Dura-ace chainwheel failure

Check with a machinist. They may be able to make and fit a new axle made of better quality and thicker walled (and gasp!!! slightly heavier) stronger tubing which won't corrode.

Better buy a new one dude, that looks screwed!!!

It doesn't look like corrosion caused that, but it's hard to tell: the broken part is blurred and facing away from the camera..! It would be interesting to see a focused (macro?) picture of the broken edge.

Either way, I'm sure everybody else is right - that axle doesn't look replaceable.

no difference that you'll notice...

it's allunminium isn't it? Shouldn't corrode.

I'm not positive but I believe the spindle is pressed and/or bonded to the crank. Two different pieces of metal. That is why I think you should ask a machinists' opinion as to weather it's possible to make a new spindle. Not only will it be cheaper but possibly a stronger replacement can be made. Bicycle manufacturers are always looking for ways to save a few grams on parts sometimes at the expense of durability. Most crankshafts in car engines use pressed in pins to hold them together so there are people out there able to do such things and in the case of a bicycle crank would be considered a very low stress part compared to something producing 300+hp. As for replacing the 172.5mm arm with a 170mm. I think some people may not notice but most probably would. I sure can tell the difference between different length cranks and wouldn't want to use a different size.

scherrit wrote:

A bike crank would be considered a high stress part as a human generates close to as much torque as an average 2 litre petrol engine. higher hp/wattage for the petrol engine is made up by turning a lot faster than a bike crank, but forces from a human are surprisingly high, and your DA equipment is more Ferrari than toyota in terms of over-design.

Buy a new one.
Good luck,
S.

Agreed about the high stress but nowhere near the stress that an engine spinning at 5,000+ RPM, pushing 2,000+Lbs of weight, developing 120+ horsepower and generating huge amounts of heat can produce. Not to mention that a car crankshaft is expected to last well over 120,000 miles of this. In any case the point is that small engine shops have the capability to do a similar job as what is needed for your crankset. Why waste an expensive, repairable crank without checking out a fix that may be an improvement over the old spindle?

Looks like a fatigue failure to me, but a picture of the broken surface would make it obvious.

rockmount wrote:

will3 wrote:

Looks like a fatigue failure to me, but a picture of the broken surface would make it obvious.

So long as you have your camera attached to one of these, that's what we usually use to determine failure modes in metal components >>

really? seems a bit OTT to me, fatigue failures in hardened materials usually leave signs pretty obvious to the naked eye, eg polished area of original crack followed by 45dedgree fast fracture area etc etc. Imean, sure, to be absolutely certain, but given the highly cyclic loading etc, you'd be in with a good chance of being right just by looking at it.

rockmount wrote:

...and you'd be able to determine the cause of fatigue initiation, not to mention counting striations with the naked eye to determine cycles to failure etc ?

Did I say I could?
No merely fatigue failure/not fatigue failure. Sure your microscope will spot the inclusion, void, or what have you that might have been the root cause.

Rockhopper wrote:

I'd still be going back to my dealer with it - a chainset shouldn't fail like that after two years use.
Getting a new one made and pressed in would probably cost more than a new chainset from Shimano - someone would have to do a very accurate drawing of it first.

+1 I doubt getting it fixed is an option. Now warranty, I'd see about that and I'm sure Shimano would be very interested in the broken one(quality control and all that).
Just a guess on my part but I would say it was a freak defect and not something that happens with any regularity.