Not sure the new SRAM Eagle Transmission is for you - What other options are there ?

SRAM recently released the New Eagle Transmission system.

Rumoured and talked about for 4 years and shown in Patents for 6 months now, ridden extensively by Nino Schurter last season, it's finally released into the real world. Originally announced as a modular direct mount derailleur, its evolved to be a complete high-end transmission.

2018 Direct Mount.png

2018 Direct Mount Patent
Nino Scott.png

Nino's Bike Last Year
Patent Full.png

2022 SRAM Patent
Patent Rear Full.png

2022 SRAM Patent

An AXS derailleur based system with easier setup, smoother shifting, improved longevity and a considerably stronger and a more robust derailleur design which does away with the traditional hanger :eek: Everything is new and the only items which are backwards compatible are the AXS controllers (The older controllers will also work with this system). The AXS batteries are also the same as previous systems.

The lower Jockey wheel and cassettes are all now Narrow/Wide, except for 7th gear, which is used for setting the system up. Even the "Flat top" chains are new, so don't expect to throw a cheap spare on to get you home.

The derailleur is also more repairable and parts can be swapped between different models in the range :

Eagle Transmission.png

Differentiating from previous systems, SRAM made the move to name it "Transmission" instead of "Drivetrain".

The new system, with parts marked as "T-Type" for inter-compatibility :

The previous Eagle range :

There's no arguing that the new system is stronger and considerably more robust than existing derailleur derivatives, only time will tell if this does or does not translates to damage to triangles. Rotating UDH's (which normally have 60 degrees of allowed movement) have already caused damage to some bikes :

UDH Frame Damage.png

It's possible, with the new sandwich clamping of the derailleur using the axle, that this will be a thing of the past. Alternatively, will an incident as shown in this picture, which can tear aluminium apart, result in the axle tearing out of the triangle instead ??

Broken XT.png

All the reviews and tests so far are positive, but as we all know, once these things end up in the hands of us real world morons, anything can happen !

Meanwhile it looks like Ratio Technology (probably better known for their Ratchet Kit which enables you to run SRAM Eagle/12 Speed cassettes with 11 speed shifters) are also developing a direct mount derailleur conversion kit, which will be compatible with UDH hanger bikes and enable you to run with some existing derailleurs.

Ratio Technology Direct Mount

Ratio Technologies Direct Mount Prototype

Are there Alternatives ?

If shiny is high on your priority lists then there's the new TRP EVO12 in Gold or Silver Finishes - but no electronic shifting.

If you have to have direct mount, but maybe have an older bike and don't find the SRAM transmissions price attractive, then you can go back intime exactly 20 years and grab a Shimano Saint M800 Direct Mount Derailleur :

Screenshot 2023-03-25 16.23.37.png

Meanwhile, Shimano seem to be developing a new system themselves, but we don't really know that much about it yet :

Shimano Three Pulley
Shimano Three Pulley Derailleur

There was also a direct direct mount system patented by Shimano last year :


TRP (Tektro) also appear to have a direct mount system in development :

TRP's seems to show a battery and a motor, so presumably will be electronic and wireless.

If you don't want to go with an existing derailleur system, what other options are out there now or in development ?

To start with, there's the usual suspects in the form of internally geared hubs (IGH), these can be driven by chain or belt. They're also something you can actually install on your bike now rather than having to buy a new bike based around an alternative transmission system.

The main ones being the Rohloff 14 526%, Kindernay 14 543%, Shimano Alfine and also the 3x3 - Nine 554% (HB Hightech) .

The 3x3 being the new arrival to this party.

Rohloff Hub.png
3x3 nine hub.png

The Rohloff and Kindernay with 14 speeds have a 13-14% difference between gears. The 3x3 Nine, with only 9 speeds has a 23.8% difference, so is more suitable for EMTB use than MTB use. It's also rated at 250nm of input torque, compared to 130 and 160 for the Rohloff and Kindernay respectively. The 3x3 is grease filled rather than oil, so theoretically has a 25,000km service interval, 5 times more than the Rohloff. The 3x3 also has a wireless shifter option.

There's already lots written about these together with the advantages and disadvantages, so I won't delve further here.

