I was driving my 5 speed Volkswagon the other day and my 19 year old was looking at me like some type of dinosaur mated to manual technology (I just can’t get to the point of driving a car with auto transmission, mainly for nostaligic reasons because i realize that automatic transmissions today are not like what my buddies had when I was driving my 1985 Honda CRX “go cart”. In any case with the jump to electronic shifting, 1x (maybe that is a step back) etc etc, it seems like the next step would be some type of continuous variable transmission, but how does that take into account a rider’s natural cadence/crank torque preferences, because you don’t know what type of engine you are getting mated to the wheels and gears…with a car you know the exact engine and torque/rpm profiles.
Could this be done with bikes somehow and would it even make sense. I am thinking about some shaft based systems with finer granularity of gearing that runs through the area that is currently the chainstay and then turns the back wheel. The chain and chrainrings would dissappear and everything would run internal to the frame back the the rear axle. Build some electronics into the thing to sense the optimal range of torque and RPMs and gradually change the gain to the rear wheel appropriately…or with some kind of manual wireless override so you could pick the gear inches you want and it takes care of things inside the stay housed gearing mechanism.
Are there patents for this type of stuff out there already?
Wow great thought! Something that never even came to light. However the way
technology is going I am going to guess we would eventually see something in relation
to you what you posted.
I do not know of any patents yet…
Paul…we have chatted in the past and my number one goal still is to get our triathlon
world of results/pictures in one central place. Still plugging away and will not stop so
we will see. It is really sad that since the beginning of tri that we do not have a complete
account of results/articles/pictures of our sport.
Great thought as technology is growing so fast…I.E. cloud/storage
growing at over 50% per year!
Wow great thought! Something that never even came to light. However the way
technology is going I am going to guess we would eventually see something in relation
to you what you posted.
I do not know of any patents yet…
Paul…we have chatted in the past and my number one goal still is to get our triathlon
world of results/pictures in one central place. Still plugging away and will not stop so
we will see. It is really sad that since the beginning of tri that we do not have a complete
account of results/articles/pictures of our sport.
Great thought as technology is growing so fast…I.E. cloud/storage
growing at over 50% per year!
Scott
Funny you mention Cloud Storage…I am on a flight to the west coast and am talking tomorrow to a VC guy about some storage technology innovation stuff!!!
In relation to my thought about, I think it is a solution searching for a problem, however, I think most of the pieces are there to do it in a lightweight and reliable fashion (both the mechanical and electrical side). The question is what premium would one pay for this capability (if required at all). I actually view electronic shifting as a very rudimentary innovation in the sense that it does electronically what our hands and cables do. But an internally housed continuously variable transmission with some kind of a manual override could dramatically change how we pedal bikes. Perhaps you could set your preferred cadence range with a manual override and then the CVT takes care of things inside it. examples of this is when you stand up to get out of the saddle and suddenly the road gets steeper…if the CVT just gradually changed the gearing so you don’t get bogged down, that would be nice. Or imagine winding it up for a sprint and you hit that point where you are spinning out and then have to shift on the fly and lose that moment when you have to shift. Hard to envision all the use cases, because the CVT would have to work in a different cadence range for a standing punchy climb, vs in the TT position, vs in a stage finish sprint.
IAre there patents for this type of stuff out there already?
Yes. Google for ‘rohloff speedhub’ and ‘Shimano Alfine’ and you will immediately understand why it is not yet more mainstream.
IAre there patents for this type of stuff out there already?
Yes. Google for ‘rohloff speedhub’ and ‘Shimano Alfine’ and you will immediately understand why it is not yet more mainstream.
Similar, but not what the OP was suggesting; continuously variable transmission implies a gearing system that is continuous as opposed to discontinuous (stepped gears).
It’s a neat idea. The biggest advantage is that you can get the EXACT cadence you feel you need. But, it have some problems:
-Limited gear range.
-Very heavy (moreso than a speedhub).
-Limited control options.
-Unknown efficiency (most likely lower than a speedhub, given that the company refuses to publish efficiency stats).
IAre there patents for this type of stuff out there already?
Yes. Google for ‘rohloff speedhub’ and ‘Shimano Alfine’ and you will immediately understand why it is not yet more mainstream.
Similar, but not what the OP was suggesting; continuously variable transmission implies a gearing system that is continuous as opposed to discontinuous (stepped gears).
It’s a neat idea. The biggest advantage is that you can get the EXACT cadence you feel you need. But, it have some problems:
-Limited gear range.
-Very heavy (moreso than a speedhub).
-Limited control options.
-Unknown efficiency.
