Heading out the door? Read this article on the new Outside+ app available now on iOS devices for members! Download the app.
Canyon’s latest Aeroad aero road racing bike debuted to much fanfare when it was released to the world this past October, and for good reasons. It was considerably more aerodynamically efficient than the already-fast previous model, it was up to 160 grams lighter and 40% stiffer, it was just as comfortable, and incorporated a novel multi-piece handlebar system that allowed for easy disassembly for travel. Combined with Canyon’s usual pricing advantages over other major brands with more traditional distribution networks, the redesigned Aeroad should have been a home run.
But instead, it turned out to be a nightmare, with seatpost-related wear issues popping up shortly after launch, and — far more disastrous — an extremely public catastrophic handlebar failure suffered by a star rider.
In an exclusive interview with CyclingTips, Canyon officials explain in detail not only what went wrong with the Aeroad and how the company plans to fix it, but also how they’re hoping to earn back the trust of their customers.
A fateful day in March
Le Samyn wasn’t supposed to be a particularly impactful race. An early-season one-day event traditionally held in Belgium in late February or early March, it’s typically considered more of a tune-up or scouting event for the bigger contests held a few weeks later. Belgian media sometime refer to it as the “Little Paris-Roubaix.”
But with just a few kilometers to go from the finish, viewers were treated to a different kind of spectacle: Alpecin-Fenix superstar Mathieu van der Poel leading out the main peloton with his hands on the tops and his driveside brake lever dangling in the breeze, the rest of his handlebar conspicuously missing after he tore off the remnants and unceremoniously tossed them aside just minutes earlier.
“When this happened, of course, we were sitting internally directly after the race, and everybody was shocked,” said Canyon’s chief technology officer, Michael Kaiser. “The safety of our customers and our riders is always our first priority, so the question you have to ask is, ‘Can you exclude that this will happen again?’ And then we said that we cannot 100% exclude that this will maybe happen again, so we made the decision very quickly, within thirty hours, that we would inform our customers and also our pro teams that they should stop riding the bike until we have clarity about what’s going on here.”
Speculation as to the cause of van der Poel’s handlebar failure was rampant, fueled by the new bike’s novel three-piece cockpit design, which features drop sections that are detachable to either adjust the width or collapse the width of the bike for easier transport.
It didn’t take long to conclude that system wasn’t at fault — the failure didn’t happen there, after all — so then what was the cause? Had the bike been involved in a crash? Was the lever overtightened in preparation for the cobbles? Was the handlebar’s design somehow inadequate?
Ultimately, what Canyon determined is that while no singular event would result in that sort of failure, the combination of all of the above proved more than the handlebar could manage.
As many on social media had already determined — and as van der Poel’s father, Adri, eventually confirmed — the bike indeed was involved in a crash at an earlier race. Crashes are common in pro road racing, of course, but while it’s often standard practice for teams to replace questionable carbon fiber handlebars involved in some sort of major impact, that didn’t happen here. According to Canyon chairman (and company founder) Roman Arnold, van der Poel even apologized to him directly for creating “all this stress because I didn’t change the handlebar.”
It would have been easy for Canyon to simply chalk the bar failure up to the previous crash and issue a release warning Aeroad owners about the dangers of failing to replace a handlebar after such an incident. In fact, such advisories aren’t at all uncommon. But that’s not what happened here.
“The reality is that, in a crash, the handlebar has to survive at a certain level,” Kaiser explained. “That needs to be the target. It also needs to be robust enough to withstand this kind of misuse cases.”
In other words, while no carbon handlebar can be reasonably expected to be indestructible, the sort of catastrophic failure van der Poel experienced — a complete structural separation while riding — should never happen under any sort of typical conditions. Van der Poel obviously managed to stay upright despite the fact that this failure happened while he was riding in the drops on a cobbled section of the race, but remember that we’re talking about one of the most talented riders in the world; the rest of us likely wouldn’t be as fortunate.
