The Ag2r Fiasco [Updated]

The Ag2r Fiasco [Updated]

If you have any relationship to social media then by now you have probably caught the video of the Ag2r rider who crashed in a time trial after he hit a speed bump and his bars separated from his bike. Fortunately, it’s not a horrific crash, like say the way Jens Voigt crashed on his face on a descent during a stage of the Tour, but it’s enough to make the entire front half of your body squirm.

The good news is that thanks to social media we know the verdict: people have decided that the bike maker and/or the bar maker are responsible.


In 2008 David Millar, then riding for Garmin, was in a two-man breakaway at the Giro d’Italia and as they approached the sprint, outside an ancient castle, the chain on Millar’s bike broke. He lost the stage, natch, and in his frustration he threw his bike away from the road. It made for the most spectacular video of that day’s stage. Garmin was sponsored by Felt and Shimano at the time and the throngs of bike racing fans around the world concluded that Felt was responsible and the Irvine-based company must be making some pretty crappy bikes.

This would be where I’d like to reiterate that the chain on Millar’s bike broke.

I think this illustrates the predicament Factor (the bike sponsor for Ag2r) found themselves in once that video began to circulate. Clearly, Felt was not at fault when the chain broke. And while chains do break on occasion, it’s so rare I would be hesitant to conclude that Shimano’s product line is somehow substandard. I’d be more inclined to wonder if something might have been amiss in the installation of the chain. Pro team mechanics, after all, have a hell of a lot of bikes to work on and very often stunningly tight deadlines.

I mention the Millar story because even before a representative from Factor had a chance to make any sort of public statement, I was reading a multitude of opinions, mostly on Facebook, but also on Twitter, about how shoddy the work from that company is. What made the social media kings such experts on the topic, I’m still not sure. People decided that the steerer sheared, that the stem broke, and a whole host of other alternatives, all of which were wrong.

Let’s get one thing out of the way, first: nothing went wrong with the Factor bike. I got that directly from Factor’s top man here in the U.S., Richard Wittenberg.

Here’s another detail not widely circulated: neither the rider nor the mechanic were part of Ag2r pro team. The rider was part of the U23 team, and the mechanic was brought in on contract; he was not an Ag2r team mechanic. Prior to the TT, the rider asked mechanic to add a single spacer to his setup to bring the bars up. Stack height on the Factor bar is variable by 60mm. If you watch the video carefully, you’ll see that the aero bars break away from the base bar. In adding the spacer to raise the aero bar, the mechanic decreased thread engagement significantly in the four bolts that hold the aero extensions to the base bar, so when the rider hit the speed bump the bolts ripped out of the base bar, where they thread in directly. This was a change made virtually on the start line, so he didn’t have the support of the service course for the team. And while we initially reported that the maximum stack height was exceeded, upon further inspection by Factor staff, that did not occur, but the mechanic should have used longer bolts to achieve full thread engagement.

Factor is a company that has made bikes for the best of the best in the industry. Their testing facility is so world class that other companies go to them to test their carbon fiber products. Wittenberg, Factor’s man on the ground in the U.S., told me, “Our number one concern in making any part is making sure it is safe.”

He went on to tell me that because their TT bike, the Slick, they sent a representative from Factor’s factory in Taiwan to the Ag2r service course to train the team mechanics on assembly techniques, torque specs and more. They have since increased the education they will give mechanics in the future.

Wittenberg also noted, “It’s a wake-up call to the industry about education. This isn’t an old Cinelli stem where once you know how it works you know how every other stem out there works.”

Factor has since released a statement on their site about the incident. You can read it here.

Had the mechanic known more about the bike, he would have known to tell the rider that inserting more spacers wasn’t an option. With bolts too short to thread all the way through the threaded insert in the base bar, that bump asked what threads were engaged to take more load than they were meant to.

So, for those really concerned with why the rider went down, it was, to use that now cliched phrase: a perfect storm of opportunity. We saw the result of a borrowed rider on a borrowed bike being worked on by a borrowed mechanic.


Update: Factor supplied us with photos of the parts in question.

This shot shows the spacers stacked up and gives you some idea how long the bolts need to be to achieve full thread engagement. The top 10mm spacer is the one added on the start line.

This shows the inserts in the base bar into which the bolts thread. This is a top view of that first, 4cm spacer. The bolt must pass through the aero bar mount, then any spacers before fitting into the inserts.

