• Andy Logan

    My life span on carbon frames isn’t very good, I am on my 5th bike in 5 years. I admit mainly from crashes while racing. Perhaps I am the outlier

    • Jayson

      Perhaps stop crashing then.

      • lol

        Says some guy that doesn’t race

        • Jayson

          Who says I don’t race? I’ve been racing since I was 14 years old and I’m now 44. You do the math (if you can)

          • Mathemajician

            I got 30

            • Jayson

              A smiley stamp for you then.

              • awesometown


                • Jayson

                  Lay off the redbull pal.

          • HeTotallyDoesNotRace

            You realize putting on lycra and commenting on the internet isn’t racing right?

            • Jayson

              Nope. News to me. Thanks captain obvious.

          • Wake

            Then perhaps you should explain why the pro racers crash and you don’t? The very top pros riding alone crash. but you don’t. Can you explain that for us?

    • Not at all. Crashes are perhaps the biggest x-factor influencing the lifespan of a carbon frame. Nothing can be done to avoid a crash except staying off the bike.

      • Andy Logan

        Last one wasnt from racing, had someone in a car pull out in front of me at 60kph. I can assure you an impact like that, no bike would survive that regardless of what it was made out of. E.g. my front forks snapped and the frame absorbed a lot of the impact but shattered at the bottom bracket from the accident.

        I suspect if you never raced and didnt have any crashes then carbon would indeed last longer, but for me, it seems that a lot of the impact that a bike takes in a crash while racing is often side on for example, I am not expert and this is purely anecdotal evidence but the frames that I have damaged from crashes in racing it is often the seat or chain stays that have cracked due to a side on impact for example.

        • jules

          I had someone ride into the back of me in a race (my fault) and the downtube buckled. I wasn’t even particularly hurt but the frame was a gonner. Their weight as they fell on me seemed to do it.

        • Toad734

          No, no other bike would have survived that crash but let’s say that bike and then replacing it cost $7000+, you could have been at less than $4000 for steel or aluminum…so there’s that.

          • Dan

            you can get A good carbon bike for 4000

            • Toad734

              Yes but the point was the initial bike and then the replacement is going to be well over $7000. If I total my steel frame I’m out $1400 for the old one and $1400 for the new one…Of course, you won’t know I totaled it because I am like 400 yards back at the bottom of the hill behind everyone with a carbon frame…And a Specialized Tarmac comes in at what, 19 lbs…Bianchi Intensio at 18 lbs, Lapierre Sensium 18.8 lbs…I probably can’t get my steel frame below 20 lbs but for some of those low to mid range priced carbon frames, you really only save 1-3 lbs. So $1600-$2000 for 2 lbs…And it’s far less likely that I will have to replace my steel frame as well.

              Don’t get me wrong, I wish I had a carbon frame every time I climb any sort of hill at all but for some carbon bikes that are on the heavier side, you might be better served with a good steel frame and spending the extra money on better wheels than buying the carbon frame, spending more and riding the stock wheels.

              There are some exceptions obviously, A Cannondale Synapse at about 15 lbs and $3000 is sort of a no brainer….If you are just going by weight vs cost.

      • OGS_SD

        I ride quite a bit, 200 miles a week, year-around, in So-Cal. I race but mostly not cries, longer road races etc.
        I have crashed once this year, hard crash at 27 mph and my carbon bike frame survived the impact just fine.
        I had had more minor spills before, but nothing major. I had my first carbon frame for 7 years before I got my current bike bought 2 years ago. Mostly because of lower weight, better performance and aesthetics. I still have my old frame and I ride it on occasion. I crashed on it a bit, and probably mistreated it – but nothing too major.
        7 years or 9 years is not lifetime, but I believe that with proper care and barring a major catastrophic bone-breaking accident a modern carbon frame can last a lifetime.

        I think maybe a proper analogy is your skeleton. It weighs about 15 lbs, the weight of a very good road bike, and it is not that delicate – it can survive a lot of bumps and tosses and be just fine. It doesn’t “disintegrate” when exposed to light, heat or bumpy roads. I have yet to break a single bone in my body and I am 39 (knock on wood) and have been doing all sort of sports activities for about 35 of those 39 years. Is it possible to break a bone? yes, for sure. Many people do it in many ways, some do it multiple times a year, in dramatic fashion. But bones should serve 99% of population just fine. If you had steel or titanium frames, you would break it too.

        • Ragtag

          Which frame were you riding?

        • Gundhi

          A human skeleton for a 73 kg man weighs about 8 lbs or 3.6 kg, not 15 lbs (measured using a DXA Scanner). Our bones also start to demineralise after about 40 years of age which is partly why we get frail with old age. Unfortunately our skeleton is not like carbon fibre and won’t last forever!!

        • jules

          how and why are you exposing your skeleton to light?

      • slartiblartfast

        Agree with this in that both times I have totalled carbon frames were during crashes. One of these was going straight into the side of a 4WD which pulled out in front of me. The top tube, seat tube, chainstays and seatstays all broke but these crumple zones actually limited the damage to me. Another carbon frame was damaged through being overclamped at the LBS but this was able to be repaired. Most likely the stiffness and lightness which we value in carbon frames limits impact resistance.

      • Wake

        Actually it is loading on the high stress areas that are every bit as fatal to carbon fiber as crashes. Considering the condition of the roads in California I’m surprised we don’t see more of it. On the other hand, in our group of 20 or so riders there are only three of us that descend at high enough speed to put significant stress on these joints. I rode on the flats at 28 mph for two miles on one occasion and at 28 mph on another occasion for 5 miles but in both of those cases the roads were almost new and hadn’t any potholes or cracks. So hard strikes of ANY sort can cause the start of catastrophic failure of CF framesets.

        Once the resin is cracked it begins the process of delamination. This is usually inside of the laminate and invisible so that you cannot discover it by visual inspection methods. At some point enough of the resin around the carbon fiber is loose enough that the carbon fiber which is nothing more than a series of threads woven together will tear and catastrophic failure occurs suddenly.

    • Jeff A

      Which frame material, common or uncommon, would survive those crashes?

      • Wake

        I don’t know how fast you descend and in what sort of condition the roads are on which you ride. But carbon fiber frames built to be strong enough are so rigid that they corner like crap on bad roads. Aluminum is better but only just. Steel frames constructed with the top grade steels have just enough “give” to them that they corner MUCH better than anything else.

