2̶5̶k̶W̶ 5kW Electric Pitbike / Small Motorcycle

update: I got the power calculation completely wrong, the bike actualy is only 5kW. Turns out although the controller says it's 350A, in reality it only draws 90A maximum from the battery, so after some losses and such, the final power is actually around just 5kW. It still feels crazy powewrful, but now I got all new respect for actual hi-power ebikes.


Good day everyone, finally after years of asking dumb questions and lurking the forum, I got something to show for it. Behold my first proper hi power ebike build:


Video here: https://youtu.be/KjFG3U_XsnU



Thank you John in CR, amberwolf, Dauntless,"Dui, ni shuo de dui", icherouveim, flippy, bamitsram, john61ct, billvon, Voltron, Juwe, ripperton and many many others I forgot, for your unrelenting helpfulness.

Some tech specs:

Battery
10 Gen1 Nissan Leaf modules with 65% SOH

Controller
Kelly KLS-N 350A Sinewave, Bluetooth programmable with proportional regen. https://kellycontroller.com/shop/kls-mn/ (I love it, except for slow takeoff from standstill - there isn't anything better at that price anyway)

BMS
no BMS currently, I bought a bluetooth ANT ebike one, but I killed it by electrostatic discharge when I touched it accidentally. Cells stay perfectly balanced nevertheless! I will probably not get a BMS again, just use a cheap RC charger for occasional balancing (that will probably not be needed anyway since the Leaf cells are so disciplined on their own)

Motor
QS Motor 3500W rated 13" scooter hub. I run it as fast as I can drive uphill downhill all day, and it gets slightly uncomfortable to the touch, but can still keep my hand on it.

Charger
You'll love this one: it's 6 Lime scooter chargers. Each does 42V/2A, so I got 3 parallel, 2 series. Gives me 6A@84V. And it works! Those chargers are built well, they are cheap because people are getting rid of them (juicing get sless profitable)


I really love the bike, it's so powerful and capable. Also cheap and practical, will destroy anything in the city traffic, and is also not too slow for small countryside roads. And wasn't expensive to build - I probably spent less than 2000 USD for parts, including material and parts I didn't end up using or broke.

3,500w rated but you're getting 25,000? I really need to understand this.

Well, electric motor watt rating is just "recommended", you can actually run it on whatever you like as long as it doesn't overheat and the isulation holds.

More discussion about this particular motor here: https://endless-sphere.com/forums/viewtopic.php?f=30&t=99709

Cool build - Looks like fun!

I'm guessing somewhere in NZ going by the scenery?

What throttle did you use? Does the throttle feel pretty durable? The cheap scooter ones from ebay dont feel like they can take much abuse or bumps, not something you want to feel on a homemade contraption

I actually used a cheap throttle, this one for $10
https://m.aliexpress.com/item/32368229312.html?pid=808_0000_0101&spm=a2g0n.search-amp.list.32368229312&aff_trace_key=8700fa52d2b44cbc819677a64ace9be6-1567415867115-07715-bPi5BEcs&aff_platform=msite&m_page_id=5758amp-inLTyJO-q00Nmvp8cT_FVA1585708586875&browser_id=cd6067f6003e412bafbf8f3fac4ab444&is_c=
But it feels really good, better than the standard ones.

Awesome build, kudos!

Just be aware that your actual power is more around 10-11 KW than 25KW. Reason being that your Kelly controller is rated at 140A.
I know it is very tempting, but you can't just take the phase current value and multiply it by the battery voltage, because max phase current will actually happen at much lower voltages during motor startup (at least that's how I understand it).

11KW with such a small and lightweight bike is already pretty awesome!

Also, nice video! be careful of those hills though, they can really destroy your motor. You can't rely on touching the outside casing of the motor to get a feel of how hot the phases are, because the stator is not in contact with it. So basically, what I mean is that your stator can be blazing hot in just a few seconds even if the outside casing is still cold to the touch. Do you have a temperature sensor on your motor?

If you can, get your hands on a cable actuated throttle, they are really cheap but they do have a much better feeling.

