Dura Ace crank crank fatigue

[drewski]

I was on a group ride this morning. I was climbing a hill 25 minutes into the ride, when the non-drive side of the crank cracked cleanly.
Shimano DA 7800 crank.

I was taken to the Hospital about 3 hours ago and as precaution they took a cat-scan luckily just bruises.

I wonder what the dang cause of the fatigue on the crank was.
Should I consider a new brand?

https://picasaweb.google.com/andrew.israel3/SerottaForum#

[avalonracing]

I was on a group ride this morning. I was climbing a hill 25 minutes into the ride, when the non-drive side of the crank cracked cleanly.
Shimano DA 7800 crank.

I was taken to the Hospital about 3 hours ago and as precaution they took a cat-scan luckily just bruises.

I wonder what the dang cause of the fatigue on the crank was.
Should I consider a new brand?

Too much power there big guy.
The 7800 are built like tanks and can take the fastest sprinters in the world laying down every watt that they have. You obviously had a defective set. You should contact Shimano (even if you didn't buy them new) to let them know. Maybe you'll save a life.

[jpw]

Did you have it from new, and did it previously ever have a hard impact?

[Kontact]

I would also assume that it is a random forging defect, rather than simple fatigue. That crank would have had to have been flexing pretty far to fail due to fatigue, and failures of this type with all cranks seems to be pretty rare.

[regularguy412]

Would be interesting to know where on the arm the crack occurred. Possibly it was weakened from a previous crash or accidental drop/impact?

Pix would help here, I think.

Mike in AR:beer:

[Fixed]

glad your ok docs are good now nothing hurts cos they do everything through your wallet now .
no warning like strange creaking sounds?
cheers i have some campy record cranks and ultegra cranks in new shape if you need them
cheers

[djg21]

Would be interesting to know where on the arm the crack occurred. Possibly it was weakened from a previous crash or accidental drop/impact?

Pix would help here, I think.

Mike in AR:beer:

I cracked a DA-7400 crank years ago where it had rubbed periodically against the front derailleur, and had been slightly scored. I was a little surprised, but didn't go down, and chalked it up to equipment wear due to operator failure.

[11.4]

Photos, please. A crack in a 7800 series crankarm is almost unheard of. Even with crashes and other impacts, 7800 crankarm fractures are almost nonexistent.

Older crankarm designs, particularly the solid designs without the current metallurgy, occasionally cracked because the metal simply couldn't be engineered with the quality they achieved once they adopted hollow axle designs. I don't think I've ever seen a broken 7700-series track crank either, which was the first adoption of the completely forged hollow crankarm design with the new metallurgy. I've seen one 7900 crank damaged, but it was in a car crash where the bottom bracket shell was actually crushed, so you can imagine the forces applied. But still no cracked ones. Crankset failure used to be commonplace with solid cranksets -- I saw them over and over with old Campy Nuovo Record -- but no longer.

[Kontact]

Photos, please. A crack in a 7800 series crankarm is almost unheard of. Even with crashes and other impacts, 7800 crankarm fractures are almost nonexistent.

Older crankarm designs, particularly the solid designs without the current metallurgy, occasionally cracked because the metal simply couldn't be engineered with the quality they achieved once they adopted hollow axle designs. I don't think I've ever seen a broken 7700-series track crank either, which was the first adoption of the completely forged hollow crankarm design with the new metallurgy. I've seen one 7900 crank damaged, but it was in a car crash where the bottom bracket shell was actually crushed, so you can imagine the forces applied. But still no cracked ones. Crankset failure used to be commonplace with solid cranksets -- I saw them over and over with old Campy Nuovo Record -- but no longer.
http://www.bikeforums.net/archive/index.php/t-464389.html
http://www.flickr.com/photos/timlupfer/4984429840/
http://www.cyclingcrowd.com/Uwe/Forum.aspx/technical/19097/Broken-Dura-Ace-7800-crank

No point being made here, but the fun part is that a quick search for "broken dura ace crank" pulled up all 7800 failures, rather than a mix of 7800, 7700, 7410 and 7400. However, that's only 5 that came up without much digging (including the OP's), so they can't be all that fragile.

[Charles M]

It's a percentage thing like anything else...



Show pics... Talk history / life of the crank / milage, past crashes etc.

anything can be broken. Almost nothing fails without warning (not all warnings are noted).

It's a whole other thing to talk about exactly why rather than speculating.

[oldpotatoe]

I was on a group ride this morning. I was climbing a hill 25 minutes into the ride, when the non-drive side of the crank cracked cleanly.
Shimano DA 7800 crank.