        The carbon fiber frames can have catastrophic failures. Granted for sport riders this isn’t very common simply because sport riders do not usually ride extremely hard and hence do not put the strains on a frame that others do.

        Aluminum hardly ever fails. This is because you are forced to make the tubing thicknesses much thicker than would be possible with the strength of the material but you are stuck welding the joints So in order to keep the stresses down low enough not to cause hardening and cracking you use more than necessary.

        Steel was experimented with for so long that bikes like the Eddy Merckx frames and the steel Colnagos and other racing bike are simply build for safety limits far beyond what the everyday rider is going to put into it. Today’s Waterford is probably the finest and safest frame available.

    • Wake

      My Colnago Star fork cracked from hitting a sharp bump at high speed causing me to lose control and crash in the next turn. The man that was leading me on a C40, also hit that bump and two days later the ENTIRE head tube fell off the bike luckily while he was entering a bike trail at only 5 mph or so. The C40 broke right through the lugs in a catastrophic failure. Luckily his only injury was a small right finger broken in two places. In another 5 minutes he would have been descending a hill at between 30 and 40 mph. There were no visual signs of any damage of any sort before he started that ride.

      Colnago states that they recommend that teams only ride these bikes for one or at the most two seasons and then destroy them.

  • RayG

    In answer to the question in the title – until it’s no longer fashionable? I raced a Giant TCR for 6 years, but it wasn’t fashionable in the first place.

  • Coach

    The elephant in the room is the lack of data showing the qualities of frames as they age compared to when they are new. I find it implausible that someone in an industry the size of the bike industry hasn’t looked at this. That we haven’t heard makes me suspect I know what they found.

    I did hear from a friend with a Sarto frame that they were conducting the tests. No word yet, but they are a custom frame builder not using mainstream asian build techniques so likely an outlier.

    • By age, I presume you mean a used frame? So the test is comparing the resilience of a used frame with a new one? Not sure how you’d control that experiment.

      • Coach

        I understand that it wouldn’t be easy, but I’d argue it’s well possible. Say a big manufacturer had a frame set where they knew the QA parameters (flex, strength etc). They could then make a program where they would recall some frames after some time or KM points for testing. If you did it with even a small number you could say something like “after two years on average our frames retained 95% of their strength and resistance to flex”.

        You are asserting here that in theory carbon frames should last forever – but if you exclude frames that have had normal wear and tear from this then you’re saying “carbon frames will last forever if they are not used” which isn’t as compelling.

        As someone who knows lots of sports cyclists my experience is that if carbon bikes are used heavily (ie 10-15k a year) a very significant portion of them do fail within the first couple of years. The frame is covered by the manufacturers, so the owner gets a new bike, and the bike shop gets to build it (sell a new or upgraded groupset, or at least get the labour) and this business model works. It would appear to me that the manufacturers know that only a small portion of the bikes they sell get this kind of treatment, and many bikes see little use or are sold (ending their warranty obligations).

        Can you find out for us what the frame replacement rates are? If they aren’t keen to do the testing then it seems this kind of data would validate or invalidate their argument that “carbon frames should last a lifetime”.

        • Alex L

          The inherent assumption in the initial premise of the article and a lot of these comments seems to be that the frames (especially CFRP ones) age and thus lose some stiffness or other property. Why would you assume this? What is the mechanism by which they age?

          Materials don’t magically age. There are known deterioration mechanisms and then there are failures. In the case of fibre reinforced polymers (FRP) the conclusions from the article and the opinions of the engineers interviewed is that known deterioration mechanisms (such as debonding of the epoxy resin due to UV exposure) has been accounted for and is not significant.

          In the case of ‘shops replacing frames’ this has nothing to do with CFRP magically losing stiffness and becoming flexible or weaker, and everything to do with mishandling, material flaws or an under-designed frame in the first place. Given the stresses on a frame, the design emphasis on minimum weight and the mass-produced nature of bicycles, yes, some will fail. This doesn’t prove that CFRP frames are somehow aging and becoming inferior to metallic frames. It demonstrates that either there was not enough strength in the design for a given condition, or there was a mistake made in manufacture, or the frame was abused. None of these have to do with the long term durability of a material.

          • Coach

            So, on a recent ride with some mates I was asking around about this. 4/4 carbon frames (all from big name manufacturers) were warranty replacements and none of the originals had lasted more than two years. I understand that this does not prove that CFRP is not an awesome frame material. I would argue however that it proves that CFRP has significant issues surviving normal use when used for bicycle manufacture.

            • Alex L

              That doesn’t necessarily say anything about the material though, it just says something about the whole package. Perhaps those frames were not sufficiently strong in the required areas, or had taken a few too many impact/out-of-plane loads, or there were manufacturing defects that should have been picked up during QA/QC (obviously highly unlikely that all 4 were QC problems).

              When discussing the lifespan of CFRP frames we need to discuss the whole package – material, design philosophy, manufacture and intended usage. Putting a number of failures down purely to the material doesn’t describe the actual problem.

              The performance of CFRP, much more so than metals, depends very highly on the layup design. CFRP doesn’t have homogenous properties. So if the sheets of fibres aren’t adequately arranged at key stress points than that joint is subject to failure, e.g. the bottom bracket shell is designed to flex primarily vertically, but out-of-plane forces such as side loads or twisting could damage the joint. Especially if that joint uses a minimum amount of material to achieve a weight goal.
              Metallic joints have the advantage of behaving roughly uniformly to stresses in all axes (not strictly true because tubes are shaped to take advantage of stresses/flex in a primary direction). So the design of metallic frames is simpler.

              As with anything, outliers will always garner more attention. The one guy in the club who has had multiple frame replacements gets discussed much more than the 300 others who’ve got perfectly functioning frames.
              Given the widespread adoption of CFRP for bike frames, the advantages of CFRP as a material, the level of engineering research and advanced design (FEA being employed in the design of modern framesets), I don’t think that there are significant issues in using CFRP for bike frames or the longevity of those frames.

              I do however think that there’s a vocal element who dislike CFRP frames, either because ‘steel is real’ or for the anti-consumerist sentiment. And that’s fine, there’s nothing wrong with an opinion, but it doesn’t fundamentally change the advantageous material properties or manufacturing benefits of CFRP.