One of the many Rohloff threads :

Similar are the CVT (Continuously Variable Transmission) hubs such as the Nuvinci N380 or the Enviolo :

nuvinci hub.png

The Nuvinci, first created by Fallbrook Technologies in 2006, doesn't ever seem to have gained that much acceptance, especially in the EMTB community. The Enviolo appears similar, because it's basically identical. However, they don't actually propose it for MTB or EMTB use only "trekking", they do offer heavy duty versions of the hub which could be suitable. It also shares the 380% ratio with the Nuvinci N380.

Some of these may look impressive and modern, but it should be kept in mind that your MTBing ancestors were already trashing Mech's when racing home from the coal face. Hence the invention in 1890 of the Velocipede (less catchy than MTB) with hub gearing, belt drive, manual 150mm dropper post, mudhugger "long", tubeless tyres with inserts (they were solid), 62.5 degree head angle (if the diagram is accurate) ! and front suspension which saved you if you rode into a Fox :

Belt Drive Ancestors MTB.png

I don't like Chains !

Maybe you don't actually like the whole idea of a chain or a belt, are there any options here ? Yes, but not really anything which is backwards compatible with your bike, so from here on most of the options discussed are going to involve "New Bike Day".

It seems the idea of the chain has pissed some people off for a long long time. This is the Hildick chainless drive from 1898 (probably nicknamed the Nut Cracker). You'll notice the main gear rotates around it's own hub, just like the lower "Magic Wheel" on SRAM's new Eagle Transmission ... Maybe they should have saved weight and completely skipped the inner hub portion of the "MagicWheel" then where would be no problem with a stick getting caught in there in the first place and more importantly it would look cooler ! Like the Absolute Black Hollowcage.

Direct Drive.png

How about a chainless "Digital Drive" system ? We like motors !

At the crank you have a generator, this then transmits your power to a motor in the hub.

For example : Schaeffler’s Free Drive system

Freedrive Digital Chainless.png

In terms of practicality for EMTB use, how would this work ? Probably not that well. For starters, there would be no gearing and the hub motor would need to provide the same power as an existing EMTB hub motor plus the extra power generated by the rider - meaning it would be larger and heavier. For people using EMTB's in flatter areas or for rugged commuting, it still might be an option.

Another option in development is the hydraulic drive. One example being the OYO from BC bikes with infinite gear range :

OYO Hydraulic.png

This one was up for crowd funding a couple of years ago, but not a lot seems to have happened since then. I suspect the system weight could have been a problem and real world efficiency is/was possibly less than expect. In theory, you could have 95% efficiency, but in reality hydraulic systems often operate at about 75% efficiency.

The other chainless drive option is the shaft drive. Successfully implemented on motorcycles, but rarely on bicycles.

Bicycle shaft drive.png

There are a couple of problems with shaft drives on a bicycle and especially an MTB. Firstly, the pinion gears are small, so the amount of torque being transferred through the system is enormous. It's possible to engineer around this, at the expense of weight, generally a shaft system will be at least 1kg heavier than a chain system.

Another problem is efficiency. A chain is approximately 99% efficient. A belt 98.6% efficient. Whereas a shaft is only about 92% efficient. On the flip side, a shaft will always be about 92% efficient, where as a chain or belt system will drop considerably in heavy mud conditions.

For MTB use, a shaft isn't really practical. There's too much movement between the crank and the axle so it becomes impractical to maintain perfect contact through the pinion gears. This can be fixed with CV joints, but then you have less efficiency, more complexity and more weight.

However, there has been some recent development work on shaft drives. The Driven from Ceramicspeed, which has been spun off to Driven technologies :


The system is showing promise, though they still have work to do overcoming the torque problems. The main system works slightly differently from a traditional shaft drive by using a series of cartridge bearings intermeshing with two circular pinion arrays.

Will this ever work on an MTB, probably not as the rigidity required in the rear triangle is incredibly high.

The system would normally have covers on, so wouldn't be this exposed in real life.

