Anyone seen this yet?

mwkeefer

1 MW
Joined
Jun 14, 2009
Messages
2,263
Location
Malvern, PA USA
Has anyone seen this yet?

Anytime-folding-bike.jpg


It appears to me as if this addon unit is driving the crankset chainguard friction drive style? Am I missing somthing? Not good for high power / torque applications but for 250-500w I would think it should suffice... and maybe be easier than other methods (lower weight without addtl chainring sprockets, chain, etc).

-Mike
 
First thing that sprung to mind was "its a baby version of Miles L337 e-bike" :D

Closer inspection though its nothing like it ... Interesting idea though good inner city commuter bike
prolly? and very stealth...

KiM
 
It could have attributes of lightweight, excellent torque, high efficiency, and variable motor gearing all through 1 simple setup. Its clever, I like it. :)
 
Not too much info on the specifics of the electrics but here is something from the site:

The 13th International Bicycle Design Competition winners were announced on the opening day of the Taipei International Cycle show that was held March 17-20. Folding bikes made a significant presence at this year's competition comprising 11 designs in a field of 20 finalists. Taking 2nd place in the competition was the "Anytime" folding bike designed by last year's IBDC winner, Larry Chen. The bike features a removable power assist unit called the "power pack". The externally mounted power pack design integrates a motor, control unit, battery and tail light.

http://www.foldingcyclist.com

Calling the power pack the "power pack" was quite imaginative.
 
I did some enlargement and pixel enhancement and it seems I was wrong, this is not a friction based setup !!!

In my opinion after figuring out the idea behind this, I retract what I said previously about not being viable for higher powered setups - with proper components, it could very well do the trick - in fact it gave me another idea which I hadn't yet thought of (more later..)

So instead of a roller on the chainring guard, the chainring is actually exposed... the thingy which looks like a roller is not, it is (I can't be sure of material but it is comprised of composite and somthing else, likely aluminum) a gear / sprocket?

Basically it's about 3/32nds (or chain link width) and it is in essense (it's hard to count) a 9t inverted and widened sprocket. Based on the coloring of the surface I would actually go so far as to suggest their assembly method is the inverse of epoxying magnets into an outrunner - more specifically it appears to be a hollow tube (to be mounted to output shaft of an internal reduction unit perhaps, more likely at the power level they are using its a single stage reduction) which has teeth machined, cut or otherwise formed from ?delrin - (I assume to prevent chainring damage and reduce noise of engagement) and then glued into slots or position on the aluminum tube to form splines... there may be retaining rings but I can't make those out.

So basically it's a low noise, low friction, long life 9t gear for driving a normal 3/32 bicycle style sprocket.

Assuming a chainring (I'm in the 20" world for this idea) of 58-60t which is the largest I have been able to locate for BCD 5bolt style (if you know of larger please let me know, in process of sourcing best possible gearing for maximum speed on a 20" with a custom 3,4,5 or maybe 6 speed cassette built up specifically for use with eBiking).. That gives you a reduction of 5.27:1 - 6.667:1 right at the crank and reduces weight of adding additional metal sprockets, belt pulleys, etc.

I figure it also means a single chainring which freewheels independent of it's pedals would do the trick nicely... this way at above the maximum gearing speed at 80RPM cadence of 25mph you can hold your pedal feet still and the motor can continue to drive - ie, pedals disconnected from driveline.

Torque would be the only issue I can think of for a HIGH POWER setup but then that brings me to my next idea... Why do we use chains and sprockets and belts for this purpose... why not a pair of gears... drive gear could be small (depending on the strength and material - steel much less prone to torque damage than others at weight disadvantage) say 8-9-11t and then a second chainring bolted on is a 3/32" thick sprocket of say 60t or better (smaller diameter due to pitch and engagement) located on the inner side of the pedal chainring)... Now you have possibly even higher ratios (I need to check into this) for instance I know a #219 would let me run 11t to 76t and the 76t would still be smaller in dimension than the 60t chainring - what if we can find appropriate gears capable of handling the torque loads - it still eliminates the chain addition and in most likely cases the weight of the 2 gears would be less (at 3/32 - 1/8") than the weight of equivilent bike or #219 chain.

