An Electric Bicycle

[vtsteam]

Hockey pucks? boat trailer bed rollers? etc.

Like I mentioned earlier, freezing rubber makes it machine or drill easier.

[S. Heslop]

I glued the next tyre on and it seems to be going okay. But now a new problem; the motor is turning the wrong way! So i'll have to change the ESC settings.

In this heat just going up and down the stairs is killing me. It rained this morning too so the humidity is up.

The ammeter still isn't responding, so I think the whole thing is duff.

[S. Heslop]

Well, I figured the tyres would wear out quickly from slipping but I didn't imagine they'd do this. All the melted galled rubber made the thing bump every time it turned.

For a very short while it was actually moving under its own power, except not too fast since it was still slipping. The rubber is a fairly soft rubber.



The first test it didn't move at all (but did help a bit going up a hill), so I put on a much stouter spring. It might need a heavier spring yet. Not sure if i'd want to stick with the soft rubber for the sake of gripping or switch for a harder rubber that i'd imagine wont gall up as bad as that as it wears? Or maybe make a thicker tyre that can squash a bit to give more contact?

I'm beginning to wonder if this idea will even work at all though, but i'm not ready to give up yet.

[vtsteam]

Well it looks like you've got the motor going in the right direction, now, but usually you don't have to re-program the ESC just to change direction of rotation. Just switch any 2 of the 3 motor wires. Bullet connectors make that easy.

The melting of the tires is probably the result of slip-- and maybe a low melting point compound. Better rubber will help plus the tension should be sufficient to prevent slip. That spring doesn't look like much.

Since other roller motors seem to work on bikes, maybe investigate their rollers.

Sounds like you're close to getting this working! Which is pretty amazing considering it is only the first iteration! Check my sig....

Ammeter is essential. Can you get a universal replacement automotive ammeter? Often pretty cheap.

[S. Heslop]

Well it looks like you've got the motor going in the right direction, now, but usually you don't have to re-program the ESC just to change direction of rotation. Just switch any 2 of the 3 motor wires. Bullet connectors make that easy.

The melting of the tires is probably the result of slip-- and maybe a low melting point compound. Better rubber will help plus the tension should be sufficient to prevent slip.

Since other roller motors seem to work on bikes, maybe investigate their rollers.

Sounds like you're close to getting this working!

Ammeter is essential. Can you get a universal replacement automotive ammeter? Often pretty cheap.

Had no luck finding automotive ammeters cheap, and the thing that puts me off them is that you need to wire them in series I think. To have that ammeter at the front of the bike would be alot of cabling. I might just try ordering another ammeter of this style since they're reasonably cheap. Hopefully a 50amp fuse will be enough to 'monitor' the circuit though, and blow before anything overloads (with the motor rated at 65 amps). I'd still like to see the current draws out of curiosity though.




For the wheel I think i'll try something like this. A step and 2 washers holding a solid rubber wheel. The blue part represents threads! Except finding solid rubber like that might be difficult. But it'd be much easier to change the wheels for new ones this way. I hadn't really considered what i'd do about the original wheels wearing out...

I tried rubbing a skateboard wheel against the rim of the bike and found it got a good amount of grip, despite being very hard rubber. It's not my skateboard wheel to use though, but I think I have a set hidden somewhere. Chances are they're buried in that box of useful looking stuff. I imagine skateboard wheel rubber is probably designed to wear away... gracefully.

Also in that one video of a similar motor system, it looked like instead of using strong springs to grip the wheel it used the force of the motors turning to pull them into the wheel and grip. That said, I haven't found a single video of it actually pulling a guy under load.

[vtsteam]

I have a feeling we're going to be seeing a video of a motorized bike pulling under load, soon! 

[vtsteam]

An audible ammeter would be kind of cool -- maybe a tone that rises with amps -- like a metal detector does! Wish I was more of an electronics guy.

Or a remote pickup ammeter -- that way it could be up in front and only the pickup wire going to the back, to keep batt cables short.

