Friction drive outrunner setup. New DJ bike.

[EVTodd]

I'm so impressed with this design of yours that...

I want to try a build like yours, but I'm limited not able to do the bearing or motor shaft installations. (I don't own or have access to any machining equip -much less experience doing that skill set.) Would you consider doing a mount if I send you the roller and parts for some $ ? (Or, anyone else that can do this -please- pm me.)

What brand/model of bike rack is that?

Thanks!

Hi deVries,

Thanks for the kind words. Sorry I didn't get back with you sooner via PM. I've been out of town for a bit.


Something you may be interested in are these bearing supports Timma2500 talked about in a different thread:

These would be perfect for the EV Warrior rollers. They are the correct size bearings for the shaft and it would be very easy to make a top plate to hold them together. http://banebots.com/pc/MOTOR-ACC/PB-S5002-BB

As for the rack. I'm not sure what brand name it is but it's the same as the Schwinn ones Walmart sells:

http://www.walmart.com/catalog/product.do?product_id=9207799

 

[EVTodd]

Went for a shakedown ride today with my gps. It was showing a top speed of 28.5 mph (I could have gotten more but I keep running out of street ). Comparing the numbers it looks like my cheap cycle computer is reading about 1.5 mph fast. I probably don't have the tire diameter set exactly right in it.

I have to say... Anything over 25 mph on a bike is scary. I guess if I had a DH mountain bike it wouldn't seem quite so fast but I can't see myself using the top speed all that much.

I tried a couple of hills and the motor pulled great. I have one long hill that isn't all that steep I was able to go up at 25 mph. The motor was pretty warm at the top but the controller was still cool. I'll have to hurry up and make a fan. I'm sure some of the magnet problems people have are from heat. I will be hiding my setup in panniers so I'll need to do some creative cooling.

I got my diy throttle working too. I'll post some pictures of that tomorrow.

 

[oofnik]

Awesome! Glad to hear it's running so well.
What voltage did you say you're running at? My motor gets quite hot at 36 volts after a few hard miles, almost too hot to touch. A fan sure would be nice

 

[EVTodd]

Awesome! Glad to hear it's running so well.
What voltage did you say you're running at? My motor gets quite hot at 36 volts after a few hard miles, almost too hot to touch. A fan sure would be nice

I'm at 36 volts too. I can see it getting too hot to touch if you pushed it for an extended period.

I did one funny stupid thing. I thought the controller was losing sync from time to time. Turns out I had the lvc set at 36 volts. I'm surprised it wasn't cutting out all the time while accelerating.

I will get my butt in gear as far as the fans go. I think it would make a huge difference.

 

[Grinhill]

Todd, have you seen this post about heatsinking on my build thread?

http://endless-sphere.com/forums/viewtopic.php?f=28&t=9172&start=60#p162309

Also, as recommended by liveforphysics, if you sand off the anodising from the motor mounting face and use thermal transfer paste you can improve conductive heat transfer.

I think you've come up with the ultimate build for simplicity and low parts count. I think a lot of people (including myself) are fair-weather only cyclists, so the loss of friction in the rain wouldn't be a huge issue.

 

[EVTodd]

Todd, have you seen this post about heatsinking on my build thread?

http://endless-sphere.com/forums/viewtopic.php?f=28&t=9172&start=60#p162309

Also, as recommended by liveforphysics, if you sand off the anodising from the motor mounting face and use thermal transfer paste you can improve conductive heat transfer.

I think you've come up with the ultimate build for simplicity and low parts count. I think a lot of people (including myself) are fair-weather only cyclists, so the loss of friction in the rain wouldn't be a huge issue.

I did indeed see your post about adding a heat sink. I think that's a great idea. In fact, I added silver thermal compound between my motor and the mount. I may have to play around with additional metal like you used. I'll still need some kind of forced air since about half of the can of the motor will be inside my panniers. I think I can get the temps down with a fan and a plastic scoop or two forcing air into it.

 

[EVTodd]

Quick picture of the throttle I made:

From the top it looks a bit like the throttle Matt made for his PK. I never saw the underside of his but mine uses a brass gear reduction to the potentiometer.

