Though likely not unanimously agreed on, it seems trailing spokes should be heads out. After breaking a hub flange where the trailing heads were heads in, I'm a little spooked seeing my front hub's trailing spokes are heads in too. It's been on loaded touring for thousands of miles and the marks where the spokes have been seated are very promiment.

Is it better to keep the "incorrect" lacing or make new deformations by correcting the lacing?

  • The Kindernay swap cage is very different from a normal IGH hub shell and I suspect (but don't know) that caused the failure. Most IGHs are very large OD, so braking forces are easily shared between the flanges, and drive forces are transmitted into the shell somewhere in the middle by a gear on the core engaging teeth on the shell. Both are completely different on the Kindernay design: the NDS flange is probably handling all the loads both from braking and driving. This one wasn't built according to their instructions and it failed, and I suspect there's an engineering reason there. Dec 23, 2023 at 16:29
  • 1
    "trailing spokes should be heads out" - Well, no. The spokes with the largest transient loads should be heads in. Simply because being heads out puts additional stress on the spoke's bend, and that stress should not be added to the max transient loads. I.e. for a rear wheel, the accelerating spokes should always be heads in (they work harder on steep ascents than when braking). For a front wheel, you only have directional transient loads if you have a disk or hub brake, and in that case the decelerating spokes should be heads in. That's the opposite from the rear wheel. Dec 23, 2023 at 19:46
  • @cmaster-reinstatemonica Interesting. I follow the logic a bit, but are decelerating spokes leading or trailing?
    – Laoshi
    Dec 23, 2023 at 20:22
  • That depends on how you define "leading/trailing". I define an "accelerating" spoke as one that works extra hard when you accelerate, and a "decelerating" spoke as one that works extra hard when you brake with a hub/disk brake. And I define "works extra hard" as being under extra tension load. I find that fairly unambiguous, whereas I'm always confounded by whether "leading/trailing" refers to the hub or the rim. Dec 23, 2023 at 20:35
  • 2
    That would be an accelerating spoke in my parlance. They drag the rim forward, so to speak. So, with that your wheels were laced exactly in the way that I would want a rear wheel to be laced (= robust in acceleration). For a front wheel with a disk brake, those trailing spokes should be heads out because they are not stressed when braking. Here you need the leading spokes that try to hold the rim back to be extra robust (= heads inside the flange, bend on the outside). Dec 23, 2023 at 21:24

1 Answer 1


For most hubs and situations, the answer to this is no, don't rebuild it if there are no issues occurring. "Double grooving" a hub shell creates a lot of unnecessary stress risers and they've been known to fail because of it. Though rare, this is a real thing that happens, and the risk tends to outweigh any benefit of switching the lacing. For most applications, the main practical import of which side gets the head has to do with the risk of running out of clearance between the brake caliper and/or rear derailleur and the spokes under braking/drive load, and if these aren't issues on a particular bike (which they usually aren't), there is no benefit in changing.

The Kindernay swap cage you broke is an unusual animal. It appears that all of the braking and driving loads are going through the non-drive-side flange only. Kindernay explicitly recommends the NDS trailing spokes going up the inside of the flange.

enter image description here

It wasn't done this way and it broke. I'm intrigued by that but can't explain it. However, flanges breaking is not the normal consequence here, and I suspect there's an engineering reason specific to the Kindernay design that it happened in this case.

  • 1
    For those of you wondering why Nathan mentions some niche Norwegianly designed product out of the blue, I alluded to this hub's flange breaking in a different question: bicycles.stackexchange.com/questions/91758/…
    – Laoshi
    Dec 23, 2023 at 17:11
  • Yes, the first time I broke these it was beacuse the spokes did not first cross over the cutout between each pair of spoke holes, and the second time was presumeably having the heads in (although making a minimalist shell out of cast aluminum didn't help). Some people never learn. The flanges tilt slightly inward, is this a common feature or one that may neccessitate a particular spoke head orientation?
    – Laoshi
    Dec 23, 2023 at 17:17
  • 1
    @Laoshi It's good they tilt inward, this is called canting (canted flanges) and is a feature some of the very best, most expensive hubs have. My understanding is it adds significant machining complexity. It's done to "aim" the spokes at the rim, which reduces the manual work needed by the builder to correct the spoke line and adds to the probability of success (good because gauging it visually can be an imperfect process, so built-in advantages are welcome). The more shallow the spoke angle (big flange, small rim), the more you want this. Dec 23, 2023 at 17:33
  • @Laoshi out of curiousity, I wonder if you wouldn't mind posting a picture of the broken flange. Dec 23, 2023 at 20:18
  • Really appreciate the interest! Popping the pic in the question part might confuse people. imgur.com/a/0Rnxjss. Then after adjusting the adjacent spokes to re-true the wheel, it broke two more times over about 400 miles
    – Laoshi
    Dec 23, 2023 at 20:52

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.