Kepler Friction drive comes of age.

Keen to test a skateboard hub motor with the drive so I have put together a variation of the drive mechanics and installed onto a folder. 90% of the drive is standard but with the addition of a clamping block and modified swing arm. Keen observers may also notice that the endstop plates a spring tension parts are mounted on the outside of the the main mounting plate.

Mechanically all ready to go. With get the speed controller and battery fitted up on the weekend. Will just use a stock 6 Fet ebike controller for starters. Should be interesting.
 

Attachments

  • 20170305_124818.jpg
    20170305_124818.jpg
    34.8 KB · Views: 5,229
  • 20170305_125233.jpg
    20170305_125233.jpg
    43.6 KB · Views: 5,229
  • 20170305_132005.jpg
    20170305_132005.jpg
    58.8 KB · Views: 5,229
  • 20170305_132238.jpg
    20170305_132238.jpg
    60 KB · Views: 5,230
  • 20170305_132215.jpg
    20170305_132215.jpg
    70.3 KB · Views: 5,229
Interesting. Since my frame cannot accommodate a seat tube mount at that position, do you think a flatter chunk of plastic bolted to the kickstand mount would work?12608.jpeg
 
Hi Kepler, i am interested like Payder to know the price for the swing arm without motor ?
thanx a lot.
 
hi all im new here but have been reading avidly many interesting ebike stuff here. as a motor mechanic and now an environmental scientist i wanted to get out of the car pollution world . I am interested in one of kepler's wonderful drive kits just the fancy bracket of his i think i will be well able to put it and the electrinics together given my electrical and mechanical experience . i have loked at many designs of ebike drives and have decided that simple is best and want to go the friction drive route for the moment, later when i have time as im studying full time at the moment, i plan to build a mid drive device using one of these new drone pancake motors that have 48 poles and a large diameter that should give excellent low rpm torque needing less stages of reduction. So when and where can i order one of Keplers wonderfully simple gadgets ? please let me know. i think ill go the vesc route as it looks to be the best option and being famillar with linux and its intricacies that should not be too hard.
 
mechhead said:
hi all im new here but have been reading avidly many interesting ebike stuff here. as a motor mechanic and now an environmental scientist i wanted to get out of the car pollution world . I am interested in one of kepler's wonderful drive kits just the fancy bracket of his i think i will be well able to put it and the electrinics together given my electrical and mechanical experience . i have loked at many designs of ebike drives and have decided that simple is best and want to go the friction drive route for the moment, later when i have time as im studying full time at the moment, i plan to build a mid drive device using one of these new drone pancake motors that have 48 poles and a large diameter that should give excellent low rpm torque needing less stages of reduction. So when and where can i order one of Keplers wonderfully simple gadgets ? please let me know. i think ill go the vesc route as it looks to be the best option and being famillar with linux and its intricacies that should not be too hard.


ah I misunderstood, you aren't trying to use a drone motor with the friction drive.
 
He said that he will start off with Kepler's friction drive, then latter on he will experiment with using a new drone pancake motor to make his own ... Mid-Drive .

Mechhead, there will be as much interest in your drone pancake motor mid drive as Kepler's friction drive. We are very much in Need of a Simple/Lightweight Mid-Drive System !

Example : For a light weight Road Bike setup , Kepler's new Friction Drive looks like the best that is being/will be , offered on the market right now .
( For those of us riding on Dry Pavement Roads )
However
For Mountain Biking, especially for going up 1-3k feet in Elevation ( 300-100 Meters ) or more on each ride , with Knobby Tires, crossing streams, etc. A light weight Mid-Drive would be Best. 8)
I have been waiting for someone to design/make a light weight RC motor , Mid -Drive system. I have known it is possible for a while now and have been watching and researching , but China is not doing it , nor do they seem to have an interest in doing modern / innovative DIY Kits.

So Mechhead, know that many of would like one as soon as possible ! Please PM me so I can keep informed.





mhaskell said:
mechhead said:
i plan to build a mid drive device using one of these new drone pancake motors that have 48 poles and a large diameter that should give excellent low rpm torque needing less stages of reduction.


large diameter motors are not needed with keplers brackets (or possible). A 50mm motor works perfect for me. What do you mean by "stages of reduction", this is a direct drive setup.
 