              • Coach

                So why don’t they offer the statistics saying that there are only x% of their frames replaced under warranty? That would be a very effective way of illustrating durability. You aren’t addressing that. Afraid of something?

                I thought about what you’ve said – but then I realised of the people I know who ride a lot it would be easier for me to list off the ones who haven’t had at least one carbon frame replaced.

                • Alex L

                  1) The data is probably not that easy to get and collate. The big companies have their frames distributed across the world by various distributors which are then on-sold to bike stores. In order to make the data meaningful the precise cause of failure would have to be recorded for each case, something that takes additional effort from the distributor or local store, who each have a limited interest in spending that time for no return. It’s probable that bicycle manufacturers conduct some sampling to get an idea of the number and types of failures, but it wouldn’t be extensive, it doesn’t need to be.

                  2) There’s no impetus for bicycle manufacturers to release this data. If they have this data it would be used in house, there’s no advantage in releasing this data to the public. By releasing that data you invite armchair critics to ‘assess’ your products, or you potentially give a competitive advantage to competitors. We’re talking about multinational capitalistic companies, not universities or charities.

                  3) There’s very few people actually calling for this data. If the majority of people didn’t want carbon bikes or were worried about their durability, they wouldn’t be buying them.

                  I’m not afraid of anything. I currently own two carbon road bikes (and a steel track bike and an alloy CX bike) and I ride both as hard as I can. I’m not about to be a princess with either bike. If I had any concerns about the durability of the frames I wouldn’t have bought them.

                • I wonder if the issue here is related to mass production rather than carbon fibre?

                  • Dave

                    It shouldn’t be.

                    If Apple is serious enough about protecting their brand to demand a dead iPod have its problem documented AND the dead one returned to them, companies like Spesh and Scott can damn well do the same with carbon fibre road bikes they sell for far greater prices.

                    The logical reason is simple – the business they stand to lose by releasing the figures would be far greater than the business they would gain from having a reputation for transparency. This is something your piece would have picked up on if you had also interviewed a composites engineer instead of just regurgitating the manufacturers’ spiel.

                    • There have always been problems with mass-produced bike frames, regardless of the material used. While mass-produced carbon bikes have a reputation for cracking, mass-produced steel frames were never straight. Same applies to aluminum frames, but then they also seem to have a habit at cracking right next to the welds.

                      If you’re disappointed by big brands, then there are alternatives.

            • Don Salamon

              Typically a reliability study is done for the material in question. You consider the cyclic loadings, figure out the endurance limits, and consider the elements that can cause the material to become weaker. I imagine is that there was some sort of fault with the layup allowing stresses in places where stresses are not supposed to be resulting in a catastrophic failure. The material is very brittle, so there will not be any indication of a failure mode before it is too late.

              I have to agree with the above poster, issue was probably manufacturing not material.

            • Wake

              After crashing 3 weeks ago from hitting a bump at high speed and the resulting crack in the fork forcing me to lose control and hit a stone wall at 25 mph or so I began wondering. Two days later my friend who also hit that bump at speed had his entire head tube fall off breaking directly through the oversize CF lugs.

              Now granted these bikes were 10 and 14 years old but they were the very top of the line. They were always inspected. And they had top notch components on them.

              That sold me. I’m changing back to steel.

        • Matt DeMaere

          There is little to no accounting for false positives when it comes to frame failures, so you should assign a degree of uncertainty to your anecdotal evidence.

          The average punter has learnt, through engagement in the community, that it is possible to label a frame as “cracked” and when labelled as such, obtain a replacement. Since the industry is constantly moving on, a replacement might even become an upgrade — a fairly enticing situation for some. Because of the economics at play, very few of these frames are subjected to non-destructive testing in which to confirm with certainty. What is accepted instead is sufficient cracking of the surface, which may or may not be a true failure, their relationship with the shop and their ability to kick up a stink — err skills of persuasion.

          That is not to say that frames don’t fail, I have personally seen some where it was obviously the case, but not all of those which get replaced have failed.

          I personally have had a replacement, but this was not a failure but an alignment issue. The frame was nearly 3 years old and had done 40,000km, been ridden up/down Jacob’s ladder in TAS, raced regularly etc, etc, all with a rider floating between 80-88kg. It did not have a sheltered life and was still going strong.

          Excluding abuse or misadventure, early failures will be manufacturing defects. This is the same thing seen in data centers, where the deployment of so many commodity parts in a precise environment allow companies significant statistical power.

          As was outlined above, there are many QC issues to be dealt with, not all of which are the source of the frame maker, but can just as well be due to variability of the resin or textile, or the interplay of all.

        • Notso Swift

          I bet they have done it with the pro bikes

      • Lyre_bird

        In a sense, Kevin Steegman at Polytube Cycles in Germany has done the work for you. He has measured the torsional rigidity of a lot of frames in the process of repairing them and reports that commercial frames cluster around a value of 100^-1. He ensures that they have similar values after his repairs and I’m sure he would have noticed and commented upon any decline in that value with age of frame. Caveat: I haven’t asked him this specific question, I’ll pass his details along if you are interested in asking.

    • nicklothian

      Pretty sure if any vendor had actual data showing that you need to buy a new frame every 5 years they would be sharing that as widely as possible!

      There is no reason why the carbon fibers themselves will change over time. I could see logic to suggest the resin maybe a problem, especially with UV exposure etc. But the evidence just doesn’t back that up.

      Carbon spars (masts and booms) have been used in sailing for decades, in much harsher conditions (in terms of UV exposure) than bikes. They are regularly seeing 20 year+ lifespans. Resins are much more advanced than they were 20 years ago.

      • Peter

        Are carbon spars made thicker, though? It stands to reason that a carbon construction made from thick layups of carbon and resin would last longer, but bikes are made to be light and stiff in some places, but flexible in others. Spars are made to resist the forces of the wind, so would need to be much thicker and stronger, surely. So can they be used in a comparison with a bike frame?

        • slartblartfast

          Interestingly, carbon yacht masts tend to be more durable as well as lighter than alloy ones although I agree the parallel with bike frames is remote.

          • Winky

            I some ways I agree, but yacht masts and spars, like bikes, are built to a compromise of lifespan, stiffness, strength, weight and cost, and are subject to complex and random cyclical loading through their lives. Failure of spars and bikes are both very bad things, but well heeled enthusiasts mean there is a market for products that seek to optimize the compromises.

