Patent Reveals How E*thirteen's Anti-Pedal Kickback Sidekick Hub Works
We got hands-on with a new hub from e*thirteen labelled "Sidekick" at the last round of the DH World Cup in Poland. It seems to have a "soft catch" when the freehub engages, similar to how an O-Chain feels, presumably to minimise pedal kickback by allowing the freehub to rotate relative to the hub shell, providing some extra chain slack to help the suspension compress unhindered. It also coasts silently. At the time we could only speculate on how this worked, but searching around revealed a patent called "Hub system, method and device with adjustable deadband" filed by The Hive Global, Inc (e*Thirteen's HQ) that may reveal what's going on inside.
The pawl pusher (140) sits axially inboard of the freehub body (102).
The pawl pusher has fingers (144 a-c) which have ramps to push the pawls radially outwards when the freehub is rotated clockwise.
The patent shows a freehub ratchet with three pawls and an outer ratchet ring - so far, so normal. However, in a normal hub, the pawls are pushed outwards by a spring to engage immediately with the teeth of the outer ratchet ring when the freehub is turned clockwise; in e*thirteen's design, the pawls are held inwards towards the freehub body and away from the ratchet ring. This explains the silent coasting, but how does it engage when the rider starts pedalling?
The hub contains a component you won't find in a conventional hub - referred to as a pawl pusher (140). This is housed concentric to the hub axle and inboard of the freehub body. It's connected to the hub shell with a sprag clutch - a one-way bearing which allows the hub shell to rotate clockwise* relative to the pawl pusher, but not the pawl pusher to rotate clockwise relative to the hub shell. So in other words, the pawl pusher has its own one-way mechanism just like the freehub, but the pawl pusher is not directly connected to the freehub and cassette.
*if viewed from the driveside
The pawls engage with the ratchet ring (180) while the pawl pusher is held inside the sprag clutch (42) which prevents it from rotating clockwise relative to the hub shell (32).
The pawl pusher has three** ramped "fingers" (144) that interlock with the three pawls of the freehub. When the rider pedals and the freehub rotates clockwise, the pawls are pushed into contact with the ramps of the fingers on the pawl pusher, which cannot rotate clockwise due to the sprag clutch that connects it to the hub shell. These ramps push the pawls radially outwards until they engage the ratchet ring, which allows torque to be transmitted to the hub shell and the wheel. When the rider stops pedalling and the wheel rotates clockwise relative to the freehub (coasting), the pawl pusher moves clockwise due to the inevitable drag in the sprag clutch until it contacts the "pusher stop surfaces" (136) of the freehub. This creates a "deadband angle" (A) through which the freehub must rotate relative to the hub shell before the pawls engage again.
**There are three depicted in the diagrams, but the patent says there could be any number of fingers and pawls
A cross-section of the freehub (102), pawl pusher fingers (144) and ratchet ring (180) viewed from the drive side. In the coasting position (left) the pawls are retracted and the pawl pusher fingers are clocked such that they touch the pusher stop surfaces (136) of the freehub body. When the freehub is engaged during pedalling, (right) the freehub rotates clockwise and the pawls are pushed outwards by the ramped fingers of the pawl pusher until they contact the ratchet ring.
During a potential pedal kickback event, where the suspension is compressing very quickly (resulting in a rapidly growing upper chain length) and the wheel is rotating very slowly, this deadband angle could allow the freehub to rotate a little further before it "catches up" with the wheel rotation speed, thereby spooling out a bit more chain and reducing the chances of pedal kickback. The concept is similar to the O-Chain device that many DH racers are using but at the freehub rather than the crank. The obvious downside is that there will be more lag (slower engagement) when getting on the power.
When the deadband adjustment key (108 ) is installed, the deadband angle (B) is reduced.
The patent also mentions how a "deadband adjustment key" (108 ) could be inserted between one of the pusher stop surfaces (136) on the freehub and the pusher finger to reduce the deadband angle. This would reduce the effectiveness of eliminating pedal kickback but also reduce the engagement angle when pedalling.
Why not just use a slower engaging hub?
You might think you could achieve a similar result by using a freehub with fewer points of engagement, but the problem with that is inconsistency. If a hub had just one point of engagement per revolution, sometimes you'd have to turn the freehub a full 360 degrees before it engaged, but sometimes it would engage straight away (depending on the prior position of the tooth and pawl). The same is true in a pedal kickback event - sometimes the cassette would be able to rotate freely and sometimes it wouldn't, depending on the relative positions of the freehub and hub shell at the time. With e*thirteen's design, the deadband angle is consistent, not random. It can also be adjusted to suit personal preference. It may reduce coasting drag slightly too, if the friction from the sprag clutch is less than a conventional ratchet.
But if you want my humble opinion, pedal kickback is a corner case that rarely happens in the real world - and when it does it's not necessarily a problem. Snatching pedal turns here and there is something that may happen several times during a race run, and increasing the deadband angle could hamper this at least a little.
Author Info:
Must Read This Week
Sign Up for the Pinkbike Newsletter - All the Biggest, Most Interesting Stories in your Inbox
PB Newsletter Signup
We've had this hub in development for 3 years now and Dakota Norton was on the sidekick for most of last season. Ronan Donne won Hardline earlier this year on it along with the WC DH race this past weekend. They (and other WC pro's who have tested it) have come away extremely impressed and we have other non-e*thirteen sponsored riders who have reached out to request testing.
