Cracked carbon fiber frame - repairable?



danthesaxman

New Member
May 31, 2012
2
0
0
Hey guys,
Yesterday I was loading up my Avanti Cadent 3.0 2010 roadie on a steep hill and all of a sudden I heard a *crack*. I stopped and noticed that the front inner chain ring had snapped in half! As a matter of fact, what had happened is that a few of the bolts holding the inner chain ring on had obviously worked themselves loose over time (something I hadn't even thought to check regularly) and so eventually the chain ring had too much pressure (on the hill) and simply buckled and cracked. That would be fine if that was the only problem, but it turned out that a small 2 inch piece of the chain ring broke off and lodged straight into the carbon frame, creating a thin hole about half an inch wide right at the base of the frame. I have done a photoshop mockup of the location and approximate size of the crack on one of the avanti black frames as I don't have any photos of it. The white crack indicates the approximate size of the crack which has gone right through into the middle of the frame. See the link: http://i219.photobucket.com/albums/cc273/3DanTheSaxMan3/Bikecrack.jpg

As you can see it is right at the base of the frame.
Now this kind of problem isn't naturally covered on the warranty so I have been given the option to pay for the replacement Cadent frame which will cost be around $1300. Seems like such a loss of money since I'll be getting back the same bike as before in essence. Anyway, my issue is what to do with the old frame? Is there a way I can repair it myself? Or would it be better to sell it on to someone who might be happy to fix it and use it? Should I repair it myself and hold off on getting the $1300 replacement frame?

Cheers
Dan
 
I'm no expert in these matters, but I know a little about carbon fibre construction, 2 pack epoxy and the like. If the damage is fairly minor i.e. not weakening the structure of the frame in any obvious way, and due to the fact that its out of sight I would be inclined to try and fix it myself and save the 1300 bucks. If this doesn't work you've lost very little as you have the option of getting a new frame.

You could try a repair by rubbing down the area and using some fibreglass sheet and resin, then paint it to match in with the frame. The repair needs to be only very small. Guess it depends on how confident you feel doing this, or if you have a boat yard or car repair place nearby, see what they see.
 
I wouldn't attempt to repair the frame, especially if the damage is what you described.

Firstly, the damage is at the bottom bracket shell/downtube. This is a high-stress point on the bike, especially when pedaling out of the saddle as in a sprint or during climbs. The pedaling and rocking motion on the handlebars of the rider flexes and twists the frame from the headtube to the BB shell thru the downtube. Bike manufacturers have gone great lengths to beef up that area to increase frame stiffness/rigidity (BB30 anyone?).

Secondly, carbon fiber is a high-tensile strength material. It is very strong - but only in one direction (unidirectional weaving). To make anything made from carbon fiber stiff in a variety of directions, the CF cloth is laid up in different directions before molding it in epoxy.

From the damage you described, that 2-inch piece of chainring embedded itself into the frame and cut some of the carbon fibers. This has greatly reduced the CF's ability to resist torsion and flexion.

No offense to the Doc, but I don't think the CF repair kit will bring back the structural strength of the damaged fibers.
 
And it's scary when CF breaks. It doesn't give you any warning - unlike metals that bend and stretch before failure, they just SNAP!
 
e_guevara said:
And it's scary when CF breaks. It doesn't give you any warning - unlike metals that bend and stretch before failure, they just SNAP!
This is absolutely not the case all of the time and maybe not even most of the time.
 
Simple. There have been numerous accounts of people with damaged CF frames noticing the failures because of either inspection or because of a new noise. Obviously that the OP had a failure without that dreaded "SNAP" indicates that CF failures don't always result in catastrophic failure. There also a multitude of images on the web showing wheels and other CF bits that have failed under ProTour riders that didn't completely fail. If CF did only "SNAP", then no wheels would pass the UCI wheel test.
 
Steel/Titanium/aluminum has decent strength in tension and compression. Yield to destruction is often a fairly tame event when compared to carbon.

