Originally Posted by
sstorkel
Just like aluminum? Or modern thin-walled steel tubing?
I'm not sure if this was ever true, but it certainly isn't true with modern carbon fiber construction methods. At least for bike frames; I don't know about handlebars or other parts. A modern carbon lay-up uses multiple layers with different compositions and orientations. Scratching the clearcoat or cutting through a small number of fibers on the outermost layer isn't going to compromise the structure. If you somehow manage to cut through multiple layers, then you should probably start worrying...
I could be wrong, but, with regard to Aluminum, it seems the most common failure mode is work harding that leads to a crack before catastrophic failure occurs. While there are certainly other failures, it seems the majority are either near weld beads or in any of frequent high flex (drive side chainstay just aft of BB). These "usually" provide a crack, flex, creak, before letting go completely. I realize there will be plenty of stories of JRA when something let go. But, generaly speaking, the majority of aluminum failures provide some warning signs if you know how to recognize them.
With regard to carbon. I'd caution not to generalize all "modern" designs together. There are a lot of differing techniques in use and degrees of engineering. Everything from lugged carbon tubes, to glued together modular carbon sub frames. And, although some manufacturers refer to their frames as monocoque. I'm not aware of any that truly are. Layup schedules can and do vary greatly. The issue with carbon failure from scratches and chips is not the total number of threads that have been interrupted, but, that a stress riser may have been created. Image machining an inside radius from a block of aluminum. The longer the length of the radius on that inside corner, the less likely it is to develop a crack radiating outward. With the interruption of just a couple carbon threads the "crack" so to speak is already started and then propigates (sometimes very quickly) through the remaining material. This is enough of a concern that some manufactures prescribe the repair of even superficial gelcoat damage in order to mantain the integrity of some components.
It's not that carbon can't be immensely strong. Just that we need to be careful about selecting appropriate components. There is a reason why manufacturers have starting prescribing weight limits to carbon components.