Tuesday, February 16, 2010

25 Spokeless Bicycle Wheel Design From Yale

Whether this inspires you or irritates you is your choosing. A group of nine mechanical engineering students from Yale University leaked pictures of a spokeless wheel design onto the internet for all of us to see. Thank you, Gentlemen. And thank you Bicycle Design for the tip-off on Twitter.

The bicycle as you see here is a single speed setup. Because of the lack of the traditional spokes and hub, the boys had to find a way to transmit pedal power to the rear wheel. Which they did by dividing the gearing into two stages - a 53-13 combo, output to right side crank, from where another 53-13 sprocket combination yields final output to wheel. It also looks like the output hub mates with a toothed profile on the inside curve of the wheel similar to what you would see in a timing belt.

What they also found out was that the cost to produce a single wheel was huge, so they decided to just make one spokeless rear wheel to see if it'll work. The wheel is made out of 6061 Al alloy, which is cheaper compared to the aerospace version 7075.

6061 has a density of 2.70 g/cc, an ultimate strength of 310 MPa and a yield of 276 Mpa, showing about 12% strain at rupture. If I draw a stress-strain diagram of this stuff, the area under the curve is somewhere around 30-35 Joule/cu.m. Multiply that with the thickness of material used and you get it's energy absorption at rupture per unit width, in Joules/sq.m. By the way, this basic calculation is really handy if you want to make bicycle fenders and chain guards!

Having been through an engineering school already (phew), I think I may be qualified to speak on what exactly the driver might be behind this work. This bicycle design is very likely for an engineering design class, either the initial MENG 185 - Mechanical Design, or the MENG 471/472 Senior Level Engineering Design Course.

There, students form groups of 3-4 (or more) and under the guidance of an advisor will seek to realize a design idea. That may either be for an innovative new product or something which may offer genuine improvements to an existing product. Some great ideas that can be made may even be granted patent towards the engineering department. Other ideas don't make the cut for patents but nevertheless, all these students will receive 4-5 credits for the class so they pass and get a mechanical engineering degree.

That said, what would still be interesting is to learn the rationale behind choosing a spokeless wheel to work on for a project. Just for being cool or something else?

The bicycle is near a state of perfection and like we always yell again and again, the more you try and fiddle with what exists, the more you sacrifice what's already there in terms of design trade-offs. I suppose that's the thing with design. It's tricky. Gain some, lose some. Gain-gain is rare. What's better in terms of taking loads, transmitting motion, and reducing friction - spokes, ball bearings and an axle or a two extra cranks, chains, and a timing-belt kind of setup?

Spokeless wheels for bicycles aren't anything new. Most folks who think of this idea have always maintained the inconvenience of the occasional tensioning of spokes in prior art as a reason for going spokeless. What that entails is a new system from them with more complexity, more moving parts, more weight and more friction. That's not much improvement.

What I would have done if I were them would be to visit the nearest patent office to see what was done by individuals in the years past in spokeless wheel design. Were they successful? Did the idea SELL? Did it actually make someone's life better? Perhaps spokeless wheels would be better for wheelchairs because that way an operator's hand wouldn't be at risk of going in between spokes? Just my two cents.

Good work anyway guys! "A" grade for now.

* * *


  1. Anonymous11:52 AM

    Now that patents and many designs are so readily searchable, there is really no reason not to include this as part of some sort of initial study prior to the Design Phase of a project.

    Maybe Universities need to add a class focusing on Design & Patent Research?

    - Ryan

  2. Ryan : Well, I'm pretty sure that in design classes, advising professors will either themselves talk, or bring a patent expert to speak on how the patent process is essentially run. It is also likely that students will be shown how to do a basic search for national and international patents. I can't see that as not being part of a design class.

  3. Anonymous12:43 PM

    "The bicycle is at a near state of perfection"

    Close the patent office...

    I disagree, the crude chain/derailleur/shifting system with 100's of moving parts requiring precise adjustment, moving a chain between sprockets, constant maintenence & incomprehencible to the vast majority of users is perhaps the WORST design for its intended application.

  4. @ Anon

    Interesting. It'd be nice if you name the exaggerated "100's" of moving parts within the transmission system. Are you proposing spokeless, hubless wheels with shaft drives, and derailleurless?

  5. Anonymous2:07 PM

    Bike Boy

    Exaggerated..hardly..how many pieces are there in a chain moving, sliding & turning all waiting to separate? How many parts are there in an STI shifer?

    You'd have to admit a chain/derailleur on mountain bikes(mud rocks water)is a horrible application?

    What I am proposing is a simple infinitely variable chainless drive without the inefficiency of shaft drive.

  6. to Ron re: Patent classes. Yep, this definitely happens. I completely my mech eng degree in 1999 in Melbourne, Australia and the patent lecture in about 4th year was still one of the most memorable ones. I think the example the guy used detailing a patent for a contraption which will freeze the balls off a brass monkey got us interested.

