Carbon Forks - No Way

In No Carbon for Me, Thanks, I laid out my problems with carbon fiber in general, as used for frames, forks, and components. In this post, I want to focus just on the forks. The fork of a bicycle is a particularly vulnerable part of the frame system, absorbing a lot of the force of many kinds of impacts -- often like the "front line" of an assault. In addition, the fork is like a large lever, which can magnify the forces put upon it. Many bikes today come with carbon forks -- including many bikes that have frames made of steel, aluminum, or titanium. To me, the notion of making the most vulnerable part of the frame, and a part that is particularly important for controlling the bike, out of a material as sensitive to damage as carbon fiber (especially when that damage can be so difficult to assess), is ludicrous.

In researching for this post, I happened upon this article in VeloNews. In it, Lennard Zinn answers readers' questions about carbon forks. In response to a question on the life span of carbon forks, Zinn got answers from several manufacturers. Here's one from True Temper:

True Temper’s . . . test is a ramped load, meaning the load is increased periodically until failure occurs. Starting at 180 lbs, the load is increased 45 lbs. every 5000 cycles. Every fork will eventually break. Strong forks will last more than 10,000 cycles with a load of 270 lb. But our minimum standard begins at over 15,000 at 315 lbs. for road forks and 18,000 for cross forks and tandem. But our production forks are stronger than that, often going into the 20-25K range and beyond at loads of 360-405 lbs.

Even I have to admit, that sounds pretty impressive. In lab tests, carbon beats everything -- even steel. Then again, there's more to the story. The rep from True Temper continues:

Obviously, crashes are uncontrolled events and it is not easy to guess what loading was applied to a component by the speed or violence of the crash. After any crash it is important to thoroughly inspect the frame and components for visible cracks, dents, and bends. An Alpha Q fork that has been damaged (usually evident as a crack) should be replaced.

The question anyone should be asking is what about the damage one can't see? Short of getting the fork X-ray'd (and even that might not reveal anything) how is one to know that a carbon fork that appears to have completely "shrugged off" an accident is actually sound? A steel fork will bend, or it may show a crack that will spread slowly. The carbon fork may show that crack as evidence of structural damage -- but it may not. Then days, maybe weeks, maybe months later, it can break suddenly.

Here's the response from Deda:

Carbon lasts longer than metal. Only love is stronger than carbon. Bonding is a different story. I believe that a good glue (epoxy) can last for 2000 hours of work, or about 800 days, not in continuous daylight, and below 35 Celsius. Whenever a carbon “part” has crashed, even if you cannot see a failure, if there is any reasonable doubt about having surpassed the elongation limit, the part must be replaced.

This isn't making me any more confident. Keep in mind, when they talk about "bonding" it goes beyond joining the fork blades to the crown, or joining the crown to the steerer (not that that isn't totally critical -- it is). The carbon fibers themselves are bonded with some form of epoxy just to hold them all together and give the tubing its shape. What does that mean, "a good glue (epoxy) can last for 2000 hours of work, or about 800 days, not in continuous daylight, and below 35 celsius"? I mean, now we're not even talking about crash damage -- we're just talking about too much sunlight, or temperatures being too hot or cold, breaking down the integrity of this very critical frame component. And 800 days? That's barely over two years! I can only hope that this is just some language barrier, or maybe I'm misinterpreting the answer -- because this doesn't sound so impressive.

I liked this answer from Look:

There is no limitation because carbon has a natural flexibility. It can be used a hundred years while maintaining the same stiffness.

That kind of ignores the whole "composite" nature of carbon fiber as it is actually used in the manufacturing of a fork, frame, or other component. It also sidesteps the real concern -- carbon itself may not lose stiffness, but what keeps it all from de-laminating, which is where the plastic/epoxy/resin or whatever it is "lets go," leaving the carbon fibers free to come apart? Remember, without that plastic "matrix" holding the carbon fibers together and giving the tubing its shape, it would be like trying to ride a bicycle made of string.

In further reading, I found this blog post by Isolatecyclist. In his argument for carbon forks, he cites the above-mentioned Zinn article from VeloNews pretty extensively. He then goes on to say:

If carbon forks are so prone to catastrophic failure, why are there no high profile carbon bike or fork class action lawsuits circulating on the internet? In a litigious society like the U.S. at least one carbon fork manufacturer would have been sued out of existence by now. Instead, many carbon fork manufacturers offer a lifetime warranty on their forks. Just from a business standpoint, if the failure rate was high, they would go out of business.

Now, I don't want to sound like some kind of conspiracy theorist whack job, but the lack of news about lawsuits in relation to broken carbon forks isn't that hard to explain. All one has to do is look at the examples from other, much larger and very successful industries, such as the auto industry. Somebody gets hurt because of a defect in your product, pay them whatever it takes to settle any possible lawsuit, then have them sign a non-disclosure agreement as part of the settlement. In terms of the bike industry, obviously I have no proof whatsoever that they take this route (and I'd like to think they don't), but it's what a lot of other industries do, so why wouldn't they? And if they did, the whole point of non-disclosure agreements is that people don't talk about them, so there is no proof.