 

[EVTodd]

I FINALLY got out for a decent ride tonight (about 10 miles). I still can't get over the amount of power this motor puts out. I tried it on a fairly steep hill (not sure what the grade is) and I'm simply amazed at how hard it pushed me. I haven't been looking at my battery voltages after my short rides but I'm sure I'm using up a LOT more power than I did with my Kollmorgen. I may need to buy a fourth bosch fatpack to make up the difference. Then again I could try to be a little easier on the throttle.

The motor didn't get all that hot this time either. Temps in my area are down quite a bit today (around 75 for the ride) but it gives me hope that it won't be that hard to keep the motor cool.

I also tore the motor down to check the magnets. So far so good. The epoxy seems to be holding (knock on wood).

 

[etard]

Todd,
Awesome work! You have inspired me to dig out all my friction drive parts and continue my rc roller drive. This has got to be one of the most efficient transfers of power available. Have you worked up any numbers concerning the total system efficiency? Also, maybe its in your thread somewhere, but how much pressure are you putting between roller and tire? One more question, can you pin the throttle without slipping the roller for pure acceleration? Thanks for showing us all that friction ain't so bad..... remember all those naysayers in my "why not friction drive" thread? I think they just got served. LOL

 

[EVTodd]

Todd,
Awesome work! You have inspired me to dig out all my friction drive parts and continue my rc roller drive. This has got to be one of the most efficient transfers of power available. Have you worked up any numbers concerning the total system efficiency? Also, maybe its in your thread somewhere, but how much pressure are you putting between roller and tire? One more question, can you pin the throttle without slipping the roller for pure acceleration? Thanks for showing us all that friction ain't so bad..... remember all those naysayers in my "why not friction drive" thread? I think they just got served. LOL

Thanks Etard. I hope you do give your roller drive a try. I think you'll be surprised.

As far as calculating the total system efficiency, I honestly wouldn't know where to start. If someone wants to point me in the right direction I can try to get some numbers for you. The only thing I try to keep track of is how many miles I ride and the voltage at the end of each ride. I haven't even really started doing that yet with this build. I'm kind of afraid to see how much more juice I'm using with this motor.

Pressure between the roller and tire... There's no real science going on here either. I just slide the rack down until the roller sits on the tire and tighten the rack. I then pick the back end up and give the cranks a spin to make sure the tire still spins freely. I then go for a quick test ride. If the roller slips just loosen the quick disconnect on the rack and slide it down a hair. It really doesn't take as much pressure as you might think. I've also found that if you mount the roller forward of the rear wheel axle you will get rid of a lot of slipping. In this position the roller is really trying to climb the tire and will get tighter as you accelerate.

Can I go full throttle without the roller slipping? I really haven't tried it yet since I'm trying to make sure my motor's magnets are going to stay glued. I'm guessing the answer would be no. With the amount of power these motors produce you could probably make it slip. Then again. You can always put more pressure on the roller. It might not be as efficient but it would work. I will say though, that it accelerates very rapidly even though I haven't pushed it all that hard yet.

There are a couple of things I think you HAVE to have with friction drive... I think a floating roller is really important since no wheel is perfectly straight. If the roller doesn't float on the roller shaft I think it could walk off the tire. On that same point, you should make sure the motor/roller mount is perfectly straight on the rack and on the tire (for the same reason). Again, the floating roller helps a bit here but you want to get it close. You should also consider some kind of struts to make sure the rack can never move from side to side. I've never had it happen but I think it's a safety issue. I'd hate to have the motor slam into the tire at 30 mph.

I'm also really liking the cruiser tire I recently put on. It has a nice wide flat tread that gives the roller a lot to bite on.

 

[spinningmagnets]

DeVries, I've always felt a friction-drive should be a viable option for enthusiasts who only need modest speeds (27-MPH/44-KPH max) that live in fairly dry climates. I just thought it hadn't been "done well" yet, so thanks for posting this, Todd ! (PS I live where its fairly dry most of the time)

For someone who wanted a little more friction, I've thought from the beginning that having two rollers (toothed-belt drive on the opposite side to the motor) was a possibility.

 

[EVTodd]

DeVries, I've always felt a friction-drive should be a viable option for enthusiasts who only need modest speeds (27-MPH/44-KPH max) that live in fairly dry climates. I just thought it hadn't been "done well" yet, so thanks for posting this ! (PS I live where its fairly dry most of the time)

For someone who wanted a little more friction, I've though from the beginning that having two rollers (toothed-belt drive on the opposite side to the motor) was an option.