The issue remains - even on a 10S pack with 100A peaks the torque an Astro 3210 10t would produce at the crank would be 277.880858 ft lbs after reduction to provide for maximum crank RPM of 60 RPM @ LVC (20mph).

Granted the the legal power mode operation with a current limit of 25A (750w @ 30v) the output torque would be 69.49 ft lb of torque and I don't really know too much about torque generated by a human rider but I will assume that we have less HP (hence the est of 100-200w for most riders, 250 for lance armstrong (not me) and if you assume a bicyle indended for a rider weight of 225lbs max lbs and has 2 power strokes accounting for (guesstimate) approx 80% of the possible duty cycle so 225 * .80 = 180 ft lbs torque expected / designed for.

Please if anyone understands how to compute these torque distributions and capabilities better, I am dying to understand more - I assume from TDC to end of stroke 10% is wasted on each side as stroke transfers from one foot to the other - hence the 180ft lb torque continuous rating I assume.

If that were the case - then the most you would want to drive the crank with in terms of torque would be about 65A @ 10S reduction level and 80RPM max crank (more reduction would increase the torque so voltage or current increase would exceed the value too far - even here were talking about 180ft lb or so continuous - actually I suppose that's perfect and will match human torque?

In either case... Though my initial observation of this was wrong, friction could be the best possible manor to setup a lower powered to legal powered eBike... even using cheap motors such as the tower pro 5330 9/10t or many others with a kv between 180 and 235 or even better with an Astro (94% efficient at legal 750w power levels)...

A bit of slippage could be good - spring tension engage/disengage?? (Steal other friction methods here from evTodd, etc)
Chainguard material - my folder has cheap plastic chainring guard (might be fine with a rubber coated drive spindle) - I am thinking version 1.0 aluminum (because I can finally make them here at my home) replacement (it is the kind that wrappers the chainring and is just a tad larger diameter than a 58t chainring.

I don't know the exact diameter but I do know that with a generic friction roller (alumimum tube with shrinkwrap or delrin tube epoxy bonded and then mounted to motor output shaft with set screws the reduction possible right there is about 6:1

This design eliminates the complexities of the cyclone style kit which requires the use of a large chain idler and a very non-effcient chainline path (too many bends to engage the motor in the driveline.

-- with regards to Miles build;

It took me quite some time to read all of miles folder build posts and cross thread info and even longer studying the pics of the actual bike (and I still don't get the little tiny mounts to the frame rails?) but the beauty of Mile's setup is the synergy between e and bike when he wants it... I think 8 useable gearing ratios for the pedal side and 3 for the motor is much better than 8 for the pedal side an 0 for the motor!!! Remember going up a hill - your motor wants to spin as fast as the voltage it's getting would permit but it can't, that's inefficiency... also consider a 20" bike - though you may have 30-50ft lbs of torque available if your single speed drive motor to rear - you may have the abiliity to shift pedal side for more efficient (or surviveable) required force at lower speed but unless you can do that for your motor... going up huge hills where you could be pushing 138.94 ft lbs through a reduction of 1.36:1 which would increase the final reduction from the 1:1 and give you more torque.

I am inspired by miles work (and that darn bike - hey bud, want to sell it when you upgrade? Dibs?) but what I am more trying to achieive is a simple method to drive the crank directly (yes IPS is required unless cadence is kept 60-80 RPM which most people can handle - I do) both efficiently and with as little required weight and visible "e" bike stuff.

Thanks for the feed back everyone... sorry to be so long winded but though i can't say I like the bike (the brakes are cool) I found it too inspirational and just had to figure out how it worked for legal power levels and what / if we could modify it for higher power useage without breakage.

