I've broken 2 aluminium cranks!
Here's one I just cracked, though I'm sure it would have broken if I hadn't noticed the damage when cleaning the bike.
View attachment 504321
[*** THIS is to save
@Yellow Saddle having to tell me that I had overtightened the bolt!
***]
Is this a challenge? I am now gonna tell you that you didn't overtighten the cranks....in a manner of speaking.
Cracked cranks don't really crack the way most people think they do. That's because it is impossible to crack a crank by overtightening the bolt in one go. If you install a crank right now and turn that bolt way beyond the recommended 40N limit and on and on, the bolt will break but the crank will be perfect. To understand how a crank eventually does break, I've designed this little analogy which I've been using in the Yellow Saddle Labs for years.
Look at this photo.
The brush represents the crank and the dark surface represents the crank spindle. At this stage the crank is just installed loosely on the spindle and the bolt isn't in yet. As you turn the bolt, the crank quickly settles in a position where there is no more space between the crank and the square taper spindle and the fit is perfect. The straight-down bristles represent the fact that there's no friction between the crank and spindle yet and therefore no lateral force (left to right) between the crank and spindle. Everyone is happy, it's just nice and snug.
Now, we torque the bolt and this happens.
The crank is being forced into a place where it doesn't want to go. The friction between crank (red) and spindle (grey) creates a force between the two that wants to push the crank to the right. Unfortunately for poor Mr Crank, he cannot move to the right because the bolt is pushing it left.
Eventually you settle on 40Nm of torque and the crank and bolt and spindle is in equilibrium with no more movement. The crank still feels the pain because internally, it is being pushed to the right. The fabric of the crank feels this ever-present force that wants to push it off.
Now you go for a ride. When you ride, you push against the crank and the square hole in the crank pushes on the spindle and the whole thing moves forward. But what's happening inside the crank is what's so interesting. At the four corners of the square taper, we're now changing the pressure pushing between crank and square spindle. At the leading edges of the square the crank is pushing harder against the spindle and at the trailing edges, pressure is slightly relieved.
Have a look at this sketch.
The spoon shaped thing above is a nice Campagnolo aluminium crank. The square in the middle is the square taper crank spindle.
As the crank rotates cockwise, various points between crank and spindle change pressure. The arrows pointing towards the spindle indicates an increase in pressure and the arrows pointing away, a decrease.
Now imagine the crank interface as the brush above. At the beginning, before riding, all the bristles were bent backwards. But as soon as you put significant force on the crank, some of the bristles relieve pressure and return to the position in photo 1.
If you now stand on the crank when coasting, the opposite sides get relieved and so on and so on.
The nett effect is that the crank moves away from the crank bolt and further up the taper. That's because it is the only direction it can move to decrease pressure - the bolt prevents it from moving the other way and besides, the built-in pressure wants to be relieved in a certain direction only.
The crank crank thus tightens itself with use. Now Johnny DIY comes along with his fancy new torque wrench and during routine maintenance "ckecks" the torque on the crank and guess what, he finds it is only 20NM, not 40NM like in the beginning. The logical conclusion is that the bolt has loosened itself. However, it hadn't turned one iota, the crank distanced itself from the bolt and not it feels loose. DIY Johnny obliges, again and again, until the crank cracked.