I think a dual roller is a great idea but I'm not sure you really need it. If you simply have more pressure on the roller you can get rid of most, if not all, of the slipping. Of course it's at a cost (efficiency) but I imagine you would have the same losses with 2 rollers and a belt drive.

I personally don't mind it slipping a little if you use a lot of throttle from a dead stop. It saves wear and tear on the motor and controller.

 

[D-Man]

EVTodd:

From what I've seen online, stock ev warriors friction drive bikes were driven by 2 motors. (one on each side) It seems that there are 2 parts that telescope inside the roller. They look like driveshafts. Do these driveshfts rest inside the bearing and are you using them?

More ignorant questions: How did you get the motor shaft to index into the EV Warrior roller? How did you control side to side endplay on the roller so it does't create friction when turning?

 

[EVTodd]

EVTodd:

From what I've seen online, stock ev warriors friction drive bikes were driven by 2 motors. (one on each side) It seems that there are 2 parts that telescope inside the roller. They look like driveshafts. Do these driveshfts rest inside the bearing and are you using them?

More ignorant questions: How did you get the motor shaft to index into the EV Warrior roller? How did you control side to side endplay on the roller so it does't create friction when turning?

D-man,

The ev warrior drive roller has one solid hardened shaft which floats in a one way bearing that is pressed inside the roller. The motor (or motors in the case of the ev warrior) shaft slides into holes machined in the ends of the hardened shaft. The ends of the shaft also have slots machined in them so you can simply machine a hole in the side of the motor shaft and use a pin to index it inside the roller assembly. It's kind of hard to explain without pictures and I'm saying the word 'shaft' way too many times. :lol: I'll try to get some more posted soon.

I'm not sure I totally understand the side to side endplay question. But with my setup the roller shaft is supported by bearings on both sides. Side to side play of the shaft is stopped by the indexing pin on the motor side and a pin I pressed into a steel shaft on the other side. Again... I'll have to post some pictures to really explain it. It sounds a lot more complicated than it really is.

As long as you have the roller centered you don't have a problem with it hitting the sides of the mount. I just floats right in the center.

 

[rkosiorek]

the drive roller on an EV-Warrior has a one-way roller pressed into its inside surface. not 100% sure but i believe the ID of the roller clutch is 0.500" (12.7mm) the shafts of the Warrior motors are 8mm metric in diameter. near the end of each motor shaft ther is a hole with a 3/32" diameter roll pin driven in to act like a drive key. taking up the space between the 8mm motor shaft and the 12.7mm ID of the roller clutch is a 1/2" OD hardened steel pin. the pin has a 8mm ID thru hole. at each end is a 3/32" wide slot. the motor shafts fit into the pin and the 3/32 roll pins key into the slots. that pin locks the two motor shafts together. the roll pins are long enough that the drive roller is trapped between them.

some people also call roll pins by different name either split pin or spring pin.

rick

 

[etard]

I think pictures are necesary here. I have the roller and that explanation lost me. Another source for EV type rollers are the ones used by gas stand up scooters. I have two different sized (.700" and .850") rollers with a very rough finish on the metal. You can also get them with two different levels of grip. They are very small, but if you have a high KV motor or a small diameter tire, they could work. Also, BMX pegs could do the job as well.

 

[astrid]

Hey Rick can you send any to Oz?

 

[EVTodd]

Etard,

I'll get some pictures of my mount and it's parts this weekend. I know my explanation was pretty much useless without them. I've been slacking on my project a bit this last month since I've been down with an injured shoulder.

You mentioned rollers for gas scooters... Do they have one way bearings inside like the EV Warrior rollers? If so do you have a link?

 

[rkosiorek]

the problem with putting the pressure up to high it that it seriously affects the freewheeling friction of the drive when using pedal power alone.

the EV-Warrior overcame this somewhat by not having the motors rigidly mounted. the motors are clamped facing each other to a bar that is above the motors. the motors face each other and are a CW/CCW matched pair. the brush plates are differently timed and the motors do have a significantly better performance in one direction of rotation. when not under power the motor assembly hangs from the mount bar and only lightly presses against the tire.

