How to evaluate different carbon technologies/frames? 2018 edition

[tylercheung]

I've always ridden steel frames, and I think I have a reasonable understanding of the technologies in them - tubing characteristics, parameters, geometries, etc, welding and brazing techniques.

But, it's hard not to hear about carbon fiber/CFRP frames and how they can do all sorts of things from a materials-science point of view that other materials can't achieve and all that. The thing is - it seems like CFRP technologies are changing dramatically year over year, and the typical marketing wording is unrevealing about exactly what has changed or what specific frame characteristic has gotten better - frames are always described as "stiff", "fast to accelerate" but also "more stable" and "takes bumps better" which just seems to be what every frame is supposed to do. There is a plethora of marketing about black box tech/science on tube shaping, compliance, etc. The best I can tell is from some of the press at say, Pez or Cyclingtips through factory tours of certain manufacturers, or by word of mouth. For a while, it seemed the Taiwan (Giant, Merida)/China has had advantages based on scale and R&D innovation, but there seems to be more recently the growth of at least partial or complete in-house CFRP manufacture of mid-size brands (Ridley, BMC), and now local, in-house CFRP shops, Parlee being the pioneer that has always done this for years, but also Sarto, Alchemy, and Allied.

I guess the question is - how can one evaluate the state of CFRP frames as of 2018, the quality of each manufacturer, the viability of the different techniques and technologies used, etc?

[dustyrider]

Ride them all!

[eddief]

with exactly the same dimensions and parts and tires and pedals.

Good luck.

[Kontact]

I don't think there are really "new technologies" being used. The standard carbon is just being laid up in better ways with more sophisticated designs.

Evaluating them is a little tricky since no one is in complete agreement about what makes a bike fast or efficient. You can point to BB deflection tests, but that only tells you how stiff the BB is in a certain way, not whether it makes the bike "better". And people really don't have a natural sense for what is best over any longer period of time, so they generally favor stiff tires, rigid forks and hard tires until they are told different.

So other than reading ad copy, I don't know how you would really evaluate anything about the actual construction of a CF frame.

[bicycletricycle]

From a ride quality stanpoint you just have to ride them.

From a safety/reliability/durability standpoint I think it is really important to consider the likely manufacturing variables/errors that coincide with different “technologies” regardless of their theoretical advantages.

Example 1.

Monocoque vs. tube to tube

Monocoque frames are theoretically better but the complex joints they require often lead to voids that go undiagnosed (almost no one X-rays or ultrasounds). These voids lead to lower performance fragile frames.

Tube to tube is a much lower tech way to make frames but the filament wound tubes “never” have voids and even the most basic reinforcement method at the joints (wetting the joint with epoxy and wrapping some unidirectional around it till it is a big ball and then grinding it down later) ensures good compaction and plenty of strength.

Example 2.

High modules fibers

The higher the modulus the more brittle the fiber. Technically a higher modules carbon is higher performance but it also severely limits the shapes that can be achieved, hi-mod fibers in tight bends can snap when being forced into shape and loose a lot of strength. Even if you just use them in the straight part of the tubes they can yield a wall so thin that even though it is strong enough under normal conditions it may be highly susceptible to punctures by debris or just normal “shop wear”.

Cool info from this guy-
https://youtu.be/kJl5V_KTRzc

[tylercheung]

Yeah, trying to distill things into different variables.

The Taiwanese bikes, i.e Giant/Merida + contracted models from the Europeans, all seem to be monocoque. Maybe they've gotten good enough, or maybe at $800 for complete bikes, they don't care as much.

The in-house layups at Parlee, Sarto, Alchemy, Allied all seem to be tube-tube; the modularity approach probably adds a few grams but lets them fine tune ride characteristics a lot better.

But yea, I guess you'd have to ride them all, but you'd have to find a good bike shop w/ a liberal test ride policy or spend a small fortune...

[Kontact]

Originally Posted by tylercheung

Yeah, trying to distill things into different variables.

The Taiwanese bikes, i.e Giant/Merida + contracted models from the Europeans, all seem to be monocoque. Maybe they've gotten good enough, or maybe at $800 for complete bikes, they don't care as much.

The in-house layups at Parlee, Sarto, Alchemy, Allied all seem to be tube-tube; the modularity approach probably adds a few grams but lets them fine tune ride characteristics a lot better.

But yea, I guess you'd have to ride them all, but you'd have to find a good bike shop w/ a liberal test ride policy or spend a small fortune...

I don't think most of those bikes are properly "monocoque", or one piece formed over bladders. The standard is really the OCLV method, where subassemblies are die molded and joined via tapered overlaps. This method looks like the old monocoque Kestrels, but is functionally more like a tube and lug construction where the lugs are pre-formed. The parts are slip jointed together.

[GonaSovereign]

Originally Posted by bicycletricycle

Tube to tube is a much lower tech way to make frames but the filament wound tubes “never” have voids and even the most basic reinforcement method at the joints wetting the joint with epoxy and wrapping some unidirectional around it till it is a big ball and then grinding it down later ensures good compaction and plenty of strength.

--> That's not how tube-to-tube joints are made. Grinding through fibers would negate the benefit of the material. You might be thinking of the fillets builders often create between the tubes before they wrap the joints with CF. The joints, made out of DP420 and similar, are sometimes sanded into shape prior to applying the CF.