I was taken to the Hospital about 3 hours ago and as precaution they took a cat-scan luckily just bruises.

I wonder what the dang cause of the fatigue on the crank was.
Should I consider a new brand?


Where did it break? Pedal hole, at the spindle, in between?

[1happygirl]

OMGosh Glad you're okay!!!!!!

docs are good now nothing hurts cos they do everything through your wallet now

cheers

Fixed hahahahah

[bike22]

friend of mine sheared a 7800 crank clean off where the ds crank meets the crank/bb spindle.

[avalonracing]

No point being made here, but the fun part is that a quick search for "broken dura ace crank" pulled up all 7800 failures, rather than a mix of 7800, 7700, 7410 and 7400. However, that's only 5 that came up without much digging (including the OP's), so they can't be all that fragile.

Everyone has a story of how something "has never broken" and how "something always breaks". At least it is easy to disprove everything on the net.

A few years back I was about to purchase a Bontrager Race Lite MTB frame from a friend. Another friend who worked at a shop said, "Sweet! You know there has never be a report of one of them breaking." Hard to believe on a MTB with a skinny little monostay. Anyway, a week later my friend called to say that he couldn't sell me the bike anymore because he found a crack on the monostay when he was tearing it down.

[mjbrekke]

A friend of mine had the same thing happen a couple years ago while starting from a stop light at a busy downtown intersection. He was out of commission for two months recovering from the fall. Fortunately, he was wearing a helmet, but still had a headache for two weeks. He had an older Shimano crank and vowed never again to install a used crank of unknown history on one of his bikes. I'm not going to tell him about the OP's incident. Sometimes, ignorance is best.

[Louis]

Should I consider a new brand?

I wouldn't.

I doubt that there is a cycling manufacturer out there who hasn't had some sort of failure with their parts. If this were a newbie manufacturer just getting started I would say yes, but given Shimano's sterling reputation, I see no reason to switch based on this one data point.

[drewski]

Photos will be coming once I get the bike back. I am glad to have survived
with bruises and scratches.

The crack occured a about 3 inches from the bb spindle.



Andrew

[shiftyfixedgear]

At a shop I worked at I've seen a broken 7800 road crank that was just like the OP. Not raced, just a recreation rider Walter Mitty racer-type.

I've also replaced two broken 7700 road arms on repair bikes. Not crashed, just fatigue from use. One broke across the middle of the arm, and one cracked at the top near the axle forging.

One crankarm I've never seen broken is the Ultegra 7 & 8 speed 6400 series. All of the 7400 series Dura Ace arm designs, yes. All of the older Campy Record and C-Record designs, yes. All of the various Frenchy makes, too.

But never an Ultegra 6400. Weird.

[cnighbor1]

Save all the broken pieces. by looking and running tests shimano can tell what happened
charles

[RPS]

I cracked a DA-7400 crank years ago where it had rubbed periodically against the front derailleur, and had been slightly scored. I was a little surprised, but didn't go down, and chalked it up to equipment wear due to operator failure.
I thought it was coming from the shoe rubbing on the pedal.

Not saying it happened in either case above, but in general any kind of unusual wear that resembles a deep notch in the metal should be avoided because it can serve to concentrate stresses, which in turn expedites fatigue failures.

[dlflemingos]

The first part of fatigue crack initiation shouldn't happen and was probably caused by a flaw or some damage (scratch). If it was in a high stress region it would lead to crack growth till it reached critical crack length (and failed catastrophically). If you can get pictures of the surface good enough to get measurements on the growth and failure (It should be fairly obvious) it would help. You may even be able to see some sort of flaw at the initiation point. A scanning electron microscope would help

How many miles, hills, how much do you weigh?

[Louis]

Was there an EB weld anywhere near the crack?



(that's an inside joke for dlflemingos)

[AngryScientist]

glad you're relatively OK drew, of all the things in my drivetrain, an alloy crank arm is practically the last thing i expect to catastrophically fail. would love to see the pictures.

if it were me, just out of superstition/principal, i would look toward another crankset.

[ultraman6970]

Happy you are ok, in general i don't see or highlight a crankset as unbreakable. I'm a campangolo guy and I'm sure campy cranks snap, haven't heard any case of the new UT ones but back in the day the aluminum ones did, Shimano ones did also. Remember the edco ones, darn those if you were too strong snapped after a few months only.

Pictures pleasee....

