An Electric Bicycle

[S. Heslop]

Bit late but look what i found..
http://www.petrolscooter.co.uk/spare-parts/electric-scooter-hobby-parts.html

Lyn.

That's not bad. If this RC speed controller explodes I think i'd be able to replace it with one made for electric bikes on there.

[S. Heslop]

Ugly flash photo of current progress.

Cutting that aluminum bar was a chore, and i've got to do it yet once more~

It took a terrible amount of time to then turn the bar down. Alot of it was spent trying to make a tool that could break the aluminium chips since they were getting kind of dangerous (one whipped my face!), which I finally managed. Didn't take any photos of it yet though.

I'm still undecided on how to attach the wheel to the shaft. It's a light press fit as it is. I'm thinking either loctite (the kind you can heat up to release it if needs be), set screws through the thicker part (with the holes covered with the tyre), or a pin through the thinner part. I'm liking the idea of pins the most since it won't require buying anything, and i've learned from experience that grub screws are hard to find (and i'm sick to death of ordering stuff online).

Unfortunately i'm kind of getting to the point where the project doesn't seem so exciting any more. I think my favourite part of any project is the 'solving problems and learning new stuff' part, and the rest sort of feels like going through the motions. That's not to say I wont finish it, it's just i'm thinking about other stuff most of the time (right now I can't stop thinking about turret tailstocks, and how i'd build one without a mill! It's part of the banjo plan).

[vtsteam]

It will be pretty exciting to sit in that seat and strapping on that war helmet, shoving off and switching on, though! 

[S. Heslop]

It will be pretty exciting to sit in that seat and strapping on that war helmet, shoving off and switching on, though! 

Yeah that's true! I could probably have it together for a test ride in a couple of days if I get my act together. Still got to build the whole control side of things before its complete though.

As a side note, I was at the boot sale on Sunday and spotted 4 different big tubs full of laptop batteries. If there's as many next Sunday, I might buy them all. I'd only really given up on the idea of assembling a battery from old laptop cells after finding out they cost a fortune on eBay, but if I can get them cheap enough it might still be preferable to spending more on RC batteries. Plus lithium polymer is kind of frightening!

[S. Heslop]

No shortage of aluminium chips!



Here's the tool. It only really works on facing cuts, but it's ideal for hogging out the bulk of the material. It's a fairly deep groove. It might be old hat to some people, but i've been trying to make a tool that can break aluminium chips for ages with no success till now, and it certainly makes life easier.

I also found I could cut the aluminium bar with one of those little TCT blades I got at Aldi a while back, the saw it came with has little power so it tends to stall way before it would risk kicking back, so I could keep rotating the bar in the vise and cutting as deep as I could till it stalled, then rotating the bar and starting again. It's not ideal, and maybe a little bit dangerous, but preferable to spending several hours hacksawing through the thing. A powered hacksaw is still on my list of things to do at some point though.

[S. Heslop]

Got the contact rollers on the arms. All I need to do to test it is to mount a spring and hook up the electronics. Hopefully that fuse box will arrive tomorrow.



I also borrowed this brake lever from one of the old bikes rotting in the back yard. I was worried i'd have to make some sort of mechanism to do exactly what this does. I'm also glad this bike came with those little extensions on the handle bars for mounting stuff like this! The idea is that the arms can be released from the wheel like a brake caliper in reverse so I can pedal without working against the motor. Forseeable problems are the heavier motor side not wanting to lift as much as the lighter idler, away from the wheel.

[S. Heslop]

Thank god the suns setting and it's getting cooler. Today was too hot!

I went shopping for various electronics bits and made a box.



It's not very discreet but should do for testing! Fuses should be enough to 'monitor' the amperage normally, but i'd like to be able to see what the current draws are when going up hills and stuff at first. I'm not sure if the analog volt meter will be accurate enough to gauge how much power is left. I'm also not sure how well these meters will handle the vibrations from riding.