Here's an example pulled from the internet: that's the epoxy fillet prior to CF application.

Originally Posted by bicycletricycle

Cool info from this guy-
https://youtu.be/kJl5V_KTRzc

-->Agreed - Luescher is a great source.

[Kontact]

Originally Posted by GonaSovereign

--> That's not how tube-to-tube joints are made. Grinding through fibers would negate the benefit of the material. You might be thinking of the fillets builders often create between the tubes before they wrap the joints with CF. The joints, made out of DP420 and similar, are sometimes sanded into shape prior to applying the CF.

Here's an example pulled from the internet: that's the epoxy fillet prior to CF application.



-->Agreed - Luescher is a great source.

No, he's talking about a method Calfee uses on odd joins and with their hemp jointed bamboo bikes.

Like this one:


After grinding they look like this, full of interrupted fibers by plenty strong due to compaction and shear mass.

[bicycletricycle]

I chose the worst tube to tube method to illustrate the point, grinding through those fibers is defintely harming structure but the amount of extra material present in these types of joints seems to negate these problems. (My evidence being not hearing of or seeing any joints like this fail, I’m sure they do but voids in badly compacted monocoque frames seem to fail way more often even though those frames are “higher performance”)

Lots of ways exist to reinforce tube to tube joints.

Originally Posted by GonaSovereign

--> That's not how tube-to-tube joints are made. Grinding through fibers would negate the benefit of the material. You might be thinking of the fillets builders often create between the tubes before they wrap the joints with CF. The joints, made out of DP420 and similar, are sometimes sanded into shape prior to applying the CF.

Here's an example pulled from the internet: that's the epoxy fillet prior to CF application.



-->Agreed - Luescher is a great source.

[GonaSovereign]

Originally Posted by Kontact

No, he's talking about a method Calfee uses on odd joins and with their hemp jointed bamboo bikes.

Like this one:


After grinding they look like this, full of interrupted fibers by plenty strong due to compaction and shear mass.

I had blocked that from my memory. Not exactly elegant. Calfee has done much better work...

[Bob Ross]

Originally Posted by bicycletricycle

Monocoque frames are theoretically better but the complex joints they require often lead to voids that go undiagnosed (almost no one X-rays or ultrasounds). These voids lead to lower performance fragile frames.

Captain Pedantic would like to point out that there's no well-documented correlation between voids in carbon layup and "lower performance" in bicycles. There is certainly the presumption that those flaws [sic] should lead to a lower-quality structure...but so long as the "performance" refers to how that finished structure functions in its intended role (i.e., as a bicycle) rather than as some ideal perfectly fabricated object, the data just isn't there. Carbon bike frames don't necessarily fail because of voids in the layup, nor do bikes without voids ride any faster, longer, or more comfortably than those with.

If it's any consolation, Luescher often seems to make the same presumption that there is correlation...but afaik never backs it up with any evidence.

[Kontact]

Originally Posted by Bob Ross

Captain Pedantic would like to point out that there's no well-documented correlation between voids in carbon layup and "lower performance" in bicycles. There is certainly the presumption that those flaws [sic] should lead to a lower-quality structure...but so long as the "performance" refers to how that finished structure functions in its intended role (i.e., as a bicycle) rather than as some ideal perfectly fabricated object, the data just isn't there. Carbon bike frames don't necessarily fail because of voids in the layup, nor do bikes without voids ride any faster, longer, or more comfortably than those with.

If it's any consolation, Luescher often seems to make the same presumption that there is correlation...but afaik never backs it up with any evidence.

A bike is a pretty complicated idea to talk about with simple ideas like performance being more or less on some scale.

But an individual element of a composite structure could certainly be looked at, and if it has voids it is either going to be weak in those areas (leading to failure) or it is going to have to be overbuilt in general to contend with the probability of voids, making it heavy and/or less supple.

So it doesn't seem absurd to say something with flaws in it is going to give up "performance" in use or longevity to a relatively unflawed example. It just won't matter much if it isn't already at the pinnacle of the technology because the cheaper stuff has already given up performance to the high end designs.

[fignon's barber]

Originally Posted by Bob Ross

Captain Pedantic would like to point out that there's no well-documented correlation between voids in carbon layup and "lower performance" in bicycles. There is certainly the presumption that those flaws [sic] should lead to a lower-quality structure...but so long as the "performance" refers to how that finished structure functions in its intended role (i.e., as a bicycle) rather than as some ideal perfectly fabricated object, the data just isn't there. Carbon bike frames don't necessarily fail because of voids in the layup, nor do bikes without voids ride any faster, longer, or more comfortably than those with.

I'm picturing Ernesto Colnago, cross section of C50 in hand, pleading for all to look inside his frame, while The Big Bike guys pull further and further away in revenue.

[benb]

Just figure out what fits you and ride stuff.

A huge amount of the hype is just marketing. There’s a lot less science and engineering going on than what marketing claims. Most of the smaller companies don’t even employ any engineers. The big companies do. There’s more tuning going into the frames from the big companies but the question becomes who did it get tuned for?

It’s like musical instruments or something. There are tons that are really good, but not all of them are going to feel right for you. The bike doesn’t go down the road by itself, it’s what you do with it. So all the marketing/engineering doesn’t matter compared to whether it feels good to you. And no one online can tell you what feels right for you.