Heading out the door? Read this article on the new Outside+ app available now on iOS devices for members! Download the app.
Canyon isn’t the only WorldTour bike supplier to have suffered a catastrophic failure recently. Several weeks ago, at the Omloop Het Nieuwsblad, Israel Start-Up Nation rider Tom Van Asbroeck was pictured on the side of the road, the stem and handlebar of his new Factor Ostro VAM aero road bike awkwardly dangling off the front of his bike after the steerer tube sheared off. Like Alpecin-Fenix rider Mathieu van der Poel, Van Asbroeck thankfully (and somewhat miraculously) didn’t crash.
Shortly afterward, team riders were quietly switched from the Ostro VAM to other Factor models that had a longer and more proven service history, including the older O2 and One. Factor issued no public statement or explanation at the time, with the company instead opting to investigate behind the scenes until they came to a conclusion.
After a long delay, Factor director of engineering Graham Shrive described to CyclingTips what Factor believes was the cause of that steerer tube failure, and how it’s being addressed.
As is almost standard practice for high-end aero bikes these days, the Factor Ostro VAM features a dedicated cockpit with fully internal routing for the control lines. In the case of the Ostro VAM, those lines feed through the bar and stem before taking a downward turn along the front side of the steerer tube (and inside the upper headset bearing) before making their way into the frame. In order to create sufficient space for all of this to happen, the Ostro VAM uses a D-shaped steerer tube profile instead of the conventional round one.
A special aluminum insert then slides into that space so that the stem has a round cross-section on which to clamp, and a special D-shaped headset compression plug is used in place of the usual round one. Both of those are a generous 7 cm-long, which, in many cases, should extend down past the upper headset bearing and — at least in theory — provide more than enough reinforcement to prevent a catastrophic failure from ever happening.
Except that it did.
According to a recent post on Factor’s web site, Van Asbroeck “encountered an issue with his bike which ended up with his steerer tube being broken inside the frame after striking a curb,” — and similar to what Canyon conveyed to me during a recent interview, none of Factor’s in-house testing or previous team rider experienced revealed any issues in this area.
“At no point in our development and testing to date, on either the Ostro or the integrated O2 VAM, had we seen a failure of this type, including the 2020 Tour de France where riders like Andre Greipel rode the Ostro,” continued the statement on Factor’s site. “Therefore, it took some doing to ensure that we could repeatedly replicate the failure in a controlled environment, and then assess what the root cause of it was.
“Impact followed by fatigue tests are relatively common in the industry (ISO has the reverse steerer impact followed by fatigue and a steerer torque test, for example), but in this case, we were a bit flummoxed as the steerer system had passed our internal standards which go well above these minimums with no issues.”
Upon the conclusion of Factor’s investigation, Shrive says the issue boiled down to out-of-spec compression plugs, not an inherent problem with the carbon fiber steerer tube itself.
“We had a batch issue with our compression plug that saw the outside surface receive a clear anodize coating over the specified sand blast to white metal,” he explained. “This caused the surface roughness to be too low and led to preload issues with the headset. The team responded at camps and early season races by torquing these progressively higher, and finally by bonding in the plug while tightening it to ensure the headset stayed tight.
“This outward force caused a discontinuity in the fiber path and a distortion in the tube that led to Tom’s failure by concentrating the compressive bending stress at the point of discontinuity, instead of allowing the steerer to more uniformly distribute that stress over the length of it. I dealt with a similar issue to this with [UK Continental team] One Pro on the S5 in about 2015 [Shrive was formerly the director of engineering at Cervelo — ed.] when they used expansion plugs against recommendations and sheared their steerers.”
Compounding the issue was an unexpected change in the plug’s taper geometry that altered how it expanded inside the steerer tube. Instead of exerting uniform force along the inner wall of the steerer tube, the plug was expanding more at the bottom than at the top, which Shrive says made sense given that Van Asbroeck’s steerer tube failed right at the lower edge of the plug, not at the split ring on the upper headset bearing where you’d normally see the highest stresses.
