Replies to This Discussion

Permalink Reply by Jeremy Bloyd-Peshkin on October 5, 2016 at 6:52am

The real scoop is that it's complicated. But first, there's something to get out of the way about that video. The frame in question had been damaged in a crash, and so was not structurally sound when it met its fate. Now with that out of the way...

Carbon fiber is a tricky material to get right. It's very strong, but not in the way you typically think of when you think about strength in a material like steel. The things we need to look at when we say a material is "strong" is its density, its modulus (how stiff it is), its yield strength (how much you can pull on it before it stops springing back when you let go), and its tensile strength (how much you can pull on it before it snaps). When we look at the picture of a material as painted by those four factors, we see strength. I'm going to do some comparisons, so I'll lead off with a disclaimer: there are lots of steels and lots of carbon fiber composites. I'm going to take some averages and make some serious generalizations here. Tensile strength is the easiest way to visualize how much you can beat up on a material before it breaks. It's measured in psi, which is the pounds of force per square inch of cross-sectional area on the section you're looking at. 1000psi is abbreviated to 1ksi. A basic heat-treated cro-moly steel has a tensile strength of ~166ksi. Generic Carbon has a tensile strength of ~110ksi. That's lower than the steel, so why do we say it's stronger? Factor in the density - the same volume of carbon weighs ~1/6 as much - and you end up with a specific tensile strength 4 times higher. Again with the caveats though: the steel has that strength any way you pull on it. Carbon is only that strong in the direction of the fibers. Off that axis it is much weaker. But that's point of failure. What we care about is the point at which we put so much load on it that it deforms past the point of springing back. That's yield strength. Generic cr-mo sits at ~161ksi, generic carbon at ~164ksi. Those numbers are both pretty similar, but you'll notice that the yield strength of the carbon is much closer to its ultimate strength. Not much margin for error - exceed the yield strength and failure is right around the corner. The next thing - and what we care about the most for bicycles - is modulus. How much does a bar of the material deflect for a given load? How stiff is it? Our generic cr-mo has a modulus of ~27ksi, generic carbon ~40ksi. And again, it's 1/6 the weight. That right there explains a lot of why bike frame manufacturers love carbon. We crave stiffness and hate weight.

With these constraints in mind, we can paint a picture of the practical differences between carbon and steel frames. Steel frames are durable and springy. Carbon frames are stiff and light and very strong - but only when they are loaded in the way they are designed to be loaded. Carbon is not forgiving of being outside its designed use case. Much of this is due to the alignment of the fibers in the direction of anticipated load. But a lot is also due to removing any excess where it is not needed. Since a custom mold is made for every frame produced, it's easy to have completely arbitrary material distribution to put material only where needed. An entry-level frame will be overbuilt to tolerate some abuse, but a full race frame will have no excess. Carbon is also extremely notch sensitive - once chipped (or if a small crack appears) that area becomes very weak and will soon fail. Fun fact - titanium is also very notch sensitive. If a frame is damaged in a crash, it doesn't take much further abuse to cause catastrophic failure.

So basically this was a long-winded way of saying that carbon is strong and carbon frames are safe, but if you crash a purpose-built carbon track bike and then subject it to a lot of force in a way it was never designed to handle then it will break. That's exactly what happened in the video.

As a follow-up - is carbon strong? Yes. Is it safe? Absolutely. Is carbon right for you? Probably not.

• ▶ Reply

Permalink Reply by David P. on October 6, 2016 at 8:40am

This is an excellent response and is correct. carbon-fiber composites are just another material, and you design around the constraints of the material in the same way that you do with steel, aluminum or titanium.

• ▶ Reply

Permalink Reply by daWillinChi on October 7, 2016 at 2:59am

What Jeremy said ^^^ :)

Truth is all materials have a shelf life. Some maybe longer than others and some can withstand damages better than others as well.

If I were a racer, maybe I'd buy and ride carbon for a year or two at the most for the performance and light weight. But I wouldn't spend too much money on it because I'd want to get rid of it after a couple of years, at the most. I also don't want to sell it to other people without fully disclosing the age of the bike and every little accident/scratches/fall/etc - thus further lowering the price. With that said, I also wouldn't buy used carbon bikes, ever.

I would get carbon forks, on the other hand, because I know I can afford on replacing worn ones and won't cost an arm and a leg. I'd probably just send the used ones to the manufacturers for recycling instead of selling.

• ▶ Reply

Permalink Reply by Yasmeen on October 6, 2016 at 3:23am

I love my carbon frame road bike. I prefer it over my other bikes for long rides because it's light and easy to ride. My next road bike will either be carbon or titanium. I go back and forth about it.

• ▶ Reply

Permalink Reply by curt(is) locke on October 6, 2016 at 5:27am

Jeremy Santucci, loved him in "American Psycho."

• ▶ Reply

Permalink Reply by Juan 2-3 or more mi. on October 6, 2016 at 6:28am

Plastic has a markedly shorter lifespan than metal. Those frames use various 2 part mixtures of resin (with expiration dates) to make carbon fibers stiff.

Good for racing at the moment but not to be ridden for a lifetime.

• ▶ Reply

Permalink Reply by Jeremy Bloyd-Peshkin on October 6, 2016 at 6:31am

Yes and no. The resins do have an expiration date before curing, as the reaction that cures them won't go off properly after that time period. They are stable after curing though. The main difference in terms of lifespan is that resins will be weakened by exposure to very high temperatures and UV light, whereas steel will not. Painted and cared for, I see no reason a carbon frame shouldn't last a lifetime.

• ▶ Reply

Permalink Reply by ad on October 6, 2016 at 8:02am

True.  I ride steel, and it's nice to know I could keep riding it forever if I take care of it, but many people (myself included) end up picking up a new bicycle well before a carbon frame would ever reach its shelf life (which I've always heard placed around 10 to 15 years based on the older frames, not sure on the newer ones). For me, it's always the itch to try something new, which sets in maybe every 5 to 7 years or so. 

The sad thing is the carbon frame just dies when it wears out (or reaches its supposed shelf life), rather than ending up with a second life under someone new (like many old steel frames).   I've bought 20 year old steel frames.  I wouldn't buy a 20 year old carbon frame.

But that's not to say you can't get your moneys worth out of carbon---even non-racers.  Many of us aren't going to be riding the same bicycle the rest of our lives anyway.   

• ▶ Reply