Need Help Choosing Motor, Controller, etc.

[bdiermeier]

Hi!

I am just starting on my project for my independent study at school, an electric go-kart. I've spent hours and hours researching what electric motor to use, and how to hook everything up, but it seems like every time I learn more, I find another issue.

Attached is a picture of what I have right now. It's since been stripped and cleaned up a little. Sorry for the poor picture, it's the only one I have, and the kart is at school. The rear end is a fixed axle, with about 4" useable on each side.

I am looking for something to do probably about 20mph, or more would be great. It doesn't have to go super fast, but I'd like it to be a little fun

My original plan was to run two 1000w 36v motors, one on each wheel, but I've found that that is not a good idea due to imbalance of load on the motors. I thought about converting it to a live axle, but I've heard that's no good on pavement, which is where it will be used. So what would be the best setup for this kart? It MUST be electric, it is a requirement as part of the budget (it will be used for demos).


Thank you all!

 

[Poboy kartman]

All that time and effort. ...and all you had to do was come here........help is on the way. ........

Welcome to the forum! !!!!!

 

[bdiermeier]

All that time and effort. ...and all you had to do was come here........help is on the way. ........

Welcome to the forum! !!!!!

Thanks!


based on the countless posts I've read, everyone here seems very knowledgeable!

One detail I forgot to mention is the motor/controller budget is about $500. The total budget is $1000, and chassis parts are adding up fast! (wheels, hubs, steering parts.... everything needs replacing)

 

[itsid]

well.. *taking a deep breath*
Well live axles will work very nicely on pavement if you take care of proper Ackermann geometry, not to hard to achieve.. if you want to run two motors that's a good start.

two individual motors on a dead axle, are worse than two motors on a licve axle, since they will still push both wheels, but at a different rate if bad comes to worse.. so no..
do NOT mount two motors on individual wheels until you want to go super fancy (reads: expensive or time consuming)

Now, let's calculate abit
and there you go 1000W will be most likely enough to reach your desired speed, but a bit more will help you getting there.
1200-1400 would be ideal.

my1020s are nice and cheap... if you find the 650W being cheaper than the 1000W you can those, otherwise take the 1000W version, not too much of a difference.

but in any case ask the seller to get two motors of the exact same batch (consecutive serial numbers if possible)
manufacturing tolerances are in most chinese factories.. you want the motors to be as similar as humanly possible!

Anyways, now comes the hard part.. the controller!
you have basically two options..
one controller and powering the motors in parallel
(never try to power them in series )
or using two individual controllers.
the second option can be the cheaper one, since all you need is common e-scooter tech (a controller is as cheap as 20-30 per piece) while a bigger more powerfull controller that you'd need to power both will cost roughly 100 at least.
the problem will be synchronising the motors..
can be done in different ways, easiest I can think of right now is to get one tach per motor and a few lines of code to read and sync the controllers externally.
(like with an arduino or any other microcontroller)
IMHO the best option is get ONE motor with ~1250-1500W and one controller..much less hassle to set it up properly

there are some "bicycle" motors in that range.. some more expensive than others..
just take a firm look on the internet

or the easiest way.. find one of these:
http://citygreen-group.com/content/?606.html
(maybe used )
All in one package known to work properly

If you have more question just ask

'sid

 

[bdiermeier]

Thanks for the AWESOME reply! Do you have more info on what it would take to sync the motors/the code required? That may be the way to go, since as part of this project I am supposed to learn basic programming.

I originally was hoping to go with one motor, which would be really nice. But I looked, and couldn't find any motors over 1000w, or a controller to handle it. Do you have any recommendations? I am confident I could sort out two motors, but I'd prefer not to if I can avoid it

 

[itsid]

great!

well I can give you clues, but I haven't done such synchronisation, so I don't have any code to share.

But it's basically done like this:

you need to set up three input ports,
how is a question of what sensors you use.
a hall effect throttle for example needs an analog input,
a simple (also hall effect) tach sensor will require a digital input.
you can also try to read back emf of the motors (so you do not need any extra tach signal)
that also would require an analog in...
etc.

And you need two analog outputs.

Say you have In_TachA, In_TachB and In_Throttle.

And your In_Throttle is set to 3.2V.. which is
-since most controllers ask for a 0-5V throttle input-
64%..

Now you would think you only need to set both outputs to 3.2V and you're done...
but in fact it's unfortunately not that easy, especially not if you take a corner or your weight distribution isn't perfect (it never is )

what you need to do is.. read the tach signal to estimate the motors load.

to do so you need a prior Voltage value per output,
one you do not have if your kart is standing still,
let's assume you figured out how to get a quick 0-100% value for your tach signals
it's not too hard, it's only important to keep the code short!

since the kart wasn't moving at all, both tachvalues should be zero.
no need for further calculations,
set both outputs to 3.2V, store that value individually per output channel and loop over to reread tachsignal throttle etc.

let's say the next time you come to this point your throttle is still at 3.2V (64%) and your tach tells you one motor is running at 51% the other one at 49%
Now you would think you need to increase both voltages to reach 64%.. but that's NOT it..
you set the powerdraw with a throttle, not the rpms..
So you want to level those values out.

there are more than one way to do this, but since you want the fastest possible code (most checks/settings possible) we simply ignore the slower motor.
and try to reduce the speed on the faster one (yes that way around!)

we then either check a reference table or do some calculations to see what
voltage we need to set, to reduce the speed from 51% to 49%
(this again is a percentage.. which we multiply by the known voltage )
set, store new values, exit loop... repeat.

