Single swingarm (cheap, quick and dirty) friction drive

[Fietsbel]

This is my first post here, first have been reading a lot about a geared down rear wheel belt drive, got all parts for that, and than stumbled onto Adrian's friction drive concept, which is way more stealth and freewheels a lot better!... So took the motor I already had (50-65 turnigy) and the 190-200a HK SS controller, some pieces of scrap aluminum and bolted it in my MTB frame with a new smooth rear tire, with a nice flat surface in the middle, and it worked.

At first I made 2 supporting swing arms, with extra bearings, though.. due to the overconstraining by these extra bearings, they almost immediately ran warm/hot and died.. (took 600watt at that moment, while going 21kpu)
This single swingarm works like a charm, no extra bearings required.. keeping it simple is key here..

Battery is made up out of old, though matched, laptop battery cells, giving a 4s or 5s27p 47AH pack. (huge, though quite some range, and can deliver the req. amps)
Cruising speed at 12 amp~ 30kph (4s) and 38 kph with the 5s pack config, pulling 25 amp. Those 8km/u extra take quite some watts.(maybe it slips?)
This all still with the terrain front tire, better figures expected when the front road tire arrives.

At the moment I am running without friction sandpaper, just the can of the motor directly on the rear wheel. My best guess is that the tire profile allows this, as it is very nice and flat in the middle of the tire: vredestein dynamic city. I have some good cloth sanding belt, though I'm not sure of what glue will hold. (and that is available in the Netherlands)

Just wanted to share this with you, and thank Adrian for the wonderful concept I kinda stole .
Further I am looking for an a123 18650 6s10p pack... though what ebay seller can I trust? I ordered some cells at Cammy, and some cells at ohmygod, both performing quite well.. any other good sellers I am overlooking?

pictures

 

[adrian_sm]

Awesome work. Well done.

It is so great to see someone else take on the humble friction drive.

5s47Ah. Wow. That is almost a kwh of capacity you have there.

4s 12amps = 4v * 4 * 12 = 192 watts @ 30kph
5s 24amps = 4v * 5 * 24 = 480 watts @ 38kph -> that does seem a bit too high, were you pedalling less?

As for glueing the cloth sanding belt on, look for contact adhesive in the hardware store. You should be able to find something that will suit metal and fabric.

Again fantastic work, it is great to see the friction drive concept evolve as each of us incorporate our own ideas.

And welcome to ES, great first post.

- Adrian

 

[Erogo]

Hey,

regarding the sanding belt sticking to the can, I wonder if this could be less of an issue than you suspect? Or could be made so? I know folks have had experience stripping sanding belts off of their cans, which suggests I'm wrong again, but ...

I had to sand out the inside of a hole I'd drilled with a hole-saw in some plywood. My holesaw was 70mm diameter, and i drilled three holes to get a 100mm hole in the wood. Those three 70mm holes left three peninsulas of wood sticking into the 100mm hole that I wanted to sand off.

I wrapped a sanding belt around my holesaw, revved it up in the 100mm hole, and held it against the outside of the hole. Friction against the wood tightened the belt onto the holesaw, it gripped and bob was my uncle.

Now admittedly if I took the holesaw off of the side of the hole the belt lost its grip and flew across the garage, but there might be some inkle of a principle there that an actual clever person could use to advantage?

Catch you later,

E

 

[adrian_sm]

From a PM question

What are your thoughts on using only a 5ah pack? (6s a123)

Depends on two main things power, and range.

Pack must be big enough to reliably deliver the power you want.
Pack must be big enough to travel as far as you want, without reaching, while still having some spare capacity for side trips, and pack deterioration.

So calculate your peak power, and average power for the speed you want to travel, then do the maths.

- Adrian

 

[Fietsbel]

Thanks for the input, 5ah will do in my case, but mainly because its long cycle life and charge/discharge rate being lifepo4, and offcource that ist light cheap and small!
About the self-fixing sanding belt, this works because of overlap, though, in a drive roll that creates a slight bumb at the egde, seems not to be good for efficiency imho..

As I suspected slipping of the can on the tire, i glued some sanding belt to the can. Though, with no better result at 500 watt, besides enourmous tire wear.. thats going off again.. I hope. :lol: Still 37kpu at 500 watt, NO pedaling..

Further got the new front tire installed, that did not make a huge difference in drag, though it does in sound and comfort on the road..

So far for now, more to come with new smaller battery, instead of this huge... thing. See photos for the testing setup..





Thanks for watching!

 

[Erogo]

Re: bumpy overlap
Yeah, that's true. Your machining and workmanship is on a different planet from mine. I'm "pretty rough" I guess.
E

 

[adrian_sm]

As I suspected slipping of the can on the tire, i glued some sanding belt to the can. Though, with no better result at 500 watt, besides enourmous tire wear.. thats going off again.. I hope. :lol: Still 37kpu at 500 watt, NO pedaling..

