Tuesday, 1 April 2014

Two big bike problems

I've been trying to get bike.4 ready for use. Bike.4 is a 20 inch folder, bought as an e-bike, but I retired it when the axle holding the pedals became much too loose.

Then, much later, I bought a replacement bottom bracket, fitted it, and so the pedal problem was solved. But it didn't have a back rack, which I now find vital.

And then I found a back rack that would fit it! So I put on a new controller (Infineon 12 FET 4110), and it worked very well. But it didn't need such a big powerful controller, so I got a 6-FET 3077, and put that on. But when I ran the bike, it was OK for the first several hundred yards, and then it started stuttering; the motor would go, stop, go, stop and so on, with a cycle time of about one second. I asked the folks at em3ev, which I got the controller, and they told me that the problem was that I was running it at 33 volts, and it needs 36, or it thinks "battery is low, I'll cut out".

What confused me, was that A) I'd set the battery cut out (LVC) to 25.6 volts, and B) the same things worked fine with the 12 FET 4110. But this controller is hardwired to an LVC of 32 volts minimum, unless I solder in a new resistor (which I plan to do, of course). Whereas the 12 FET 4110 isn't so hardwired.

Oh well, I thought, I'll give it more volts. So I gave it 50 volts (the 4-cell packs I use are 16.8 volts each, so I either use two or three of them). And it ran fine! Mighty fine! It cruised up the hill at the bottom of my road like it wasn't there.

And then disaster; a horrible grinding noise from the back. I pedalled it home, and found that the back wheel rotates freely in one direction (because of the freewheel) but is completely jammed the other way. I think the internal gears are messed up; I'll have to take it apart to see. If so, it's probably not easily fixable. I can only be glad it happened while I was testing the bike, and not out on some lonely fen in the middle of nowhere.

The second big bike problem was with bike.1, my usual ride. It was doing a stop-go sort of thing; not on a regular pattern like bike.4, but it was behaving like there was an intermittent connection somewhere. I pushed wires, I tugged wires, no joy. So I used the bike all day in that condition, which means that I did a *lot* more pedalling than I like. And then I noticed; the blue LED lights that are supposed to indicate fullness of battery, weren't working. And neither were the headlights. So I was pretty sure that the problem was up that end, not at the battery or motor. I decided not to try to fix it in the middle of a Cambridgeshire fen; these batteries pack a lot of punch, and if you do something wrong and cause a short circuit, you get burned wiring.

When I got home, I had a good look. It turned out that the connector that connects the headlight/keyswitch was loose.

The headlight is pretty poor; it's OK for ensuring that other road users see me, but it doesn't give much of a beam; I use my headtorch for that. Which means, I really don't need a headlight. There's a horn there too, but I prefer to use a bell, it's politer. And I don't need a keyswitch on the handlebar; I have a big old circuit breaker switch down by the batteries. So I've removed the whole assembly, and the bike is now in a permanent state of "ignition on". But of course, in storage, there's no battery attached.

It's annoying to think that if I had opened up the tape around that connector, I'd have seen the problem and fixed it immediately.

Still, it means I got plenty of exercise today.

And which I was out, I remembered bike.5! Bike.5 was originally a 20 inch wheel folder by "Downtube", with Sturmey Archer hub gears, which are completely different from the gears most bikes have. I added a front wheel motor to it, but I've only ever used it once. I'm going to get it out of the shed, and see what I can do with it, because I suspect that  bike.4 is history.

No comments:

Post a Comment