[John Rudd]

An audible ammeter.....
Or a remote pickup ammeter -- that way it could be up in front and only the pickup wire going to the back, to keep batt cables short.

Could be done using an analog Hall device.....food for thought?

But why make it complicated?

[S. Heslop]

Wish I was more of an electronics guy.

Same here! I keep thinking that some day i'm going to go all out with electronics, and especially microcontrollers. Seems like there's alot to learn though.

It used to be my main interest as a kid, but I started finding it frustrating to try learn new stuff at some point. It was really hard to find information before the internet. There weren't many good non fiction books in the local libaries. Didn't help when school wouldn't let me take the electronics module and I got stuck sanding lumps of wood they cut for us (which they called Resistant Materials. Fancy words for wood'n'plastic).

[vtsteam]

An audible ammeter.....
Or a remote pickup ammeter -- that way it could be up in front and only the pickup wire going to the back, to keep batt cables short.

Could be done using an analog Hall device.....food for thought?

But why make it complicated?

John thanks -- not food for thought for me, because I can't design circuits.

The problem with running long battery wires to an ammeter in a dash panel is that I've read in a few places that ESC's are sensitive to length on those lines. If the batts and ESC are close to the ammeter, then it wouldn't be a problem. Usually the ESC and batts in a model plane are very close.

Do you have any other suggestions?

[S. Heslop]

John thanks -- not food for thought for me, because I can't design circuits.

The problem with running long battery wires to an ammeter in a dash panel is that I've read in a few places that ESC's are sensitive to length on those lines. If the batts and ESC are close to the ammeter, then it wouldn't be a problem. Usually the ESC and batts in a model plane are very close.

Do you have any other suggestions?

Myself i'm quite happy with putting in a shunt like i'm already doing. I think it's just the ammeter I got is duff. I'm gonna order skateboard wheels today at some time so I may as well order another ammeter from somewhere else.

[John Rudd]

You could just wire the shunt in to an op amp circuit to give a direct reading feeding a voltmeter and read off the value as Amps......no worries about long wires then if you need the meter within eyesight.....

[S. Heslop]

You could just wire the shunt in to an op amp circuit to give a direct reading feeding a voltmeter and read off the value as Amps......no worries about long wires then if you need the meter within eyesight.....

I appreciate the suggestion but this stuff is way beyond my understanding. It'd be a project in itself trying to learn it all!

I ordered another ammeter along with skateboard wheels to give a plain ammeter another shot.

Another thing I could do temporarily is buy some banana plug leads for my multimeter and wire it into where the ammeter would go. I left space on that panel for some multimeter jacks so I could get more accurate readings than the cheap meters would give. I'd need to read up on the maths to figure out how much the reading would drop with a 75mv shunt though.

[John Rudd]

Is there any other info with the shunt?

Basically inserting the shunt from ....Ohms law...V=IR,

75mV=I*R, so if the resistance is 75mOhm, a curent of 1 amp will flow....

Whats the range of the ammeter supplied with the shunt?

[Bluechip]

Don't forget the shunt resistance should be matched to the  meter coil resistance...

You can't just poke any shunt on any MC  meter ...

Were they bought as a pair to be used together ??

Dave

[John Rudd]

Dave,
Unless I've interpreted this wrongly, the meter is an electronic one with a led display?.....
But as you say the shunt and meter need to be matched for an mc type.....

[Bluechip]

Dave,
Unless I've interpreted this wrongly, the meter is an electronic one with a led display?.....
But as you say the shunt and meter need to be matched for an mc type.....

Oops, sorry! Should have read the whole thread ...  was looking at #106 ish ....

My excuse is I've just been stung by a soddin' wasp ... barsteward !!!

Anyway, back under my stone. After I've checked for snakes etc ... 

