Fiction drives? Why not? Where did they go?

I'm wanting a friction drive for a very simple one speed bike. The Eboost site exists but it appears the kits are no longer made. (waiting on an email response).

Why aren't they popular? I saw a version on one of the start up sites, however it is more than I think I want to spend. The boost looks infinitely simpler. Not as "slick", but I find it an appealing design.

Pointers please.


Tom

I don't think they are the best choice for every application, but I am also mystified as to why there are not a few more friction drives in the builds around here. They are easy to build (see link in my sig below). Perhaps because they are limited to about 1,000W, and experience slippage somewhere above that?...

Here's a thread that's become an index for ES friction drives:

http://endless-sphere.com/forums/viewtopic.php?f=28&t=14403

Lately I've recalled how EVTodd mentioned that in wet conditions, if a tire with a flat tread profile is used, the power applied to the wheel can be greatly increased. I think that also applies to dry conditions.

I'm still thinking about friction drive design, also. The motor would chain-drive the roller, with the freewheel being on the roller sprocket.

I'm planning to try to make a friction drive roller out of materials besides metal. (pictures from http://www.electricscooterparts.com/sprockets25chain.html )

For example, the big freewheel/sprocket assy could be put onto to a plastic or wooden roller whose diameter is the same as the freewheel mount of the sprocket. Then grip tape would be applied to the roller. The drawback is that the clicking would tend to be loud because the roller spins so fast on the tire compared to if it were spinning once with the wheel like it usually is made to do.



With the smaller freewheel/sprocket assy, it would be mounted onto a 5/16 shaft that goes through the roller. This one would be a tougher to do. The force has to transfer from a small diameter to a large one, and it seems hard to prevent the small shaft from stripping inside the big roller.

I am using a tube that has an interference-fit over two roller clutch bearings (they roll free in one direction, and lock up in the other). The shaft is a case-hardened 1/2-inch shaft.

http://endless-sphere.com/forums/viewtopic.php?f=28&t=21365&start=25#p313906

I think answer is pretty simple. They are conceptually simple and elegant, but require eternal tuning for optimal performance. And while terrific in some situations, they aren't so good in others.

I have pretty good pitch discrimination. Can build wheels successfully, tune guitar successfully, have even built guitars from pile of wood successfully. And I'm pretty obsessive compulsive on details So, I'm a tuner at heart. And maintaining a good friction drive system is a tuner hobby. If you aren't a tuner, they probably aren't so good. Even if you are a tuner… they are a pita, albeit a fun one.

I built a friction drive based on Adrian's "Commuter Booster" kit couple years ago. Love it! But, it sure isn't my "daily driver".

Maintaining the exact geometry for optimum performance is finicky. Tire pressure drops a bit? Not so good. Drive moves a couple mm out of best location? Not so good. It rains? Not so good. Tire wears? Not so good. Heading up a steep hill? Not so good.

"Not so good"… just means you need to re-tune the system.

But when I just want to ride to the store for groceries on a Saturday, I'll take a different system.

When I want to go for a "wow this is amazing, so much kick from so minimal a system… just wait a minute while I tweak this factor…" friction is tops. When I want to just jump on a bike and go for a ride, other approaches are more forgiving.

footloose said:

I think answer is pretty simple. They are conceptually simple and elegant, but require eternal tuning for optimal performance. And while terrific in some situations, they aren't so good in others.

I have pretty good pitch discrimination. Can build wheels successfully, tune guitar successfully, have even built guitars from pile of wood successfully. And I'm pretty obsessive compulsive on details So, I'm a tuner at heart. And maintaining a good friction drive system is a tuner hobby. If you aren't a tuner, they probably aren't so good. Even if you are a tuner… they are a pita, albeit a fun one.

I built a friction drive based on Adrian's "Commuter Booster" kit couple years ago. Love it! But, it sure isn't my "daily driver".