-Mike

PS:

I've seen I think some DH bikes with gears in behind the pedals which then feed a chainring - if anyone knows, I am wondering 1.) what kind of ratio or virtual chainring can you get with something like that? 2.) Does anyone know where I can find gearbox stuff like that? 3.) The problem is with a 20" (god forbid 16") tire on a folder (or other smaller commuting bike) - you can dump tons of money and time engineering solutions to suspension issues, noises, balancing, proper fit, etc - you can even go so far (as I know it's the limit really) as to implement the 60t chainring / 11t high gear for 20mph at approx 62 RPM cadence - but I've made my folder exquisite in terms of alignment, balance, tire, suspension upgrades, joint tightening and finally getting it to the point where it feels better to me than my 26" - the issue is I can't pedal along more than 20mph for any length of time... I can do 25mph or 80 rpm cadence but only for a short while so I am hoping to be able to find a pedal system which would give me a 2:1 increase taking my input RPM and providing 2X output via a chainring and then onto the normal system?

Thanks in advance - know some is off topic

-Mike
 
Reminds me of donkey-trains with cogged wheels that had to climb unusually steep grades. I'll look for a pic, back in a minute...never thought of directly driving the chainring with a cog, no reason it wouldn't work....You'd only need a single large-diameter chainring that freewheels (rather than two chainrings for a Cyclone copy). No worries about rainy-day slippage from a friction-drive!

Doesn't look like the chainring freewheels, so perhaps its a "pedelec" which seems to be a growing trend globally.

13abt.jpg
 
Neat drive idea. It should be possible to replicate the drive "sprocket" (if that's what it is) by machining up a couple of discs and fitting chain rollers at the right spacing between them, on small bearing pins. If the hub of this caboodle was fitted with a sprag clutch/one-way bearing, then it might make for a neat pedal-assist-only type set up. The clutch would allow the bike to be pedalled normally with the motor off, which might be useful.

It doesn't look to be too difficult to make, especially for anyone with access to CNC tools.......................

Such a drive might well go well with Matt's reduction drive unit, just replace the final drive pulley with the special drive sprocket and make up a bracket to hold it in position against the chainwheel.

Jeremy
 
Do I know precious little about bikes or is that saddle too far rearward? Almost directly over the rear axle seems good only for wheelies, and any incline will require walking the bike up, or are you supposed to always stand up on the pedals while riding?
 
I am immediately drawn to the idea of making a drive cog out of two aluminum discs with the engagement pins (the reverse of a toothed chainring) being 9 bolt shafts. There is often space between the seat-tube and the front of the rear tire to allow a small RC motor for direct drive (no belt or chain) or I could install a 24" wheel in a 26" frame to increase this space. What is longest length RC motor that will fit and not hit the left pedal-arm, and the fattest for max power potential?

An outrunner "reverse-cog" sleeve could be made, but that would increase cog diameter. A separate cog allows an outrunner or inrunner, with inrunners having shell-fin and cooling fan options.

A recent trend in full-suspension frames is for the BB to be a part of the frame rather than the rear suspension arm, so a frame-mounted RC direct-drive cog-onto-chainring has a lot of potential....
 
Wow, try to find a good photo of a "Lantern" pinion gear on the www.
It never occured to me to drive a sprocket with a lantern pinion- the geometry is identicle to a chain.
I can only imagine the growel of steel on steel. it would put a lot more strain on the driven sprocket tooth....but I would think a composit chain wheel driven by a steel pinion may sufice...
http://ericfreitas.com/images/12-C.jpg
 
I would assume that the motor has a freewheel or one way bearing - given the size of the drive unit (appears to be 6-8s max) and the apparent UK, Asia market it is intended for - the max Cadence is likely 60rpm and yep pedalec (since the cvontroller is in the same unit so it appears no throttle.

The crank could be freewheeling though since most only require pedal once every minute to keep going - this is a rather cool idea no matter what it has great potential

-Mike
 
Hi,

Main site:
http://www.t-onedesign.com/anytime.html
Description
Any Time is a folding bike with removable power assistant unit called [power pack]. The external mounted power pack design integrates a motor, a control unit and a battery. It allows users to remove and install with a quick-release system. One can have the bike with power pack for daily commuting, or with out power pack for weekend exercise. The front and rear wheels are positioned paralleled to each other after fully folded, of which also allows for towing the bike whilst walking. It is a commuting and exercise equipment perfect for people who live in big city and don't have big rooms at home.

