I think it would be cool to have a tube of ball bearings instead of a chain to turn the wheel on a bike. It would probably be hard to change the gears, but I'm sure some smart engineers out there could figure it out. What do you say?
There is no string connecting these balls, they just compress each other around in the tube. There is a break in the tube where the pedal gear inserts its teeth, and also where the rear wheel gear inserts its teeth.-- lawpoop, Oct 18 2005 Like this? http://images.amazo...01._SCMZZZZZZZ_.jpgPerfectly doable. [ldischler, Oct 19 2005] Or like this? http://www.zamek-ry...yna_grafik/ball.jpg [normzone, Oct 19 2005] Oh, you mean like this... http://static.howst...if/bearing-ball.gifA bearing ball, not a ball bearing [normzone, Oct 19 2005] And not like this... http://alaska-korea...veimage/eskimos.jpg...a Baring ball...got that strait? [normzone, Oct 19 2005] No, no, no, like this! http://nerdcraft.net/ball_bearing.jpgThe balls don't have a string running through them. They just touch each other, like billiards. [lawpoop, Oct 19 2005] Version 2.0 http://nerdcraft.ne...bearing_motor_2.jpgThis one has gears with teeth that fit in the empty space between balls. [lawpoop, Oct 19 2005] Pictures of different type of bearers http://www.myhimala...image/porters_1.jpg [normzone, Oct 20 2005] Bear balling? http://www.dongetty...ownbearsmating.html [bungston, Oct 21 2005] Bear bearing ball? http://i.cnn.net/cn...er.bear.ball.ap.jpgBut lets not have any links for baring bare balls, [normzone]. We need some semblance of dignity for this fine idea. [bungston, Oct 21 2005] Hydraulic drive http://home.mira.ne...tech/222basics.html [FullRaw, Oct 21 2005] It might have to be a string with some form of universal joint between balls, but it's impractical and would require sprocket and shifter redesign, so [+]-- normzone, Oct 18 2005 [Normzone] Why do you need a string? Are you talking about a shifter?-- lawpoop, Oct 18 2005 Are we talking about ball bearings or bearing balls?
I read this as a tube of bearing balls, through which the power is transfered by compression, rather than tension (as with a chain).
The balls would recirculate, like in an, erm, recirculating ball bearing (which is a type of linear bearing).
Complicated and pointless though...so bun, I guess.-- Texticle, Oct 18 2005 Bearing balls, not ball bearings. I meant power being transferred by compression.
Is that system really that much more complicated (in terms of manufacturing and physics) than a chain?
As far as pointlessness, I think this system would break less than a chain. The tube might rupture, but make it out of metal or something ;)-- lawpoop, Oct 18 2005 I'm just trying to figure out what modifications need to be made to shift this device from gear to gear. If it's a fixed diameter loop, the sprockets may need to be on cams.-- normzone, Oct 19 2005 [ldischler] No, -- it has no connecting string, but rather a tube that houses the balls. The balls push each other by knocking into one another. There is no string pulling them.-- lawpoop, Oct 19 2005 I think it would work, but the friction would be very high with large loads.-- Ling, Oct 19 2005 Ok, lawpoop, I get it. It works opposite a chain. With a chain, one section is in tension and the other one slack, but here you have a stack of touching balls in compression that drives it. But for this to work, both straight sections have to be full of balls (as you've shown). And this means, as the cog tooth tries to push between adjacent balls to engage a ball, it stretches the retainer, unless a ball is being released in exactly the same way. So stretching is going to be a problem. A big problem, because any stretch is going to screw the system. The first time a tooth misses a ball on the return side...that's it. Another problem: the balls would have a great tendency to pop out of the tube on the drive section, since they’re under compression. Still, it’s at least conceptually workable, and bizarre enough to get a +-- ldischler, Oct 19 2005 Not a bad idea at all. You could drive some serious torque with that sort of system. (Maybe suitable for other applications too?)
Yeah, friction of the balls pressing against the sides of the tube would be a killer.
A solution could be that the balls could have two opposing flat surfaces machined into them (truncated spheres?), so when you have a line of them (flat to flat to flat etc), the compressive load would more or less flow directly through the centre of each (ie perpendicular to the flat surfaces), and not force them against the wall of the tube so much. There would still be a slight tendency for "buckling" to occur over such a long "chain", but nowhere near as much as round balls. Add lube too...
If you want gears, you could just use trusty old skool internal hub gears. I think some well known bicycle component manufacturers still make them, with more ratios than you can poke a line of steel balls at.
