Phaser g-boxx style middrive

crossbreak said:
Guess this is not so dramatic, since the chain length is enlarged and there are only 3-4 sprockets.
Don't forget we're using heavy duty 1/8" bike chain.... But I agree that, with 3 sprockets, shifting the chain shouldn't be a problem.
 
It's not g-boxx anymore, since it doen't need a g-boxx frame. This drive has a lot of potential, so why should we not call it the omnipotent ES drive :D

I'm still not sure if the 2-stage reduction easy to build. The space on the left side is quite limited. I made a CAD model of a very similar drive once, but I can't find it atm, but I needed at least a 153mm BB.
 
crossbreak said:
Made a plausible specification for the MAC/BMC/PUMA:

crank to jackshaft (or macshaft :mrgreen: ) : 48T to 13T
front sprockets: 14-20-27-34, increase rates: 43%->35%->26%
rear sprocket: 25T
highest crank to rearwheel ratio: ~5:1, lowest: ~2:1
average macshaft rpm: 295 @80rpm crank cadence, motor: 1772rpm

Interesting....
 
crossbreak said:
I'm still not sure if the 2-stage reduction easy to build. The space on the left side is quite limited. I made a CAD model of a very similar drive once, but I can't find it atm, but I needed at least a 153mm BB.
That's why I was thinking to have the motor separate, with the first stage to the crankshaft. 3 shafts including the motors.
 
I played around with some numbers. If one can trust the e-bikes.ca simulator, the MAC/BMC should be capable of about 100Nm peak torque. It could be 200Nm at the wheel, making it possible to climb every hill up to 50% gain.
 
The derailleur is a very tricky part. Thought about modifying a regular one, turning it around, so that it moves correctly to the gears. The cage could be sawed apart and a new one could be screwed on, being turned to the other side.
 
I actually played around with the cassette on the motor as well, but it leaves very little room for the chain to flex due to shifting sprockets between the motor and the idlers and I think in actual practice will not be very stable. I like the idea of moving the crank ring to the other side like you did but I decided it works best traditionally. I have not had any problems with my freewheel crank in over 1000 miles so far and YES it makes incredible torque in low gear! I believe I could climb a 45* slope easily if I could keep the front wheel down! This is the steepest hill I can find locally and it had no problem climbing it pulling only 1200 watts and motor stayed cool after three climbs!

https://www.youtube.com/watch?v=tzA-CThP0ps&feature=youtube_gdata_player

https://www.youtube.com/watch?v=bM63WOAAQy4&feature=youtube_gdata_player
 
Ref: http://www.endless-sphere.com/forums/viewtopic.php?f=28&t=16329
:

file.php
 
Very cool! I would like to try something like that!
 
I actually played around with the cassette on the motor as well, but it leaves very little room for the chain to flex due to shifting sprockets between the motor and the idlers and I think in actual practice will not be very stable.
I think i get what you wanna tell. Idlers dont have to be axially fixed, since they only idle (note the 4th pic below). That's why I see no problem. I used them once in my tongxin drive, ones that came with an axial chain alignment (alignment disks on each side). Worked fine. Guess theses alignment disks aren't even necessary, that's why I dropped that in the CAD pics.

Played around a bit with various chain tensioners. Guess the one I chose now is the best solution, keeping the distance between idlers as small as possible. Chose 11T idlers and 55mm axle distance.

With the last pic I try to show how it would look like if a small pitch chain was used for the crank to jackshaft overdrive.
 

Attachments

  • 55mm_right.png
    55mm_right.png
    72.7 KB · Views: 2,145
  • forth_left.png
    forth_left.png
    63.6 KB · Views: 2,145
  • fourth_top.png
    fourth_top.png
    41.5 KB · Views: 2,145
  • fourth_small_chainring_32.png
    fourth_small_chainring_32.png
    58.6 KB · Views: 2,145
the motor in the pics is the Bafang SWXB front motor. It's a 24Volts one run @37V. It's an 20p/18t outrunner.


motor to jackshaft planetary reduction: 5.42:1
crank to wheel overdrive: 229%, 1.82-4.17, max speed: 17.5-40km/h @ 80rpm cadence
motor speed @80rpm cadence: 1467rpm
primary overdrive: 44/13=3.38
front sprockets: 14-20-26-32
rear: 26T
idlers: 11T
chain: 120links; 1524mm; only 3/32 to make using four hyperglide gears possible.
 