In this position the friction roller only lightly touches against the wheel. that added to the one-way roller inside of the friction roller means there is very low resistance when pedaling. most owners stupidly mis-adjust the drive unit to provide a ton of bite into the tire in this position. then they complain that the bike is too hard to pedal, tire wears out too fast, motor bearings wore out, etc.

when you add throttle the friction roller will bite into the wheel and pull itself into the rearward drive position. the drive position is slightly over-center of rotation so the assembly locks into the drive position.

off throttle the tire will push the roller back out into the freewheel position. there are rubber bumpers both front and rear of the motor to limit travel of the assembly. in the rear position the roller bites completetly into the tire. about 1/8" at the correct tire pressure is about the amount that you need. more than that and you are just abusing the tires. most owners were also oblivious to the fact that tire pressure plays a major role in proper operation. usually loosing traction could be fixed by adjusting the tire pressure. wet roads will still have a major impact.

there is the secret of the Warrior drive mechanism. when adjusted properly in dry weather it was amazingly effective. using 750W Bosch motors (instead of the stock 450W) and a modified 24V 120A controller i could pull wheelies with 26" tires. with the rear wheel unloaded i could smoke the tire.

rick

 

[rkosiorek]

Hey Rick can you send any to Oz?

if i had any i would.

at the moment qwerty has purchased some from Industrial Liquidators in California. since they will not ship outside of North America i am acting as a go-between for his purchase. this means that qwerty will be paying for shipping twice. once to send them to me and second for me to forward them to him.

Industrial Liquidators is difficult to deal with. they discourage shipments even to Canada by using UPS ONLY. UPS likes to charge all kinds of extra fees and crap for brokerage, storage, handling, etc.

i'm still waiting to hear from someone in Canada who may have a couple of these rollers. if he does i will pass on the information.

rick

 

[John in CR]

Rick,

Thanks for that explanation of how the motor pivot works on the EV Warrior. Once I finally do the friction drive, I will imitate it, because that really is an ingenious method. I had the machine shop make me roller with a larger diameter using 2 EV rollers. Each motor has it's own shaft and one way bearing inside a common roller, so either motor can run independently or both at the same time. The assembly is just too wide and heavy with the wheelchair motors we made it for, so I need to make different shaft adapters to run a pair of Kollmorgen's or RC motors.

John

 

[rkosiorek]

the mechanism was so simplen that it was lost on most people. i never did try multiple rollers. i sort of skipped over that and went to a traction belt/friction drive. that was the best. my choice of belt was wrong though and it wore out too quickly. but while it lasted it was great.

here is the point in time when i discovered hub motors and i started experimenting in that direction

rick

 

[EVTodd]

I really like the pivoting motor. So simple yet such a great idea. I never knew that detail of the ev warrior drive. That gives me a couple of new ideas for my drive.

 

[John in CR]

... i never did try multiple rollers. i sort of skipped over that and went to a traction belt/friction drive. that was the best. my choice of belt was wrong though and it wore out too quickly. but while it lasted it was great.

Mine has just 1 roller, but it has 2 shortened EV Warrior rollers in it, one on each side, enabling use of L R or both motors to drive.

Please explain the traction belt/friction drive. Do you mean something like typical on a riding lawnmower deck, with a lever to engage the belt with the pulleys?

I'm having a hard time letting go of the absolute dependability and silence of my direct drive hubbies, especially since weight isn't a big concern in my builds. I would like to get some any paved hill climbing ability regardless of incline. Maybe a simple friction drive with a spring loaded lever to engage, and that slick biting into the tire to roll into the ideal position is just what I need. Now that I think of it, probably a non-freewheeling roller would be best for me. That way as I crest the hill and let off the throttle for the helper motor, the rolling wheel will automatically kick the helper friction drive out of engagement. No belts, no chains, no freewheel, no extra losses at all in regular drive.

That 1.3" diameter EVW roller in my parts box that I've been bitching about being too small, seems just right for the high torque, low speed needs of steep hill climbing. I don't have to give up my 50mph hubbie, and all I have to do is get the pivoting linkage right for this friction drive. Even just a pair of stock Kollmorgens, modified only for opposite rotation on one, may be ideal for hauling a big load up very steep hills at 8-10mph. I don't ride in the rain or mud, and the friction drive wouldn't be my primary drive anyway, so it may be a good option for me to try.

John

 

[rkosiorek]

I really like the pivoting motor. So simple yet such a great idea. I never knew that detail of the ev warrior drive. That gives me a couple of new ideas for my drive.

most people were too busy dismissing the idea to pay any attention to little details like that. for it's day and using mostly off the shelf parts, it really was a well engineered bike.

rick