In my very rudimentary “internalized” CVT “pre concept” (It is probably several person months of engineering before it would even qualify as a concept). I was thinking that if you could remove the chain, chainrings and derailleurs and internalize everything into a CVT system that is electro-mechanical system, then you also gain on the aeroness factor and bikes could be totally re designed around that. Rear axle could basically end up being slightly larger than current front axle since you get rid of the entire cassette since the rear wheel would be turned by a single gear wheel at the end of a CVT shaft mechanism that mates with the rear wheel
Something like this below. The CVT would have different stages of gain up the shaft to the crank arm axle.:
A “true” CVT such as the Nuvinci hub. The primary problem with this is efficiency which in some cases drops below 90%. That’s a brutal hit.
An automatically actuated planetary gearbox. There are lots of ways to do this. Planetary gearboxes are already out there now such as the Pinion, Rholoff, and Shimano Alfine. You would just need some means of actuating it automatically. The problem here, again, is efficiency. It’s not as big of hit as the Nuvinci. Actually, it’s variable because depending on which gear you’re in you might only have chain losses (like riding a fixie).
Use Ultegra/DuraAce Di2 with the Baron Biosystem. It’s just a prototype now but basically the computer takes power (you have to have a power meter) and cadence measurements and determines what gear you should be in and shifts for you automatically.
That last system is probably as good as it’s going to get for us for the foreseeable future. Chains are a really efficient means of communicating power.
The only “change” I’d like to see in current systems is to move the one-way clutch (free-hub) from the rear hub to between the chainrings and the crank. Make the rear hub fixed that way when you’re descending you can move through the gear range while keeping your feet at 3/9 o’clock so that when you do finally go to move the pedals you’re in the right gear.
Just pointing out that most cars today do not use CVTs. They use the rough equivalent of an automatically shifted manual transimission.
But I don’t see the need for a shaft system. Overly complex and less efficient. You just need a variable-sized cassette that automatically changes diameter while maintaining appropriate tooth spacing. There are existing variable-sized chainrings - I just think cassette makes more sense.
But I don’t see the need for a shaft system. Overly complex and less efficient.
Exactly, a shaft is just going to lose the effeciency battle with a chain. Do you want to lose 3%-5% of your power just to have a shaft? No way do you gain 3% less aero drag with a shaft drive. This is not something where there could be technology to improve the shaft drive to get it closer to a chain, this is just the fundamental nature of the shaft system. There is a reason bikes use a chain drive. Although a belt drive is theoretically more efficient than a chain, testing shows that current belt drives are not as good as a really good chain.
I have ridden a bike with a CVT transmission. Your pedaling used a chain to drive a hydraulic pump with a variable stroke volume. Then the pump drove a hydraulic motor that powered the rear wheel. The setup was you squeezed a lever to increase the gear. So no squeezing was the easiest gear, then the harder you squeezed the lever the harder the gear. It was pretty interesting to ride (it was a 3 wheel recumbent). But it is was heavy, inefficient (even though it used some very expensive high efficiency pumps and motors), expensive, and required the system to bled often.
Cool idea but efficiency will be the killer. Current bikes are astonishingly efficient, a well-maintained drivetrain operates at something like >98% efficiency. Even the best hub gears like Rohloff lose a few more percentage points in comparison and hub gears have been around for a long time so you’d expect them to be pretty well refined. Hard to see how CVT could be anything but less efficient than current hub gears given the extra complexity. In a car or motorbike (and it’s worth noting that CVT is only really common on small motorbikes and scooters/mopeds), sacrificing a bit of power in exchange for greater simplicity (for the driver at least) is a trade-off worth making in a lot of cases. In a bike where the rider has to generate every watt themselves, then even a small loss of power is unacceptable for most people. Maybe you could claw back a few of those lost watts with aerodynamics, but doesn’t seem like it would be enough.
As you say, feels like a solution in need of a problem. Most drivetrain improvements that have caught on have offered significant improvement to the rider with negligible downside. Indexed shifting made for less thinking and quicker shifts than a friction shift, particularly for an inexperienced rider. Brifters meant shifting without moving your hands down to the frame. More rear gears has meant more increments to enable you to always have the “right” gear, while also increasing the range of gears available has led to reduction in weight and complexity by largely eliminating triples at the front and now making 1x possible. Electronic shifting is probably the most incremental benefit to date, in that it arguably adds complexity without really offering much improvement in shift quality over a well set up mechanical set-up operated by a skilled rider, and shift quality doesn’t offer much tangible speed improvement anyway. I was a big doubter about electronic shifting and still think it’s largely a luxury that will be confined to the top end of the market for the foreseeable future, but the sheer quality, reliability and coolness of the electronic experience does now at least make it a justifiable luxury for those with the money
I simply can’t see how CVT is going to offer me any benefit, or how it could do so without adding some negatives. Could see it being launched on some cool (and mega-expensive) urban bikes. Maybe as a project with somebody like Mclaren or BMW just to attract some column inches and demonstrate what their engineers are capable of. Can’t see it becoming mainstream though. I’m still surprised we haven’t seen a lot more bikes with belt drives paired to hub gears, to me that seemed like an obvious next step for commuters or tourers who are prepared to sacrifice a few watts in exchange for not having the hassle of cleaning and lubing a chain, cleaning and adjusting derailleurs, etc. But they’re still pretty rare, I keep an eye out for them but they only seem available on a couple of production bikes.