It works like this (with a 48t chainring it would be 6:1 reduction). I wonder how noisy it is and if alignment or flex are issues:
0808084541-a.jpg


0808084541-c.jpg


More images here:
http://www.t-onedesign.com/anytime-photo.html
 
Cool idea. Would be fun to try a system like that using a freewheeling crank.

The more I look at it the more I like it. Very simple way to drive through the gears. I bet flex wouldn't be that big of an issue with a strong freestyle bmx chain wheel. Not sure about noise but I'm thinking using small bearings to cog wih the gear would work. Heck, you could even make a chain ring and actually use a chain that mounted with machine bolts instead of pins. Designing that so it wouldn't rip apart would be the big challenge. Although the more I think about that it probably wouldn't be that hard either, just have another ring on the outside to support the bearings or chain. Hmmmmmmm.
 
From the moment I saw the first picture, I had assumed it was 9-10 little miniature ball bearings fixed in the inside race to the base of the drive sprocket, and with the outside race sized to match the chainring.

From how it looks in the lower pictures, it seems they are just letting it have a large amount of sliding friction between the chainring groove and the pins, which seems like it would be asking for longevity problems.
 
liveforphysics said:
From how it looks in the lower pictures, it seems they are just letting it have a large amount of sliding friction between the chainring groove and the pins, which seems like it would be asking for longevity problems.

I don't know... Would that be a big problem? A chain doesn't fit tight inside a chainring (thinking about it though, it does have more surface area touching many more teeth). I would think it would just roll along the groove if it uses bearings (looks like it).

I'm kinda wanting to try it just for giggles. It wouldn't be that hard to set up for a little test. I bet it would produce an interesting sound with bearings, metal on metal, and an rc motor. lol
 
AussieJester said:
Im think D9 Caterpillar track noise kind of sound :mrgreen:

Lol. I actually know what a D9 is. Man, if it's that loud people on the bike trail will get the hell out of the way without me yelling "on the left". That might be a good thing. :D
 
--- Removed lots of stuff others had posted.

I spoke with a friend who has 16 (he bought 20) of the 10t bike style drive sprockets at about 1/8" thick and flat (no collar or lock anything) each... he has volunteered to let me bind them with a weak resin and make a full... pourable casting of them... So that will be maximum stock (initially) of 2" per casting. I figure I will be ready to pour them by mid next week. I have been working on the empty mold to ensure it would take the "template gear" and when placed into the mold... would be exactly 1/2 depth (to make boring dead center with nothing more than a drill press easier) - the material I am using for the first batch is a bit scary and I doubt it will last too long - A long time ago a GoPed dealer going out of business was selling a bulk lot of the chain tensioners (I think nylon but perhaps delrin - need to check, different melting points and cooling rates)

Luke,

Deflection and deformation could be an issue - but I would be willing to bet 10-15lb of torque at the chainring, and 250w maximum power - how many torque inches could such a motor (even if it's internally reduced 10.52:1 for 25mph at 32.5lvc on 10S) produce? At 17A and 32.5v an Astro is 94% efficient and couldn't put more than 185.351395 in oz of torque - factor in the reduction to the crank to maintain a minimum crank RPM of 80 for 25mph (420 rpm) and you have 10.1557119 ft lbs of torque available.

Upgrade to 1hp model (750 / 32.5 = 23.077 A) and the numbers jump a tad but not much...
in oz torque @ 23.077 A = 251.609067 in oz of torque or a gain of 66.25767 oz in which might seem alot unti you factor in the reduction for a total of 13.77466 ft lbs of torque at the crank.

I think in normal circumstances the design should be quite fine and deflection (even with alloy crank) shouldn't be bad... I do agree now that those must be bearings mounted to the race because the friction of a non free rolling surface would cause heat and warpage (even at 60RPM and such low torque numbers).

Now take my personal configuration intended (and man am I glad to see from you guy's comments I'm not the only one this spurred a project and improvement / alterations on the design for what I like to think of as the most efficient DD method I have encountered yet - without the added weight of coupling chain):

Astro 3210 or Tower Pro 5330 (actually both but on differrent builds space constraints - looking into a 1700hp max inrunner and possibly rewinding it for lower kv.