While the hub is still in final stages of development we do hope to be able to share some more news later this summer. It's one of the most exciting projects we've worked on.
For the pawl pusher to move back fully when you stop pedaling there needs to be something moving it, this has to be drag between it and the hub shell through the sprag clutch mechanism. So that the onward movement of the hub-shell carries the pawl pusher forward and allows the pawls to retract.
If you had zero drag in the sprag then when you stopped pedaling the only thing pushing the pawl pusher back would be the inward movement of the pawls and it would not be inclined to move past the tip of the pawls themselves so you would never get more "gap".
You could actually save a fair bit of money here by replacing the sprag with a simple drag spring like we use in BMX Clutch hubs, though I suppose this might then introduce some patent conflicts...
The E13 Patent number is US2024157728A1 in case anyone wants to read it.
(For those not in the know, G-Sport is the epitome of functional high-quality design in BMX)
A low engagement hub where there are 10 degrees between the engagement points, if the suspension rotates the cassette 5 degrees there is a 50% change the pawls won’t engage (as all the rotation happens between the 10% engagement), but there is a chance that the pawl is literally just past the engagement point as you land a drop and the pawls would engage immediately.
Essentially if you did the same drop 10 times, half the time you’d have unaffected suspension, the rest of the time you’d have various levels of suspension performance.
With this system and the O-Ring, there is always the equivalent of 10 degrees (unknown actual value) of rotation required prior to the pawls being pushed to engaged + the random factor of the engagement points, so you can have high engagement and get free suspension. The engagement would also always be consistent (like a high engagement hub) just delayed by a fraction.
In this case the pawls are off the rachet until the initial rotation has occurred meaning no drag.
The OChain works the opposite way round so once the chain engages there is a spring that makes the system soft, in a ratio that barely affects the suspension.
Basically;
OChain normal engagement then soft range to allow mostly free suspension.
This the suspension moves free but there is always a delay (consistent) to engagement.
Normal hub = random level of suspension freeness but quick engagement.
What's holding the pawls inward I don't know maybe a small magnet (as it wouldn't need much). So I feel e13's claim is valid
So this and the pawl based freecoasters (ignoring the planetary ones for now) start out the same.
Pawls are held retracted by default (by spring).
There is then a subtle variation.
This hub "says":-
IF driver rotates forward relative to HUB-SHELL: THEN ENGAGE Pawls
While the pawl based freecoasters "say":-
IF driver rotates forward relative to HUB AXLE: THEN ENGAGE Pawls.
With this hub, there is no distinction between the driver moving towards the shell or the shell moving backwards towards the driver, so this would not function like a freecoaster in reverse.
You are not going to beat Nino at an xc race on your Enduro bike. But he may beat you in a Enduro race on a xc bike.
I've never used one, but I've heard good things about them
I did buy this off of a sponsored pro (this wheel was on the bike when she won Leadville). But I doubt it is a special hub.
pretty rad for an 80 year old
@2004hyuandielantra my o-chain doesn't make any noise. It rotates silently, so there's something wrong with it if it's loud.
This E13 hub design seems cool, but IDK if I'd trust E13 to reliably manufacture such an intricate design for a very high-stress component.
Also, I feel like getting a chain ring for any set of cranks if your OChain goes south is going to be way easier than a small part for a freehub.
I guess since it’s a gravity oriented component you can try to Gwin it when your chain hasn’t snapped but your pawls, drive ring and/or pusher have snapped.
Even Shimano (inventor of the modern freehub) has gone to a ratchet ring design.
I wish e13 the best of luck, but like SRAM brakes or Crank Brothers stuff, they’ll need 3-5 years of bulletproof reliability to overcome a long history of bad products.
You can choose to ride slow on smooth pathways with old-tested technology for ultimate reliability or try to go fast on rugged terrain and innovative tech..
Running 4th chain on this setup!
All I've replaced is the BB bearing and one spoke on front wheelset of which the BB bearing went under warranty no question asked!
Or what if I just want to remove it for cleaning? Now I gotta worry about loosing pawls and springs.
It's just not good.
Your sister is ratchet, your mom is a pawl pusher and your dad plays in a deadband.
We have resolved nearly all the issues, geometry, kinematics, reliable dropper post, durable cranksets, high performing predictable and tunable suspension.
Manufacturers are desperate to bring solutions to problems that virtually don't exist and may providing a riding improvement of like 1%.... probably less impactful then having a clean bike, or fresh pedal pins.
No thanks!!
Then explain why a large number of profesional racers and mechanics are choosing to run o-chains after real world testing? I know who i'm going to listen to. Also the random nature of engagement almost prooves it is an issue that will affect you more times in a ride than a quick pedal stroke, how many times does your suspension compress in a ride? A lot and some of those will line up with the ratchet cuasing instant engagment and therefore pedal kickback.
Jokes aside, this is a neatly packaged design and hope it sees success!
There are bit of bro-science out there saying this is stupid and very low engaging hub is better...thankfully there are people put there who explore things further than just eat the words of some influencers.
Chain tension does affect suspension performance. I turn off the clutch on all my full sus bikes for this reason
the sidekick uses a sprag clutch as part of the system to enable the pawl lifter, but only after a short delay for the sprag clutch to engage plus another delay for the pawl lifter to force the pawls up.
I haven't spent a ton of time on them to comment on how that relates to kickback.