Chain Reactions, Part 2

Well the previous tensioner had too many little problems, primarily that of chain riding up onto the edge of the teeth and coming off, then the tensioner springing up and jamming into the motor sprocket. :-(

So I first tried a nylon (maybe teflon, not sure) spacer with washers either side instead of the jockey wheel, and it worked a lot better, but not perfectly, because still the chain could come off and allow the tensioner to jam in the sprocket.

So I moved the tensioner to face the other way, with the spring pulling it against the rear sprocket instead, and it works much much better, but it is so long that it rides up against the actual sprocket teeth, causing a lot of noise.

The next best solution would be to make a shorter arm or move the pivot point, but the *best* solution would be to tighten the chain by moving the motor or shortening the chain. Shortening it means I would have to have the motor about 1/4" closer to the rear, as it is about a half link short of reaching with a shorter chain, or one and a half links too long a chain as it is now. You can see the sag clearly below, even with the tensioner.


Just to give an idea of the torque problems, this is the top of the motor mounting plate, which has stayed perfectly flat with all the pull of the other motor for months:

You can clearly see the front bolt hole is pulled and dimpled downward. the rear plate area under the large washers (much bigger than those I used on the other motor) is also bent from the torque pulling the motor backwards down away from this plate.

It's worst at the right rear hole, because of the large hole in the plate for the clutch lever of the old motor to pass thru.

While I was experimenting with this stuff, I also decided to try Don's suggestion of 28 teeth for the front gear, to give me back my top speed of 20MPH, but not at full throttle--at about 3/4 throttle because that is possibly (likely) where the most efficient point for this motor happens to be, based on some other similar 4-pole motor dyno data. I know that I don't get much extra oomph out of this motor when I go beyond about that throttle point, so it's likely reaching some "saturation" point, though it may not be actual core saturation.

I welded the 28 tooth to the unused side of the hub, since bolting it thru the teeth of the other sprocket was not firm and probably would cut thru the bolts, based on the amount of torque this thing puts out. This puts the 28T sprocket about 1/4", maybe a bit more, farther out (left) than the original chainline, but it is not enough to worry about derailing if the chain is under good tension.

Now back to the tensioner problem--Even though I went from 21 to 28T, the number of links needed still doesn't add up to a tense enough chain; I end up still with too much slack, and need a tensioner, so i started thinking again about which way to fix the tensioner I already started with.

Then I thought of another solution. Since the front of the motor mounting block is not against the bike frame's mounting plate anyway, I can just change the spacing on the front bolt point so that the mounting block is just a bit farther down, pivoting the motor back very slightly and shortening the chain.

I changed the spacers I had between the motor and the bike by removing that bolt (which requires removing both front batteries), then shortening the chain two links (really one link; two parts). Then I put the bolt back in, passed thru the now slightly thinner spacers (different washers), tightened down the bolt, and now the motor chain is as tense as it was when it was on the other motor, which worked fine for months.

While I was at it, I also swapped out the chain from my DayGlo Avenger, as it is the newest chain I have, and removed the damaged rear 7 speed cassette and replaced it with a non-worn 6 speed, which I think came off the Kensington bike from way back in the beginning of this project. I tried to take pics of the damage, but it is so small and reflections within the cassette mess up the focus and/or lighting, so I could not capture it.

Basically the tips of several teeth on the smaller chainrings are stripped forward, as if peeled up and bent over, by the force of the chain slipping across them under load. Not a failure I've seen before, but they weren't designed for this kind of torque and load, I think.

Hopefully the newer chain and sprocket are enough to prevent the problem again, as long as I also remember to gear back down for all startups and not hit the throttle hard no matter what gear I'm in. I may even have to change the ramping timing in the 2QD controller to slow the startup current change, to prevent drivetrain damage by accident!