isn't realistic, and some of their boats aren't even used in the ocean. However, this is a worst case scenario test that gives them a good understanding of how the bolts will perform over time."
Regarding braking force and friction, Sweeting said "you cannot deny the first law of thermodynamics. Energy in equals energy out, always. That means that with 1200 watts input from the motor, every rim sees and must dissipate that 1,200 watts."
On the subject of braking friction, Sweeting said that in the second phase of testing (which only the Alto wheel made it to), lever load was increased from the initial 7lb. "In phase two, lever load is increased to 9 pounds and the rim spins slower than any other rim in the test (19.3mph) prior to pad glassing," he said. "That means the coefficient of friction between the rim and the brake pad is higher than anyone else's. However, the rim still runs for 20 minutes without failure and dissipates the heat at a faster rate. I would like to hear their reasoning for how this is possible, especially if their only argument is regarding friction."
"As far as brake tracks, and stopping power in general: We found it interesting that a textured track does not automatically indicate good stopping power," he said. "The Bontrager rim has the smoothest brake track in the test and they had the best stopping power of any rim."
Sweeting added that the test was done blind. "For anyone who believes that we designed our test so that we would win, I would like to ask them how this is possible. We had one rim from each brand, supplied by Spark Wheel Works, and we had one shot to test them and film them. We had no knowledge of how they would perform, nor did we know the coefficient of friction of each rim, because none of them have ever published actual results," he said. "So we went into it with the test protocol and no other information."
What comes next
Alto's test got the attention of other wheel brands, for the insinuations on safety as much as the competitive ranking.
"We are currently discussing the Alto test at the ASTM and WFSGI levels to prepare a coordinated industry response," said Knight's Quan. ASTM is an international safety standards organization, and WFSGI is the world federation of the sporting goods industry. "It's clear that this topic is important to the public and we need to do a better job at explaining our testing procedures."
At Trek, Applegate said Bontrager is following up with Alto. "Ultimately we want to learn from them about what they found, and what may or may not be learned from that test. We have more questions than answers at this point. It starts with more testing on our end, where we can control all the inputs. Tests give you information. That’s the whole point of testing."