“We went immediately into the analysis of this issue, and what we figured out is that, with our normal testing conditions, in our test laboratory, it’s not enough to more or less simulate the real world,” Kaiser said. “We were able to create this type of breakage in the lab by over-tightening the shifter fixation on the handlebar in combination with an exceptional load on the drop bar end, such as a road hole or maybe also a pre-damage from a crash.
“On Mathieu van der Poel’s drop bar, we were not able to measure the fixation torque after the incident; the handlebar was too strongly damaged. Our requirement is that a drop bar, as a safety relevant part, has to withstand these kind of combined load cases, even if they seldom occur.”
There’s a concept in failure analysis called fatigue damage accumulation. In essence, what it says is that for fatigue-related structural failures, it’s usually not any single loading event that ultimately causes the part to finally break, but the complete history of all the loading events that occurred up to that point.
In this case, the crash itself wasn’t enough to break the handlebar, nor was over-tightening the lever clamp or an unusually powerful rider traveling at race pace over Belgian cobbles. But when you add all three together, well, we all know what happened here.
Playing it safe
In reality, that confluence of events would be exceedingly rare for the vast majority of riders and racers, and indeed — at least according to Canyon — Van der Poel’s is the only handlebar to break out of about 1,500 new Aeroads out in the field (including both consumer and team bikes). But again, Canyon got lucky here, and even though this sort of failure was unlikely to happen to someone else, the company wasn’t about to bank on having luck on its side again, either.
“We were going here to the limit,” Arnold admitted. “And in the end, we had to learn this handlebar is better with all these circumstances to not save weight in this area. Even with misuse, it still has to work.”
The solution? Make the handlebar thicker (and a little heavier) in that area such that it could simply absorb more damage and, if it eventually came to that, fail in a less dangerous fashion.
“What we figured out in the end here is that, especially in the drop bar area on the front side where the shifter is positioned, that the wall thickness is not strong enough,” Kaiser said. “The solution to solve it is that we have already produced reinforced drop bars that are much stronger in wall thickness in the clamping area, and these are already passing the tests. Everything looks very good, so the idea is that we now go into mass production as soon as possible, and we want to exchange the drop bars from our customers as soon as possible. I think we can solve this problem for our customers in July, and they can go back to the bike, because I think this is the most important issue, and that we have a safe product.”
But what about that custom lever clamp? As several people have noted (myself included), Canyon uses its own multi-piece clamp for Shimano and SRAM levers, a requirement prompted by the original handlebar’s non-round cross-section. According to Kaiser, in-house testing hasn’t uncovered any problem with those clamps whatsoever, and he also pointed out that the same clamp has been used on “thousands” of the Canyon’s carbon fiber Grail CF gravel bikes, supposedly without issue.
Nevertheless, the new handlebar drops will go back to a conventional round cross-section, and Canyon will revert to stock clamps as supplied by Shimano and SRAM. While this might come at the technical cost of some fraction of aerodynamic efficiency, the long-term benefits in terms of standardization alone make the move worthwhile.
As for the adjustable handlebar design, Canyon has absolutely no plans to table it.
“[The failure] has nothing to do with that system,” Arnold insisted. “There’s nothing we can find wrong with it. It will be our standard system for many other bikes in the future, and we will continue to build with this because we think it’s a great solution. It’s just the reinforcement in these other areas.”
What about the seatpost?
Adding to Canyon’s pain were already-existing issues with the Aeroad’s novel seatpost design. Aero carbon fiber seatposts with deep cross-sections such as what’s found on the new Aeroad might be efficient in terms of drag, but they also usually bring with them a rough ride. To combat that, the Aeroad uses an unusual two-chamber stepped shape. Only the forward half of the post is truly structural, while the trailing half is essentially just a more flexible carbon cap to complete the cross-section.