            Yacht spars generally don’t get “crashed” in the same way that bike frames do, though. That carbon yacht spars are apparently outlasting the heavier (and cheaper) aluminium ones is interesting, I think,

            • Gordon

              The major reason carbon components outlast alloy one – in marine conditions – is that they don’t corrode. Fittings don’t come loose around the pop rivets, etc etc ..

        • H20

          Sailboat (and windsurfer) masts are also designed to be light. The lighter the mast, the more sail you can carry for the same heel (lean) and therefore the more drive you can create. The boats also bounce less through the waves when the rig is light. And because mast bend is used to control the sail shape, it’s vital in some types of sailing craft (like windsurfers) that the mast is stiff in some places but flexible in others.

          Pressures are pretty high; skiffs can have over a tonne of static pressure on one wire; in offshore yachts it can be 10 tonnes or more. Add to that the loads imposed by the rig and then dropping the whole bundle off a 3-30′ high wave, and then add in the fact that top boatbuilders say that good carbon boats have apparently no limit on their lifespan, and carbon starts to look fantastic.
          Having said that, there have been massive failure issues such as the one that saw windsurfer pros breaking 20 masts (or more) at $1000+ a pop. On the other hand, in other classes carbon masts will last for 20 years without a problem. It gets down to what the rules and equipment designers allow, which is food for thought when people start complaining about the UCI’s 6.8kg rule.

    • H20

      Isn’t it highly likely that the results of any such testing was seen to be commercially sensitive and therefore kept as company IP? It doesn’t seem any more suspect than the fact that coaches like you don’t seem to provide your coaching plans and your charges’ stats on the ‘net.

      In 2000 I asked two top-line builders of carbon sports equipment, costing up to $10 mill per item, the same question about carbon’s durability. Both of them came up with the same answer as in this article – that is, there was basically no limit to the lifespan of high quality composite construction (unless it was abused, of course).

  • George

    Ti or die!

    • Sean

      I’ve cracked a Ti frame… and I kept it as proof!

      • Peter

        Yes, Ti is very fragile if the welding is not done right.

    • Hippy

      Or Tie Dye

  • Cam

    Please remove this information before my wife sees it.

  • GutterBunny

    Look around at the used bike market, desireable used steel and alloy frames are selling for what their initial costs were (adjusted for inflation) if not more than original retail price. I see very few used carbon bikes for sale – and I suspect the reason isn’t that they are out lasing the other materials. If that was the case there would be no need for articles like this one.

    By the way the fibers will last forever, but it’s just a cloth (ie the newspaper in paper mache), The epoxy (ie super glue) is what holds it all together. So by focusing on the fibers themselves is extremely misleading you can have the best wrapped and layered frame ever made but without the best epoxy it’s worthless. The fiber really isn’t nearly as significant as the glue. Which by the way is petroleum based and isn’t recyclable -but then again the fibers (again petroleum based) aren’t really recyclable either.

    I’ll stick with steel thank you.

    • Alex L

      I think the sale price of second hand steel/aluminium/titanium bikes has far more to do with perceived value and market scarcity than change in material properties over time. As stated in the article, carbon is now the choice of material. So steel or aluminium bikes are either cheap, or high-end or ‘custom’ builds. And obviously unusual high-end frames maintain a good resale, coupled with the fact that there aren’t many on the market in the first place.

      I think the frequency of second hand carbon frames for sale is purely due to the consumerist nature we now have in the bike industry. We are being encouraged to want to continually upgrade, and naturally very recent secondhand frames are popping up. I have a 2013 Cervelo S5, which I absolutely love and have no intention of selling… except the new model came out and it’s x% lighter, y% stiffer and a different colour. And despite there being nothing wrong with my current S5 or its performance (hot tip: the bike is not the limiting factor in me not winning races), I want the new one.

      What does recyclability of the material have to do with anything?

      • Ronan O’Connor

        The issue with second hand carbon is that you cannot stand it up to a visual inspection to know if the frame has a fundamental structural issue. You can do this to an extent with other frame types.

        • Alex L

          Cracking tends to initiate at the outside surface due to bending stresses, so generally speaking you’ll see the cracks, regardless of material. Where CFRP differs is disbonding or delamination can occur inside the material, and these are hard or impossible to see. If the CFRP has been made correctly and with good QA/QC, you shouldn’t have any delamination issues, and they shouldn’t form in service.

    • Steel

      But the fibre is what gets loaded when a force is applied to it along the axis of the fibres, not the resin. Unless, of course, you load it incorrectly – impacts.

  • Neo

    “It is truly a matrix”.
    Told you.

  • mzungu

    Carbon composite may have infinite fatigue life, but the adhesive(epoxy) joints that hold the pieces together does not.

  • Steel

    As a mechanical engineer with a strong background in structures and mechanics of solids, I’ve always been perplexed by the resistance to carbon fibre reinforced polymers – carbon.

    Tennis rackets are an interesting comparison. In the early 90s when carbon tennis rackets were introduced, makers of the products saw their sales fall away. Carbon rackets had become so strong and so durable that people had no reason to replace them. The makers had put themselves out of business.

    Having solved this problem for the market, they had to look at creating new problems for which they could create solutions for in the form of newer rackets with more fancy solutions. So we saw holes in racket heads, different racket head shapes and sizes, string layouts etc… These rackets give you a faster swing, better sweet spot etc… (sound familiar)

    The same thing is happening with bicycles. Having found the near perfect material for construction, the frame makers now need to keep our interest with other innovations such as aero tubes, micro suspension solutions, integration of disc brakes. These innovations are perfectly valid but are more at the margins of performance – hence we see the big sale pitch for something that seems identical to the previous iteration of the product – see Spesh Tarmac 4 to Spesh Tarmac 5.

    I’ve got a 2008 Trek Madone and can see absolutely no reason to upgrade it. It’s been driven in to Maccas drive throughs, dropped it (not crashed), worked on with my shitty home mechanics skills, and done more ks on the roof of my car than I actually get to ride on it. It has not missed a beat in that time.