Hi mhaskill and scooterman101
To clarify firstly i am after one of keplers friction drive motor brackets because im time poor at the moment with doing honours at uni and having a three yo daughter.
and his setup looks simple enough and cheap enough i can put it together in my non existant spare time from easy off the shelf parts .

I have the idea to incorporate a brake lever pull system and brake switch into the vesc and keplers friction drive unit to incorporate regenerative breaking that is activated before and as well as the standard friction breakes that are fitted to the bike and if the drive motor provides sufficent breaking then maby loose the rear friction brakes entirely, although it doesnt hurt to have a backup where breaks are concerned particulalry with how friction drives perform in the wet.

the drone pancake motor idea involving the new agricultural drone pancake motors i will start a new thread for . i will be brief here about my idea I have to combine somthing simmilar to keplers friction drive with a mid drive setup as a means effecting regenerative breaking on a mid drive without and extra motor/generator and still keeping the freewheel systems on the rear wheel and crank set
 
Kepler said:
Keen to test a skateboard hub motor with the drive so I have put together a variation of the drive mechanics and installed onto a folder. 90% of the drive is standard but with the addition of a clamping block and modified swing arm. Keen observers may also notice that the endstop plates a spring tension parts are mounted on the outside of the the main mounting plate.

Mechanically all ready to go. With get the speed controller and battery fitted up on the weekend. Will just use a stock 6 Fet ebike controller for starters. Should be interesting.
Rock on! Amazing. Using a controller I have 3 of is exciting.
 
Of all the parts in the kit, what would you think would get the most wear & tear? The spring? Though I'm thinking a bit of rebending of it would allow it to maintain its function again. That way we could order extra of those. I'm talking parts you typically wouldn't find in a hardware store.
 
There really isn't any wearing parts as far as the mechanical components are concerned. Mechanically, the drive is bullet proof. Motor and electronics would be the failure points. Thant being said, I have never had a motor or speed controller fail with any of my testing to date. That's not to say it couldn't happen especially if people start to push the envelope with regards to power output. I always kept the power level no greater then 500W which no doubt is the main reason for the zero failure rate of speed controllers and motors.
 

Attachments

  • 20170319_203341.jpg
    20170319_203341.jpg
    98.9 KB · Views: 4,873
  • 20170319_204129.jpg
    20170319_204129.jpg
    76.8 KB · Views: 4,873
Also have now tested the the skateboard hub motor. Runing it on 7S 18650 and as expected, it is slow. Max speed was 18kph. That being said, it was quite smooth and queiter then an RC motor even using a standard ebike controller.

I will volt it up to 10S 36V and see how it goes. Should be be good for 25kph on 36V whick is not a bad speed for a folder. A 48V setup should see around 35 kph. I have a Vesc X on its way so I will try this motor out on 48V in FOC mode. It also has hall sensors however the startup for dead stop is fine so not sure if I will bother with the added complication.
 

Attachments

  • 20170305_132247.jpg
    20170305_132247.jpg
    121.8 KB · Views: 4,874
  • 20170305_125233.jpg
    20170305_125233.jpg
    87.2 KB · Views: 4,874
The case is CNC machined so i am sure the same drawing file could be used on a 3D printer. However, no heat sinking then. Really for the amount of work in this case, I think $38 USD is very reasonable. The Vesc is a quality item and a quality item I think can justify the cost of a quality enclosure. Ends up around the same price as Vesc X. However the Vesc X isn't water proof. To be honest, I dont think those CNC cases will not be available long term for that price. I cant see the seller making much profit out of them. So I am quite pleased to have picked one up.
 
I have finished the 3D printed design of the VESC-X, high/off/low switch, and battery bracket parts.

The bracket mounts to bottle boss holes, and allows the use of any size milwaukee M18 batteries, from 2AH to 9AH. Batteries just slide in and lock just like a normal power tool, and are no wider than a standard water bottle.