The idea here is that the design makes the post more flexible than you’d expect for a smoother ride quality, which still offers the aerodynamic advantages of a deep profile. In practice, it works as advertised; the new Aeroad is, indeed, unusually comfortable to ride. However, because the clamp point is located deep inside the seat tube — meaning some of that flex happens below the top of the seat tube — a number of Aeroad owners have reported dramatic levels of seatpost wear due to abrasion (the frames thankfully seem to be holding up just fine).
Unlike the handlebar, this isn’t a safety issue since the seatpost would still fulfill its structural role even if the back half was worn through entirely. However, it’s obviously an unacceptable outcome nonetheless.
As it turns out, the alternative seatpost design that was published by CyclingTips recently isn’t the intended fix (although the company does plan to use it for the upcoming Endurace redesign).
In fact, Canyon isn’t planning on changing the original seatpost at all, and insists that the design and concept are sound. Instead, the company is working on a way to seal the area from water and debris.
“We saw we have problems especially in countries with a lot of rain,” explained Kaiser — one benefit of how Canyon’s direct-to-consumer business model provides the company with more reliable information about its customers. “You have some particles between the seatpost and frame, and it can happen that you have some abrasion in this interface. So we want to offer our customers a sealing solution, which is already being developed and tested. The next step is to go to the industrialization phase, and the plan is to have them ready at latest in the fall of this year.”
“Generally, yes, there is always some risk there,” he continued. “But you always have some micro-movements everywhere. The question is only what is possible and allowed, and what is not allowed, and you have to bring it to a level that you don’t see it on the surface and that’s acceptable to the customer. We have already run a lot of tests here, and what we see is that you really can prevent it. We developed a new test machine that can simulate all the dirt coming off from the tire, and we really stressed this interface a lot. It’s really working out very, very well. The sealing is not so simple, I have to say. I think when it’s ready, it’s not like just putting a hat on it or something on the top side of the frame. It’s a bit more complicated of a design. This will keep any dirt from getting in, and it will also reduce the friction between the seatpost and frame as much as possible.”
Canyon will, of course, provide new replacement seatposts when these seals are ready to roll out. But will this system really do the trick? And perhaps more importantly, will owners feel satisfied with it? How finicky will it be? What will it look like?
We’ll find answers to all of those questions soon enough, I suppose.
A bumpy rollout
Let’s get back to the handlebar, though.
Much confusion has surrounded Canyon’s recent announcement that all affected Aeroad owners should immediately stop riding their bikes, and that the company would be replacing handlebars. But how? When? By whom?
Arnold admitted that Canyon’s official communications on the matter to date have been hastily crafted, and not nearly as clear as they should have been. Initially, the implication was that all of those bikes would have to be physically sent back to Canyon headquarters in Koblenz, Germany, but in reality, that’s not the case. Although the company is still ironing out details, Arnold says the goal is (obviously) to make the process of making these bikes safe and rideable again as painless as reasonably possible for the people who have shelled out an awful lot of money for these things.
“Right now, we are thinking about three different offerings: offering number one — I’m not sure from a legal point — but we think most of our customers can even do it by themselves. We’ll just send the new bars to them. Second option is Velofix; we will pay for it, of course. Third option is to go to a local dealer, and we will pay for them to install it for you.”
While that certainly sounds a lot better than what was initially assumed, remember that Canyon doesn’t expect to have replacement handlebars until some time after July. To save you a look at a calendar, that’s more than three months away, and three months without the bike most owners were looking forward to using during some of the best riding weather of the season.
Canyon, unfortunately, can’t accelerate that timeline any more than it already is, and while Kaiser says warranties will be extended to include the amount of time owners weren’t able to ride their bikes, the company is also offering a financial compensation to help dull the pain: US$1,200 for Aeroad CF SLX owners, and US$1,500 for people who bought an Aeroad CFR (or the rough equivalent in other currencies).
Money always helps, of course, but that still doesn’t change the fact that buyers are stuck with bikes they can’t use. Canyon is a bike company; couldn’t it instead offer to provide bikes (of similar cost to the company as the offered payout) so that customers have something to ride?