I've found that one of the pins on the servo tester had melted off, which must've been what caused that spark. Hopefully everything still works since a short like that could have wrecked the speed controller. But that short made me think twice about just taping a big mess of wires onto the handlebars for the sake of testing, and I may as well try and do it properly the first time.




I like having a table saw though. I've tried making a box with full miter joints like this in the past with just a circular saw, and it was practically impossible to make it accurate enough. It's rounded over since i'm probably going to paint it all black. The visible gaps are from the poor quality plywood!

[vtsteam]

Still following Simon and it's impressive how much you've progressed these last few days. The motor rig looks great on the bike.

Does that ammeter have a shunt in it already? Seems like it might if it is calibrated on the front in amps -- unless the shunt you showed earlier is packaged with it, Or it's spec sheet calls for an external shunt.

Those meters look like the sensitive type rather than the automotive type. I hope they will handle the vibration levels of a bike.

The cause of your short isn't explained but must have hit the servo tester hard to melt a pin. I would guess it is dead. Do you have an R/C servo to test it with as just a servo tester? I believe you were into R/C cars?

Do you know anybody locally who is into R/C who could check your wiring out? Is there an R/C club nearby?

[S. Heslop]

Still following Simon and it's impressive how much you've progressed these last few days. The motor rig looks great on the bike.

Does that ammeter have a shunt in it already? Seems like it might if it is calibrated on the front in amps -- unless the shunt you showed earlier is packaged with it, Or it's spec sheet calls for an external shunt.

Those meters look like the sensitive type rather than the automotive type. I hope they will handle the vibration levels of a bike.

The cause of your short isn't explained but must have hit the servo tester hard to melt a pin. I would guess it is dead. Do you have an R/C servo to test it with as just a servo tester? I believe you were into R/C cars?

Do you know anybody locally who is into R/C who could check your wiring out? Is there an R/C club nearby?

Yeah the ammeter came with a matching 75mv shunt.

I think my wiring should be fine. I'm no electrical engineer but i've checked the specs of all the cabling and connectors i'm using to make sure it capable of carrying the current and has appropriate insulation. As for the wires going to the right place, I've made a very professional diagram to help keep things organised.

I do have some servos but they're buried in my 'big box of useful looking stuff', which itself is currently buried.

[vtsteam]

Simon, question -- the battery on the servo tester in your circuit diagram -- does the ESC you purchased have an onboard BEC, and isn't the servo tester powered by that through the ESC connector?

Could that be the cause of your short?

If you use a separate battery for the servo tester, you may need to break the RCVR power connection from the ESC if it has an onboard BEC. Generally receiver power (~5V) is supplied through the red central wire in the ESC connector, and the central pin.

The white or orange wire is usually the control signal, and the brown or black wire is generally the ground.

[S. Heslop]

Simon, question -- the battery on the servo tester in your circuit diagram -- does the ESC you purchased have an onboard BEC, and isn't the servo tester powered by that through the ESC connector?

Could that be the cause of your short?

If you use a separate battery for the servo tester, you may need to break the RCVR power connection from the ESC if it has an onboard BEC. Generally receiver power (~5V) is supplied through the red central wire in the ESC connector, and the central pin.

The white or orange wire is usually the control signal, and the brown or black wire is generally the ground.

That could be it. I cant find anything that says it has onboard BEC, and the sheet that came with it is generic to a whole range of ESCs. But looking at that picture at the top of page 2, there is a group of 3 wires leading out of the ESC.

I'd prefer to feed the servo tester with external power so I suppose i'll break that red wire.

So far today i've been cable-tying stuff onto the half finished carry rack to have a go at testing the bike. But i'm still fussing about how to build a throttle. I'm not so sure if I want to be sitting on the thing trying to control the speed with a loose potentiometer. The pot inside of the servo tester reads 0-10kohm, so i'm gonna try using a 0-100kohm pot i've got since the throttle lever wont be able to go through the pot's full motion. Although I think I once read that potentiometers are often used instead of rheostats since they're kind of two-way. I just hope that using a higher value potentiometer won't muck up the servo tester's function.