  • winkybiker

    I crash rarely – about once every 10 years on average. Obviously, I don’t race (mainly because I don’t like to crash). I have been riding carbon frames since the first Trek OCLV frames in the early 90s and have never broken or worn out anything at all made of carbon. I have broken many steel bits over the years (frame, steerer and numerous spokes), a couple alloy bits (handlebar (in a rare crash), crankarm and a spoke) and had an alloy sleeve come unglued from the BB shell in a Trek 5500. For me carbon has been 100% reliable and infinitely long-lived. But that’s just me.

  • lefthandside

    This is very interesting – I asked a friend this question on a long ride yesterday. There is reference to ‘overcooking clamping’ which makes me think about things like carbon dropouts on frames. I realise that a key point to take from the article is that the fatigue properties of carbon frames should be distinguished from the other risks / wear and tear on carbon frames. But thinking of carbon dropouts it is obvious that when we talk of ‘the frame’ there are quite different parts to think about – some are heavily reinforced, some are entirely solid (dropouts), others are quite thin because they bear less weight. I wonder if the general wear and tear susceptibility of carbon frames (impacts especially, but also rubbing, or over clamping) has prompted a response from frame designers to those key parts of the frame that are most at risk?

  • Simon

    OT but some typos stand out and can drastically change your intended meaning. The “k” is missing in the third word of your disclosure statement.

    On topic, a good article, thanks.

  • Michael David Robres

    If you ride your bike a lot, smarter to go with Alum. Just go fast & hard with no worries. I don’t have time to check all over for cracks every ride. Just to save a pound or 2.

  • Carlos Ferreira

    I had 2 scott bikes. Loved them but both of Those bikes fatigued. I don’t understand when they say it won’t fatigue.
    I just got a Cannondale Super Evo6. The bike is clearly heavier than my scott Addict team but it seems stiffer.
    scott is sending me another frame becUse the team one cracked. They are now sending me the lightest version possible (SL). The complete bike should be at 12.5pnds. I am excited but I have the feeling that my normal 190-198 pnds body will fatigue this frame faster than it fatigued the team version.

    • Winky

      What is the evidence of the “fatigue”? Did they crack, fail or just get soft?

      • Carlos Ferreira

        Do you know that feeling when you ride a stiffer and lighter bike? It is the same but the opposite. I had a funny sensation that my wheel was flat, the sensation would go away and come back. For a while I thought it was the wheel, whoch flexed too. When sprinting I felt a wobble… It may sound exaggerated but it in fact happened. The BB was also corrupted to a point it had a play. When paddling my crank would make noises and literally move.
        I am getting an even lighter replacement which sounds nice but let’s think. Is the technology really there that the expensive and lighter bikes are just as durable? I do not believe and I have experienced the issue.
        Having that heavier Cannondale Super Evo 6, I will use this bike for flatter bikes and save the Scott addict SL for climbing days… hopefully I can get more use out of this bikes

      • Carlos Ferreira

        Got sofr

      • Carlos Ferreira


  • Craig

    The real answer to the question is ….How long is a piece of string.
    Lab testing is fine but In the real world bikes break even JRA.
    Nuf said!

  • Roeland

    Scott and Specialized are twisting the truth when they say cfrp shows unnoticable levels of fatigue. Sure, carbon does not fatigue in the way metal fatigues. However, there is the mechanism called ‘damage accumulation’. Anyway, a quick google search would reveal to anyone what CFRP fatigue is a real thing and is being researched. Take a look at this paper (a free one I could find)
    The graphs showing the decrease in stiffness and strength of a piece of CFRP subjected to cyclic loading are telling. A 10% stiffness decrease is possible. And then these guys are telling you ‘unnoticable’. At the same time, they’re marketing their new frames by saying those are 6% stiffer.
    That is excluding water absorption effects of the epoxy (all plastics absorb water).

    It might be that at the load levels of bicycles, the effects are less pronounced. But Scott and Specialized are throwing no data in the mix, are not telling how they test frames (do they ride them for 5 years and then re-test them or just pull a new one apart after a month in a shaker?), but just that all their frames easily pass their tests. If that’s the case, your tests aren’t any good now are they? After that, they’re all like ‘nah, those frames will last you a lifetime, don’t worry about it’. They haven’t shown one iota of data that would back up their claims that you shouldn’t worry about it. That’s a reason to worry about it.

    Statements such as ‘better than any aluminium or steel’ with regards to composites are also strange and borderline lying. Aluminium isn’t all that good in fatigue, while steel is exemplary as steel has a defined fatigue limit, unlike cfrp.

    cyclingtips, I love you guys, but really, in this article you’ve been blown off by people who didn’t want to answer your questions properly. Please talk to someone who has no vested interest in selling frames. Any university with a good composites lab can help you.

  • Il_falcone

    writing an article and citing only guys working in the industry for companies who, on the higher end of their product spectrum, only sell carbon frames is either pointless or barely hidden advertisement. While it’s fair to ask them for their point of view – although we all knew beforehand what they would say – you should have also done some research by asking bigger cycling clubs, shops, guys specialized in carbon repair, whomever else in order to offer a more balanced, a more realistic view of that matter.
    And while it might be accurate that many well-made carbon frames outlast any metal frame under laboratory conditions where they are subjected to the very same precisely defined loads over and over again, the real-world riding reality shows that most carbon frames fail earlier than any high quality steel, aluminum or especially titanium frames do.
    When a high-end carbon road bike frame can develop a crack at the down tube because someone riding close to you hits a stone with his wheel and shoots this stone across right onto your frame’s down tube, then you know that your frame is designed for the laboratory bench test and for winning the stiffness-to-weight category but not for real-world riding.

    • Alex L

      Seems like a conspiracy theory where bike companies are trying to sell us faulty frames in the hopes that we’ll waste our time, their time and our local distributor’s time getting warranty replacements!

      Or the problems are overstated and there’s a strong case of confirmation bias here.

      • Il_falcone

        I’m not sure what you want to state, please elaborate if you want me to reply to it.

        It possibly helps to understand what my point is when you consider that many of the carbon road frames which the pros actually use are reinforced versions of what you can buy in the shop. That does (or has) at least apply(ied) to Trek, Ridley, Cervelo, Canyon, Giant to name just a few in recent years. Those frames look the same because they use the same mold but are considerably heavier because of the added material. Why? Not because of the weight limit since some of those pro bikes are even well above the 6.8-kg UCI weight limit – not the ones of the GC hopefuls and climbing aces obviously – but simply because those production frames failed to withstand the (ab)use a racing frame sees. Given that frames ridden by any professional team are only ridden for a maximum of 12 months – which typically only applies to their training bikes but not the ones which are actually raced -, this can be interpreted as a strong hint as to why companies try to sell us ever lighter (and stiffer and more comfortable …) carbon frames. The considerable numbers of failed frames (warranty replacements) is already calculated into the price you happily pay for your carbon bike.