Something is not right. It should not be wearing your tire at any noticable rate. Either the angle it is making contact with the tire is too steep, meaning it just slips on the tire and not climb the tire, or the maximum engagement you have is not enough for the power you are applying.

Have a look at the setup guide for my drive, it may give you some useful tips.
Step 5 of Mounting the Commuter Booster
https://sites.google.com/site/commuterbooster/assembly-instructions

Hope that helps.

- Adrian

 

[Fietsbel]

Yeah though.. what is not right? I guess its not slipping, as with or without grip sanding belt, the max speed and watts is the same, only pressure could be a bit lower with grip material.

Material was p360, not p40 as you are using.
Wear was very clear, after a 500m testrun. (before tire was brand new, now very clear rubbing signs)

So took of the sanding belt stuff, quite hard that was (bison tix contact glue)
Did some test runs: (with the bare can)
230 watt: 27kph
450 watt: 37,8 Kph
Both no pedaling at all, smooth road.

I am thinking the efficiency of the motor just goes down steep .. remember it is only a 30$ 50-65 http://www.hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=5204&aff=2791
What other explanation could there be?

at 20v, 0-load of the motor only is 1,5 amps
at 20v, 0-load of the motor spinning the wheel at full throttle is 4,5 amps... that is almost 100w loss, though still 27kph is achieved at 230 watt, that is not so bad?

 

[adrian_sm]

The power figures for those speeds actually look fairly normal. Remember it is not a linear relationship between speed and power.

Were you actually getting a lot of tire wear, or did the surface of the tire just look a bit different?

The power losses for spinning the tire seem a bit high. Depends on a few things, but engaging the motor too much with the tire will waste power. That is one of the reason I use the sanding belt stuff, the higher coefficient of friction allows me to reduce the amount of engagement with the tire. Resulting in better efficiency.

I think both will work. Bare motor can on tire will require more pressure on the tire leading to lower efficiency, and traction issues in the wet.
Using the belt sander stuff can improve the efficiency (due to less rolling resistance from less tire deflection) may increase tire wear, but should be better in the wet.

The key difference is you would need to adjust the amount the motor pushes into the tire for each setup to get the best out of it.

- Adrian

 

[Fietsbel]

Hi Adrian,

The tire really wore quite a lot at some places, the rough profile got even completely stripped at some places. Maybe my sanding belt stuff was too fine, bigger grit might help out here.. might try that some other time.

For now i'm waiting for the twist throttle (hall) that I ordered, and the a123 18650 5ah 6s battery & balance lead. Ordered at ohmygod, the cells I got from him earlier performed quite well. Cammy did not offer spot welding..

Picked up some nice saddle and frame bags, to play around with stealth storage of the batt, servo tester etc.

About those power to speed figures; i compared them to the graph on your site, considering the graph I am at the higher side in terms of watts.
I am aware that this is the bigger, more efficient motor, what can you say about comparing the small motor I am using, and the bigger one, at this < ~500w power level?

 

[adrian_sm]

The power graph is just indicative. You are pretty close to it. Don't worry too much.

I am more concerned about the tire wear. Still sounds like something is not right...

 

[adrian_sm]

You may want to try the static torque test I did, to check you have enough engagement.
http://www.endless-sphere.com/forums/viewtopic.php?f=28&t=22187&p=391322&hilit=torque#p391322
This will tell you roughly how much torque is required before the motor slips on the tire.

But based on the no load power consumption for spinning the wheel, it doesn't sound like you have too little engagement...

 

[Fietsbel]

0,28 meter --> 0,2 kg
so that gives 0.56 Nm

If i Did it correctly.. sounds like it is really low.. though, what amp draw/speed are we looking at now?

Kt (N-m/amp) = 9.5478 / Kv
So that makes 9.5478 / 270 = 0,0342
0.56/ 0,0342 = 16,37 amp.. Makes sense, that this influences the efficiency in the higher amp draw situations.

So, more pressure, or gripping stuff on the can

 

[gtadmin]

Hi guys, I have zero experience with friction drive, but to wear the tyre like you are describing sounds to me like a combination of too fine a grit for the paper and cranking the power on too fast at too low a bike speed. Another thing to consider is the tyre pressure, is it too high? Of course I could be wrong ...

 

[Fietsbel]

I guess you might be right, the grit used was quite fine, and brand new/sharp!.. (and maximum tire pressure)

Though as I was pushing back the axle through the endbell, I killed a bearing... the smaller one(s) at the back of the motor... BUMMER!

