First EV Build: Brushless/Lipo Ninja 500

[Phate]

I've got a thread over at E-S, and felt like joining up over here.

I'm just about done with the "Hoard a pile of components" stage of my build, and have done a good chunk of the CAD work.

Before I go into the specifics of my build, I'll get the basic goals and why I'm doing this out of the way.

- Easy to register for road use. In NY State, as long as the donor bike has clean paperwork, I pretty much just change the fuel type to "electric" instead of "gasoline" when I go to register it.
- I like small, lightweight bikes. An entry level full fairing standard/sport tourer would be perfect.
- I want to commute on it when the weather is nice. 25 miles will get me to work and back, but I can charge the bike at work if I have to. I can build a larger battery later.
- I'd like it to be highway-capable, geared for a 75-ish mph top speed. I don't intend to take it on the highway on a regular basis, but it would be nice to not feel like I'm gonna die if I have to take it on 55+mph roads.
- Strong acceleration at low speeds. Wheelies would be hilarious, but it at least needs to be faster than a 250, at least up to 45-50mph.

I'm a mechanical engineer working in the controls industry, and lately I've felt the need for a new project that will stimulate my need to build things. Aside from computers, I mess with 3d printing, RC cars and real cars (Honda's and Subaru's).

I don't really have enough room to work on any more cars, and I already wanted to get a motorcycle, and since I got back into RC's I've been amazed by the performance you can get with cheap lithium/brushless combos.

The Donor
I managed to find a 1995 Kawasaki Ninja 500 that had thrown the chain, which wrapped around the front sprocket and punched a few holes in the crankcase.

$400 later and it was mine. Brakes are great, tires are decent, engine runs but has a hole in it, but most importantly it had clean paperwork.

It's definitely fallen over (like all bikes, hence the dent in the tank), but it's never been crashed.

The plan is to use an aftermarket set of full-cover fairings to clean up the appearance of whatever mess of battery pack/motors/controllers/wiring/etc. The fairings will also help aero, but mostly I just want it to look cool.

Example:


Battery Pack and Charging
I wanted to be able to play with the full performance capabilities of the bike a lot more than I want it to go far, so this pack is pretty small. I'm using salvaged SPIM08HP Lipo cells that were originally in a hybrid city bus. They're standard 4.2v/3.7v lipo's with an 8Ah capacity, rated for 200A continuous. I bought 88 of them anticipating a few bad cells, the plan is to run them in 21s4p (20s4p if I there are too many bad cells). At 290g each, the energy density is pretty trash but the power density is nuts.

Most 18650's weigh around 50 grams.
A high energy 18650 like an LG MJ1 has 3.5Ah capacity, and is rated to push 10A. That comes out to 3.4g/Wh and around 1.2g/W
A more balanced 18650 like an LG HG2 has 3Ah capacity but is rated to push 20A. That comes out to ~4g/Wh and .59g/W
The Bus Lipo's weigh 290g each, have 8Ah capacity, and are rated for 200A continuous. That comes out to 8.6g/Wh and .34g/W

With the 21s4p bus lipo's, I'm looking at a 60-ish lb battery pack that comes off the charger at around 2.8kWh, which is not great at all, (a comparable pack of HG2's weighs well under 30lbs), but I basically don't need to worry about current draw, since the pack will happily push 800A. The pack of HG2's will only push 200 or so amps.

I figure I'll bomb around on it with the current pack, and when I want to go farther I'll buy a bunch of 18650's to build a second pack that only needs to supply enough power to cruise at speed, and put some form of current limiting on the second pack.

I'm getting some aluminum plates cut to connect the tabs of the cells together, and the cells themselves are going to be attached together with 3M tape before they go into an enclosure (2020 aluminum T-slot frame with plastic sides).


An aliexpress 8-24s bluetooth smart BMS will be used for pack monitoring as well as balancing. The shunt in the BMS is only rated for 300A, so the controllers are going to be wired directly to the pack with a separate 750A coulomb meter being used to monitor power draw and charge state.

For charging I went with a pair of Meanwell HLG-480H-42A's in series with the voltage cranked pretty much all the way up. I'm only charging from 120v outlets at home/work, so I don't want to pull much more than 1000W from the outlet. Eventually I'd like to charge faster using 240v outlets, including J1772 at public charging stations. For fast charging, I'll probably end up just adding some larger Meanwells in parallel, with a couple relays to switch between 120v/240v charging. The bigger charger will become more important once I add the second pack.


Powertrain
Brushless madness.


