Roller Pinion + Shaft Drive

[NATURAL ANKLING]

Hi,
I would like to see the chain disappear too.

Interesting but too many flaws, and getting beyond themselves with swapping pinions.

Badly made parts with way too much slop.

Weak overall support and fragile parts, claims of 99%...................lets test that on the bike with a 80 kg rider pushing..........

Gear changing with software........................more flaws in the hands of even a five year old excited rider.

[mercalia]

This prototype chainless bike suggests the future of cycling could be ceramic

https://www.digitaltrends.com/outdoors/ceramicspeed-chainless-bike-drive/

[Brucey]

I do think that they would have been better served to have announced this slightly over three months ago; it would have been more widely taken in an, uh, 'appropriate fashion'.

cheers

[NATURAL ANKLING]

Hi,
Last pic.
Be interested to see how far the bike was ridden before something broke / the roller rode up on the front rear cogs etc.
Like Brucey says axial loading will destroy the thing.

How is the rear cassette supported from flex, the whole cassette would need a thumping great thrust bearing with a heavy backing plate........unless its fixed gear, sorta eliminates changing ratios on the move.
If rear cassette is not made from aluminium then it will weigh a ton and will still need substantial support, even if its aluminium it will in current design flex way beyond serviceable use, before the teeth strip.

The video has vague comments about sliding rolling friction.................the design has all the elements of a roller chain minus a derailleur..................there is still sliding friction in roller parts on the pinions.

Also the rollers will not track true on the teeth, even with barrel shaped rollers, I am not sure you could eliminate axial thrust on the rollers......................even with special shaped teeth too.............if you used balls instead of rollers you could eliminate...more so contain axial thrust on the components but there would still be compound thrust on the mating parts.

Rather than eliminate friction you now have potential for more.

Like Brucey says, friction losses in gears is far more than any saving in bettering a chain.
Roller chain only has to deal with radial sliding friction (rolling friction exist in practice but in theory is minimal) on the roller and pin, bushing etc, this design adds more elements of friction.......let alone forces you don't want.

[NATURAL ANKLING]

Hi,
Just came out of the shower and whilst there it hit me that should of been blindingly obvious.
The largest design flaw by far.
Insertion depth of rollers in the cog pinion teeth whatever.
Only one roller has full insertion depth with a say design clearance of say 2 thou at any one time.
Once you move away from this even with a few degrees of turn of the shaft your insertion depth (roller and teeth mating) grows till meshing is no existent.
Looking at the video, at one time all the force is taken on one tooth on front pinion!
The maximum teeth in mesh at any time is two but they have less than perfect mesh with increased clearance.
Pause video to see this when the mech is being activated by hand.

All this adds up to a pulsing of the drive where meshing clearance varies from good as designed one tooth to bad two tooth clearance.

Add the different rear pinion roller count and the pulsing from varying clearance as roller moves in and out of mesh, two unsynchronised pulsing pinions...........................forget the human two pistons........................

[Geoff.D]

The bicycle as a basic design of which we have now, was invented a century ago.
They still haven't improved that design despite many many ideas.

This is just yet another dead end IMHO.

I saw this report at the beginning of the week, and felt it was a solution looking for a problem. But, nonetheless, I rather like it that people play around with ideas, test them out and decide if they have a future or not. It seems to me to be part of the human condition (inquisitiveness, creativity, play, etc). It doesn't bother me that they are playing with a concept. If it did then I'd have to restrict my own "play" in the garage to maintenance and repair tasks, and nothing else.

However, I don't agree that the basic design hasn't been improved upon. I see the development of the recumbent as an improvement, for all its manifest advantages. I see the advent of electric assist on a trike as an improvement. Some types of suspension (eg the Moulton) are an improvement. Adapting bicycle technology to inner city delivery is going at a pace in Europe (eg the Armadillo). Folding bikes (eg Brommie) are an improvement. Improvements for commuting (such as the Radnabel with its lwb, load carrying, leading link suspension and weather protection) are also manifest in Europe.

Naturally this is a selective list, and I could add more off-shoots from the original safety cycle. It's very subjective, because one person's "improvement" is another's bag of bones. Nor would I personally own each of the iterations that I consider improvements, because they just might not fit for my purpose.