  7. Rob Woods4:07 PM

    75mm OD Matex planetary should do the trick. Now if they put a small planetary in that thing then they could do it in a single chain stage. Put clutches on it and you can get a reverse out of it too:)

  8. Rob Woods4:09 PM

    Hubless and spokeless is old news.

  9. @Anon: you're right, there are 100's of parts, but how can you say it's the "worst"? It works well. Its been around in roughly the current form for 60-70 years (i.e. derailleur gears) and 4 large companies (Shimano, Campag, SRAM, Suntour), and countless smaller companies have competed in trying to get market share by making their implementation better. So it's been refined pretty well you'd have to say. That's not to say that its impossible to improve a bike transmission, but it seems apparent that we are approaching a local maxima of efficiency and effectiveness.

  10. Anonymous10:04 PM


    I was responding to the use of "perfect" & the the implication that there isn't room for improvement.

    What you are describing is a Porsche 911. A "refined" version of an inherenetly flawed design.
    Sure it works "effectively" but it is far from the ideal solution.

    Chains are not well suited to outdoor environments. Derailleurs de-couple power on each shift & require adjustment. Few understand gear duplication. Chains brake & derail.

    Could an automobile manufacturer sell a vehicle that required regular rear-end adjustments...why is it acceptable on a bicycle?

    We are approaching the end of developement of the current system...but we are nowhere near the end of bicycle transmission developement.

  11. Anonymous11:14 PM

    I say worst because had any other system had the same developement effort over the past 70 years, it would be superior...

  12. Anonymous @ 10:04 : Chains are not well suited to outdoor environments? Well you need to bit more specific here rather than being blunt. Like what environment conditions are not "well suited" for a chain? If its mountain biking you're worried about, know that it has worked well all these years. Specify some ambient conditions for which chains are not well suited. Chains are pretty much the most durable, cheapest thing (most bang for buck) out there. It has been used several years, everywhere from tractor, trolleys and lawn equipment to power generation. The derailleur has been refined ever since 1905 and it's modified parellogram movement is a reliable, simple design. The de-coupling of power is probably for a very minute duration, a micro second or little more at best with each shift. I think more power is lost when riders just don't know how to pedal in circles efficiently. So what's more relevant? Very minute numbers on the bicycle or other bigger variables from the rider that can be improved?

    Sure I can see a problem riding a bike in snow. A belt drive or something is not going to be an effective solution, belts don't work with multiple sprockets. Belts work on friction, you need to keep it away from dirt and moisture. Also, I calculated that the tensions seen by a belt are way higher than what each link in a chain would see. There is a significant difference there.

    I agree with you on the CVT, but it already exists as the NuVinci hub. I agree we'll have to refine that and bring more innovative designs there, as the CVT will be ideally suited for capturing infinitely variable gearing shifting along sync with rider's leg input. Still, you would need to encase the CVT and protect that from the elements or it's not going to work efficiently.

    I feel the modern bicycle is in or definitely has exceeded the period where one could take a design and put in maximum benefits in it. What you see these days in designs are mostly just marketing, for the purposes of bringing additional revenue to a company. Minimal changes may or may not improve. In light of the spokeless wheel we're talking about here, I think it is clear to all of us what the tradeoffs have been in its design. Question is, can those be compromised..you know.

  13. @ Rob Woods : Incorporating a planetary MAY be better. It'll get rid of the multiple out of plane 2 stage reduction for sure, and get that all into one compact unit, lessening vibrations and power loss. But again, the planetary is an additional 5-10 pounds of weight. Plus, how are you going to restrict the hubs from axial movement sideways? Gotta think of all that.

  14. Anonymous2:15 PM


    RE: Outdoor chain use

    "It is no surprise, then, that some owners, from one end of the scale to the other, simply use the "do-nothing" alternative. They accept more friction, less efficiency, more noise and more frequent replacement as they only minimally maintain the lubrication of their roller chains. This is a classic "trade-off".

    My experience contradicts your "worked well all these years" statement. I have suffered numerous broken chains (last week in 25ยบ temp ..grr), prematurely worn chains/sprockets, broken derailleurs/hangers, dropped chains, sucked chains, constent maintenance, debris in cogs, broken/bent cogs...and on & on.
    Is this a proud testament to 105 years of refinement?

    NuVinci is only an incremental improvement. It still uses a chain, is complex & is quite heavy. Curious how it has had little acceptance in the bicycling community even with substantial financial backing.

    Disclaimer, I do have a dog in this fight, as I have been working on my above mentioned chainless IVT, so I'm not without prejudice. I just believe we have learned to "tolerate" the many shortcomings of the current system & I'd hope that especially on this forum, the readership would at least acknowledge the need for new ideas.

  15. Heh, it's funny how this concept keeps rearing its head, and then once someone builds it (it's been done a few times before - not for bicycles though, as far as I know) it recedes into the background for another decade or so. Because it's a silly idea.

    As far as derailleur/chain goes, it is a very efficient drivetrain -if- you know how to tune it and keep it clean. I just spent a while yesterday doing some seriously fiddly adjustments on the front derailleur of my road bike yesterday trying to get it to play nice with my new freewheel...it's satisfying to make mechanical stuff work, but it's definitely not immediately apparent.