This is one of the four bolts that pulled out of the base bar. Prior to adding the 1cm spacer, the full 15mm of threads were engaged, but after the spacer was added, less than 5mm of threads were secured in the mounts.


, ,


  1. Colin

    The arm-chair quarterback comments on social media regarding incidents like this are always frustrating. It’s always funny when someone such as yourself actually gathers facts from the people involved in the situation to better understand what happened. Thank you for sharing with us.

    Very rarely do I ever look at these incidents as a representation of the bike or part companies involved. My experience has taught me that more often than not, issues seen out on the road are a case of user error rather than manufacturer defect. Of course that’s not always that case but it usually is.

  2. Shawn

    Hmmm… As a self taught bike mechanic with the advantage of an ME degree under my belt, it seems rather fishy that a contract pro mechanic didn’t realize the amount of thread engagement he was obtaining. Considering that four bolts engage to hold the aero bar in position, the mechanic must have been quite out of touch with the amount of thread he was obtaining. What I wonder, is that in the haste of bike preparation for the TT, that perhaps the mechanic over torqued the bolts and thereby weakened the thread strength. Add in a rough road surface and a sudden speed bump, and voile’, TT rider a la superman…

    Let this be a lesson to striving-to-be bike mechanics, lean in, open your eyes, and listen wisely to your local bike shop mechanic. Learn from their wisdom and the humorous stories of their mistakes…. and try not to change your bike setup minutes before the start of a race unless your really sure of what you are doing…..

    1. Nick Sanders

      I think another lesson is an exaggerated version of pleasing the customer in the moment at the expense of better judgment. At the start line, pro rider X asks green stand-in wrench to “just add a spacer.” I can hear the debate in my own head, magnified by the time pressure, career moment, eagerness to get’er’done, etc. Glad the rider was not more hurt, and hope the wrench got a wake up call and not a permanent exile.

      Nice investigative bit, Padraig!

    2. Author

      Have a look at the photos just added to the post that were supplied by Factor. Remember that this was a rushed adjustment made at what I’m told was the last minute. Because the bolts don’t visibly protrude, it’s hard to tell just how much thread engagement you’re getting. Your observation about not properly torquing the bolts is likely as well, but the problem, based on their own investigation, is insufficient thread engagement.

    3. SummitAK

      One issue contributing to the potential for this failure is that this is a mono-pedestal. So there are only two bolts used to hold the aero pads and extensions to the basebar not four.

    4. Pat O'Brien

      “When you rush a project to completion, you ruin what was almost ripe.” Tao te Ching translation by Stephen Mitchell.

  3. Chipps

    Nice to see some decent research in a bicycle article. It’s good to remind ourselves sometimes that we’re bike journalists and not just press release re-fluffers…

  4. Shawn (a different one)

    The new pics are confusing. I assume the 40mm “spacer” is threaded to receive the bolts. I also assume the aero bar slides over/around the “stack” and is clamped to the stack, like a stem is clamped to a steerer tube on threadless headset applications.

    Under these assumptions, if the pictured bolt threads into the 40mm spacer, then the bolt had at least 10mm of thread in the 40mm spacer. That should be adequate thread.

    But even if the assumptions above are wrong, the pics don’t make much sense.

    If, instead of threading into the 40mm spacer, the bolts are supposed to thread in to the flat “base bar”, then the pictured bolt, which is about 45mm long from base-of-head to tip of threads, is too short to thread at all.

    If the aero bar does not slide over the stack but is instead inserted into the stack and bolted down, then the stack will actually be larger than pictured, since the height of the aero bar must be added to the stack height. In that case, the pictured bolt is too short to even thread into the 40mm spacer (much less the flat bar beneath it) at all, unless the aero bar mount is less than 10mm thick, which doesn’t make much sense.

    More confusion: the bolt head in the last pic does not look like the bolt heads in the third-to-last pic, showing the assembled stack.

    I agree there’s no reason to blame the frame manufacturer for a component failure, but are these pics really helpful?

  5. Shawn

    Details on the unique mounting system, Pics:×400.jpg×400.jpg

    From this Bikerumor article, where the single-mast mount aero bar mount is reviewed:

Leave a Reply

Your email address will not be published. Required fields are marked *