An Advanced Friction Drive System

After putting together a 100V hub motor setup on my Merida TFS500 Hard Tail Mountain bike, I figured it was time for a new project and an entirely new direction. I really enjoyed the high power hub motor setup but felt my good quality super light 10kg bike was wasted on this setup and really missed just riding it as a normal bike (except up hills). My 100V hub motor gear is going into a cheap dual suspension platform but that’s a story for another day.

So my main criteria were for the new system to have as little impact on the standard bike as possible. It would be used primarily as a hill assist with the bike ridden as a normal bike most of the time. Target checklist is as follows.

=> 2 Kg in weight including battery
=> No drag at all on the bike when throttled off
=> Very quick release and re fit (30 second target)
=> 6S LiPo configuration to utilize cheap low Voltage speed controllers
=> 1000 Watt max output
=> 40 kph top speed
=> Very compact configuration

From the title of the thread, this obviously is a friction drive setup. However this is a friction drive with a difference taking in all the best documented design ideas but adding a few new tricks which makes this setup a little unique.

The prototype is built and under test at the moment with the results very encouraging so far. Stay tuned. Photos and design details to come.

ive been contemplating an rc friction drive for a while now but ive been trying to figure out what the best system would be. i was looking at the turnigy 63-64 motors and similar motors to that size using an ev warrior spindle. im still contemplating this.

let us know how it goes. ill be interested in seeing the outcome.

So what's different about this friction drive? Firstly, the drive is directly off the motor casing as apposed to using a driven roller. Also the drive completely decouples from the tire when you throttle off. The motor hangs from a spring loaded pivot that holds the motor a few millimetres away from the tire. The spring is weighted in such a way so that the torque of the motor starting is enough to slightly swing the motor pivot allowing it to just knock the tire. The spinning action of the motor then causes the motor to climb the tire where an over centre action locks the drive positively against the tire. There's no slip what so ever and as soon as the throttle is released, the motor springs back completely decoupling from the tire until throttle is applied again.

Now I certainly know the over centre locking action on the drive is not new, I have just tried to utilize this system in a way that is as compact and efficient as possible.

The motor is a Hyperion 40 series 320 kv unit that was selected because of its small diameter (48mm) and the fact that it has massive bearings supporting the motor casing. The motor is very well supported on both sides a uses an extra support bearing fitted to the prop driver side of the motor. With a rating of 2000 Watts, my current limited 1000 Watt setup is well within the motor's capabilities. Speed control is via a Hobby King 100A 6S ESC modified with extra large heat sinking. The ESC is then controlled using a Thottleizer and hall effect half twist throttle.

In relation to meeting my original design criteria, things have worked out reasonably well.
Actual total weight including a 5000mah 6S LiPo => 2.3 kg
No drag when throttling off => Check
Very quick release and re fit (30 second target) => Check
6S LiPo configuration to utilize cheap low Voltage speed controllers => Check
1000 Watt max output => Around 1200 Watt peaks
40 kph top speed => 45 kph on flat ground
Very compact configuration => Check

So now its time to get some kms on the setup and see how it all holds up. So far I have run 20ah through the setup and all is holding up very well. The motor is hitting 55 deg C peaks which I think is acceptable especially considering this is purely a power assist as apposed to a continuous drive system. Below are a few pictures of drive during the build. I will post some close ups of the drive once its all finished.

Very nice set-up. Makes me want to try something similar.

My ebike based on a hardtail MTB weighs in around 28kg, and feels twice than when the electrics give out.

I bought a road bike recently just to see what a decent non-electric bike is like, and love how efficient it feels. With something as light as your set-up, I might just try bolting it on to my new road bike. Hmmmm...

- Adrian

P.S. Just noticed that you live close by.

Brilliant design, and a great addition to the knowlege base!

If I bought one of these, the only addition I can imagine is a replaceable wear sleeve over the motor case. And as long as I would be making a sleeve, perhaps one with a waist that matches the tire profile. I'm sure I would wear out several tires before any metal case wear became apparent (without using any sleeve, just the motor case), but by then it would be too late to get a smooth fit if I chose to add one.

I also really like how it easily adapts to a variety of frame styles, its mounted away from the road, and has the potential to be very stealthy (hidden inside cargo-bags/panniers). Some enthusiasts like the wail of an RC-motor, but I suspect they would not be too hard to significantly muffle this inside a thoughtfully designed housing...