That option might sound good in theory, but the reality of the current supply situation made that untenable. And so Canyon made a calculated bet.
“There are no bikes,” Arnold explained. “We all know bikes are the new toilet paper. I can not ensure that I have a bike for all of these customers. And when I think a little bit further about it, if it’s only about riding, most of us — at least for the usual SLX or CFR customer — have more than one bike in the garage. So this was the reason we didn’t offer it. We also discussed internally if we should instead give a voucher. Finally, we decided that there would be no voucher, because Canyon does not have too much stock on accessories. So we went for cash. Of course, the financial guys were asking us why we aren’t just giving a voucher to the customer.”
There is one catch, however: owners won’t get the money until after their bikes have been fixed. According to Arnold, this isn’t an attempt to delay the hit to the company’s balance sheet, but rather a way to ensure that there aren’t any leftover first-generation bars still floating about in the wild.
Multiplied by the 1,400-1,500 bikes out in the field, Arnold figures the cash payout, together with the process of retrofitting affected machines, will cost about US$2m.
Potentially dramatically inflating that number, however, is the prospect of customer returns.
Canyon normally offers a 30-day return policy for its bikes, provided they’re still in like-new condition. That period has long passed for most Aeroad customers, and virtually none of them will still be in “like new” condition. However, Arnold confirmed that if someone really insists on such, and because of the unusual circumstances, Canyon will take their bike back and refund their purchase price.
Naturally, though, Canyon’s hope is that the reasons buyers opted for a new Aeroad in the first place will still apply, and that owners will want to keep those bikes despite the early headaches.
However all of this shakes out, Canyon is about to spend a hefty sum to fix the Aeroad’s woes. The cash payouts, the manufacturing costs for the replacement handlebar and seatpost seal, the testing time and equipment expenses, the reimbursements to Velofix mobile providers and brick-and-mortar shops, heaps in freight charges — it all adds up a big pile of money.
Whatever the total financial cost of the Aeroad issues ultimately ends up being, though, that will likely pale in comparison to the long-term damage done to the reputation of the Canyon brand. This is no small recall or flaw, after all; it’s a critical safety issue warranting a total stop-sale and stop-ride situation at a unique time in history where affected owners have almost no good alternatives available to them, and at a time where social media provides everyone with a megaphone.
Now that the technical fixes are apparently well in progress, all Canyon can do now is humbly and profusely apologize, and hope for the best.
“I, and also Canyon, apologize a lot for what has happened,” Kaiser said. “It’s a really big disappointment and this is not to the Canyon standard, or what we want to deliver to our customers. We really do a lot of intensive testing, both in our test laboratory with test machines and also in the field, and we never saw this issue. In the end, what I have to accept is that, if you are an innovative company, sometimes you have to move some limits. In this case, we were a little bit over.”
“From the customer side, I really regret that these two things have not worked out like they should be, and I deeply apologize,” added Arnold. “I would not say the Aeroad was a failure; it’s still the fastest bike in the peloton with all the aerodynamic issue, so after we fix it for the customer, if a customer still wants to have it, it’s a great product from my point of view, and still the best product you can get on the market. But we have to make sure it’s 100% right.”
Ultimately, Canyon might consider this whole debacle a lesson more in communications than engineering. Although obviously undesirable, technical hiccups aren’t entirely unexpected with initial-release products. However, it’s far more rare that such an oversight renders something completely unusable — even if only temporarily — especially at this very pointy end of the price spectrum. Either way, the only thing that matters most from an end-user point of view is how well the company is perceived to be handling the problem, and how well it’s taking care of its customers.
That Canyon made some major R&D errors on the Aeroad is no longer in dispute, even if part of that error was simply underestimating how much a bike like the Aeroad might see stresses that fell outside of the predicted limits. At this point, it sounds like technical solutions are well underway, and only time will tell how well the related issues will shake out long-term.