And the thing is, there ARE lawsuits against the manufacturers. But since bicycling is still, in the big scheme of things, considered by many to be something of a "fringe" activity, the lawsuits aren't going to make big headlines. Nevertheless, here's an article about one such lawsuit: Carbon Fiber Forks on Bicycle Can Hold an Insidious Danger. In it, one can read the actual settlement (spoiler alert: the plaintiff won).

Regarding those lifetime warranties, considering all the disclaimers about crash damage and abuse, they don't have to honor the lifetime warranty if they can show that the fork was either in an accident, or the owner didn't follow specified guidelines for use. What kind of guidelines? How about these, from Trek, released after a string of fork failures?

1) Always use a torque wrench, and never over-tighten stem clamp bolts.

2) Always use spacers above and below the stem. Although less obvious than correct torque, a minimum of 5mm and a maximum of 40mm spacers under the steerer, plus a 5mm spacer above the stem are required. Riders should factor in these spacers when sizing their bike.

3) Use only the stem brand and model that came with the bike, because not all stems will work with carbon steerers. Often the lighter the stem, the less chance it will be compatible with a carbon steerer. Weight-relieving cutouts on the stem clamp and steerer interface can create stress risers.

So now a rider can't even change stems unless he uses the exact same make and model of stem that the bike came with from the factory. Otherwise the warranty is voided. Great. It's nice to know that carbon forks are so much better than steel. Such an improvement. And only love lasts longer.

Isolatecyclist (please don't think I'm attacking him -- he's not an expert -- just a blogger like me) also pointed out this interesting note about the reports of broken carbon forks:

Most of the "unexplained catastrophic fork failure" reports are located on personal websites or in bike forums. There aren't well-known consumer or professional cycling websites warning consumers about the "dangers" of carbon forks.

Of course, the "well-known consumer or professional cycling websites" aren't going to take a shrill "sky is falling" tone about carbon fork dangers (remember who actually pays them and keeps them in business) -- but the stories are there. The reports are more than just anecdotes from faceless, nameless nobodies on forums. How about this story from VeloNews: "Steered Wrong? Racers Concerned About Broken Carbon Steerer Tubes. In it we get to see Trek blame the owners for their broken forks because they were using the "wrong" stems (i.e., not the Trek-owned Bontrager ones) with their carbon forks. Then there's this one, also from VeloNews: Cervelo Founder Calls for Industry-Wide Fork Steerer Test. Here's a quote:

The cycling industry isn't doing enough to prevent fork steerer failures.

That’s the opinion of a man who would know: Cervélo co-founder Phil White, whose company suffered a large fork recall in 2008 following fork failures that resulted in multiple crashes and injuries.

Cervélo is not alone in that sorry episode.

The cycling industry has seen a rash of fork steerer-related problems in the last few years, from nearly every major brand sold in North America. Big brands with big test labs have experienced consumer injuries and expensive recalls that the manufacturers say resulted from a mix of engineering and manufacturing flaws. 

Clearly, the industry is aware of the problem. But they keep chasing the chimera of lighter weight, knowing full well that weight-obsessed roadies will pay a premium to be used as crash test dummies. Many of these riders will try to dismiss the reports as unimportant, unconvincing "anecdotes" from faceless nobodies who didn't take care of their bikes, or who crashed and ignored the obvious signs of damage. They'll remain certain that their own forks are sound. They'll hope or expect that engineering or manufacturing flaws, such as voids buried deep in between the layers in the carbon layup, only happen to "other" brands, but not the one they've chosen. But the problems seem to be happening with every brand.

Here's another sign that carbon forks are known to be problematic, and have been from the beginning. Look at the way the dimensions of forks have changed so much in the last few years. Note that steerers for steel forks were 1-in. diameter for generations. Then 1 1/8-in. after the advent of mountain bikes. It wasn't necessary for steel forks on road bikes to go larger, but when steerers started showing up made of aluminum, the additional diameter was needed for strength. But the junction between aluminum steerers and carbon fork crowns was problematic. Different materials with different properties. Add salty sweat into the bonded joint between those different elements and it becomes battery acid -- galvanic corrosion could eat at the joint. Then the manufacturers started making the whole fork out of carbon fiber, but the strength was still at issue. The "step" or transition between the crown and the steerer was thought to be a weak point -- so we now are seeing tapered steerers with 1 1/4-in. at the fork crown, and 1 1/8-in. at the top. They're still more prone to failures. It makes me wonder what we'll see next.

Of course, the ads will never say "We had too many of 'em breaking before, so now we're making the steerers bigger." But the strength issue and failure rate is why carbon forks are evolving in the way we're seeing. There isn't really a better explanation for it.

Can steel forks fail? Yes. There have been recalls on defective steel forks due to manufacturing errors. But it's a much rarer thing. Accident damage with a steel fork can be seen and diagnosed much more readily. Damage that goes unseen tends to develop more slowly, giving more time and more clues before failure. I'm not trying to be some kind of Cassandra, stirring up irrational fears of carbon -- but for me, the safety and reliability of a good steel frame and fork are worth a lot more than a couple pounds.