Maintaining the exact geometry for optimum performance is finicky. Tire pressure drops a bit? Not so good. Drive moves a couple mm out of best location? Not so good. It rains? Not so good. Tire wears? Not so good. Heading up a steep hill? Not so good.

"Not so good"… just means you need to re-tune the system.

But when I just want to ride to the store for groceries on a Saturday, I'll take a different system.

When I want to go for a "wow this is amazing, so much kick from so minimal a system… just wait a minute while I tweak this factor…" friction is tops. When I want to just jump on a bike and go for a ride, other approaches are more forgiving.

Click to expand...

Why do I always get sucked into these conversations about friction drive? lol

Interesting perspective about friction drive but that's not my experience at all. I have several thousand miles on mine and use it as my everyday bike. If you look at my threads you'll see that I have a sliding mount. I don't think it's quite as finicky as a pivoting mount.

One correction though. I've never thought friction drive was good in the wet. It just isn't. If you need a bad weather ebike I would consider something else.

As for all of the other issues people bring up about friction drive. Doesn't work uphill, wears tires out, can't ride on a dusty road or the motor will explode, blah blah blah. I haven't had any of those problems.

I keep thinking about switching to a different system just to try something new. My plan was to get a bafang mid drive this year but I really want to stick with 36 volts to save weight and money on my battery options. Unfortunately the 36 volt bafang doesn't look too exciting to me. In fact, I don't think there are many 36 volt ebikes that can match the performance I get from the drive I'm using. I'm talking about drive weight, top speed, hill climbing ability (yes, friction drive can work great going up hills), etc. I'm sure a bafang mid drive is more efficient but I can live with that considering the huge price difference.

I know a lot of people flat out don't believe friction drive can work so well and that's ok. I think some of those people have spent a lot of money on their bikes and don't want to believe it. I'm not selling drives and I'm not gaining anything by promoting it. I'm just sharing my experience.

You could PM Kepler or Adrian SM.
http://www.electricbike.com/friction-drive/

I copied Todds mechanism, although the drive I ended up with is larger and heavier than his because he is a machinist and can make things precisely the perfect size. I used off-the-shelf parts and modified them to show it's possible to make with a drill and hacksaw. Turns out it's a great design. The tire profile is a circle (of course), and since the drive slides in and out in a straight line, the more load you put on it, the deeper into the tread it pulls itself.

Mine is mounted to the seat-post, so I flip the seat-post clamp lever and lift the seat/drive/battery off to carry inside. Once it's clamped into place on the seat-post, it's not finicky...I drop it into the seat-tube, flip the clamp-down lever, and it just works.

I do use a square cross-section tread beach cruiser tire as Todd suggested. Traction on the roller is great. It does slip some if I run through water...at least for a few seconds until the tire tread dries off. I don't ride my bike in the rain.

The drive struggled with the steepest hills here at 24V (worked great on flat land @ 24V), but I switched to 36V and it runs the hills fine. The 63mm diameter motor has plenty of copper mass to manage the heat, and the 295-Kv works well for what I use it for. If someone wanted to run 44V, there are plenty of Kv's in the 63mm diameter motors between 150-295 to choose from.

footloose said:

I think answer is pretty simple. They are conceptually simple and elegant, but require eternal tuning for optimal performance. And while terrific in some situations, they aren't so good in others.

I have pretty good pitch discrimination. Can build wheels successfully, tune guitar successfully, have even built guitars from pile of wood successfully. And I'm pretty obsessive compulsive on details So, I'm a tuner at heart. And maintaining a good friction drive system is a tuner hobby. If you aren't a tuner, they probably aren't so good. Even if you are a tuner… they are a pita, albeit a fun one.

I built a friction drive based on Adrian's "Commuter Booster" kit couple years ago. Love it! But, it sure isn't my "daily driver".

Maintaining the exact geometry for optimum performance is finicky. Tire pressure drops a bit? Not so good. Drive moves a couple mm out of best location? Not so good. It rains? Not so good. Tire wears? Not so good. Heading up a steep hill? Not so good.