        In the US-based titanium frame industry where most manufacturers offer a lifetime warranty against material or manufacturing failures a failure rate of 0.1% is considered the achievable gold standard. So only 1 out of 1000 frames is expected to be returned to the company at some point in its long life because of a frame failure. If as a titanium frame manufacturer you have a considerably higher failure rate than 0.1% you have done something wrong (flawed material, gone too light, poor process, whatever) and should reevaluate what you do.

        Carbon frame manufacturers can only dream of reaching failure rates like this. Most would be happiest if they were to achieve something in the range of 1% at some very distant point in the future.

        Obviously not all is bad at the carbon bicycle front, though. There are some examples of rather compact structures with less complicated load patterns like handlebars for instance or forks which, if well-made, do actually last longer than their aluminum counterparts. So as long as you don’t crash heavily very often a carbon handlebar might actually be a very good choice not least because it won’t corrode beneath the bar tape because of your sweat which many aluminum bars do.

        • Alex L

          What evidence do you have to support the idea that many pros ride beefier frames? I hadn’t heard this one. As for pro team frames only lasting 12 months, this is rubbish. They might be raced for a season, but many frames are used as spare bikes for the next season (especially low budget teams) or last season’s race bike becomes a training bike, or are on-sold to the public (Garmin was selling ex-pro team R5s with new groupsets), or are given to friends/family etc (I saw one of the Etixx boys and his dad in Adelaide during the TDU, his dad was on his son’s last season race frame complete with DA and carbon wheels).
          The idea that an entire industry would knowingly make consumer products that are severely under-designed and at-risk of potential catastrophic failure is ridiculous. A couple of serious lawsuits from people who had crashed due to poorly designed CFRP frames and the major manufacturers would be running for cover.

          Many CFRP bikes have ‘lifetime warranties’ too. A lifetime warranty doesn’t mean forever in the legal sense. Additionally, perhaps the Ti frame market does have a failure rate of 0.1% (this seems unrealistically low), the very small volume of Ti frames being produced means there’s naturally less exposure to total number of failures. How many Ti frames are produced for every CFRP frame? Unsurprising that failures of Ti bikes are much less discussed. There’s probably also a case to be made that less people race Ti frames competitively compared to CFRP frames, which means the stresses and load cases aren’t directly comparable.

          Aluminium bars corroding because of sweat is a borderline case. Aluminium bars are usually anodized black, which builds up the passivating layer which makes them highly resistant to corrosion, and sweat is not that corrosive in any case. In borderline cases you could have concentrated pockets of high-salt content sweat building up which creates a corrosion cell and may lead to localised corrosion of the bars. You could also have a combined corrosion/wear failure mechanism where rubbing away of the passivating layer leads to accelerated corrosion or pitting. In general, aluminium bars don’t and shouldn’t corrode.

          • Coach

            In my experience most people that buy Ti bikes do so because they want a frame to last and or have broken too many CFRP ones. I find this more impressive than (they aren’t ridden competitively). For instance, a couple in our club run bike tours in Europe and their Ti Frames were up to 160k last I spoke to them. They just keep adding groupsets when they wear out.

            I’m sure you’ll say “back it up” and I’ll just say we have Scott and Specialized here saying our frames will “last a lifetime”. I say if that’s the case – don’t just talk theoretical maybes – show us your warranty data (which they definitely have).

            • Lots of great discussion here guys, thanks for all you input. As has
              been noted, carbon fibre literally has its strengths and weaknesses. And
              while bike manufacturers have developed testing protocols to satisfy
              safety concerns, marketing needs and quality control, there is no test
              that will ever prove how long a bike will last an individual (regardless
              of the material it is constructed from). Even if we were provided with
              data on frame returns and warranty replacements by every manufacturer,
              it would still do nothing to forecast the lifespan of a frame for an
              individual (nor would it provide any comfort to those that experience a manufacturing defect).

    • Jonathan.

      Yep. Not much talk about day-to-day durability from the manufacturers. That stray stone flicked up, or a clumsy mate knocking it over at the bike shop, or a minor tangle in a bunch ride could all quite easily lead to terminal damage on a carbon frame.
      I know that CF frames are the latest and greatest, being all laterally rigid and vertically compliant and all that, but I can’t help but think that the average punter would still be better served by a decent lightweight steel frame. If you’re not getting paid to race a bike and actually put your own money into what you ride, steel or Ti makes a whole lot of sense.
      But then cycling is fashion-dictated and every Fred wants the latest and greatest. Not many people buying bikes with the long-term in mind.

      • Alex L

        Except nice light weight steel frames have very thin walls and are easily prone to denting – something that could happen to the average ‘Fred’ at a coffee shop.

        Every material used in bicycle construction has upsides and downsides. CFRP has too many upsides not to be the dominant material of choice in the [mass-produced] bicycle industry.

        • Jonathan.

          True, I guess. Though I’d put my money on the steel frame being rideable with a ding for a lot longer than a CF frame with a nice crack.
          But that’s just one guy’s opinion.

  • PieterVP

    Just for fun …
    I had a 2011 Scott Addict for 2 years, then discovered this crack just below the front derailleur.
    Bike had not been crashed and went in a hard case when flying.

    BUT I had used the bike on a wind trainer at least 2 hours a week for 2 years (200 hours perhaps).
    A carbon repair shop guy told me he was 99% sure that the stress of the wind trainer was the cause of the failure and that he’d seen plenty fail like that. Scott said that was unlikely to be the cause.

    Scott did warranty the frame – thanks Scott – but it took 4 months to come through, which I felt was a bit slow.
    In the meantime I bought a Cervelo.

  • We need to find a material that forms of the circular economy, i.e. is fully recyclable. At the moment, carbon is simply too deleterious on the environment to be considered a long term, sustainable material.