The compressive strength of carbon fiber is...not so much.

"Efforts were made to estimate the axial compressive strengths of carbon fibers from the fiber fragment lengths produced by subjecting to a strain greater than the fiber ultimate strain for PAN-based and pitch-based carbon fibers. The estimated compressive strength of carbon fibers decreases with increasing temperature in a temperature range from room temperature to 100°C. This decrease in compressive strength may be accounted for by a decrease in the radial compressing force. The real compressive strength, determined by extrapolating a linear relationship between the estimated compressive strength and the radial compressing force, is approximately 25–60% of tensile strength for PAN-based fibers, while it is approximately 10–35% for pitch-based fibers."

No, carbon failure is NOT 'always' spectacularly catastrophic...but in a hellova lot of the incidents it is.

Likewise, the complete failure of steel is NOT 'always' gradual and with warning...but most of the times I've seen or had it happen there is sufficient warning signs to avoid the explosive failures seen with carbon.

Aluminum falls somewhere in the middle, with a more rapid crack propaogation than steel in most cases....alloy dependent,wall thickness, shape and location of course, but the T6 tempered alloys exhibit rapid crack growth in most instances. All materials can be engineered to work and all will offer strengths and weaknesses.

As to repair, there are professionals that can effect a safe, cost-justifiable repair.

Damaged:



Repaired:
 
Originally Posted by alienator .
Quote: Originally Posted by e_guevara .

And it's scary when CF breaks. It doesn't give you any warning - unlike metals that bend and stretch before failure, they just SNAP!
This is absolutely not the case all of the time and maybe not even most of the time.

Sorry, but I was trying to over-simplify.

Originally Posted by alienator .
There have been numerous accounts of people with damaged CF frames noticing the failures because of either inspection or because of a new noise.
Technically the frame hasn't "failed" yet when cracks appear or when noises are present. Frame failure is most often (but not always) associated with a fracture (separation of the component parts; breakage) which is what we often see as "catastrophic failure" IMHO. The presence of a crack (crack formation) is a sign of impending failure.

Typical of materials with high moduli is very little deformation and rapid crack propagation before failure, thus giving little or no warnnig at all. CF is one of them, but same can be said of high-carbon steel and high-modulus aluminum.

Yes, not all CF failure is catastrophic as you (alienator) and CAMPYBOB have illustrated (nice photo BTW), because it mainly depends on the failure mode the material is subjected to.

Peace.
 
Originally Posted by alienator .

Simple. There have been numerous accounts of people with damaged CF frames noticing the failures because of either inspection or because of a new noise. Obviously that the OP had a failure without that dreaded "SNAP" indicates that CF failures don't always result in catastrophic failure. There also a multitude of images on the web showing wheels and other CF bits that have failed under ProTour riders that didn't completely fail. If CF did only "SNAP", then no wheels would pass the UCI wheel test.

Pot, meet kettle. That is nothing close to the extensive scientific evidential proof you so often demand of others here.
 
jpr95 said:
Pot, meet kettle.  That is nothing close to the extensive scientific evidential proof you so often demand of others here.
That's because extensive "proof" isn't needed to show that something doesn't always follow a given sequence to produce a given result every single time. It only takes one example to show that is true. I'm sorry you don't get that.
 
Originally Posted by alienator .


That's because extensive "proof" isn't needed to show that something doesn't always follow a given sequence to produce a given result every single time. It only takes one example to show that is true. I'm sorry you don't get that.
I'm sorry but scientific inquiry requires that results should be repeatable and measurable (within a given level of confidence) given the same procedure every single time.

In the laboratory where conditions are controlled and tests are carried out by machines and computers that can replicate the procedures exactly every single time, this holds true. Manufacturers do subject their products to rigorous scientific testing (usually until failure) and collate the data to determine the "safe" operating conditions for the product. And most of the time, this is where we base our idea of failure on IMO. However, due to the infinite amount of variables and combinations thereof, one cannot test for every conceivable condition that can happen. Therefore tests are done with a "controlled" set of variables.