A closer look at the aftermath suggests other discrepancies on Van Asbroeck’s bike relative to the intended stock Ostro VAM setup; discrepancies that Shrive didn’t address directly. Whereas the design of the standard compression plug results in a compression bolt that’s offset slightly from the usual location — similar to what Ridley uses on its F-Steerer design — the cap on Van Asbroeck’s bike has a bolt that’s clearly concentric with the headset like what you’d find on a bike with a round steerer. And as for that long aluminum reinforcing insert that is supposed to be placed on the front of the steerer tube and securely clamped by the stem, it’s nowhere to be seen, although it’s possible that it’s obscured given the image angle.
As for Israel Start-Up Nation, why has the team been off of those bikes for so long? Doesn’t all of this mean Factor could have just supplied new plugs and called it good? In theory, that could have potentially been a possibility under different circumstances, but in an effort to squelch the headset issue once and for all, team mechanics eventually bonded the plugs into the steerers, and since there was then no way to determine which plugs were overtightened at some point and which ones weren’t, Factor decided to just replace all the forks to be safe.
“The team is off the bikes because the bonding left us unable to change the plug, and even if the forks were not bonded, we’re worried they could be permanently damaged after an overtorque occurred, leading us to make the decision to replace all of the team forks,” Shrive said. “With having to supply new forks, anyway, and having pretty well zero stock with the high demand, we built the replacements from scratch and bonded in the M6 preload thread ourselves at the factory in order to totally, completely, and visibly deal with their preload concerns, which they are understandably still worried about.”
Team riders will apparently be racing on the updated setup within the next few days.
Ok, but what about everyone else?
Speaking of regular riders, this is obviously an issue that is potentially affecting everyday customers, too.
“[Factor company owner] Rob [Gitelis] and the sales team have been working with any customers having preload issues to get them the revised plug with the white metal surface, but it’s not a widespread issue with consumers,” Shrive said. “For example, I’ve been riding the clear anodized plug with no preload issues since last year. What we are finding, though, is that there are a number of riders unclear about the setup requirements, which I have to take on the chin as I could have been more proactive with making some more specific manual illustrations. In most cases, with guidance from customer service, issues people are having are cleared up quickly.”
Factor has recently — just within the last few hours, in fact — published a rather lengthy blog post to its web site on this topic, detailing the investigation process used to “replicate and repeat the failure mode in the lab and isolate the variable affecting the outcome.” In short, the main message Factor is trying to send through the post to consumers is to stick with the recommended torque for those compression plugs no matter what, and that Factor will supply new ones at no charge if needed.
“We have had limited reports of Factor riders struggling with preload but for anyone that does report this we are supplying the replacement plug to them immediately via their shops, distributors or directly,” read the post. “However, at its root, if there are no preload issues, there should be no overtightening, which will avoid any further issues. This all being said, there are many bikes happily in use without any preload problems, as the batches prior to this troublesome one followed the original drawings.
“If you find yourself unable to keep preload on the headset of your Ostro or VAM, please contact Factor for a replacement compression plug, but do not attempt to resolve the issue with overtorque.”
The fact that the company is being particularly cautious with team-issued forks by permanently bonding threaded inserts in place nevertheless raises another question. If Factor is making the decision to be extra-conservative here, should everyday riders be similarly cautious? After all, Cervelo — in partnership with then-fork supplier Alpha Q — pioneered the use of bonded-in steerer tube inserts as a way to provide better “resistance to clamping damage due to over-tightening of the stem,” and given Shrive’s long history at Cervelo, you’d think that might have been presented as an option here, too. At least for now, that’s not on the cards.
There’s another unanswered question at the moment: if you’ve been having issues with the headset loosening on your new Ostro VAM and have already overtorqued your plug, the situation might be a little more complicated. Although Factor says no consumer steerer tubes have broken in the field, that doesn’t mean some consumer steerer tubes might not already be damaged — and given what happened to Van Asbroeck, it seems reasonable to think that a replacement fork might be in order there as well. However, who is responsible for the cost and whether you should continue riding that bike is a matter you’ll likely want to take up with Factor customer service.
In the meantime, we’ll continue to dig for information from Factor and will update this article as needed.
More information can be found at www.factorbikes.com.