"Not so good"… just means you need to re-tune the system.

But when I just want to ride to the store for groceries on a Saturday, I'll take a different system.

When I want to go for a "wow this is amazing, so much kick from so minimal a system… just wait a minute while I tweak this factor…" friction is tops. When I want to just jump on a bike and go for a ride, other approaches are more forgiving.

Click to expand...

Was recently thinking about the belt idea where a belt stretches across the tire tread. If the belt is wider than the tire, it could solve the problem I imagine of how a belt would want to fall off to the side of the tire. The tire should form an indentation in a wide, thin belt to help keep it centered, so I'm thinking that the tire should nestle into the belt.

spinningmagnets said:

The tire profile is a circle (of course), and since the drive slides in and out in a straight line, the more load you put on it, the deeper into the tread it pulls itself.

Click to expand...

SM -- question please -- with this design, how is the roller-to-tire contact initiated? Is it set up such that the roller is always in slight contact? Or is there some event that initiates the contact, which then becomes self-reinforcing as roller digs into tire?

The Commuter Booster design pivots motor into tire based on the initial torque "kick" of motor startup, which is very cool but a little bit finicky to get geometry just right.

During this past winter I built trike to be adaptable to several different types of e-drives. This is my first e-trike, I did a brush motor friction drive because I didn't understand how brushless motors and controllers worked. From an electrical standpoint, it is a simple setup. The left brake lever operates a cable to pull the motor/roller assembly into the tire, simultaneously I push the left thumb shifter with activates a relay to turn on the motors. It has a 1.5 inch aluminum roller which works great.

I am using 4 turnigy 4s 5ah batteries arranged 8s2p. It only goes about 14mph on it's own, with a little pedal effort does 16mph. No problem with hills or slipping (I don't ride in the rain). For me the advantage was I got started on e-bikes with something I fully understood.

Now that I understand brushless motors, I will be replacing the friction drive with a mid drive for a little more efficiency and speed. I wouldn't need to replace it, but as they say "It's OK to fall in love with an idea, but you don't want to marry it because a better one always comes along"

footloose said:

spinningmagnets said:

The tire profile is a circle (of course), and since the drive slides in and out in a straight line, the more load you put on it, the deeper into the tread it pulls itself.

Click to expand...

SM -- question please -- with this design, how is the roller-to-tire contact initiated? Is it set up such that the roller is always in slight contact? Or is there some event that initiates the contact, which then becomes self-reinforcing as roller digs into tire?

The Commuter Booster design pivots motor into tire based on the initial torque "kick" of motor startup, which is very cool but a little bit finicky to get geometry just right.

Click to expand...

The roller always makes light contact with the tire. It has a one way bearing inside that allows it to freewheel when the motor isn't being used.

Yes to what Todd just said. Whether it "seems like" it should or shouldn't work well in theory...it works well.

You can pre-order a commercial friction drive, the Rubbee, but it's $1200 bux on QuickStarter!

Forum member 'Keplar' has explored an ultra-light carbon commuter with a matching light-weight hub-drive. It's still human-powered, but capable of 35 kph on 12S LiPo with good hill-climbing performance. Read the whole saga here: Kepler’s Carbon “Super Commuter”

EVTodd said:

The roller always makes light contact with the tire. It has a one way bearing inside that allows it to freewheel when the motor isn't being used.

Click to expand...

Got it. Thank you. Thought that might be the answer when I spotted the one-way bearings in SM's original bill of materials.

I've been trying to figure out how to add power to a Bike Friday Tikit 16" folder (http://www.bikefriday.com/bicycles/commuter). Every approach so far has been dead end for one reason or another. Looks like some variant of this approach might possibly work.

I'm also working on a pivoting friction drive on my folder. The "kick into drive" system seems to do its thing, I'm just waiting for some slightly more robust shunts at the moment
http://endless-sphere.com/forums/viewtopic.php?f=28&t=53365