Now that works! I like it! Floating idlers, what a great idea!
 
thanks for your objection. It made me think of that further. Floating idlers have the drawback: the the axle, they ride on, is always faced to dirt :? Guess new materials like teflon make it possible that they wont squeak :D
 
one could set the idlers on the derailleur bracket. This way the derailleur would need get a stronger grip-shift switch on the handlebar, but that is something which could be solved. I made a 3-speed thumb switch for a sachs internal gearbox once cause it broke (bad lubricated bowden cable). The hole job took me only an hour and it was better than the original one :D
 
this is the simple version. still has it's right to exist due the reduced drive loads and thus reduced wear while keeping the possibility to pedal at high speeds. Mabye we should just get comfortable with chain guards :D

Gear reductions kept the same from the recent design.
 
I've been reading the thread but I'm a little confused about the render. So the motor is now direct driving the rear wheel through a one stage reduction?
 
the final aim of this project is to reduce those common drawbacks of middrives:
- complexity
- limited power, limited due bicycle drive train (usually about 1000W)
- increased wear at the drive train
- lower efficiency on flat roads due drive train losses
- there is no need for a freewheel crank, which tend to fail.

The bicycle freewheels in these freewheel cranks are always over stressed, even the expensive ACS freewheels break form time to time. Mine broke within the first 2000km in my tongxin middrive bike although I'm a lightweight person.
 
fizzit said:
I've been reading the thread but I'm a little confused about the render. So the motor is now direct driving the rear wheel through a one stage reduction?

The used motor is a converted geared hubmotor. It has still it's build-in reduction. Boosted by the gears to the wheel.

Gear reduction: 73:17T+1=5.294
gear reduction to rearwheel: 1,78-0,83
motor reduction to rear wheel in first gear: 10

for comparison, a GNG drive has a first gear reduction of 16.2 when using a 44T front and a 34T rear sprocket. The dilemma with the GNG drive (and cyclone and so on) is that the wheel chain is over stressed and will wear quickly. This is avoided by speeding up the chain to get a higher power rating and still get a longer lifespan for chain and gears.
 
Hmm, I. Like the simple design! If I calculated correctly, I could actually get lower gearing than I have now with my 20t motor gear and dual 42t chain rings on the cranks feeding my 11/34 cluster and get even better efficiency since I'm not going down then back up with the reduction AND I can keep the cadence where I want it on the other side IF I can find a simple source for a backwards freewheel assy. Boy that's really worth trying! Any idea where to find a backwards freewheel for the left side?? Cheap is good here too since there is no longer any power being transferred there either!
 
http://www.staton-inc.com/store/catalog/Multi_Speed_Sprocket_and_Roller_Chain_1_2_Pitch_x_3_32_Width_or_thinner-53-2.html
bottom of the page they have left and right hand threaded :mrgreen:
 
Excellent so that will thread onto my right hand side crank arm to use on the left side or will I have to find a different one of those as well? I never was good at threading info.. THX!
 
the thread can be either. You definitely need a freewheel adapter to mount it on the 12mm shaft of the modded hub. You can chose yourself in which direction you want to mount it. I would stick to the right hand DICTA ones, they are cheep and quite durable.

In my build I wanna use a 13T (special 30mmx1 thread) one with more than 13T is a bad choice IMO. A roller type freewheel would be nice, maybe we can take a normal 13T Dicta and get a CSK-17 into it.


I found no adapter on this page that can fit. First they have no 30mmx1 threaded ones, second their bore is too big. We need 12mm, that's less than 1/2inch. We've got to build our own ones.
 
Back
Top