Do spokes exclusively fail in shear at the little hole through which they pass in the hub? To clarify, have you witnessed spoke failures in other areas?

Perhaps to material defects and thus excess tension causing necking, or shearing at the rim from a torque, or some other failure mechanism?

This question is purely academic. (And I would guess that necking is highly unlikely, hence why I say by defect, because these things can take very large tensile stresses as I understand it. Also, have you ever witnessed any severe radial loads that caused a spoke to buckle?)

Straight gauge spokes usually fail at the elbow where it exits the hub as this is where the spoke is subject to the greatest fatigue stress and this increases greatly when the spoke is not properly bedded against the hub flange.

Butted spokes will often fail in the middle where they are thinner while the thicker butted ends tend to resist breakage more.

With a well built wheel spoke breakage should not be a concern if the spoke tensions are maintained.

Spokes are only loaded in tension, so there are no compressive loads to cause buckling.

Wouldn't the spoke have to accept an impulse normal to the contact patch? Or is that absorbed entirely by tire & rim? That seems unlikely, because they have no "anchor" or support to react to the force. That's what I mean by asking. Because impulses get transferred to the seat in some way, and the only way I can think of is through the hub...and to get there from the rim/tire, it has to pass through the spoke. If there is reaction at the hub, that's a compressive load.

No, the reaction is caused by tension in the spokes above the hub, not compression of the spokes below the hub. That's a simplified version.

Right. Makes more sense. Blame not looking at the big (sensible) picture.

Thanks for the responses! I didn't even realize there were different spoke designs.

No, but elbow failures are most likely.

Spokes fail due to fatigue, with the number of cycles (about 750 per mile) dependent on both mean stress and the magnitude of the variation.

When the elbow is formed not all of the spoke is taken past its elastic limit thus leaving areas with high residual stress far beyond everyplace else in the spoke so non-stress-relieved wheels at sufficient tension always fail first at the elbows.

With insufficient tension you might get spokes going slack and bending near the nipples for a high variation and failures there.

Personally I gave up on wheels built by machines or shops with potentially marginal mechanics about 15 years ago and haven't had a broken spoke since then. Jobst Brandt (author of _The Bicycle Wheel_) claims over 100,000 miles on 15/16 gauge double butted spokes which are properly tensioned and stress relieved.

Threading , rolling the threads in , they are not cut, does create a stress riser ,,
and then they fail at the threaded end.

A reasonably healthy spoked wheel is a pre-stressed structure. Means it's already under tension, even before there's any load put on it by the bike.
Spokes between the contact patch and the hub reacts by losing some of that pretension as a reaction to the load.
The other spokes will see an increase, but it's so well distributed between the rest of the spokes that it can actually be hard to measure. The loss of tension in the contact patch spoke(s) however is much bigger, and far easier to measure. Somewhere here the misleading phrase that a wheel stands on its spokes (as that's the most visible difference) is born.

On top of that a spoke is too slender to be able to carry any significant load in compression.

No.

Yes. Running high cross patterns on high flange hubs can cause the nipple to sit at an angle relative to the spoke out by the rim, something which the threaded part of the spoke doesn't particularly appreciate.

Particularly rear wheel, and hub braked front wheels can experience fretting at the spoke crosses, which can be bad enough to act as a stress risers and eventually become failure points. Dropped chains and poorly adjusted rear derailers can also chafe against spokes and prompt a failure.
Occasionally there are reports about bikes in coastal regions suffering spoke failure induced by the corrosive environment.

Never seen necking. With the bend, the nipple interface, and usually the rim too being weaker points there's really not much chance to get up to the loads required to cause that kind of deformation.

All the spoke fatigue failures I've ever seen were at the elbow or, more rarely, at the nipple threads. The only spokes I've ever seen break in the middle, butted or straight gauge, were those damaged by a thrown chain or other road hazard.

Anyone care to make a free body diagram of the hub?

I just repaired my older daughter's front wheel where three spokes failed. Actually the heads of the nipples sheared off opposite where the rim and a curb collided at speed. The failure didn't show up immediately, but rather several miles later.

Brad

Too many engineers on BF.

If you're into the science, engineering mechanics, aerodynamics, materials, and human physiology of bicycles and cycling, check out "Bicycling Science" by David Wilson (MIT Press). Good fun reading, if you're into that kind of stuff.

The most common place (by far) for spokes to break is right at the elbow. That's usually due to inadequate tensioning.

Another common place for spokes to break is where they enter the nipple. I solved that problem on a Bike Friday triple by relaceing the rear wheel 2 cross rather than 3 cross.

The only other place where I've seen a spoke break has been due to overshifting the chain and buggering the spokes.

I wouldn't say that failure midspoke is any kind of failure that happens 'often'. Failure at the crosspoint might happen if the spokes were particularly loose or very old.

Failure at the elbow is by far the most common mode. Failure at the spoke nipple is next most common. From my own experience, I'd put the ratio at greater than 100 elbow failures to 1 nipple failure. I've never had a spoke fail at any midpoint.

https://yarchive.net/bike/wheel_stresses.html

Already have it and read it frequently. But there's just something more memorable about discussion/banter/argument...

I have seen spokes break about 2cm away from the elbow and nipple ends. These are forced breaks due to overcoming the ultimate strength of the spoke due to side-impacts and crashes. This is very, very rare compared to the most common fatigue failure at the elbows. Even with double-butted spokes, they still fail most often due to fatigue at the elbow. I have a 20-year old wheel with 14/15ga spokes with Mavic MA40 rim. Has over 60k-miles on them and going strong.

Dang, Danno- you're starting to sound like Jobst Brandt!