Hi Dui, thanks!

about power - I would have to dyno it to get accurate number, but I feel the real power figure has to be more than 11kW. If you noticed at the beginning of the video, there is a Suzuki Gixxer 150 petrol bike, which allegedly has 10.9kW, and weighs slightly more (not by much). My ebike absolutely destroys it in everything but the top speed, it doesn't come even close, not even if you really try and redline it.
I feel the 25kW power figure is more accurate, it feels close to gas bikes in that power range (I had lots of bikes in my life, from mopeds, dirtbikes to big sportsbikes)

But if it actually is as you say, then than means the battery cells don't put out even third of their potential

It's not clear that the kelly CAN'T do 300+ battery amps for short periods. If they rate like ASI with the same limits for phase and battery this man could actually be seeing 20kw briefly. Where is the water dude getting 140 amps? The specs for his controller don't have that number anywhere.

Power aside the difference between this and the gas bike is it has very high torque in lower rpm and instant power everywhere. Gas bike may not be in the right gear or RPM range and rider must use a lot of clutch to take off hard. Knowing your power is extremely simple: multiply max battery current the controller actually draws by voltage under that load, and then figure you're getting somewhere around 70-80% of that actually transferred to the ground.

The build looks extremely clean and aesthetically pleasing in any event.

Thanks flat tire Actually measuring the battery current is a good idea, I have to see if I can get my hands on a meter that can do it, at least to get a rough idea.
Originally I planned to have a BMS that had a power meter function in it, but I broke it

You can get a generic shunt appropriate for your max current value off Amazon for $10 and hook it up to a cycle analyst which lets you input the shunt's value. Then you have a ride computer that tracks your power usage and energy consumption closely, which is very useful as a fuel gauge.

flat tire said:

Where is the water dude getting 140 amps? The specs for his controller don't have that number anywhere.

Click to expand...

Not sure if I'm "the water dude", whatever that means, but the 140 Amps specs come from googling "KLS-N 350A" and litterally looking at the first image that pops up.
The spec also matches most controllers I used in the 150A range, so I'm quite familiar with them. For example, is entirely comparable to my scooter's sabvoton controller:
-150A rated current
-350/Phase.
-72V.

This is quite standard, many, many controllers have that kind of specs. Sometimes the Kelly controllers will output more than this at startup for a very brief moment because they are usually bad at current monitoring, but that's a bit anecdotic, what is more interesting is the actual sustained power.

So yeah, power should be expected to be around the figures I mentionned. The reason you're beating a 150cc petrol bike is that you have higher torque at an almost constant rate, and probably also smaller wheels. The gas bike will have very low torque at a narrow spot, plus probably a transmission so power losses and gear shift times. It just can't compete. It's less efficient and has a much narrower range of torque/power.
My motorbike has 200+A battery and 500+ amps per phase and it is comparable to a 400cc scrambler acceleration wise.
The scooter had 150A battery and 350A per phase and was able to take out most 250cc (it was lighter so that helped)

You can measure the actual power draw easily. Two ways:
-Buy a current shunt sensor. Either analog or numeric. That's what I did on the scooter, had an analogic one. It's cheap, reliable, works fine. But the shunt is big, bulky and a bit difficult to protect well (I used heat shrink tube for that).
-Give an other chance to the ANT BMS. It will tell you the current you're drawing from the battery and also calculate the actual power draw in real time, because current is just half of the equation.
I'm very surprized you said you destroyed it, that thing is pretty tough. Mine got through more than 2 years of daily abuse and several re installations, disconnects and such, it still works.
Maybe you could open a thread so we can diagnose it and make sure it is indeed damaged beyond repair?
If I were to do another build anytime soon, I'd go for the ANT BMS again, it's really a nice piece of hardware.

You really should use a BMS, otherwise you'll have trouble monitoring your battery's discharge and will have to manually do that often, which at some point you'll get lazy of doing. That will consitute a risk.

flat tire said:

generic shunt

Click to expand...

What did you just call me?? Haha, jk. I didn't know these existed, might be interesting to try!


This all means next time I should get three times as big controller as I thought. The Nissan modules can allegedly handle 500A, if I'm running them on 140 then they are underutilized .

About the ANT BMS - I checked it pretty extensively, and even gave it to a very experienced electronics guy to check - conclusion is it's bricked. RIP

Dui said:

Not sure if I'm "the water dude", whatever that means, but the 140 Amps specs come from googling "KLS-N 350A" and litterally looking at the first image that pops up.
The spec also matches most controllers I used in the 150A range, so I'm quite familiar with them. For example, is entirely comparable to my scooter's sabvoton controller:

Click to expand...