Ceramispeed shaft.png

Whilst this once again appears to be revolutionary and ground breaking, you can clearly see the same system (without the cassette) was used in the 1898 (popular year) Quadrant cross-roller gear, from the Quadrant Bicycle company, Sheepcote Street, Birmingham :

Quadrant Cross Roller.png

Not just a proposed design - a surviving example :

Old Quadrant Cross Roller Transmission.png

The Image is from this here, which is quite an interesting website if you like that kind of thing. Many of the marketing claims from today seem to be lifted straight from the same pages shown from 120 years ago ..

Mid Mount/Crank Gearboxes :

It's fair to say that most of the options covered don't really fit with mainstream EMTB requirements. The hub drives will be ok for some, but you generally have an increase in weight and some won't change gear whilst under load. The increase in weight is also where you don't want it, in an unsprung area. Anything you can do to reduce weight here will ultimately improve your bikes suspension performance and obviously moving any weight nearer to the centre of the bike will also improve handling, hence mid mount / crank gearboxes which also remove the exposed derailleur !

One of the early innovators here was Honda - 20 years ago we had this beauty :

Honda Gearbox Bike.png

The bike housed the derailleur and 7 speed cassette in a case with the crank :

Honda Gearbox Bike internals.png

Amazingly, this was the third derivative of the system. The first two were complicated CVT systems. If you want to know more about this system and it's history there's an excellent article by Alee Denham here : Inside the Honda Gearbox

Recently, there's been more of a development push in the MTB and EMTB world with Gearboxes.

One example is Pinion who've been making mid mount bicycle gearboxes for 10 years and have recently introduced electronic shifting for use with ebikes - though only hub motored ones. Effigear being one of the other main developers along with Kervelo.

Another interesting example is Allen Millyards work on the MR001 and MR002 - worth watching even if you're not that interested in gearboxes !

Combined Motor Gearboxes

One of the earlier developers of a motor/gearbox combination were Continental. Merging a motor with a Nuvinci CVT system in one housing. Unfortunately, the project was seemingly abandoned.

There's presently several systems in development and production, many of which, as you'd expect are quite bulky looking.

However, there is Kervelo's Quartz, which effectively adds a motor straight onto their existing crank gearbox :

Kervelo Quartz.png

At the moment, the prototypes don't look overly attractive, but looking at the layout you'd imagine it wouldn't be hard to come up with designs which more closely resemble something like a Fuel EXe or for that matter any of the Fazua Ride 60 or TQ HPR50 powered bikes.

Valeo have developed a system which combines an Effigear gearbox and a motor :

Valeo Effigear.png

Revonte are another company with the Revonte One, which appears quite slick looking in terms of sizing !

Revonte One.png

Backwards compatibility ?

If you're an EP8 owner and in some cases, possibly an e8000/e7000 owner, Intradrive are developing a combined motor gearbox which uses the standard EP8 mounts. Their battery will also use the standard shimano mounts. However, as Shimano has just made the change to second generation batteries with the EP801, presumably down the line they'll have to offer an alternative EP801 compatible system.

The Intradrive will come with an 8 speed internal oil lubricated gearbox with a 430% range and 23% steps.

Intradrive Motor Gearbox

The Future :

I quite like the idea of someone combing a motor and gearbox, but creating a gearbox similar to the Koenigsegg lightspeed transmission. 9 Speeds, but compounding the gears so you end up with more speeds for less gears, you do need a series of clutches to operate the system, but if you imagined the gearbox reduced to bicycle requirements, it could be tiny and offer instantaneous gear changes which are smooth and silent :

Lightspeed gearbox operation.png

And from SRAM ?

Well, if the new one isn't complicated or expensive enough for you, they are proposing a self charging derailleur which takes power from the upper pulley wheel .. Looks simple ?

About author
Self confessed Muppet and EMTB lover, based in France. Crash tests a lot.


Great post, and it definitely answered some questions for me.

However, these solutions are not for me, adding a kilo to the rear wheel (compared to my preferred 11 speed) really negatively effects rear suspension performance.

Also the need to really let off the power doesn't help.
Great post, and it definitely answered some questions for me.

However, these solutions are not for me, adding a kilo to the rear wheel (compared to my preferred 11 speed) really negatively effects rear suspension performance.