Power will come from a pair of water cooled 50kv 120100 sensored outrunners. They seem to be the same motors that are sold by Alien Power Systems and EPowerhobby. 28 poles, 9 lbs each, and rated for up to 100v/350A. The original 12mm shaft will be removed, and replaced with a longer shaft running through both motors, which are going to be bolted to a chunk of aluminum rectangle tube. A 20 tooth #40 sprocket runs from the motor shaft to a 30 tooth sprocket on the countershaft, which is supported by bearing blocks. A 16 or 17 tooth #50 sprocket on the countershaft drives a 45 tooth rear sprocket from a 1990-1995 Kawasaki ZX600 with a 530 chain. The ZX600 sprocket should bolt right on to the Ninja 500's rear wheel, and allows me to run a regular #50 industrial sprocket on the countershaft.

Total gear ratio should be around 4:1, and if my math is right I should have around 260-280lbs of torque at the rear wheel, which seems like a decent place to start. I can always throw a larger rear sprocket on if I want to trade some top speed for more low speed hooliganism.



The drive unit should be fairly light, and can be manufactured almost entirely with basic tools (like "drill press" basic). I could make it way lighter if I had access to a bridgeport or something, but oh well. At the top of the motor cans, it's only a bit over a foot wide, and the countershaft section is narrow enough that I will have no issue locating the front sprocket right where the original one was for proper chain alignment and similar-to-stock chain lengths.

Controllers
Because I'm running two motors, each one gets its own controller. Because these motors have a high pole count, I needed controllers with lots of commutation speed. I went with two Kelly KLS7320S sinewave controllers. They'll kick out 300 phase amps each for 20 seconds at a time, which should be plenty of performance. I know I'll eventually want bigger controllers, but these should get things working nicely and fit in my budget well.

I will be retaining the stock Ninja 500 twist throttle, with a cable pulling on a honda throttle body. The honda TPS provides the 0-5v throttle signal for the controllers.

[Spoonman]

dual water cooled motors (so rad required, and pump, lines, expansion bottle etc..) & controllers, with transfer box, and multiple meanwell chargers, in a GPZ500 chassis...
...it's a hell of a drivetrain dude but yeah, you're gonna be tight on battery space.

Also worth thinking about is that the original motor probably produced about that much torque at the rear wheel in first! Given you're about to dump the primary stressed member, you're also going to want to allow for some cross-bracing in there if you want to be able to actually enjoy that performance, otherwise she'll be wallowing about the place with a chassis that's just as sloppy as the stock suspension on those things. She'll be a short range beast if you can get it right though!

[Phate]

dual water cooled motors (so rad required, and pump, lines, expansion bottle etc..) & controllers, with transfer box, and multiple meanwell chargers, in a GPZ500 chassis...
...it's a hell of a drivetrain dude but yeah, you're gonna be tight on battery space.

Also worth thinking about is that the original motor probably produced about that much torque at the rear wheel in first! Given you're about to dump the primary stressed member, you're also going to want to allow for some cross-bracing in there if you want to be able to actually enjoy that performance, otherwise she'll be wallowing about the place with a chassis that's just as sloppy as the stock suspension on those things. She'll be a short range beast if you can get it right though!

The motors and primary drive unit are smaller than the look. About a foot at the widest point (motor to motor), and the aluminum rec tube is only a foot long as well. Designed it to fit into basically any frame. Should be smaller than the bike's original transmission. Battery pack is heavy but I should be able to fit it on the bottom of the cradle in front of the drive unit, it's about 8in x 10in x 14in.

That said, I expect to fill most of the space available inside the frame rails, so it's absolutely gonna be a mess for sure, which is why its getting full fairings, lol. There should be a decent chunk of space freed up after I cut the bottom out of the fuel tank.

I'm going to start with the controllers turned waaaayyy down because I don't want to die. It should at least be faster than the stock 500 was, aside from that I have no idea. I haven't really seen anybody mess around with these motors before, so I'd love to get this sorted out.

The 500 has a double cradle frame, so it's a lot less of a wet noodle with the engine removed than a 250 is. I'm going to tie the original motor mounts together wherever I can, but the bike has rubber motor mounts from the factory so the engine isn't doing quite as much as it really could be in the "stressed member" department.

I'm still pulling parts off the bike, so I haven't gotten to make my final cad model for laying all the components out, but I'm not too worried about fitting everything on board. Secondary battery pack might get interesting, but worst case I grab some hard saddle bags, lol.