However, I do think that imaginative engineers have brought many examples that are different to the original safety cycle, and "improvement" is in the eye of the beholder.

[Brucey]

re design and loading of teeth; Mr Evil helpfully provided a link in his post which contains this image;



which explains how (in a linear roller rack) multiple teeth (with a modified cycloid shape) can share the load and it is even possible to apply a little preload too. Similar considerations apply to the right angle drive, albeit there is always some sliding mainly in the direction parallel to the shaft.

However this all relies upon the clearances being perfectly maintained, which in the envisaged application will certainly not be the case, hence I agree it isn't going to work very well.

Note that the variations in shaft speed are only about the same as with a roller chain of the same pitch though.

cheers

[NATURAL ANKLING]

Hi,
I think its possible to come with a different idea to the chain but moving away from the chain will be a alternative that is already being used today, that will also be acceptable to the sort of city sit up type of stuff that's fashionable today.
Chain drives are cheap and well established.

Its a bit like the wheel, how do you improve on that.

A chain works well as its power is transmitted with tension in the plates, so it can be light weight.
Its a closed loop between two sprockets, self aligning, self retaining, needs no chassis or gearbox.
Importantly power travels in one plane so losses are small.
Power is shared over many teeth and chain links, so no one element sees all the power in one go.
The faster a chain moves for a set power the less force or surface pressure.
Manufacture can be crude and will still be very effective and efficient.


Got to admit that I never looked at the links in the OP.
I did think to my self that it would be possible to get a mesh like gears.
But as you know remove backlash and there is normally whine associated with gears.
Using anti backlash gears is normally achieved with split gears which are sprung loaded to provide preload but this is normally used for quite running.
Eliminating backlash is normally achieved with balls, on a thread.
The main problem with the rollers is that that contact area is small.
And any attempt with shaping the teeth will never eliminate the swivelling of roller on the mating faces of teeth.
Because the rollers are parallel and the meshing teeth do not move linearly in the same plane as rollers. Unlike a linear rack.
So first contact of tooth with rollers as the (to use a better word) curved rack, is not centre on roller, power transition is sliding as said parallel to the power shaft...................

This sliding does not exist at all with a chain, certainly not axial to power plane.
Even the modified curved rack shown in links will not eliminate what is now introduced, though less possibly, a now complex profile of tooth contact area is needed and I suggest that their modified profile goes only part way in attempt to eliminate this sliding contact, in angling the tooth an attempt is made to mimic a linear rack, this is done as first contact in now at a larger diameter................as the roller moves down the tooth..........the diameter of the curved rack is less.......so contact between the two parts it is staying in middle of the roller....actually a nice trick, I could model this (if I had time / inclination) and possibly all sliding friction is now gone?
To achieve this in perfection means that the tooth contact area needs to be profiled correctly so that the roller contact does not swivel from the edge of tooth face on to the face.
Imagine a strip of paper like a 12 inch rule.........form that into a tooth shape.....now twist the ends....this is what the contact area on tooth should look like on both the curved rack in video and their modified angled tooth curved rack!
I envisage a full mod of the tooth contact area means that you cannot machine the tooth profile in one pass with a milling cutter.
It would ether be needed to use a CNC milling machine / cast / forge / mould the curved rack.

Now if you introduce a barrel shaped roller.................you could keep with the standard curved rack as video and result would be same as their modified angled tooth curved rack.........

A barrel shaped roller is a cheap trick..........eliminating of course any sliding parallel to the shaft, but NO it reintroduces axial thrust in the shaft
Another trick would be to dish the roller pinion ...............as well as a barrel shaped roller, this would mean that the contact always stays centre of roller, a small adjustment to the angle of the machine cutter machining tooth profile and its done.

We now have what now resembles a bevelled gear used in gear box for 90 degree power transmission, axial thrust is always present.

[[XAP]Bob]

The bicycle as a basic design of which we have now, was invented a century ago.
They still haven't improved that design despite many many ideas.

This is just yet another dead end IMHO.

I disagree - the design has been improved a number of times for different use cases. ThebUCI, and conservative manufacturers however...

[crazydave789]

I wondered about bolting it together with a bottom bracket gear thingy, but you still have a weak point and the belt drive might remain the better option.