    • Rodrigo Diaz

      Not necessarily. You’re focusing on the end-of-life impacts, which are significant. But there are a lot of upstream costs (environmental) that may offset these. As an example: smelting aluminium is extremely energy intensive. So you really need a lifecycle analysis to determine what is more deleterious (That’s my day job, but not with materials exclusively),

      For what is worth, steel manufacturers did a comparison with carbon fibre car components. Steel came out ahead, but just so. This is of course a different case but as you might expect the carbon fibre associations complained that this failed to take into account lightweighting gains, lifetime differences, and emerging waste recovery techniques (Trek has a carbon recycling facility now, if I remember correctly).

      Just because something is recyclable doesn’t make it environmentally preferable.

    • Dave

      I’m not sure it’s a huge issue because carbon fibre is only used on insignificantly small numbers of racing bikes. Normal bikes are still largely made using steel.

    • Alex L

      To add to Rodrigo’s comments, steel and aluminium are not fully recyclable. When you recycle steel the concentration of trace elements increases. Some of these trace elements are very difficult to remove (such as copper) and adversely affect the strength, heat-treatability and welding parameters of the steel.

      Aluminium is a material with good recyclability, although it can also have problems with trace elements affecting its properties. And if you need to smelt more from bauxite, is very energy intensive.

      As Dave said below, the actual number of carbon bicycle frames in the market is largely insignificant to the total number of bikes in production in the entire world. If I remember correctly a CT article from a little while back mentioned that Giant had 7 bike manufacturing facilities, only 2 of which produced CFRP framesets. The other 5 were dedicated to metallic frames.

      • Dave

        All recycling processes have their detriments, inefficiencies and associated costs.

        Even paper does, to name a popular one not directly associated with bicycle manufacturing.

  • Who here has bought a second-hand carbon framed bike? It’s hard to tell just from inspection if a carbon frame has been crashed, or otherwise abused. I buy used steel bikes a lot, and can always sell them for as much or more than I paid for them. Sure, steel frames can have issues, but they are fairly apparent to a buyer with a good eye. I have no worries putting in a lot of miles on third- or fourth-hand steel bikes made in the early 70’s. I would not feel anywhere near as confident on a five year old second-hand carbon bike, and I suspect I’m not alone.

  • Dave Krenik

    Re: New materials. What about Graphene?

    • Rodrigo Diaz

      Yes, that’s one. And some nano-technologies as well. One company I’m familiar with (Integran) developed a metal/carbon composite material (Nanovate), which is used in the high-end Cervelo fork steer tubes to make them harder and more resistant, while saving weight. Lots of cool stuff,mostly trickle-down from aerospace.

  • SpaceKnight

    I had several failures with carbon frames, and actually all except one of them had to do with the bottom-bracket aluminium lugs that came loose from the carbon in which they were glued. (Also hate the pressfit bb30 system it will not stop make cracking sounds no matter what i try.) Last frame even came without lugs, figured it was intended and it four months to get a new one from warranty once i found out it was a production error.

  • Matthew Sheeks

    I have never seen a carbon frame not break (aside from Giant TCRs), for one reason or another (defective, operator error, crashing). I have seen all types of brands break, and with the exception of several Giant TCRs have never seen a carbon frame make it past the 5 year mark. Like I say, sometimes this was due to operator error, but if they had a bicycle of a different material, the frame wouldn’t have been so sensitive to moderate amounts of abuse in the first place. If you want to buy carbon make sure you buy it new and are the warranty holder. I would give about 90% odds that you will have to get your frame warranted at some point.

  • Robert Merkel

    I don’t doubt that carbon frames are structurally pretty durable, but any painted surface gets cosmetic damage over time.

    After five years, give a brushed titanium frame a clean and it looks like new; you have to be both lucky and incredibly disciplined for a carbon frame to remain pretty for that long.

    That said, in five years time I’m sure the bike industry will have new models with 13-speed cassettes, ABS hydraulic disc brakes, the aerodynamics of a streamlined gnat, etc. etc. etc., and I’ll be looking for an upgrade anyway!

  • John Siviour

    Informative comprehensive article. Thank you

    • Dave

      Apart from the lack of interviewing any sources without conflicts of interest.

  • Ed Zachary

    While that was a very informative article regarding carbon bicycles, I did not see an answer to the question in the title: what is the lifespan of a carbon frame?

    • Dave

      Until it breaks!

  • GeeTee

    As a non-engineer, it seems to me that the argument boils down to who is making the frame, not the material so much. Airbus and McDonnell Douglas make passenger and fighter aircraft out of carbon fibre composites, so I say the debate on the longevity of the materials is over. Like most things – its who’s using it, and how its used is the thing. Having said that – I inherited an old (unbranded) titanium frame from my uncle – it must be 30 years old now, seen countless knocks, scrapes, crashes, falling off the car rack, its bulletproof.

    • Sean Doyle

      Yes. No. The big issue is the aviation stuff has that many checks and balances that failures are very rare. In the bicycle industry the consumer push for lighter and lighter frames has meant that carbon composites have been pushed to their limits and beyond. If every frame was NDT’d before it left the factory you could have utter faith in the material. Of course then you’d be paying through your nose for even the cheapest frame. To counter that you over build the product which means it’s heavier and it doesn’t sell as well. Catch 22 for the manufacturer.

      The crazy part is the a 900g or 1kg frame is still ridiculously light and it’s the components that make the difference really. So building 700g frames is reckless IMHO. The steel frame I built with my own hands and has no weight weenie parts on it and tips the scales at bang on 7.1 kg. It is not heavy.

  • Sean Doyle

    Here is an interesting read from Craig Calfee. Granted he is pro carbon but it covers all the major points. One error that is still used in comparison of modern steel to other materials is they keep trotting out the 4130 specs. The steels used in the top shelf steel tubing have UTS that easily surpasses Titanium. The dent resistance is quite high in these tubes and of course the tube is relatively easy to replace to be 100% the original strength. One number does not make a perfect material though and all the materials properties need to be considered. The big advantage that carbon has obviously, as Calfee pointed out, is control of shape performance.

  • Carl Hemmings

    I’m still using my custom Racing bike, I used to race on with the Blackburn Racing Club, made by Farleigh in 1982, ummm, 531 steel is real. It’s not as light as new carbon bike, it’s 12 kg for a 64 cm frame ( I’m 194 cm), but with updated components it rides good.