[ss-jimbo]

Interesting pics of two different broken 7800's.

home.comcast.net~bicyclebill/DACrankset/

[johnnymossville]

A friend of mine had a 7800 crank arm crack right in half cleanly as well while riding, no wrecks, no obvious shoe rubbing. I think I've heard of at least a half dozen cases of this happening now.

Could be a few bad batches out there. Shimano replaced it no questions asked.

[m.skeen]

You guys aren't making me feel any better.

[Mark McM]

You guys aren't making me feel any better.

Maybe this well help a little bit. Take a look at the photos in the web page cited just above, especially at the first breakage (the one that broke across the pedal hole). You'll notice that there are two distinct colored/textured regions. If you look closely, the area closest to the pedal will show more indications of wear/corrosion, and the other area will appear more jagged. This is a classic progression of a fatigue crack. After a crack is initiated, it will slowly grow, a little bit at a time as it is loaded and unloaded. As the crank flexes slightly with each load cycle, the two faces on either side of the crack can rub against each other, which will leave wear marks on faces in the crack. Moisture can also seep in, which may cause some surface corrosion.

As the crack grows, and the remaining cross-section is reduced, the part will become weaker, until there comes a time when one final peak load (the "straw that broke the camel's back") will cause the remaining cross-section to fail in a brittle mode. This is leaves a more ragged surface across the broken halves.

So, what's the part that can make you feel better? The knowledge that these types of breaks don't occur without a detectable warning. If someone had examined these cranks before the breakage, they would have been able to detect that growing crack before it eventually broke. This is why periodic equipment inspection is important - many eventual breakages can be detected far in advance of their final failure, and this crank is an example of that.

[mpetry]

Good explanation. The one that failed at the pedal hole looked like a notch fracture from the threads. At least to me.

Mark Petry
Bainbridge Island, WA

[benb]

So, what's the part that can make you feel better? The knowledge that these types of breaks don't occur without a detectable warning. If someone had examined these cranks before the breakage, they would have been able to detect that growing crack before it eventually broke. This is why periodic equipment inspection is important - many eventual breakages can be detected far in advance of their final failure, and this crank is an example of that.

Thank you for the explanation..

Questions:

Is this the type of fatigue crack that could have been detected with the naked eye during parts inspection, or would it have required special tools?

How the heck do they forge these to be hollow with such a complex shape? It's easy to take parts for granted.. it is pretty amazing stuff when you take a step back and look at it.

[Kontact]

Good question about how they do this. I would bet that it is forged in stages - initial forging is done around a mandrel, with subsequent forging done away from the hollow section to close up the ends after the mandrel is withdrawn. It could also be done with a hollow casting that is forged down to final shape. Neat stuff, and almost impossible to do in small scale production.

[m.skeen]

Maybe this well help a little bit. Take a look at the photos in the web page cited just above, especially at the first breakage (the one that broke across the pedal hole). You'll notice that there are two distinct colored/textured regions. If you look closely, the area closest to the pedal will show more indications of wear/corrosion, and the other area will appear more jagged. This is a classic progression of a fatigue crack. After a crack is initiated, it will slowly grow, a little bit at a time as it is loaded and unloaded. As the crank flexes slightly with each load cycle, the two faces on either side of the crack can rub against each other, which will leave wear marks on faces in the crack. Moisture can also seep in, which may cause some surface corrosion.

As the crack grows, and the remaining cross-section is reduced, the part will become weaker, until there comes a time when one final peak load (the "straw that broke the camel's back") will cause the remaining cross-section to fail in a brittle mode. This is leaves a more ragged surface across the broken halves.

So, what's the part that can make you feel better? The knowledge that these types of breaks don't occur without a detectable warning. If someone had examined these cranks before the breakage, they would have been able to detect that growing crack before it eventually broke. This is why periodic equipment inspection is important - many eventual breakages can be detected far in advance of their final failure, and this crank is an example of that.

Informative post. My point was that I have never thought about my 7800 cranks breaking until I read this thread. Not that reading this will effect whether I ride on them or that I thought they were indistructable in the first place. I will start doing better parts inspections during routine bike cleaning though. Might be a good excuse to try out Campy though ...

[rice rocket]

Good question about how they do this. I would bet that it is forged in stages - initial forging is done around a mandrel, with subsequent forging done away from the hollow section to close up the ends after the mandrel is withdrawn. It could also be done with a hollow casting that is forged down to final shape. Neat stuff, and almost impossible to do in small scale production.

Did you see the material? Looks like a crappy casting.

[Kontact]

Did you see the material? Looks like a crappy casting.
Despite how it looks to you, Shimano advertises it as "hollow tunnel forged arms". That isn't a casting.