Use Ultegra/DuraAce Di2 with the Baron Biosystem. It’s just a prototype now but basically the computer takes power (you have to have a power meter) and cadence measurements and determines what gear you should be in and shifts for you automatically.
…
The only “change” I’d like to see in current systems is to move the one-way clutch (free-hub) from the rear hub to between the chainrings and the crank. Make the rear hub fixed that way when you’re descending you can move through the gear range while keeping your feet at 3/9 o’clock so that when you do finally go to move the pedals you’re in the right gear.
When I first read the subject line the Bioshift was the first thing that popped into my head. I do think that the mechanics could be re optimized to take more advantage of auto shifting using electronic shifters. Something more along the lines of closer gearing in the back and wider gearing the in the front. Or some other way of not having much overlap of the small ring gears and big ring gears and the front derailer is used for almost every shift.
I don’t see how you can move the clutch anywhere but the hub or chain rings and I don’t think the peddling a few times while descending is too much of a bother. But I could see how it would be a big improvement in a commuter type bike to have the gears shift while braking to a stop. Just like in cars, automatic shifting is best used in stop and go traffic.
Just pointing out that most cars today do not use CVTs. They use the rough equivalent of an automatically shifted manual transimission.
But I don’t see the need for a shaft system. Overly complex and less efficient. You just need a variable-sized cassette that automatically changes diameter while maintaining appropriate tooth spacing. There are existing variable-sized chainrings - I just think cassette makes more sense.
I agree with you on this. Variable electronically actuated chainrings exist. I’d actually like to see SRAM pursue this since they have wireless shifting. You basically have a small servo that drives a worm gear to change the size of the chainring. Changing the front and rear in unison would be ideal. You could have perfect chain line and chain tension as well as very fine (infinite in theory) gear changes.
Use Ultegra/DuraAce Di2 with the Baron Biosystem. It’s just a prototype now but basically the computer takes power (you have to have a power meter) and cadence measurements and determines what gear you should be in and shifts for you automatically.
…
The only “change” I’d like to see in current systems is to move the one-way clutch (free-hub) from the rear hub to between the chainrings and the crank. Make the rear hub fixed that way when you’re descending you can move through the gear range while keeping your feet at 3/9 o’clock so that when you do finally go to move the pedals you’re in the right gear.
When I first read the subject line the Bioshift was the first thing that popped into my head. I do think that the mechanics could be re optimized to take more advantage of auto shifting using electronic shifters. Something more along the lines of closer gearing in the back and wider gearing the in the front. Or some other way of not having much overlap of the small ring gears and big ring gears and the front derailer is used for almost every shift.
I don’t see how you can move the clutch anywhere but the hub or chain rings and I don’t think the peddling a few times while descending is too much of a bother. But I could see how it would be a big improvement in a commuter type bike to have the gears shift while braking to a stop. Just like in cars, automatic shifting is best used in stop and go traffic.
I agree re the narrow rear wide front gearing. That’s sort of how I’m set up now 52/36 front 11-25 rear. Wickwerks has 53/34 chainrings that I’m going to try at some point this year. IF that works that would be an ideal setup for me, personally.
I am not an engineer but wouldn’t an IVT (infinitely variable transmission) be the ultimate in efficiency. I had a similar thought to yours Paul a few years ago and I came across a guy by the name a Steve Durnin. I saw an interview on YouTube of his invention of the D-Drive. Below is a link to his company’s website and what they have been working on. I think something like this could be applied to commuter bikes, but I am not sure about race bikes, but who knows. It is definitely fun to think about.
I am not an engineer but wouldn’t an IVT (infinitely variable transmission) be the ultimate in efficiency.
Depends on what efficiency you’re talking about. It might promote the ultimate in physiological efficiency, e.g. always being at the most efficient cadence for a given rider.
But in terms of drivetrain efficiency it would depend on how you implement the infinite variability.
