Last year, CeramicSpeed blew a few minds with their shaft-driven drivetrain concept on a road bike. Then they made it shift. Now, they’re addressing many of the comments about the design by showing how it can work on a full suspension mountain bike, too.

prototype ceramicspeed driven shift-drive chainless mountain bike drivetrain concept

The premise is the same, use a carbon fiber shaft to connect the front drive ring and the rear “platter” of circumferential sawtooth rings.

prototype ceramicspeed driven shift-drive chainless mountain bike drivetrain concept

Both ends use a series of small bearings that roll between the teeth…

prototype ceramicspeed driven shift-drive chainless mountain bike drivetrain concept

…with the front turning the shaft, and the shaft turning the rear. This particular prototype is only showing proof of concept for how it would all move along with a full suspension bike’s rear end. The inner shaft slides to expand and contract, much like a derailleur pulley cage would move to allow for chain growth and contraction.

Behind the “chainring” in the front is something akin to a ball joint that allows the shift to pivot along with the rear triangle.

prototype ceramicspeed driven shift-drive chainless mountain bike drivetrain concept

The design is built around 13 speeds, but this model was limited to 12 (they had to machine off the smallest series of teeth) to fit the modified Canyon frame. Part of the issue is that it’s still being built around a traditional freehub’s internals, but they say it requires something else because the torque is no longer being applied along the path of a chain. Instead, much of the forces are perpendicular to the driveshaft. Here’s a video explaining the system and showing how it will eventually all work together:

How does the CeramicSpeed drivetrain shift?

That’s the million dollar question, and the area that they’ve been working the hardest on. Thus far, they’ve had a rideable singlespeed version, and this prototype shifting version…but not one that actually combines both ride-ability and shift-ability. Part of it’s resources, they’re not a complete drivetrain company after all. But part of it is the complexity of getting it all to work seamlessly with a lot of moving parts, electronics, and speed.

prototype ceramicspeed driven shift-drive chainless mountain bike drivetrain concept

Originally, we thought the rear bearing ring would simply slide back and forth. But that wouldn’t work because it could fall out of alignment with the teeth as it passed from one row to the next. So, they had to split the drive ring into two, effectively using two bearings as a sort of clutch to release tension from the row it’s departing.

So, to shift, the shaft has magnets inside to determine its position and rotation. As the “leader” half of the drive ring rolls to the outside, it slides back or forward (depending on which way you want to shift) and starts to engage the next row of teeth.

As its doing this, the leading edge of the other half is able to slide into a small gap and move at the speed of the departure row, which would be different than the new row. This effectively disengages it from that departure row, and then that half slides into the same position as the first half to realign itself and continue to drive the bike forward.

All of the electronics and a worm-drive gear are inside the shaft. All you need to shift it is a wireless set of buttons, which could end up going anywhere on the front of the bike. A small magnetic charging port sits on the outside of the shaft and also allows for software updates.

As for worries about contamination gunking up the shifting, they’ve developed a cover that protects the system from mud, etc., And makes it even more aerodynamic than the already impressive results they achieved in the wind tunnel.

Sweet, when can I buy it?

Unknown. They’re still in development, and it could be up to six more months before they have a rideable prototype that also shifts. But the most important hurdle is getting a bicycle brand to develop a compatible frame…you can’t just swap this onto any bike. How long that takes remains to be seen, but there’s no shortage of interest in the design!


  1. Matt on

    Watching them manually shift by hand with his fingers near that death “cassette” and the “decoupling” of the drive doohickey was cringe-worthy at best. Good luck!

  2. Michael on

    That´s out for half a year now, and there has not been one video of riding that system. So what?!
    Now it´s possible to build a MTB prototype that stands on the exhibition.
    I want a proof of ride.

  3. Alan on

    Good for them for being brave and thinking in a fresh new direction. I am a mechanical engineer that has long pondered alternate drivetrain approaches, but their idea is turning into reality. It is one thing to conceptualize something and another thing entirely to actually develop the idea towards a viable product. These guys are trailblazing. Nice job.

  4. JBikes on

    Seems like a great brinelling torture device for bearings,

    I really don’t see this being commercially viable even if I find the act of solving the shifting problem interesting and a fun exercise. The tolerances to make this work are too tight, and to make it work without being a clattering mess of noise, even more so. It honestly seems like a senior engineering design effort.

  5. Dylan on

    Lateral forces on the free hub bearings when this is in the big cog are going to be a killer, assuming the whole contraption can be made rigid enough in the first place not to skip under load. For the MTB application, engineering it to not self-shift as the suspension compresses will be difficult (or limit the potential to suspension designs with zero growth).

  6. Tim M on

    I give an A for effort, but that’s where it ends. The thrust forces on the rear cassette would require a dumbbell size weight to handle the loads. Its pretty, but will NEVER EVER happen in anything close to this configuration. Now, if they somehow “trapped” the rear shifting/bearing assembly mechanism between two linked plates, it might have a chance, but that creates a world of other engineering challenges.

  7. Stoli on

    Added weight, added complication, added maintenance, added cost. What is not to like about this? All they need now are some suckers who are willing to pay for it.

  8. Michael Dyer on

    Rather than machining 12 separate gears and just adding material between the gears they would have a Continuously Variable Transmission (CVT) where that could eliminate the shifting complexity. Easy-Peasy….

  9. petiot111 on

    dont get me wrong, I hate chains (100 years old transmission that has never really changed), and I admire anyone who will have a go at changing this. But this is never going to work. The lateral force, contamination issues, the wear of teeth will be a huge issue, etc. and given the need to change frame, hub, I would go with a pinion or rohloff any day.


COMMENT HERE: (For best results, log in through Wordpress or your social media account. Anonymous/fake email comments may be unapproved or deleted. ALL first-time commenter's posts are held for moderation. Check our Comment Policy for full details.)

This site uses Akismet to reduce spam. Learn how your comment data is processed.