Since I will probably build the Astro first :

Astro 3210 10t - Reterminated as Wye, Output Leads Replaced with Larger Guage PTFE insulated, Added 120 degree halls (on the test bench now, photos and detailed how to coming eventually when I get some other more pressing - ie: paying work done)

Expected Kv: 124.277457 rpm @ volt
Expected Kt: 10.9030232 in oz @ amp
Max Current: 60 amp
Power: 10S4P 20AH 37v Nominal Lipo 20/30C
Required speed at LVC: 25mph
Tire size: 20"
Front chainring: 60T (minimum - stock is 48T I think - needed because 20" folder is slow)
Rear cluster low/high: 11t/38t - (custom cassette)
Max Pedal Gear: (60T/11t) = 5.4545:1
Pedal Cadence for 20 mph: 61.60 RPM (alll this time I thought I was comfortable about 30-40, nope... avg is 65 for me)
Pedal Cadence for 25 mph: 77.006417 RPM (not simple, feels a bit odd but stable, above 80 is a bit out of balance)
Drive Sprog: 8t (made of RC car model bearings)
Output Drive Sprock to Chainring Reduction Ratio: 7.5:1
RPM of Astro at LVC (32.5v): 4039.017
Required reduction pedal cadence of 77.xxxx @ LVC = 52.34503951:1 (constant reduction through discharge so upper speeds limited via microcontroller current limiting)
Torque (inoz,ftlbs) at motor shaft @ 60A: 654.181393 in oz / 3.40719476 ft lbs torque
Torque (ftlbs) at output shaft @ 60A: 178.349744 ft lbs of torque

I am fairly certain that since a bicycle is rated for 240-300lb person, that would me (assume 80% duty cycle of human power stroke) about 200-250 ftlb of torque is useable and rated at the crank... I'm counting on this. I also fully expect to explode my cast nylonish gears quickly at these kinds of levels - but I want to test them between 250w output power and 750w first which I think they will thrive in (weight reduction, size reduction, etc) and would be great for a fairly simple bolt on kit... then I will move up to 2hp, 3hp and finally continue upping the voltage and current until I smoke the little buggers up!

One thing I want to add is a one way bearing on the output shaft within the drive cog... Then pedaling will not cause motor to turn and hopefully will eliminate any possible parasitic losses on the pedal side of things.

Other than homebrew (unless it's nice) and sick bikes, can anyone recommend a system for a freewheeling chainwheel such as a hacked up IPS system - I know there's a way to make one with a bottom bracket (old school) and a threaded cup but... those parts seem hard to source... I really suppose I could rig an offset adapter for a normal 5 leg BCD chainring mount to accept an aluminum adapter plate with threads for adding a freewheel... then attach the BCD disc to that freewheel... seems alot of work... anyone know of such a crank set with freewheel that will allow mount of standard single BCD 5 bold chainring?

-Thanks again!

-Mike


-Mike
 
Hi Mike,

A new design is great if it solves a problem. In the original design the problem this solved is a complete removable electronic drive unit. If it works well its a great idea.

The problem you seem to be trying to solve (chain weight and the weight of the second chain ring) which seem like non issues. The only other possible advantage I can see is the possibility (see below) of greater reduction. He used an 8t cog and 11t is the minimum for a reasonably quiet chain sprocket.

I don't think deflection should be a problem if the alignment is good. If it turns out to be a problem I think groves in the cogs would solve it. The potential problems I see are:
  • More precise mounting required as compared to chain drive.

    Much more limited mounting locations as compared to chain drive.

    More stress on chainring teeth (only one or two teeth in contact at a time).

    Also if there is any slop at all between the cog and the teeth the cog will bang the chainring teeth. Might bang them anyway unless there is always more than one cog engaged.
It might work better with more teeth on the cog. If so that eliminates the advantage of greater reduction.

BTW on the companies main page there is a contact email. It might be worth shooting them an email and see if they have any advice.
 