Also the need to really let off the power doesn't help.
Yes. This is one of the main issues with the IGH setups. In their defence, I think everyone I've read about who's installed one on their bike has stuck with it and hasn't complained about the weight, even some hard running riders.

There is also the Revolute Hub1, which I didn't mention as it only has 6 speeds (though 400% range) as I thought most people would baulk at 6 speeds for MTB, even with a motor. It does let you shift under load and also uses helical gears - quiet/less efficient (Alfine 11 is spur/helical combination. The tiny efficiency different is probably irrelevant on an EMTB. Its also rated at 250nm, though it won't roll backwards unless you select the Neutral gear, so you can't do some of your Danny Macaskill tricks that easily. BUT - it's still 2 kg, so 40% more than a Kindernay 7 speed - maybe that would be the way to go with the Relay ? No cassette/mech - it must be getting fairly close on weight then ?
Thanks for your post.

The main problem with current derailleurs is that they are too near the ground and thus easily damaged.

Raising the derailleur 3-4" would eliminate most of this damage.
Raising the derailleur 3-4" would eliminate most of this damage.
Thanks, I ended up writing it twice as the first one got lost somehow so a few things changed. I did cover some of that the first time through.

This image is a good visualisation of the dangly catching device in all it's glory ..


The patents for the Shimano mech seem to show something which potentially sits higher for most of it's time with a higher chainline also.
The cost! Almost $3k CDN for the complete set (cassette, chain, cranks, chain ring and derailer. If you’ve just recently upgraded anything on the drivetrain …that’s tough to swallow.

I do like the hanger-free mounting system (even though it’s mounted to the triangle) using the axle for support and the additional 1cm outboard clearance.

Good read, and I hope I’m on the same page.
Thanks, I ended up writing it twice as the first one got lost somehow so a few things changed. I did cover some of that the first time through.

This image is a good visualisation of the dangly catching device in all it's glory ..

View attachment 109945

The patents for the Shimano mech seem to show something which potentially sits higher for most of it's time with a higher chainline also.
this looks beneficial - stops pedal strikes on sharp angle uphill turns.
I do think belt or shaft drive would be great, as the faffffff with the chain cheaning/oiling would not be missed. All too often the designs seem to me as they are over complicated, they need to KISS it.

[that is keep it simple stupid, nothing to do with my ass]
I do think belt
I've never been entirely convinced about belt drives on MTB, but I think I am slowly coming around.

Obviously there are some issues, like it won't work with a derailleur, so you could only use it where you have an IGH or a gearbox.

There are some other niggles as well. The image above is a perfect installation, the sprockets at either end are at the attachment points of the chain stay, so the belt is always under the same tension, no matter where the suspension is. If the drive sprocket is in a different place to that pivot, the length will vary and you'll need a tensioner - which is something else to break/maintain/go wrong. On bikes they're also often placed at the rear, presumably as the rear sprocket will generally be smaller than the front one and you'd want the tensioner to be nearer the smaller sprocket to maintain as much wrap around as possible and probably because there's already somewhere at that end to mount a tensioner. Again though in the example above, they appear the same size, so if you did need a tensioner, maybe it could be mounted near the gearbox if you were designing the bike from scratch.

Another issue is that you can't just put a belt on any bike. If you think about the rear triangle on many bikes, they don't all have a pivot point connecting the chainstay to the seatstay -so they can't be dismantled and separated. If you can't do that, how do you fit the belt around the chainstay (below it and above it) unless you have a belt which can be broken.

The belt then needs to be tensioned, so you need some way to vary the effective chainstay length.

Then, you get a puncture and have to take the wheel off. Meaning you need to take that tension off to remove the wheel - although that's not really a big deal.

I guess one solution, as you won't need to allow the space for a cassette, would be to change the inner axle spacing. You could then mount the rear sprocket on the outside of the chainstay with the sprocket held in place with it's own mount and the axle passing through into that - a little like the new SRAM mounting above. That way the belt is easy to adjust, easy to change and the wheel can be removed as easily as a front wheel is now.