    I think that the best drivetrain for the masses is definitely a Gates carbon belt-drive and wide-range internally-geared hub. Basically requires no maintenance, works in rain, mud, snow, etc... and the marginal decrease in transmissive efficiency should be more than made up for by the fact that chain drives are only the ideal 97-99% in perfect conditions. A dry chain cross-chaining across worn chainrings and cogs is no more efficient than a belt/IGH combo.

  16. @ Anon 2:15 : I acknowledge common sense, not new stupid ideas. If you got something to show us, bring it on. I got no prejudice there. Always willing to talk about something new.

    @ Andrew : I suppose what keeps bicycling interesting is the new ideas. It keeps a lot of blogs going, they also make money yapping about silly ideas essentially. Also, I acknowledge both yours and anons suggestion that perhaps the belt drive will be suitable, and maybe combine it with a CVT for the ideal transmission, but only lots of hours of testing and documentation under all kinds of conditions will put a decision on that. For now, atleast I won't declare yet that belt drives work in mud, snow, rain, heat, cold, long hours and shock loads.

  17. I'm not sure a CVT is even really necessary, since current planetary gearsets are getting cheaper, lighter, wider range, and more robust all the time (see new Shimano Alfine 11 for introduction next year).

    As far as belt drive reliability goes, here's a fairly ringing endorsement:


    18,000 straight miles around the world, only one belt replacement.

  18. Interesting.... And that's fixed gear too, yeah? Or am I wrong? Mark Beaumont, the man who just finished a 13,000 mile trip down the longest mountain range on the planet used a chain. :) http://www.bbc.co.uk/blogs/cyclingtheamericas/

    Its pretty moot as to what is better, belt or chain. I suppose if you take care of each one of them, it's going to perform what you want it to do.

  19. Not fixed, 14-speed Rohloff IGH (same gear range as an MTB 3x9 setup).

    And I think your closing point is valid, it's just that taking care of a belt is an awful lot less work than a chain.

  20. Hi, i had this idea for a bicycle wheel in my head but i cant seem to come up with a way to make it work.
    (not a racing wheel)
    i understand that some years ago a wheel design was introduced which had weights on springs, when the wheel turned the weights would go to the outside, thus increasing the weight of the tire, somehow making it go faster, ANYHOW the idea i had was taking that one step further, but designing it is a practice in frustration. what if you could pull the weights in? to the center of the wheel, then let them out again, and pull in? on a merry-go round this pumping action between four kids sets the merry-go-round going to speeds higher than an adults hands could.
    if you could power the weights inward motion using the bicycle's momentum it would seem to me the maximum top speed would be unreachable, of course, the only energy added would be by the biker themselves,

  21. Anonymous5:54 AM

    Yes, the weighting of the wheel could technically make the rider go faster, but the momentum required to counteract the weights moving outward and pull them in, would more or less be equal to the increased extra energy gained in the first place. The pumping action on a merry go round works because the children are contantly inputing energy. Tis simple physics, to creat energy, there must be a trade off.....

  22. Riding a bicycle is not only good for your health, it's also good for the planet. Each mile you ride on your bike instead of in your car reduces your carbon footprint.

  23. Anonymous7:48 PM

    Dear Folks,
    All the research in the world about fancy electric, humungus apparatus and other R..n..D will never work. It is a complete waste of energy. The reason why this research goes on however is because the original designers never got the fundemental basics of the design perfect. They incorrectly thought perfection was,is impossible. So ever since designers have been fumbling along doing the best they can. For example, around 1970 world manufacturers woke up and started to read the rule book and found that the diagram they had been building for over seventy years didn't match the wording of the rules. For the last thirty five years the industry and the UCI people have been arguing over the matter. Recently in the last 5 or six years some people have really been studying the rules and have discovered the perfect balance point and it has crept its way into many bicycles today. But none have done research to find the last few remaining points to creat perfection.Incidently I have and know exactly what they are. And I guarentee that once perfection is achieved the general public will not want to invest great gobs of money into any thing else. Partly because perfect just feels too good and works better, is easy and cheap to build and cheap to run. I am open for business if you have the money.. I HAVE UNDERTAKEN A MASSIVE RESEARCH EFFORT TO GAIN MY KNOWLEDGE.... nab9300@hotmail.com

  24. Seems to me many have missed the point discussing chain drive, frame geometry and the like. For a spokeless wheel design to succeed it has to have a viable/viability point. It has to have a significant advantage over traditional spoked wheel design such as quicker changing of tires/wheels in a racing environment where any inherent trade-offs do not outweigh the gains. My design incorporates the quick change into the drive mechanism which also includes ball bearing chain driven components. The number of ball bearings has increased by one set and the substituted components should nearly equal the weight of the spokes and hub.

    Now that is an Engineering exercise...not just re-inventing the wheel but re-purposing it!

  25. I have enjoyed reading your articles. It is well written. It looks like you spend a large amount of time and effort in writing the blog. I am appreciating your effort. You can visit my site.
    carbon wheels


Thank you. I read every single comment.