I'm liking the idea of using the motor as the roller. Very compact. My only issue would be finding a motor with a small enough diameter to get good starting torque. Top speed should be very high with that setup though.

Is that esc working well for you? I have a theory that we might be able to get by using a less expensive esc with some friction drive setups since there is little to no load on the motor when you first hit the throttle. I may have to test that theory this spring on my girlfriend's bike.

I really really like the mounting system you came up with too. Great work!

Clean and cool-looking set-up! How hot does the motor run though? Doesn't the added friction cause additional heating of the motor casing?

MrBoots said:

Clean and cool-looking set-up! How hot does the motor run though? Doesn't the added friction cause additional heating of the motor casing?

Click to expand...

This isn't my setup so I probably shouldn't comment but there isn't enough "friction" with friction drive to cause that much extra heating. I've felt the roller (and tire) on my setup after a long ride and it's never had any heat buildup. I suppose if you had a poorly designed system that slipped nonstop it would be a problem.

Now one interesting thing would be the heat from the motor transferring to the tire. I know small outrunners can get pretty hot. I still don't think it would be an issue since with this setup the motor will never just sit on one place on the tire.

i like it. a very nice clean design. really creative. i like the use of the aircraft cable as the limit/stop. it's only got what it needs to make it work and nothing else.

KUDOS

rick

EVTodd said:

This isn't my setup so I probably shouldn't comment but there isn't enough "friction" with friction drive to cause that much extra heating. I've felt the roller (and tire) on my setup after a long ride and it's never had any heat buildup. I suppose if you had a poorly designed system that slipped nonstop it would be a problem.

Click to expand...

Even mine that used rollerskate wheels had a fair bit of slip but did not heat the rollers/tire much, just the radiator fan motors driving them.

Now one interesting thing would be the heat from the motor transferring to the tire. I know small outrunners can get pretty hot. I still don't think it would be an issue since with this setup the motor will never just sit on one place on the tire.

Click to expand...

Until you stop for a light or whatever.

Actually not even at a stop. I believe his roller is totally off the tire when it stops spinning.

i totally wasnt expecting it to be done that quick. very nice! 45kmh sounds pretty darn quick. i love how the motor is the roller. does the roller have any slippage?

adrian_sm said:

Very nice set-up. Makes me want to try something similar.

My ebike based on a hardtail MTB weighs in around 28kg, and feels twice than when the electrics give out.

Click to expand...

Exactly the reason why I thought I would try this approach. Its so nice to be on a nice light weight bike again. The bike handles and carries over bumps so smoothly again.

adrian_sm said:

P.S. Just noticed that you live close by.

Click to expand...

We will have to catch up for a chat an a ride some time

spinningmagnets said:

Brilliant design, and a great addition to the knowlege base!

If I bought one of these, the only addition I can imagine is a replaceable wear sleeve over the motor case. And as long as I would be making a sleeve, perhaps one with a waist that matches the tire profile. I'm sure I would wear out several tires before any metal case wear became apparent, but by then it would be too late to get a smooth fit.

I also really like how it easily adapts to a variety of frame styles, its mounted away from the road, and has the potential to be very stealthy (hidden inside cargo-bags/panniers). Some enthusiasts like the wail of an RC-motor, but I suspect they would not be too hard to significantly muffle this inside a thoughtfully designed housing...

Click to expand...

Thanks mate, I appriciate your comments. Good idea on the motor sleeve. I will have a look at this if find I am starting wear the motor case.

EVTodd said:

I'm liking the idea of using the motor as the roller. Very compact. My only issue would be finding a motor with a small enough diameter to get good starting torque. Top speed should be very high with that setup though.

Is that esc working well for you? I have a theory that we might be able to get by using a less expensive esc with some friction drive setups since there is little to no load on the motor when you first hit the throttle. I may have to test that theory this spring on my girlfriend's bike.

I really really like the mounting system you came up with too. Great work!

Click to expand...

I got lots of inspiration from your build so thanks for that. In regards to the motor diam, my motor is only 48mm. perfect for 6S 320kv. torque is fine but I am always moving before I start the motor.

The ESC is working fine so far. However, even on the softest start it ramps up very quickly. Using the throttleizer and always rolling above 5kph ensures the motor stays in sync though. In fact it hasnt gone out of sync once yet.

def215 said:

i totally wasnt expecting it to be done that quick. very nice! 45kmh sounds pretty darn quick. i love how the motor is the roller. does the roller have any slippage?

Click to expand...

The drive was nearly ready to test before I did my first post. Took me around 2 weeks to get it where it currenly is.

