The root problem with certifying carbon components is that there are infinitely more ways to screw up a part than just using cheap fabric, or a poor grade of yarn, or a bad resin system. This is one of the reasons that the aerospace industry has taken 15 years to print MIL-17 handbook, and now that it is out, we're still arguing about it. The reality is that poor layup technique, misalignment of unidirectional fibers, fabric wrinkles, or poor compaction could all lead to parts which are significantly weaker than design, much more so than using even the lowest grades of fiber. With otherwise perfect processing and technique you can find incremental improvements through using more exotic fibers, but the reality is that a part made with MR50 fiber (~$65 per pound) containing a wrinkle will likely be weaker than the same part made with T300 (~$15 per pound) containing no wrinkles. This is why aerospace has been so slow to really do large production with these materials. It takes longer to make the parts, and then every part has to be inspected 100% before shipping to ensure no errors were made, and when errors are made the entire part has to be scrapped and cannot be re-worked, unlike many metalic parts which can reworked in numerous ways to bring them back into spec, or re-heat treated, etc.
Furthermore, full design testing and validation cannot guarantee that future parts will be manufactured properly, so as an example with our handlebars and stems we have passed the most stringent testing at EFBe laboratories in Germany for fatigue and strength both, but the lab can only verify parts they actually test as every part is considered to be different. So we can validate the design, and through strict controls can control the production of the parts, but it's not so simple as with aluminum or steel forgings where you can test one piece from a heat treat lot and certify the batch as each part is made individually, and even though that's easy there is very little of this actually done in the cycling industry. This is similar to testing a weld, you can verify the skill of a welder, but that's not to say that he won't drink too much coffee and make a crap weld once in a while, or to say that some other issue like contamination can't creep into the process and comprimise the design. So you can validate a batch of tubing for example, but you still need to inspect every weld 100% to validate an individual frame as any given weld could have had an overheat or contamination issue which would not be present in 100% of the welds.
Lastly, with many parts and designs, there will exist no good non-destructive testing regime, so the only way to determine how strong a part will be is to actually break it... obviously this can't be done 100% or we wouldn't have anything to sell. So the next best thing is to batch parts by individual employee and material lot and perform statistical testing of 2-3 parts per 100 to ensure that the employee is producing consistent parts, again, this method cannot pick out a single bad part, but can pick out a process or individual that has strayed from design intent of the production. This is another reason that it is hard to be in the carbon business, if one person breaks one part, they are all over the internet talking about what garbage the part and the company and the engineers are, but the reality is that 99% of the time is is a very minute manufacturing glitch which led to a part failure and nothing more. Not that there aren't bad designs out there, there are, but even those are generally heavily overbuilt to satisfy some basic strength requirements, so your part may be heavy due to poor design, but it generally won't break because of it.
And bunnyman is right, 70/30 is about as high carbon content as you can reasonably acheive, as carbon content increases past this parts will actually get weaker as you will have resin starved areas within the part which act as failure zones. 50/50 is about he limit going the other direction, where you just lose too much stiffness and strength and are too resin rich to be producing a high performance product. And for the record, we use 5 grades of carbon specifically placed for strength, stiffness, impact toughness, hole reinforcement, and brake friction, so we no longer state what we use in the wheels, but the lowest strength/stiffness of all these would be equivalent to T800H/S from Toray and 1 of these materials used considered 'defense critical' and not available as a raw material outside of US borders.
cheers
http://www.SILCA.cc Check out my podcast, inside stories from more than 20 years of product and tech innovation from inside the Pro Peloton and Pro Triathlon worlds!
http://www.marginalgainspodcast.cc