Belts have certainly come along way. Some of the Motorcycle ones often last 200,000 km's now ! Bike ones often 30,000km's. I always worry though that with mud and all the microscopic abrasive materials which come along with it, and the larger bits, that these would potentially damage and wear the teeth. As we do with MTB, we'd have to accept some maintenance and wear. If a belt costs €100 euro's and you have to change it 3 or 4 times less frequently as a chain - assuming a chain will also be wearing faster in similarly poor conditions - but you don't have to change cassettes or chainrings (Sprockets presumably at some point, but certainly not as often) then economically and environmentally they'd be a vast improvement and no derailleurs to damage/replace.

Some cool solutions have already been thought about and produced for the Gates carbon drive for instance :


It sounds quite good ... Durable, cheap, takes 2 minutes to change every 10,000-30,000 km's, super easy to remove your wheel. Maybe the SRAM motor with have an integrated gearbox. Offer the new UBA (Universal Belt Attachment) with all the relevant specs for manufacturer compliant swing arms - they managed it with the UDH.

No doubt third party vendors would quickly develop bolt on sacrificial brushes which help shed mud from the belt before it enters the rear sprocket. The MudBugger .. ?!
Last edited:
BRP (bombardier) purchased Pinion gearbox about a year ago. They are currently hiring in all kinds of departments to start developing their new gearbox EMTB. It's going to be interesting to see what a company with this kind of manufacturing ability and the desire to push the envelope can come up with. I am so ready to never have to maintain or even look at a derailleur again.
AXS was already pretty damn simple to set up. And I still would prefer (even if for mental health) the knowledge that there's a breakaway part in the mix to protect my expensive frame and expensive derailleur. I'm perfectly happy with X01/XX1 AXS for now.
Currently riding an X01 Eagle without AXS as my new bike came with it and just have around 200mls on it. The new Eagle Transmission looks very promissing but at a very high pricepoint. Might look into the new one in X0 config as soon as my cassette is worn out. Hopefully prices drop a little until then.
Guess with the tiny axles and weak frames back then it wouldn't have worked properly. But that's just a guess. There should be a reason this took more than 50 years to hit the market.
I've been considering adding a Kindernay Vii and Gates belt drive to either a Pole Voima or an Orange Phase MX, but I’m to broke and into dept 😅. However, both bikes are well-suited for a belt drive since there's no need to split the frame to mount it, unlike the Nicolaj Eboxx E14.

While the Kindernay Vii is one of the lighter IGHs, I'm still concerned about unsprung weight.
Efficiency isn't as much of a concern since the motor can compensate for any loss (although range might become an issue).

Gearbox-motors seem to be the future of EMBs, and I'm hopeful that Intradrive and other manufacturers will release working products, with more manufacturers following suit.

Digital drives are also an intriguing option, but hub motors are too heavy for EMTBs, especially if they have to handle steep and technical terrain without a physical connection to the rider. They also need to be robust enough to handle the rough terrain, while also providing good control and feel. Additionally, power regulations might pose a challenge since the motor needs to compensate for the lost connection to the rider's input.

Here is a good article on Digital Drive though:
Are Chainless Digital Drive Bicycles The Future of Cycling? - CyclingAbout.

What I haven’t seen is someone putting the digital drive together in mid mounted solution? Having a motor and generator next to each other. Then just having a belt drive to the rear.

I even made a concept sketch of the thing 😅

Not sure how well it would work in practice though.


  • IMG_1433.jpeg
    226 KB · Views: 87
  • IMG_1401.jpeg
    2.8 MB · Views: 98
What I haven’t seen is someone putting the digital drive together in mid mounted solution? Having a motor and generator next to each other. Then just having a belt drive to the rear.
I think weight and size would be the problems. The generator section would need to be large enough to capture 100% of the human effort, so if you imagined that as a motor for size comparison and how much it would have to create. Whilst many of us probably only put out 500w, many put out 1000w. Chris Hoy was clocked at 2500w for a sprint race. So the generator would need to be large enough to capture that amount.

Which means the motor section driving the belt needs to be capable of producing the power it produces now and that amount.

Then there's no gearing, because we're trying to replace it with the digital solution - so we need huge torque to make up for the lack of gears.

Ultimately leading to a 8-10kg huge lump in the middle of the bike. We end up making something more complicated trying to make something simpler..