No slippage at all with this setup. The over centre action locks the drive very firmly into the tire.

Well done Kepler.

Looks like that motor is a great choice for your application.

Would you mind posting a few more pics of your bracket arrangement? I'm particularly interested in how you attached it to the frame.

Also, your pivot shaft looks quite beefy. How was that done?

Do you have a method of fine-tuning the pressure against the tyre?

Thanks Grindhill. The pivot is 10mm silver steel and is supported in 2 bearings. The hight of the pivot is set by a block between the box channel and the pivot bearings. I used shims to fine tune the optimum pressure then once I had this set, a permenant block was manuafactured and fitted. Here are a few more shots of the working drive complete with all electronics. Just plug in your 6S pack of choice and the throttle and its ready to roll.

That is sooooo slick man!! Unbelievable the stuff you Aussies are coming up with! Great design, a video would be nice too.

I'm interested to find out what kind of efficiency you are getting on this machine. Are you storing the batteries in a pack, or are they under the cover?

Kepler, I like your design. I have a road bike I think I will try putting it on. Very well thought out. A video of the parts in action would be nice to see.

The more I look at your drive, the more I realize how well thought out it is. I was especially pleased to hear that the completely disengaged motor would initially swing out to bite the tire from the starting torque alone. I had always assumed that I might employ a brake cable or lever of some sort.

Just from the pictures, I'm certain several people will attempt to copy it for their own use. Could you post some of the crititcal dimensions to ease the copy process?

Nice compact unit!

Some 35~ years ago I built a motor drive like yours using a dc motor with about a 1.5" O.D. drive tube, where I contact cemented on a section of bike inertube over the tube, the rpm of the motor was high so had to use a smaller dia. It was a on / off controller, the problem I had was the amount of losses from depressing the bike tire with the drive rollar to transmit the power. with no pressure disengagenent it was like running on a flat tire all the time.

Good thought with the disengagement. Have you checked for the amount of losses (running current) due to pressure deflection (rap around of tire to motor under full load?)

etard said:

That is sooooo slick man!! Unbelievable the stuff you Aussies are coming up with! Great design, a video would be nice too.

I'm interested to find out what kind of efficiency you are getting on this machine. Are you storing the batteries in a pack, or are they under the cover?

Click to expand...

Not sure of the best way to measure efficiency on this type of system. Open to suggestions though. The motor contacts the tire quite firmly as the tire is 1.3" wide slick. Also the motor case is relativily smooth. This pressure may effect the efficiency but not sure to what degree. I would like to show a video, but I need to figure out how to do it as the motor needs to be under load for the motor to engage. Lifting the back wheel and letting it free wheel just causes the drive to bounce off the tire. Will have to have a bit more of a think about that one.

In relation to batteries, they are mounted separately to the drive unit. This was always part of the design so I could fit what ever ah pack suited the ride. Typically I am using a 6S 5000mah LiPo mounted to one of the water bottle holders. I am thinking this will be the best weight to distance compromize. Keeping in mind the whole point of this system is to ride the bike without any assist and just use it as a hill assist. I have also tested my 6S 10ah headway packs. These a slightly lower voltage but performance was still more then acceptable.

Just use the brake to create a load. That's what I did when I made my video (which reminds me, I never did repost that video).

I think what most people don't understand is that with a system like this you rarely have the motor fully engaged. When I go full throttle the motor slides back into the tire and then backs off depending on the load (even at full throttle). If I'm going up a very steep hill then yes, it is fully engaged but 95% of the time it doesn't have much pressure on the tire at all.

spinningmagnets said:

The more I look at your drive, the more I realize how well thought out it is. I was especially pleased to hear that the completely disengaged motor would initially swing out to bite the tire from the starting torque alone. I had always assumed that I might employ a brake cable or lever of some sort.

Click to expand...

I thought the same and originally included a servo to drive the motor against the tire. I then observed when bench testing the amout of swing generated by the starting torque and came up with this concept. Setting the correct spring tension and distance from the tire then just became a matter of trial and error. I also found the over centre lock movement was quite critical. Too much and the drive wouldnt disengage. Too little and I would loose my positive drive.

spinningmagnets said:

Just from the pictures, I'm certain several people will attempt to copy it for their own use. Could you post some of the crititcal dimensions to ease the copy process?

Click to expand...

Happy to post up any info but as there are so many different bikes out there, I think the most useful information will be the angles needed to achieve a postive drive and clean disengage.