[Stevo]

The original engine isn't toast at all....the same thing happened to my brother's old Honda 750-4.... we cleaned up the holes real well with a dremel and carb cleaner, JB welded up the holes and it was as good as new again. You should be able to sell it or keep it to throw back into this chassis when you decide on a better chassis for your next build. I like your 2 motor design...looks like a very clean design.

[cmcnall]

Having had to register a salvage bike in NY, +1 for getting a clean title. It was a Pain.
I'll be interested to see how the parallel motors work out, I vaguely remember previous builds like this had issues with one motor always wanting to pull too much.
Very cool!

[Stevo]

Having had to register a salvage bike in NY, +1 for getting a clean title. It was a Pain.
I'll be interested to see how the parallel motors work out, I vaguely remember previous builds like this had issues with one motor always wanting to pull too much.
Very cool!

I don't see how that would be a problem since they are sharing one shaft.

[Phate]

The original engine isn't toast at all....the same thing happened to my brother's old Honda 750-4.... we cleaned up the holes real well with a dremel and carb cleaner, JB welded up the holes and it was as good as new again. You should be able to sell it or keep it to throw back into this chassis when you decide on a better chassis for your next build. I like your 2 motor design...looks like a very clean design.

When the chain bunched up around the front sprocket, it did more than just punch a hole in the case. The countershaft actually looks like it's not aligned properly anymore. I was originally going to do just that - clean it up and quicksteel over the holes to make it hold oil again.

Having had to register a salvage bike in NY, +1 for getting a clean title. It was a Pain.
I'll be interested to see how the parallel motors work out, I vaguely remember previous builds like this had issues with one motor always wanting to pull too much.
Very cool!

Thanks, when I started the project I originally wanted to use a dual sport as a donor (for stealthy trail riding, lol), but all of them had title issues. Eventually moved on to road bikes and found this. I wanted a 4 cylinder sport tourer (almost got my hands on an early-2000's Bandit) for the wider frame, but I'm glad I got this (especially for $400, lol), since it's so comfy.

I don't see how that would be a problem since they are sharing one shaft.

The two motors aren't even aware of each other, since they each have their own controller and sensors. They're sharing the same shaft, but they're both putting work into the driveline, so even if one motor is "trying harder", you're never gonna have one motor working against the other.


Rough calc's say that this thing could accelerate at around .5G, which would be hilarious.

[nedfunnell]

Welcome to the forum! I'm most interested in your throttle arrangment- mind sharing details? What Honda throttle body sends 0-5v?

[Michael Moore]

I think most modern EFI throttle bodies have a throttle position sensor (PTS) on them (as do some modern carbs that run power jets/power valves or use the extra data to vary the ignition map) often at the end of a throttle shaft.

[Phate]

Welcome to the forum! I'm most interested in your throttle arrangment- mind sharing details? What Honda throttle body sends 0-5v?

Thanks! Pretty much any of them that uses a cable should be usable, certainly the OBD1 and early OBD2 cars. A Honda TPS is just a 3 wire 0-5v sensor. I'm looking to grab a stock one from a 1990's civic or accord, because they're everywhere and cheap.

I've used potentiometers to fool Honda ECU's into thinking various 5v sensors exist, pretty much every 3 wire sensor on a car is just 5v power, ground, and signal. I'm quite sure there are a bunch of other options besides hondas, but those are what I'm used to working on, lol. I'll probably end up making a trip to a U-Pull junkyard and grab a couple throttle bodies with the harness connector attached.

I think most modern EFI throttle bodies have a throttle position sensor (PTS) on them (as do some modern carbs that run power jets/power valves or use the extra data to vary the ignition map) often at the end of a throttle shaft.

I think the carb on the Ninja 500 might actually have a TPS in it, but since I know the carbs are in good shape those are getting sold.

And that's exactly my reasoning. Instead of buying a potentially expensive pot box, or building something that might be less-than-automotive grade, I can just buy a used throttle body, and get rid of whatever I don't need with a cutoff wheel.

You've got a sturdy aluminum housing, with a flat mounting surface, a throttle arm/rotor, a shaft, and a position sensor. Cut off everything else (any idle control components, pressure sensors, the throttle blade itself, the portion of intake tube, etc).


If I was being paid to build something I'd find a better solution, but this is extremely DIY-friendly.


That said, I did just find a not-crazy expensive pot box that looks like it could be decent. Just ordered one to see if it feels like a total piece of crap or not, lol.
http://www.eastgem.net/throttle-adapter.html