Given that we're moving away from QRs to through-axles - partly driven by the torque needs of disc brakes, why not follow the natural progression and move the wheel bearings from the hub to the forks themselves.

Generally you want bearings as close to the load as possible. Having them in the fork is as close as one could hope. It'd also allow the axle to be very thick, which would add strength.

Given that every idea in bicycling has been tried at least once, has this been done before? What are the demerits of such an idea?

  • Do you have a sketch of what you have in mind? I haven't thought about it much (e.g. if theres any forces on the frame/fork if you move the bearings there that would cause problems/redesigns, such limiting damage on impact), but you'd want to keep wheel removal and installation relatively easy and its not clear if moving the bearings into the fork would keep this the case. Also, I'd think bicycle manufacturers would somehow use this to make wheels proprietary to their brand of bike.
    – Batman
    Dec 13, 2016 at 6:14
  • 3
    FWIW, most unicycles are essentially just like this, although there it's done so the axle (now a spindle) can be connected to the cranks. Interesting idea. Dec 13, 2016 at 7:06
  • 5
    Manufacturers will love it - think of all the existing wheels, forks and frames they can make obsolete.
    – mattnz
    Dec 13, 2016 at 9:11
  • @mattnz -- right? Which is why I'm surprised that Shimano hasn't come out with the Super-Dyna-Motion Thru Hub 3000.
    – RoboKaren
    Dec 13, 2016 at 22:04

1 Answer 1


Interesting idea but the front axle (whether is be a 15mm through-axle or a 9mm QR axle) does not rotate, the hub-body and by extension the rest of the wheel rotates about the axel via the bearings. Making such a system where the axel does rotate (e.g. unicycle fork) would require creating a capture system (which is the current function of the QR) to secure the wheel to the fork that would likely be less user-friendly that the quick-release based systems we have come to expect.

  • Through-axle clamping systems already exist. The only difference would be the location of the bearing, it wouldn't add or detract from the complexity of the through-axle clamping mechanism per se.
    – RoboKaren
    Dec 13, 2016 at 16:30
  • 4
    @RoboKaren And as I describe.. in those systems the axles do not rotate.. if they did your quick-release lever would rotate with every rotation of the wheel. So to make them rotate would mean replacing the purpose the through-axle and QR systems currently serve... clamping the wheel to the fork. If you look at a unicycle their are no quick releases... you have 4-8 hex head screws holding the bearings in place on the underside of the fork.
    – dafew
    Dec 13, 2016 at 16:36
  • 4
    With bearings clamped in quick release, the quick release would have to be extremely accurate. With solid axle, you can get away with gripping the axle with brute force.
    – ojs
    Dec 13, 2016 at 17:24
  • Specifically, with the bearings part of the quick release they have to be moved on and off the bike a lot. Normal bearing mounts are a firm press fit and that is only good for a few cycles before it wears out (10 would be a good estimate, not 1000). So you'd end up with the bearing mount being bolted/unbolted to the fork as the "quick release" process. I've done this for the "bottom bracket" on a recumbent and it took four bolts and was a PITA to build and moderately annoying to perform (but didn't require pulling the cranks off and removing the BB, which was the goal)
    – Móż
    Dec 15, 2016 at 0:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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