Why is carbon fiber inherently weak? Or is it?

Question

This question got me to thinking: If it were possible for me to use a carbon frame, I would, but the expense and my riding style keep me using steel and aluminum. (I like hauling stuff on racks and I'm not a skinny guy.)

I'm looking for a physical reason why carbon is a weak, fragile material, suitable for light bikes that will be treated delicately. Keep in mind, they make airplanes out of this stuff!

Is there some reason carbon fiber has to be treated with kid gloves? What is it about the material that resists being light and strong? Or, perhaps, is the weakness of carbon a myth, and it's all in the way that carbon bike frames are built at present?

Answer 1

Carbon fiber isn't necessarily a "weak" or "fragile" material. If you had a tube of the same diameter and thickness of typical CF as a typical steel frame tube, that CF tube would be extremely strong and durable.

Metals like steel and aluminum are isotropic materials. That means their mechanical properties are identical in all directions. If you have a cube of steel, it will respond the same way regardless which direction you pull or push on it.

Carbon fiber is a composite material. It consist of tons of little bundles of the fibers held together with an epoxy.

A block of steel is, well, like steel, but carbon fiber is like a big bundle of straws glued together. In one direction, it is extremely strong, but if you push or pull to the side, it will collapse. In that one dimension where it is strong, it is vastly stronger than steel. However, in other directions it's rather flimsy.

So, engineers have been able to exploit those properties in bicycle frames. In a bicycle frame, the vast, vast majority of the forces are primarly along a single dimension. They can make tubes thinner and lighter yet still retain the desired strength and stiffness.

So, there is no mechanical reason that you couldn't build a fully loaded touring bike or something like a Salsa Fargo with a carbon frame, and it could be just as tough and durable. And it would probably be lighter than a steel or aluminum frame. But the reason it isn't done is because of the market. Carbon fiber is an expensive material and difficult to work with, and its mechanical properties are best suited for when you demand very light applications.

When you build a steel framed bike, when you get the tubes sufficiently strong enough along their length, that because of steels isotropic properties, you get the lateral strength for free, the strength to resist things banging into it, withstanding crashes, etc.

In a carbon fiber frame, you don't get the strength in the other dimensions unless you choose to design it in. In carbon fiber bikes, where weight is a serious concern, the engineering decision has been made to not make the frames strong in those areas. They could do so, but they choose not to because its not necessary for the bikes intended purpose.

When you build a heavy loaded bike, you lose a lot of the carbon fibers advantages, and so it would be far more economical to use steel or aluminum. Especially when throwing a couple filled water bottles in your pannier nearly exceeds the weight savings.

Answer 2

First a disclaimer: most of what I know about carbon fiber fabrication comes from aircraft, not bicycles. Also note that carbon fiber is not the only composite that gets used -- just for one alternative, Kevlar fibers can be useful as well (Kevlar is stronger, but also more flexible than carbon).

Carbon fiber is strong, but does not respond well to point stresses. This is largely because it's basically cloth (woven out of carbon fibers). If you put a lot of stress at a single point, you're putting that stress on only a few of those carbon fibers. While the fibers themselves are extremely strong (for their weight), the bonding holding the individual fibers together is much weaker. For comparison, think of the packing tape that has fiberglass fibers running along its length. The fiberglass itself is really strong, but the strip of plastic and "goo" holding them together is a lot weaker. Although the details differ, the same general idea applies to carbon fiber as well.

The exact strength depends on direction as well. As I said above, carbon fiber starts out as a basically threads that are woven into cloth. The cloth is then impregnated with some sort of epoxy (the exact epoxy used varies with the application), laid up in a mold, vacuum bagged¹, then baked to harden the epoxy. You can get the cloth in various different weaves, some with the same amount of carbon fiber running in each direction, other with (say) 80% of the carbon fiber in one direction, and only 20% in the other direction. At a guess, most of the CF used in a bike frame is probably somewhere closer to the latter variety, with most of the threads running along the length of a tube, and considerably less running around the circumference of the tube.

As long as we're at it: carbon is also about twice as strong with respect to being stretched as being compressed. You'll typically have around twice as many plies where it's primarily subjected to a compressive load.