Mitch,

You do make a good point. In fact, the more I thought about it I was wondering why I found it so interesting since you can do the same thing with a chain and another sprocket. Still a novel idea though. I'm guessing it's more of a design study and I wonder if they even have a working bike.

I still may play around with making one if I'm bored some weekend. :D
 
I've got some high impact resistant castable ceramic that might be just the ticket for these "gears". It was originally for casting some ceramic pistons, so it should hold up pretty well since making the part fairly thick isn't a problem.
 
MitchJi said:
It might work better with more teeth on the cog. If so that eliminates the advantage of greater reduction.

Two cogs should be easy enough...
 

Attachments

  • anywhey.jpg
    anywhey.jpg
    25 KB · Views: 4,517
Mitch - you do have a point, my "goal" isn't to find a novel new "shiny" drive system but rather reduce the complexity and adaptability of a folder style eBike. For folders, weight is or more accurately can be an issue... Bear in mind, I see folding 20" bikes beginning at about 17lbs naked and an Astro or Tower pro throws about a pound or pound 5 to the total, even with chains of #219 weight and Amberwolf style (lightning holes) weight reductions an RC style will always be lighter than a hub but alas far more complex from a mounting perspective - especially if you want to actually "match" cadence to the motor so you can always add some assistance and even more complicated if you want a "convertible" which is exactly what this company has claimed.

All that said, I've done some bench testing (ugly so no videos) using an old BMX bike and I've managed to kill 3 normal chainrings and one plasticy type... sadly while this may be a great solution between 200-500w and with lower torque, even my tower pro 5330 10t running wye with a 45:1 reduction on 10S managed to snap the teeth of them all off.

The problem is no matter how large you make the drive "sproget" you will never have more than a single tooth of engagement *(ok, maybe you could make the sprokget so it would engage more than a single tooth at a time) and the torque load when applied to a single chainring tooth - well the result is inevitable.

Testing at low power levels shows little deformity of the chainring - I think it could handle 250w brunt and 500w with a gradual ramp up (to prevent higher torque)... this is good for legally powered eBikes such as the one I posted.

I've looked at other alternate methods, and am even looking at creating a semi circular track out of some bike chain and a few 9t tensioner rollers - the idea is to allow the chain to wrap around about 1/8th of the chainring... for a 46t chainring it could engage 7-9 teeth at a time, distributing torque but then as Mitch points out... that will require a minimum of a 28t and thus reducing the reduction to 2:1.

I still think there may be a better way to do this... another idea I'm looking into is using a drive belt similar to that on a strida... that would allow for a more... gearlike drive unit to be linked via contact and since those drive belts are about 1/2" at minimum... the proper chainring and drive gear material would be much more inclined to handle the added torque (or so I think).

In either case... I merely wanted to call attention to this because I hadn't seen anyone doing it yet, now that I've done some crude testing, I see why WE don't use this method - at our power levels, well... chainrings aren't cheap are they?

-Mike

PS:
TD - just caught your post as I was subbing this one, leave it to you to solve the engagement issue for higher torque - that's 2x 8T drive sprogets which in your illustration seems to engage a maximum of 4 teeth from the chainring - nice save!
 
mwkeefer said:
Testing at low power levels shows little deformity of the chainring - I think it could handle 250w brunt and 500w with a gradual ramp up (to prevent higher torque)... this is good for legally powered eBikes such as the one I posted.

Current mode control is the way to go, I think. My first controller was voltage regulated according to the potentiometer setting, but the one I use now that regulates the current is way better. I guess it might give a feel like pedalec without the need to have to pedal.
 
Hi,
mwkeefer said:
TD - just caught your post as I was subbing this one, leave it to you to solve the engagement issue for higher torque - that's 2x 8T drive sprogets which in your illustration seems to engage a maximum of 4 teeth from the chainring - nice save!

-Mike
It might be better to offset them so that when one cog has two teeth engaged the other one has one tooth engaged for (hopefully) a minimum of 3 teeth engaged.

Still not as strong as a chain and massively more complex to build.
 
Back
Top