[drewski]

glad you're relatively OK drew, of all the things in my drivetrain, an alloy crank arm is practically the last thing i expect to catastrophically fail. would love to see the pictures.

if it were me, just out of superstition/principal, i would look toward another crankset.


Thanks

Angry and in Pain cyclist. Having rib pain and major muscle
spasm in my back. Might be few days before I can upload photos.

:crap:

[chuckred]

only carbon could fail catastrophically?

(sorry - just kidding - but I thought it had to be said).

[drewski]

At a shop I worked at I've seen a broken 7800 road crank that was just like the OP. Not raced, just a recreation rider Walter Mitty racer-type.

I've also replaced two broken 7700 road arms on repair bikes. Not crashed, just fatigue from use. One broke across the middle of the arm, and one cracked at the top near the axle forging.

One crankarm I've never seen broken is the Ultegra 7 & 8 speed 6400 series. All of the 7400 series Dura Ace arm designs, yes. All of the older Campy Record and C-Record designs, yes. All of the various Frenchy makes, too.

But never an Ultegra 6400. Weird.

I looked at the crank last night. There is a spot inside of the crank body,
where there is a small internal rivet. The crank cracked right along the
location on the rivet. The rivet is a couple of inches down from the hole where the crank attaches to the spindle.

[old_fat_and_slow]

http://i461.photobucket.com/albums/qq332/catchernrye/jokes/thread_is_worthless.jpg?t=1302558002


....

[m.skeen]

Thanks

Angry and in Pain cyclist. Having rib pain and major muscle
spasm in my back. Might be few days before I can upload photos.

:crap:

Hope you get better soon.

http://i461.photobucket.com/albums/qq332/catchernrye/jokes/thread_is_worthless.jpg?t=1302558002


....
+1 :)

[old_fat_and_slow]

Different Dura Ace 7800 crank failure story with a few photies here (http://weightweenies.starbike.com/forum/viewtopic.php?t=18179).


Lots of broken aluminum cranks with photies and other aluminum components for yer viewing pleasure here (http://pardo.net/bike/pic/fail-001/000.html).


Of course this (http://www.bustedcarbon.com/) is always fun to read about all the latest crabon fiber failures.


..

[oldpotatoe]

Different Dura Ace 7800 crank failure story with a few photies here (http://weightweenies.starbike.com/forum/viewtopic.php?t=18179).


Lots of broken aluminum cranks with photies and other aluminum components for yer viewing pleasure here (http://pardo.net/bike/pic/fail-001/000.html).


Of course this (http://www.bustedcarbon.com/) is always fun to read about all the latest crabon fiber failures.


..


PEZ is gonna put a pox on your household.

[drewski]

https://picasaweb.google.com/andrew.israel3/SerottaForum#

Not the best pictures in the world. Here you go people.

[Louis]

https://picasaweb.google.com/lh/webUpload?uname=andrew.israel3&aid=5593965458957149585

Not the best pictures in the world. Here you go people.

Does one have to have an account to see them?

[drewski]

Does one have to have an account to see them?


Try the link I edited.

[tuscanyswe]

Wow that looks scary. Glad you are okay!

Im riding one those cranks on the crossbike. I dont think ill worry about it but its scary such a clean brake can happen without any notice.

[beer_meister]

https://lh5.googleusercontent.com/_iUlROrR7FH4/TaOUUByhjkI/AAAAAAAACIM/NS_3lnqkKZA/s1024/

It's hard to say for sure, because the photos are not clear enough, and the flash is a little too strong. From this pic, I would guess the crack initiated from the 10 o'clock position and the final break was at the 4 to 5 o'clock position. The final break is indicated by the large jagged edge. That indicates ductile fracture, and is usually how fatigue induced failures ultimately fail.

In the pic below, note the crank orientation has been rotated 180 degrees from above, what I believe to be the crack initiation site, is the dark area just to the right of the rivet. The really darkest area, on the inner-most side of the crank arm, is my guess where the crack developed.

Probably a typical Mode I type crack. Opening force being supplied by the downward force on the pedal. Hard to believe the crack appears to have propagated about 270 degrees before the stress intensity hit K-Ic.

https://lh6.googleusercontent.com/_iUlROrR7FH4/TaOULHruILI/AAAAAAAACIU/atEdIEz69Mc/s1024/metropolitan



If you can, try to take some additional pictures with
a DSLR that has a special macro lens that enables extreme close-up focusing. The flash needs to be dialed back a little too.


My $0.02. And worth every penny.

[Kontact]

Hard to imagine that failure and the rivet are unrelated.