There are some important steps you must not forget:
1) you need a fallback on top of your loop, as soon as your throttle drops to 0.1V or below, sett 0 on both outputs!
that is important.. keep a small threshold to make sure interferences cannot move your kart!
2) I cannot stress this enough, the shorter the code the better it'll work,
so keep things flat
3) it's better (safer) to reduce one motor than to sped up the second.. simple reason
a) one faulty line of code cannot make you kart go full throttle down the road with/without you in
b) a faulty motor or sensor will stop the kart and not cook your electronics

If you have a few lines of code, just show 'em and maybe we can help you debug it (or shorten it)

'sid

 

[bdiermeier]

Thanks! I think I got most of that, I've only done very basic code before, but I'm excited to try it.

With that set up could I get away with the current axle, or would I still need to switch to a live axle?

It will probably be a couple weeks before I get to that point anyways, I still need to get a solid chassis first, and right now I only have an hour a day to work on it. I figure mount the motors, throttle, etc, and save code for last.

Also- what should I do for chain? I see most sprockets run #35 chain, but all the motors I've looked at have a sprocket for smaller chain. Should I replace it?

 

[itsid]

that is for a dead axle (sorry that I haven't pointed that out)
a live axle both motors are always rtunning at the same rpm..
So the code could NOT rely on tach signals.

If you convert to a live axle you would need to read the powerdraw directly,
which is -excuse me- a big PITA!
since the controller can allow more power to flow than the motor is actually using ATM.
you cannot just get it from there but would need get some sort of simple amp-meter
to get a signal.

On a live axle, I'd say you should go with ONE controller instead of two,
that way the motors (wired in parallel!) would level each other out on their own, no code needed whatsoever.

you'd only have a problem in turns, but if you're not constantly driving in tight cicles nothing to worry about too much

I'd say get a #35 chain for a kart..
although two #25 should do, it's better to be safe than sorry (that is always true, not only for chains )

'sid

 

[bdiermeier]

Thanks! I think I'll just go live axle then, with the motors in parallel, do to time/budget/lack of experience in writing code. I may still try writing it though, and try it and some small (RC?) motors just for fun.

Would I want both motors powering the same sprocket, or two separate sprockets on the same axle? One sprocket seems easier/cheaper, plus it would make fitting both motors easier, since I would only need to run one chain.

Also, any recommendations for a 2000w, 36v controller? I think I found what motors I'll use, but I'm struggling to find a controller large enough.

 

[dcastillo]

like this
http://kellycontroller.com/kds48050e50a24v-48v-mini-brushed-controller-p-282.html

BTW, maybe I misread sid's post, but I thought he said two motors on a live axle was not the way to go... hmm, maybe I should reread it

 

[bdiermeier]

BTW, maybe I misread sid's post, but I thought he said two motors on a live axle was not the way to go... hmm, maybe I should reread it

Thanks! I'll check out that link when I'm on a real computer, not my phone.

I think he meant one controller, but with two motors. That way the motors balance each other, which makes sense to me.

Can you clarify Sid?

 

[itsid]

two motors are ALWAYS a pain...!
but two motors on a live axle (two seperate chains and sprockets)
powered by one controller (parallel connection) reduce the pain to an acceptable level IMHO.

You still want both to be perfectly identical (or at least as perfectly as possible)

I thought about it for another minute, and I think you could mechanically level the motor loads with a simple mower differential (like the peerless 100 or clone)
let each motor power one axle shaft and use the diff sprocket to power your axle..
should work I guess *shrugs*
it's bulky, it'll draw power by friction of the additional bearings etc.
but If I'm not mistaking it'll still level the motor loads just nicely

'sid

 

[bdiermeier]

Thanks! I think I'll try it with a regular live axle first, I'm trying to keep it as lightweight as possible.

I also posted another question that I don't think you saw. Should I connect both motors to one sprocket on the axle, or give them both their own sprocket?

 

[itsid]

I did and answered it in my post... just read it carefully...
Hint: second line!

'sid

 

[bdiermeier]

Oh geez! I totally missed that! Thanks!

 

[itsid]

No problem

'sid

 

[bdiermeier]

That kelly controller looks complicated... I don't even know what a contactor is. I could probably wire it up from the diagram, I'm just worried with all the extras it will add up fast.

Any other recommendations?

 

[dcastillo]

most of the lines arent needed, and I dont even use a contactor. Actually I did for a few minutes till I smelled it smoking and realized I just put 48V on my 12V contactor that cost me $100!!!!!!

Use something like this instead

http://www.thetoolboxshop.com/0-605-00-12v-or-24v-battery-isolator-switch-100a-2298.html

I have a similar one but without a key and mine can break 200A... Plus mines a push button, so I am teaching the kids, if they get into trouble, just hit the big red button

 

[dcastillo]

Attached is the simplified schematic from kelly without using a contactor, though I highly recommend at least a manually one so you can use a precharge resistor to avoid nasty sparks every time you plug in

 

[bdiermeier]

Awesome, thanks! Do you mind elaborating what a contractor or pre charge resistor is? I just went over my budget, and only have $100 left for a controller