[vtsteam]

Check out the servo tester's function first with the original pot.

Re. breaking the red (middle) wire in the 3 wire ESC lead -- you can do this non-destructively by pressing down with a paper clip wire on the square tab in the connector to release the pin and pulling the pin with attached wire out.


ps. the BEC is the battery eliminator circuit. It is meant to output 5 V to the receiver to power it in an R/C vehicle using the same batteries that power the motor (usually much higher voltage). It is often powered by a linear voltage regulator like a 7805 IC. At least in cheaper versions.

Usually the onboard microprocessor is programmed to cut the motor when battery voltage drops to a pre-set low, but to continue to supply the receiver with current through the BEC so a plane can be landed, even with the power off.

ESC's vary in function. Most can be programmed for some of their functions.

[John Rudd]

ESC's vary in function. Most can be programmed for some of their functions.

You will need to ensure that the brake facility is disabled........otherwise you might find the rear wheel will not freewheel......

[vtsteam]

True John! Though in my experience default is always brake off, out of the box. I also haven't found the brake to be very effective in real life with non-folding props. Slows them some, but it is far from a hard stop.

Still, other features may need programming after trial. Timing for instance.

Simon --bad timing is usually evidenced by a ugly sounding squeal on rapid throttle movement at low speed. Sounds mechanical like shot bearings, but isn't. Tipoff is that slow throttle advance (or smaller prop on a plane) doesn't cause the sound. Timing in that case usually can be altered by programming the ESC.

[S. Heslop]

I'll have to read up on programming ESCs. I didn't even know that was a thing!

It's been horrifically hot today, so i've tried stay out of the garage as much as possible.

So I did a bit more electronics stuff. Soldering fiddly connectors and crimping terminals. Hopefully nothing will come loose and short out.


I found when soldering DIN connectors (and wrecking two) for the commodore 64 that it's best to plug the plug into a socket to prevent the pins from moving when the plastic gets soft from the heat.




I've got no idea on the 'polarity' of these meters but it wont be a hassle to switch the terminals if the needles move the wrong way.



Maybe too much photos for such simple work, but I always get a little too excited about soldering. It's quite satisfying!


Edit: Haha, aint that typical. I've been soldering all day without a problem, and then I give myself a nasty burn heating up one of those creme brulee puddings!

[vtsteam]

Simon, the way it is usually handled is on a transmitter, the throttle stick is moved in a specified sequence to program in usually six to 12 variables. You listen for tones that the ESC emits in sequence, and then move the throttle stick to select your options. To do this you need to know what that code is. It usually is in the instructions that came with the ESC. Sometimes you have to do a little internet research if the instructions don't include this.

What is the brand of ESC that you bought, and where did you buy it from?

It's fairly likely that your ESC already is set up with usable default values, and you might not have to change anything. But you should have that documentation.

Your potentiometer will act as the throttle stick pot.

Incidentally, don't be surprised if on first trying your motor out for a bench test if the motor doesn't seem to respond. There is a lock-out feature that prevents energizing the motor if the throttle stick (or your pot) isn't set to zero throttle when you connect the ESC to power. You need to move the pot to zero to arm the ESC.

This fact is actually used to do the programming, too. Generally you put the ESC into programming mode by setting the stick to some specified positive position before connecting the ESC to power. The ESC responds by altering its initial musical sequence to let you know it is in programming mode.

[S. Heslop]

http://www.ebay.co.uk/itm/261313678316
This is the model that I bought. It was cheap enough that I was willing to take the risk.

http://iforce.co.nz/i/pm4llk0z.r00.jpg
This is a scan of all that came with it, in glorious Engrish.

http://www.hobbyking.com/hobbyking/store/__7341__Hobbyking_SS_Series_90_100A_ESC.html
A post on a German forum says 'it looks similar to this' which it does. But that's all I could find from googling around.