¹ Vacuum bagging means a big plastic bag is placed around the mold and laid-up cloth, and the air sucked out. Air pressure on the outside holds the layers of cloth tight together to (try to) ensure that when they're baked, they act as a single layer, not separate layers. This has little effect on the strength when subjected to stretching, but a huge effect when subjected to compression or bending.

Answer 3

Carbon fiber is a very strong material, but like any material it does some things better than others. From Wikipedia:

Carbon fiber is very strong when stretched or bent, but weak when compressed or exposed to high shock (e.g. a carbon fiber bar is extremely difficult to bend, but will crack easily if hit with a hammer).

Considering that a carbon fiber frame can support the weight of a rider plus all of the forces that a rider adds (which can exceed several times their body weight) it is by no means weak. All this for less than the weight of a comparable aluminum or steel frame.

But certain types of forces -- like sharp impacts -- can damage the fibers and epoxy weakening the material, something that is less likely with a metal. And a small clamp can crush a CF tube, given enough force (you can do this with thin-walled aluminum tubing too but it takes more effort).

Answer 4

I think it's also worth pointing out that while carbon fiber can be laid up to be plenty strong, it's not at all ductile, like steel or (to a lesser extent) aluminum. You can put a pretty good sized dent in a metal frame and still ride it home, but if you put a dent in carbon fiber you've probably compromised the whole tube to the point that you should probably not ride on it. It's just a lot more brittle, so deformation means breaking, where in metals it usually means something's stretched or compressed, which does comparatively less to hurt structural integrity.

Answer 5

Bit late to the party but here's my ha'penneth: As noted above a common manufacturing method of CF frames involves "laying-up" multiple layers of resin impregnated fibres of differing orientations to optimise the strength characteristics according to the expected loads and required performance of the frame (e.g. rigid vs supple/flexibe). In this sense CF may be more precisely tailored to a set of requirements for the lightest weight. As with every engineering problem there are compromises. Each layer is essentially two dimensional (think x and y axis for a flat sheet), the third dimension, thickness (think z axis) is just the accumulation of layers of fibres but does not have any fibre strength per se, only strength from the resin matrix that holds all the fibres together. So it is through the thickness of the material that CF composite structures are weakest. And a common mode of failure is known as delamination (the bond between layers fails). This can happen from a blow to the surface and any delamination within the layers will not be externally visible. Only scans can detect the extent of any damage - the low-tech method involves tapping the surface and listening for any changes in tone of the taps - it does requires a trained ear and is less obvious for the layman to differentiate between a change in tone due to a delamination versus say a change in the underlying layup (extr layers near joins etc...).

Delamination is the weak spot of CF frames and why, in my opinion, they may be described as "strong" but NOT "tough" or "resilient to damage". Since any old bang could jeopardise the strength of the frame and led to an unexpected sudden catastrophic failure. Metal on the other hand gradually yields when overloaded - so sudden failure (if correctly designed) is less likely to occur.

So the big question for me has always been - if I crash a CF bike how will I know that the fame still has structural integrity.

I speak as a cyclist and engineer who specialised in my early career in composite and bonded materials. The answer to the risk of delamination lies in composite materials where fibres also run in the z (thickness) dimension. This may be acheived through "knitted" fibre structures where fibres link/lock the layers together - the dry fibre "knit" is then held in a mould and liquid resin injected and cured. As far as I am aware no manufacturer yet uses this technique (costly - military/aerospace budget type stuff). They continue with the traditional lay-up of pre-impregnated fibres method. Some manufacturers talk of "weaving fibres together" from one tube to another in a bike frame but I don'y think this is the "knitting" through the layers of a more advanced manufacturing technique.

Answer 6

I don't actually know the full details, but I know Carbon Fiber tends to be strong and flexible in some directions, and not very strong in others. So when you build a frame out of it, you can align it just right so that the frame is bendy and absorbs shocks in the ways frames are supposed to work, but if you apply the wrong pressure to it (say, drop it sideways onto a concrete curve), it might crack.

But, as was perhaps made clear by my previous question, I'm not actually sure :)