I guess it's said not water. Yeah, you said yeah. Water is a similar word. Anyway I went to the page conveniently linked by the OP and the 350 amps is only for 20 seconds or something. Then it's only rated for 100 amps.

The user manual discusses a battery current limit settings, but the value of the numbers is never clearly stated (at first it looks like amps, and would make sense, but then they say "100%").

It does make more sense if the rating is in phase amps like you say.

Shit on Kelly for being so crazy retarded about clearly stating their specs. I hear the throttles have lag too. Does the throttle have any degree of lag?

flat tire said:

I guess it's said not water. Yeah, you said yeah. Water is a similar word.

Click to expand...

Haha, ok got it :lol:

flat tire said:

Shit on Kelly for being so crazy retarded about clearly stating their specs. I hear the throttles have lag too. Does the throttle have any degree of lag?

Click to expand...

Yeah Kelly is extremely annoying with its retarded rating system. Most of the time they are talking about phase currents instead of battery currents, I guess it sounds better marketing wise, easier to sell.
Also, they are really bad at limiting battery currents in the first place, so that's one more reason to get a current reading device, so you can make sure you are within specs regarding how much amps you draw from your battery.
In the other hand, having a poor battery current control is probably the reason why you get a more agressive kick at startup than you'd get with other controllers.

I've heard about throttle lag issues and also experienced it, but you can get rid of it after fiddling a while with the settings.
The programming interface is really a pain to deal with too, nothing is clear and most people struggle using it. But they are cheap, reliable and they do the job. :wink:

Very nice

flat tire said:

I hear the throttles have lag too. Does the throttle have any degree of lag?

Click to expand...

Default programming had a throttle lag, revs limited and I think 70% battery current limited. But that all was easy to reprogram with the bluetooth dongle.

One single thing I dislike about the controller is starts from standstill. It's kind of slow first few revs, I think there's a hardcoded soft ramp when starting from 0 RPM. It's not too bad, but you don't get a typical "getting rear ended by a train" effect.

Hey Dui, turns out you were completely right. Clamp meter only shows 90A maximum draw on the battery cables, so after accounting for some losses, the real power of my bike is actually around 5kW, so five times less than I thought.

That's pretty humbling, and gave me a new perspective and respect for builds with actual power.

Also this turns all my understanding of battery and controller pairing upside down. If my system only draws 90A, I coudl have gotten away with much much smaller battery than this and save tons more weight. Before I built a regular ebike (e-bicycle) and because I got a "30A" controller, I specced the battery to withstand 30A. But if the manufacturers use the Amp rating only for phase wires, and the actuall battery current is much lower, that means I could have saved myself a lot of weight, or, I could have bought a much bigger controller.

Set it up so you can easily swap packs in and out, and use the one suited to your planned usage that day.

Long term does not cost any more, lifespan is cycles used 1000x more than calendar aging.

And lighter not only performs better but uses less energy.

Another way is to break it down into 2-3 modular sub- packs that can be paralleled and swapped out as needed.

redmouse said:

Hey Dui, turns out you were completely right. Clamp meter only shows 90A maximum draw on the battery cables, so after accounting for some losses, the real power of my bike is actually around 5kW, so five times less than I thought.

That's pretty humbling, and gave me a new perspective and respect for builds with actual power.

Also this turns all my understanding of battery and controller pairing upside down. If my system only draws 90A, I coudl have gotten away with much much smaller battery than this and save tons more weight. Before I built a regular ebike (e-bicycle) and because I got a "30A" controller, I specced the battery to withstand 30A. But if the manufacturers use the Amp rating only for phase wires, and the actuall battery current is much lower, that means I could have saved myself a lot of weight, or, I could have bought a much bigger controller.

Click to expand...

Well If I were you I'd see that as good news!
It just means that you still have some room to go for higher power whenever you'll have a few bucks to spend on a better controller. Meanwhile, the current setup is not going to be too harsh on your motor and battery, so they should last longer and everything should be quite reliable, plus you'll max out your actual range
There's nothing wrong in having a beefier battery than needed, aside from a very marginal increase in weight (remember than most of the weight will be yourself anyways!), it's mostly advantages. Saving a few kilos won't make any real noticeable difference, you'd better keep the battery as it is and enjoy a longer lifespan, greater range and better reliability.