Dave

[vtsteam]

I believe it's a moving coil analog meter, panel type:




I'm guessing the shunt is wired in series close to the batt and ESC, and that ammeter signal wires parallel the shunt and are brought out to the meter which is located in the box, fwd on the bike? Because in earlier photos the wires in the box to the meter look to be light gauge.

[John Rudd]

Hmmmm....I obviously missed that picture when posted.........

 Ah well, never mind....
Sourcing another meter shouldnt be too difficult as long as the electrical specs are the same to match the shunt......i.e fsd of the meter and coil resistance..........

[vtsteam]

And I missed the point that the shunt could be distant from the meter when I was asking about a "remote pickup ammeter" to keep battery/ESC leads short.

 

[vtsteam]

Would a capacitor across the meter lines (if long) reduce noise?  I think that's the problem with long lines. These ESC's sense RPM via back emf in the motor to adjust timing. The input lines also see large swinging currents and back emf there from long lines can confuse the ESC.

Or I think that's the reasoning.....

Maybe it's not a problem, though since Simon has had it running already....

[John Rudd]

Well if it was me,
I'd lengthen the three leads to the motor, keeping them to a minimum( remove the existing leads from the speedo and attach the longer ones, no joints!).
Then place the ammeter shunt in series with one of the battery leads on the esc, the sense lines from the shunt go to the remote meter, and wind both thru a ferrite donut(one or two turns should suffice to reduce the effects of any noise going to the meter), but as it's an analog meter,I think the effects of any noise reaching it will be negligible .....

[S. Heslop]

Would a capacitor across the meter lines (if long) reduce noise?  I think that's the problem with long lines. These ESC's sense RPM via back emf in the motor to adjust timing. The input lines also see large swinging currents and back emf there from long lines can confuse the ESC.

Or I think that's the reasoning.....

Maybe it's not a problem, though since Simon has had it running already....

Currently it's just a temporary setup with components cable tied wherever they'll fit. I'm gonna be building a wooden box to stick the batteries, shunt, and fusebox in (still thinking about how to keep the batteries cool when inside of a box, but I need to try and waterproof this whole thing). But with longer cabling that might become a problem, but i'll worry about that when it happens I guess.

The skateboard wheels arrived today so I took a little break from taking a break and machined stuff up. Didn't have too much of a problem turning the rubber.



As you can see, its fairly low. I really mucked up my measurements, but it's not too much of a concern since I can just make a thicker washer on the bottom to move it back up. It's raining over this way though (at least it cooled down!) so i'll probably give it another break before making the new washer.

I'm thinking I might drill 2 more holes on that bar that holds the spring though, and go with two springs. I guess I want the tension to be so much that the contact roller only slips in the event of a complete stall, and the leverage of the arms really reduces the pressure of the spring where it touches the wheel.

[vtsteam]

It's been raining here, too, Simon. Can't wait to see your new rig working! 

[S. Heslop]

It's alive!




I think it might send me back in time if I can pedal up to 88mph.

Definitely needs some tidying up. I think I do too! It's just typical that the day (or few days) you don't bother to shave you end up bumping into everyone you haven't seen for a few years. People seemed pretty interested in the bike though, despite how silly it currently looks.


I doubled up the springs holding the wheel down. I think I might add a third since there was a bit of slipping, although at this point i'm a little worried about over-loading these bearings. The tyre on the idle wheel got flung off, which says to me that the wheel couldn't spin so freely, perhaps because of the load. The top bearings are 69012, 12mm inside, 24mm outside diameter, which is fairly thin. I might try boring the holes a size bigger to fit a 28mm OD bearing that can take about double the load of the current ones according to the website i'm buying bearings from.

Also with the enormous spring pressure the idea of using a break cable to release the arms from contacting the wheel seems kind of hopeless. Fortunately the motor turns fairly freely when not running, and doesn't cause any noticeable resistance.

Still, i'm quite pleased that it's actually moving now. I'm just a little concerned about how it's going to hold up in the long run. I'll need to tart it up a bit before I can take out down to the cycle path for a stress test though.