Out on the road, riding conditions are far from the "perfect" lab conditions. Your counterexample only shows that a particular condition (even if it is a slight variation from the control) was not considered during lab testing.
 
Make the repair and take the bike out for a gentle ride. See if anything untoward is going on. Then apply more strength, try a steep hill etc. In a controlled way you are stress testing the repaired frame. If it gives, you're prepared for it. If not...alls fine. Common sense prevails.
 
Originally Posted by alienator .


That's because extensive "proof" isn't needed to show that something doesn't always follow a given sequence to produce a given result every single time. It only takes one example to show that is true. I'm sorry you don't get that.
What you're saying is that all it takes is one solid example to disprove a theory such as " CF just snaps when it fails". Like the theory of gravity, or the absolute universal limit of the speed of light: find just one contrary example that can be proven and the whole principal is then called into question. When scientists at CERN recently said they recorded a neutrino that had apparently traveled faster than the speed of light, the announcement attracted huge attention in the physics community because if it were true, Einstein's Theory of Relativity would be invalid.

When talking about failures of bike frames and forks, it might be helpful to start by seperating categories of failures. Catastrophic overload failures caused by crashing or otherwise exceeding the ultimate design load limits are different than normal "wear and tear" fatigue failures.
 
For what its worth. Carbon fiber is a suitable material to build a bike frame with. If it was not the manufacturers would not use it. Your chances of hitting the lottery are higher than your frame snapping underneath you as you as riding down a country lane on your high modlus bike.

I like the picture of the blown out head tube. Looks like someones gun went off while they were riding.
 
Looks like someones gun went off while they were riding.

Hah...I hit a libtard in the dense part.

Its head.
 
Aluminum, pushed to the design limits and fatigued, beer cans and cracks/splits thru the tube wall and it has a tendency to crack at the welds (thus adding Scandium to help yield a structurally more durable weld joint). Failure can be explosive/instantaneous, but rarely is. I snapped a Campy Record aluminum crank arm in two with no warning while accelerating away from a red light. The result was the bike and me flying out-of-control across oncoming traffic...no crash or damage, thankfully.

Steel, pushed to the design limits and fatigued, also will beer can and dent, split and crack thru the tube wall. Both welded and brazed joint failure can occur, but is uncommon and rarely explosive. I had a aero steel fork blade fail when accelerating away from a red light the day after the fork impacted a Newfoundland dog (the size of a small cow) with seemingly no visible damage or alignment issues. It buckled, without warning, mid-blade on the left side. Uncommon, but it happened and the resulting low-speed fall to planet earth needed three stitches to close a split eyebrow (helmet was not even scuffed!).

Carbon, when pushed to the design limits and fatigued generally breaks. With or without warning? The only carbon piece to fail me so far was a Campy seatpost that the head came unglued from the shaft. No warning, the seat just started rotating as I set back down in the saddle after cresting a climb. I hope I never have to 'test' the carbon fiber mode of failure any further.

ALL of it can and has come undone in a hurry. ALL of it can fatigue somewhat gradually to complete failure...'IF' you get lucky. Lance Armstrong made it to the top of l'Alp d'Huez with a significantly cracked right chain stay on his carbon Madone after he snagged a spectator's mussette and crashed, with...was it Beloki?...falling over top of Lance and his bike. I watched basically the same scenario in a local road race with a low-speed fall with another rider falling over the rear of a downed Cannondale CF frameset. The rear triangle looked like a grenade destroyed it. Luck of the draw.

I think most folks are in agreement that the early CF frames and components did generate many explosive failures. I think most folks are in agreement that 'cheap' CF frames and components have carried on the record of explosive events. I think most folks also have seen some of the very lightweight CF stuff still failing explosively despite the Toray 60T plys and the best resin formulas and most monitored autoclaving. I think most folks are in agreement that despite having a female astronaut orbiting the planet, their average commercial product manufacturing quality standards and practices are somewhat...inconsistentent...putting it politely.