I think the biggest issue would be engaging the balls into the drive/driven "sprockets". Could be a problem doing this without major friction losses, as others have mentioned.-- JoeyJoJoShabadoo, Oct 19 2005 Nice drawing, BTW [lawpoop]. Good idea? Who knows. I thought that to place belt drives on motorcycles was absurd when I first saw that application, but who wants to undo that change now?-- reensure, Oct 19 2005 Now I see that stretching is not a problem, since these tubes are rigid. For some reason I thought they were going around with the balls.But there is a problem with cogging, because as each ball is released, the entire row of balls slides back. You can avoid that by using rollers instead of balls, with the center of each roller having a smaller diameter for meshing with the drive wheels. That way, the larger diameter of the rollers will be in contact all the way around the drive wheel. Thus, no falling back, and no cogging.-- ldischler, Oct 19 2005 I like this idea. I bunned it for its potential. I don't have any idea if this would be better or worse than the current chain and sprocket assembly that bicycles currently use. It seems like it could be more useful if you could generate a more effecient system but with all that friction I do doubt it. The bicycle is a darn good machine as it is. Now if you could produce this thing a heck of a lot lighter than the chain and sprocket system- people would buy it.-- Gusbus, Oct 19 2005 Hmm...if the cogs were properly designed to fit between the balls without actually having to seperate them, this could work. Don't worry about having them circulate back exactly in synch, just feed them back into a small hopper over the drive cog.
I think it might be difficult to prevent the balls from jamming in the drive sprocket, and there will probably be significant friction in the drive tube. However, that's not to say that it's not possible.
As someone else said somewher else, the halfbakery is not about practical. This could work, and there's no magic involved. [+]-- Freefall, Oct 19 2005 A little dirt would kill it.
It can't be sealed completely, but has to be sealed partially to keep the balls from popping out. Any road dirt getting in there would have a hard time coming out, and increase the already high friction losses.
Rather than propulsion, this might work as an exercise machine, with so much resistance.-- sophocles, Oct 19 2005 As far as dirt, you could encase the whole gear mechanism, and just have little greased entry points on each side for the tire or pedals to connect.-- lawpoop, Oct 19 2005 Exactly half as baked as they come [+++].-- wagster, Oct 19 2005 In one sense, this is just a hydraulic pump and motor, except the particles of your fluid are metal spheres.-- ldischler, Oct 19 2005 [Pa`ve] It isn't enough to keep the balls in a tight tube to keep them centered?
Also, can you do a diagram or drawing of your idea? I'm having a hard time understanding it.-- lawpoop, Oct 20 2005 //In one sense, this is just a hydraulic pump and motor// Could this actually be done with a fluid in a pipe? I can't think of any reliable way of moving the fluid... anyone good with pumps?-- wagster, Oct 20 2005 If the tube structure is going to be solid, you could make it structural and save weight by getting rid of the chain stays.-- Gordon Comstock, Oct 20 2005 //Could this actually be done with a fluid in a pipe?// Peristaltic pump would work similarly, gear pump would be more efficient but would need extra bits. But in theory yes, you could get some drive from the system. If you used a compressible fluid you could also get some lag between your pedalling and the bike speed (which would be cool).-- suctionpad, Oct 20 2005 I thought of a peristaltic pump (although I didn't know what it was called until now), but it requires a flexible pipe which would just balloon under the pressure. I suspect a gear pump is the only option.-- wagster, Oct 20 2005 As the bike is over 90% efficient I'm not quite sure why someone would go to the trouble of "improving" the design. But does that make this a bun or a bone?-- Jacob Marley, Oct 20 2005 Somehow it makes 2 buns. Put that in your pipe and smoke it!-- lawpoop, Oct 20 2005 Some of the best ideas around here are 'improvements' on excellent designs.-- wagster, Oct 20 2005 If you used rollers to prevent bunching up, and put cages to separate them and reduce friction, you would have...a bicylce chain. I don't really see how your idea is better, so I thought I would play devil's advocate.-- discontinuuity, Oct 20 2005 Maybe just a steel CVT compression belt?-- RayfordSteele, Oct 20 2005 Only on the HB would we seriously discuss the equivalent of pushing a piece of string instead of pulling it.-- Ling, Oct 21 2005 Instead of spokes which must protrude through the pipe, allowing the entry of schmutz, one could use powerful magnets which would grip the ball through the unbroken wall of the pipe. The bicycle has long been deficient in powerful magnets.-- bungston, Oct 21 2005 [JJJShabadoo] said truncated spheres, pushing ea. o. flat to flat, but I think rollers in a square tube with very close clearances would be a better bet: it eliminates one axis of disalignment [JJJ's idea, admittedly, eliminates both) but still permits easy cogging into the gears. Run the whole thing in a closed housing, in light oil.
You can't shift the gears, so it's back to the three-speed hub. Put another 3-spd hub in the crank for nine and you're getting a good range.-- elhigh, Oct 21 2005 random, halfbakery