[vtsteam]

Working on it. It will take some sleuthing for this particular ESC.

First hit:

http://www.rcuniverse.com/forum/brushed-brushless-motors-speed-controls-gear-drives-123/9046787-rc-smart-80-amp-opto-esc-programming.html

[S. Heslop]

That's great, thanks!

One thing I was wondering about was how i'd deal with not running the batteries flat since I think that can ruin lipos, but it looks like you can program the esc to automatically cut off before that happens.

Also when I first heard about esc programming I was a little worried i'd have to buy some sort of accessory, but it looks like i'll be able to just do it with the regular servo tester/ throttle.

[vtsteam]

Simon, yes all ESCs cut out when the battery nears the allowable limit. They often even have two styles of cut-out -- hard and soft. Soft means the engine slows down (to let the pilot know before cutoff that the motor will be cut. The hard cutoff just shuts the motor down.

When you set the programming for the number of lipos you are also setting the cutoff voltage.

Yes you can program an ESC with just tones and the throttle. It can get confusing at first, but once you "get it" it's do-able. I've never bought a programming card, and do all ESC programming "on the beep".

One more reference that might be helpful for programming this ESC, and seems to confirm is the link you gave me to Hobbyking for the "similar" esc.

If you go to that page, toward the bottom you'll see a set of tabs. Click on the FILES tab and you'll see support files for that ESC. The top two files (numbered)  are in English and are instructions for programming that ESC. Download those and compare to the link I gave earlier.

Now as for actual settings -- Suggestions:

Set the number of cells in series you will have. Set lipo. Set no brake. You want probably soft start (usually used on helicopters -- a gradual ramp up n speed). Try auto timing. If that doesn't work (loud squeal when you crank the throttle) try soft or lower timing.

But please figure out why you had that short, if you haven't already. Was it really moving a bare lead or pin to contact something accidentally, or was it wiring hookup error. If the former, then things should be better now that wiring is soldered and insulated, etc. But if wiring was wrong, it will happen again (unless it was changed).

It looks pretty definite that your ESC has no on-board BEC. It therefore does not need the center (red) lead going to the servo tester, and absolutely requires a separate battery to drive the servo tester (assuming it still works).

It would be a good idea to test that servo tester somehow separately before trying to drive the ESC and motor from it. Or buy a new servo tester to be sure. They are the least expensive component.

[vtsteam]

Few more things, keep battery leads as short as possible. Motor leads can be extended in a pinch in preference. (ESC's don't like long battery leads.)

Wire gauge needs to be sufficient to carry the expected current for the expected distance.

Make sure ESC and batts get good airflow for cooling.

[S. Heslop]

Little bit more soldering and also dug the servos out. The tester still works fine even after the short and butchering it.

I really hope it cools down a bit tomorrow because the last few days have been hell, and I haven't really felt up to doing any serious work in the heat.


Airflow for the ESC I had considered but I was thinking about putting the batteries in a wooden box. Didn't really think too much about them heating up... Maybe i'll make a wooden box with holes in it!

[tom osselton]

So when the batteries finally shut off could it not switch to another bank of batteries? And couldn't a arduino handle all that and more?

[vtsteam]

Tom it could be handled with a battery switch, or even simpler still, if cells were the same type and capacity, paralleling them for longer duration.

Simon, I get the same way when it's too hot. Small workshop jobs start to seem too complicated to attempt.

[S. Heslop]

Well it took two seconds for the tyre on the little contact roller to stretch and split. I think I might have to glue the next one down, but even then it'll probably have a short life.

But hey it seems to work. The motor turned, but i'd forgotten to insulate the three connectors between the motor and the speed controller and got a bit of arcing there. It was after that that I noticed the tyre was wrecked.

I'm not sure if glue would even work though. Might have to machine thicker rubber rollers or something.