Myself? I would rather pay the weight penalty for more material to try and gain more integrity regardless of the material. I don't mind buying 'cheap' as long as it's reasonably safe/durable.

Why?

See below...it hurts!

 
e_guevara said:
I'm sorry but scientific inquiry requires that results should be repeatable and measurable (within a given level of confidence)  given the same procedure every single time. In the laboratory where conditions are controlled and tests are carried out by machines and computers that can replicate the procedures exactly every single time, this holds true. Manufacturers do subject their products to rigorous scientific testing (usually until failure) and collate the data to determine the "safe" operating conditions for the product. And most of the time, this is where we base our idea of failure on IMO. However, due to the infinite amount of variables and combinations thereof, one cannot test for every conceivable condition that can happen. Therefore tests are done with a "controlled" set of variables. Out on the road, riding conditions are far from the "perfect" lab conditions. Your counterexample only shows that a particular condition (even if it is a slight variation from the control) was not considered during lab testing.
You're right, but the statement that carbon frames or bits don't always fail catastrophically wasn't and can't be construed as a scientific statement. Pointing out that a few didn't fail catastrophically shows that the "always" case is not true. You'll note that it's a very rare case where someone would test for an "always" result. In fact, it's the rare experiment that doesn't have some probability or uncertainty associated with it. I can think only think of one right off (it's impossible--given current theory--to measure both the particle nature and wave nature of light at the same time). Manufacturers can certainly quantify the rates of and types of failures seen via testing, and it's certainly something that many should do. Still, the statement I made only requires that one or more frames and/or carbon bike bits have not catastrophically failed when there was a failure in the CF matrix. Remember, "always" is 100% with no rounding.
 
dhk2 said:
What you're saying is that all it takes is one solid example to disprove a theory such as " CF just snaps when it fails".   Like the theory of gravity, or the absolute universal limit of the speed of light:  find just one contrary example that can be proven and the whole principal is then called into question.   When scientists at CERN recently said they recorded a neutrino that had apparently traveled faster than the speed of light, the announcement attracted huge attention in the physics community because if it were true, Einstein's Theory of Relativity would be invalid.  When talking about failures of bike frames and forks, it might be helpful to start by seperating categories of failures.  Catastrophic overload failures caused by crashing or otherwise exceeding the ultimate design load limits are different than normal "wear and tear" fatigue failures.   
You're wrong and right. The wrong part: it's beginning to appear that the CERN experiment result was due to an error of some sort. Moreover, even if true, they're experiment would not have invalidated Einstein's Special Theory of Relativity, it would only have sent notice that a modification was needed. Classical physics wasn't invalidated with the emergence of quantum physics. Instead physics was modified to that each theory QM and Newtonian Physics applied to the correct realm (small, slow vs. big, slow). And as has to be the case QM if applied everywhere at the quantum level and scaled up to the macro level gives the classical physics result in that limit. With the CERN experiment, it has done nothing to invalidate the value of 'c'. 'c' has given accurate results in physics so far. After all, we can accurately model the strength of a nuclear blast, and it is based on the speed of light, and we have experimentally shown the Special Theory of Relativity to be accurate. There is no theory that "CF just snaps when it fails". Further, there is no theory that a CF frame will with certainty catastrophically fail if the CF matrix fails. In fact there is no test that can be done to prove that a CF frame will fail catastrophically every single time the CF matrix fails at some point. The answer to the OP is that some CF frame cracks can be repaired, just like some Ti, steel, and even aluminum cracks could be repaired. It's also true that there are some cracks that render CF frames dead, just like there are some cracks that render Ti, steel, and aluminum frames dead. And this is where you are absolutely right: it's a lot more helpful to talk about the types of failures since with all bike frames, it's the type of failure and its magnitude that will determine if it is repairable or not.