So that problem needs to be resolved first.. 120 watt 0-load is a bit high and really noisy!

Found 4 bearings for 3,70 euro including shipping.. if I manage to destroy those as wel.. I will just machine the core to fit a 16x8x5 bearing with a 160 kg load rating... that will work!

 

[hjoore]

Hey Fietsbel,

I'm building a friction bike drive too in the Netherlands! Nice to see your work! it looks good already.
I am currently using a Kepler drive type, I burned my 63-54 motor after it crashed and aparantly it ran into the tyre with the stator, heating it up beyond its limits. I also have the 50-65 motor and the SS190-200 still laying around, and ordered a new 63-74 motor that has 40% more torque even than the 63-54!! Can't wait till I get it..
My 63-54 motor kept spinning without any grip tape, so I just took some RC Epoxy (Z-poxy 5 minute) and glued a 60 grit sanding belt around it. It still seems a little fine cause it can still slip at low speeds. I saw Adrian and kepler use 40 grit for max grip.

Groetjes HJoore

 

[gtadmin]

I guess you might be right, the grit used was quite fine, and brand new/sharp!.. (and maximum tire pressure)
...
So that problem needs to be resolved first.. 120 watt 0-load is a bit high and really noisy!
...

Sorry I missed this when first posted (it was 2 in the morning ) Have you tried your set up with a tyre pressure a lot closer to it's minimum? Too high a tyre pressure will, in my mind amplify both problems, those being high localised tyre wear and high 0-load wattage. But as I said earlier, I'm no expert and perhaps someone else can chime in.

Cheers,
GT

 

[Fietsbel]

@gtadmin, you did not miss it, I edited the post and added it later, im sorry for the confusion
I will try lowering the tire pressure a bit, and increase the depth the motor pushes in the tire, to see if this helps.

The high wattage was due to a broken bearing, I hammered the axle through the endbell for a test.. the drill press wasnt strong enough... Though this is fixed now, 0-load is now around 1,5 amp. at 5s li-ion, back to normal

@Hjoore,

I actually came across your video on youtube already, though now I found your thread in this forum as well! nice! If I can help you in some way with machining, let me know.
Your drive goes faster than mine, with the bigger motor, very nice. Whats your max speed without pedaling on the flat, and at how many watt?

About the 50-65 motor, its just great, nice, light and cheap. I reccommend to run it on 5s for your gf, than its a real nice pedalassist up to 40 km/u. For +-30kph just use 4s, at only 12 amps! At this amp level you dont even need sandingbelt.

I have got a twist throttle to spare, if ur interested? Just send a PM


Further progress;

Put all electrics except the ESC in a nice frame bag, and bought some saddle bags for the new ordered battery, 6s5ah A123! For low weight, high power, and short fast runs no further away than 15km round trip. (otherwise I won't take the bike anyway)
Will post photos soon.

 

[Fietsbel]

And hereby the photos.

What I forgot: Quickly created a cree headlight, aluminum housing and delrin endcap with o-ring, waterproof. The cree r2 drop in driver supports up to 18 volts of input, so it needs another regulator to decrease the ingoing voltage of approx 20 volts to 3,5 volts. An external regulator in the frame bag takes care of that, and limits the amps to 0,7. More than enough light for normal use.

Now just waiting and waiting.... for the new battery.

 

[Fietsbel]

... and new motor bearings, looks (and feels) good!

 

[Fietsbel]

All Good things come together.. it seems.

Got some balance leads, new hall sensors, new bearings, and.. the battery finally arrived! (with some dents due to rough shipping, though... the cells should be able to handle that)

Got checked by customs, but i guess they did not really know what do do with it, no letter or anything about taxes or whatever.

Off course did some test runs, runs like a charm, motor stays cool, battery is hand warm after fast discharge in 10 minutes, seems to hold up quite good. (a lot of breaking, no pedaling.. besides start up, abuse)

got 11,5 wh/km, with a lot of stops and starts and full speed runs. What do the other friction gurus think of this figure?

To end with, some pictures!

 

[adrian_sm]

got 11,5 wh/km, with a lot of stops and starts and full speed runs. What do the other friction gurus think of this figure?

Sounds about right to me.

Where did you get the battery from? How much?

 

[Fietsbel]

Hi Adrian,

The battery I got from Ohmygod. 75 usd inc spot welding and shipping.
I am in the market for another one, but want to try some other seller, to compare, maybe cammy?

 

[Fietsbel]

So i found a cheaper alternative, that is known here and has a good reputation, a123rc. Anyone ordered some 18650's from them? Too bad it looks like they don't offer spot welding..

 

[gtadmin]

Hi Fietsbel, did you solve your tyre wear problem? Really interested to hear how

GT