  • My metal Tomassini is still going well 30 years old and my Giant aluminium ocr still rides well and is 10 years old. There is lot to be said for metal frames. But if you are racing then it has to be a light weight carbon frame. I get away with it with a heavier metal frame as I only ride Sportives these days.

    • Derek Maher

      I can agree with the Tomassini,Mine is still going strong after 26 years and its been raced,Crashed and now just used for lesiure.Although I do have a carbon framed bike for those hillier routes my older legs need the 11 gears and light weight more these days.

  • Warwick Gresswell

    The Achilles heel of carbon is making something with an infinite fatigue life that becomes unfashionable and passé within 12 months of it being sold, forcing consumers to ‘upgrade’, ad infinitum.

  • Annie.

    To me, the problem lies within the fact that if a frame breaks without any exterior impact, you cannot actually prove that as a normal customer. Hence, it’s always “your fault” and you won’t get a replacement.

    My partner had a Cannondale Supersix Hi Mod Evo. One of the chainstays suddenly broke when he had just remounted and pedal uphill. When we took a closer look, you could see, it had broken “from inside out”. All in all, it was quite a scary experience: Only minutes earlier we had ridden down another mountain pass at high speed.

    He took it back to the bikeshop and was promised they’d send the frame in. So far, there was no feedback from Cannondale, so we guess he won’t get a replacement.

    Our conclusion was to never buy one of these ultralight versions of a carbon frame anymore as long as we don’t have the money to quickly replace them as pro riders do.

    • Annie.

      Update: My partner finally got a replacement!

      He didn’t get the Hi Mod Evo-version, but his bikeshop was sent a new Cannondale Supersix frameset for him by the end of the season nevertheless.

      It’s a pity it took so long as in the meantime, he had to get himself another frame, of course. But at least, they let him have the absolute rightful replacement in the end!

    • r k

      How many miles were on that hi mod evo?

  • brucegray

    The point is it is more difficult to determine in what condition a used carbon bike is, therefore I am very reticent to buy used carbon. Most avid riders I know do fall off occasionally. Steel bikes can be repaired a lot cheaper and more successfully than carbon. If a bicycle cannot be got back on again safely after several offs (without x-rays, ultrasound or whatever), then it really fails real world pragmatic use.

    There is also the matter of bottom brackets and how they are mated with bearing cups. Because of carbon’s properties, many manufacturers use an aluminium sleeve, which can oxidize and crack carbon fibers in the bb. And there’s no long lasting technique to bond aluminium or any other material to carbon.

    And how many carbon frames get trashed by airport workers? Carbon’s significant achille’s heel is it is weak when exposed to non riding forces, and it is almost impossible to tell which non riding forces a frame has been exposed to.

    • Annie.

      Good point! :(

  • prestonjb

    I wonder about other issues too than lifespan.

    My bikes are ti and steel. The tandem is 14 year old steel and had custom bottom brackets. A crack developed in the seat tube so I sent it to a builder who while it was there removed the custom bb shell and welded in a modern shell.

    My oldest ti frame has a 1 inch head tube. I love the frame and while I can still get nice forks for it I’m actually temptedd to get the head tube replaced to 1 1/8 so I can use modern forks…

    On carbon with the use of custom frame specific forks and bb and asymmetric headsets it isnt just the frame life to consider but what happens to the components like forks and bb etc. And if the idea is to keep the frame alive with modern parts can carbon frame be readily modified to keep up with the times

  • william

    dear friends,

    it is william who is the carbon bicycle
    manufactorer from china , i have many bicycle part as below.

    bmc, pinarello dogma 65.1/f8/tt ,
    specialized venge/SL4/SL5, cervelo s5/ s3, scott road frame/mtb frame, wilter
    frame, colnago c59/c60, BH, derosa frame, cipillion bond frame,and look
    685/795/986 frame etc.. if you are interested to it,
    you can contact to me, my email is

  • Aaron Stoner

    So, what is a serious impact? Will hitting a hole in the road cause deterioration or do I need to hit a tree at 15mph to see this?

  • Devis Wolowitz

    Aluminium is a metal material that contains properties that make for
    very nice bicycle frames. It is lightweight, strong, non-corrosive and
    has low density. Its biggest benefit when compared to carbon of course
    is that it is cheap to produce, easy to handle and cheap to manufacture,
    in this case, bike frames from.

  • V8 power

    I don’t think aluminium and carbon is such horrible mismatch there are many of those bikes still on the road just look up Giant Cadex 980C plenty for sale I got one too and it doesn’t give me any impression of falling apart in foreseable future, those new bikes on the other hand look very brittle and got plastic feel to them.

  • You can find the hottest and newest names in
    cycling at the website of Carbon Speed Cycle. If the budget is a massive issue,
    an online shop will have the better cost; however a local bike shop can present
    you the opportunity to see the merchandise in person, which is obliging for
    some people.

  • roadkill612

    I am surprised there is little mention of combining/bonding metal and CF. I know there is a lot of metal bonding with glues happening now. Intuitively, combining say a forged titanium headset or BB, may allow designers to cherry pick from material properties, like shock resistance.

    To stress (as mentioned here), the key watershed(s) CF crosses:

    Historically, one takes standard prefab material (pipe), and bends/welds/heats/squeezes.. it into a frame which is strong enough at its weakest points, not to break.

    The standard, same weight pipe ~must be used in the frames least weak points also, so most of the frame is heavier material than needed.

    Thats the real revolution. A material that can be perfectly assembled on a ~droplet by droplet basis, exactly where and in what quantity needed. Marry that with CAD/CAM, 3d printing of moulds, major chemical advances, 20 years of pioneers with deep pockets…. AND, it would seem, arguably much better for the task.

    So many restrictions on frame design just disappear. Is the diamond frame really so good, or is it just so due to the limits of joined pipe?

    Engineering nirvanah?

    A poor mans 9kg bike could even tempt me off my ebike, now i am a bit fitter.

    My scottish blood has always made me skeptical of the inevitable deniers that cheap mass produced can also be good, or even better. The model T was arguably better that much dearer cars of its day. Cheap bikes have perfect welds now it seems – robots? The biz model of producing junk just doesnt work anymore. An automated factory to make quality costs the same.

    I perversely, would distrust custom builds as untested.


Pin It on Pinterest

October 22, 2016
October 21, 2016
October 20, 2016