The VESC-X enclosure, was tricky. I am mounting to an aero frame (Felt AR), with has very high aspect ratio tubes, and again used the bottle bolt holes for mounting. I was able to package the PCB and wire in such a way as to tuck under the bottle cage, along the downtube, and be slimmer than a normal bottle. The fasteners are hidden and mount from the back side. To eliminate the need for a finned heat sink I replaced the thermal pad with a much higher performance version (fujipoly 13wmk), and left the heat sink completely exposed to passing ambient air.

Finally I printed an enclosure for the three position switch, tucked under the shifter assembly. It is positioned for naturally use when my hands are on the hoods, or in the drops. Currently set up for 200W low, and 400W high.

It took alot of CAD work and 3D printed prototypes to get the parts to fit well and look minimalist on my frame. A regular round tube bike would have been alot easier!

All in all the system is just perfect for my needs- great performance, quiet (FOC mode is a must!), extremely light weight, and when the motor is off you can't tell its not a normal bike.

enclosure4.jpgenclosure.JPGenclosure2.JPGenclosure3.jpgenclosure5.jpgenclosure6.jpg
 
Great work mhaskell. I know how time consuming it can be to make it as minimalistic as possible.

The CNC case that i got for my standard Vesc would have fitted nicely where your VescX is mounted but your solution is sweet also.

I really like the 2 position switch setup too. its good that you can reach it from the hoods or the drops. I think i would prefer a toggle switch too rather then a push button as it can be tiresome keeping the button pushed.

Overall a fantastic build and one very unique ebike.
 
Hello,

I am completely new to DIY Ebikes, but, I have a few RC Helicopter motors and ESC's laying around and was curious about retasking them for converting my bike into an Ebike. This seems like an excellent solution. The two motors I have are similar in size (Although using can size for motors is completely illogical in my opinion, stator size is a much better measurement of motors). However, they are 500 and 550 KV. With that being said, You suggest a 200-250kv motor on 6s lipo setup. Could a 550 or 500kv motor work all the same on 3s or 4s? Motor revolutions would be the same.

These motors are rated for 4000+ watts continuous. I'm pretty sure 250 watts at 3s will be no issue and is only 22 amps of current. I have a 120 amp ESC so even doubling that current draw is very low for the ESC.

Finally, I am not sure I understand how to tell what type of socket my bike uses. What is the easiest way to tell if this will work with my bicycle?

Thank you!

Excellent work! Very elegant and well thought out design!
 
I have never tested the drive using 3S or 4S on a higher kv motor but if your speed controller can handle the voltage and current, it should work.

I have typically used ebike controllers which only go as low as 24V so the motor selection has been centered around this. Also ebike controllers have current limiting capabilities which is an important consideration on any ebike build.

If you are planning on using an RC style speed controller, you wont have current limiting so the motor will pull what ever it is loaded too which will be a lot more then 250W and probably up around the 1000W mark before the motor starts to slip on the tire. This is where you might run into to some high amp issues with your ESC, especially under load at low RPM. Remember your phase amps are going to much higher then you battery amps at low RPM.

Best option is to look at using a Vesc with the motors that you have. They have true current limiting on both battery input and phase outputs.

With regards to working out if it will fit your bike, please refer to the manual on sales thread https://endless-sphere.com/forums/viewtopic.php?f=31&t=86961
 
Excellent,

Thank you! I'll assess if this will work on my bike. Seems like a fun project!
 
silviasol said:
mhaskell said:
I have finished the 3D printed design of the VESC-X, high/off/low switch,{/quote]

High/low switch. So there is a setting for assist now? Or what does that switch do? Is the cruise control button function working now?


The switched is hooked up to the ADC, 3.3v and ground pin on the VESC. low switch position uses a voltage divider to supply ~1.5v to the ADC pin for half throttle, high switch position sends the nominal 3.3v to the ADC pin.

I have it set now that high is ~400w, and low is ~200w. 200w is enough for 23mph on the flats, and I use high for climbing mostly.

The zipp 404's really sound good at anything over 20mph! :)
 
Back
Top