Yamaha FZR-8000 Genesis

Anders

100 mW
Joined
Mar 29, 2019
Messages
46
MPe8hEn.jpg


Goals/Background
I used to have a 72v 3000W Chinese bike (when I lived in China) with a 1.5kWh battery and a 200W charger, and while I did enjoy the bike it had some serious issues that I hope to avoid with this build. Main problems were, short range, slow charge speed and questionable legality.
So the goals for this project is to make a slightly faster roadlegal bike with a respectable range and fairly rapid charging capacity.

Specs:
Battery: Traction battery: LG M36 18650 26s30p for a total of ~10kWh, Accessory battery: 3s10p LG M36 18650
Motor: QS 273 8000W 17"
Tires: Pirelli MT-60 front and rear
Controller: Kelly KLS96601-8080H
Charger: Elcon TC 3.3KW charger
BMS: Orion 2 48 (was planning on getting a Batrium but they took forever to release their new version)


WIP pics
1m3MxJ3.jpg

This is what I started with, a 1987 Yamaha FZR-1000 Genesis rolling frame. Wouldn't reccomend this at all, I have probably already reached the price of a good condition running bike in just motorcycle parts for this build.

JHPXCCQ.jpg

One out of 8 packs of 18650 cells, going to start welding these together soon.

bBa9ve9.jpg

The charger.

NGBpCYk.jpg

Type-2 EV charging port and cable.

YSfwXxm.jpg

Motor from QS.
 
t1bOASK.jpg

Started working on the traction pack.
Cicw4yU.png

Design of the traction pack, ignore the missing cells. Idea is to use the aluminium sheet in between the cells to transfer heat to a watercooling block mounted on each end.

RKIpZaf.jpg

New brakediscs , I got a pair with the rolling frame but they were far too worn down to use. Not a huge fan of the design on these, but they were the closest to original I could find.
 
Interesting project. Any updates?
 
Been mostly busy with work recently so progress has been slow, but I've managed to mill out some vice style motor axle mounts and started modifying the swingarm to fit them.

The mounts are milled from blocks of aluminium and have a steel insert as the contact patch between the aluminium and the hubmotors axle.

H7MeZ7X.jpg

One finished and one half finished motor mount.
2KgdaBu.jpg

Motor mount halfway inserted into the swingarm.
QkEGAQI.jpg

Cutouts in the swingarm.

I've also decided to remove the direct watercooling of the tractionpack due to safety concerns with the aluminum heat transfer plates and instead go with a closed loop aircooling setup.

2prQJU8.png

CAD model of the traction pack with the Kelly controller on top

hQQlvhy.png

Section view showing airflow through the closed loop air cooler. The air will be forced through 2x 40x40x200 windtunnel style heatsinks mated to 2x watercooling blocks each. That way the air inside the battery can be cooled using a watercooling loop just like before without anything conductive getting anywhere near the cells.
Cooling performance is likely not going to be as good as before but I don't have to worry about the plates eating their way through the cell insulation and causing a short.

WOkp5Zx.png

Detail of the new water-air heatexchanger battery cooler module.

LzHg2v5.jpg

Also finished the 12v accessory pack, enclosure made from CNC lasercut 3mm acrylic. 3s10p cells with a cheapo chinese BMS module.
 
This new design is very nice, I'm pretty sure it will work fine if you use the thermal mass of the battery (ie, cooling it or warming it when it's charging). That will be one hell of a battery. I don't see the BMS in your pictures, do you plan to put it in an external box?

I just have one remark: the controller plate is bolted directly on top of your case. So, aside from the fact that it will transmit its heat to the battery pack through the plexi (though plexi is not a bad heat insulator), you probably should consider using spacers so that it's not direclty in contact. That will be important for your battery cooling system efficiency, but more importantly for cooling your controller itself, because that's the only surface where heat exchange can occur. And believe me, with that kind of power your controller will heat quite a bit.

So either just use spacers to allow air flow, or add watercooling sinks, or even better, both, so that its actively cooled and not in contact with your pack. In which case, you might want to build two separate cooling circuits, because otherwise your controller will heat the battery pack (I've built a watercooling system for my controller, the water gets around 50-60 degree quite quickly). Maybe also add a Peltier module so you can actively adjust the battery case temp, without relying on external temperature conditions.
 
The plan is to mount the controllers on spacers, and potentionally upgrade to a separate watercooling system if needed. I've ordered a fairly chunky aircooled peltier watercooling assembly that I'll use for the battery, if that works I have a spare PC water-cooler I can use for the controller.

9CGtvzG.jpg

Finally got the hubmotor mounted.
kCxS6nb.jpg

Detail of the mounts.

5EIIGmo.jpg

Part of the traction pack with a busbar with nickel strips silver soldered to it, but it seems the nickel doesn't like spotwelding after the hard soldering process so the plan is to switch over to using fusewire for the cell to busbar connection, should be much easier to solder and should give some extra degree of safety.
 
Post a pics when you have time to take some :thumb:
 
That seems like a solid design now.

I guess it's time to get your hands dirty and actually cut the acrylic now!

I would just suggest you to find a way to reinforce the attachment points of the pack. Plexiglass is brittle and it can probably fail if you encounter some bad roads. You dont want your whole pack to crash on the floor, that would suck.

So far I only see two attachment points on each side: the long plate with three holes at the bottom, and the little one with one hole at the back.
Don't be afraid of over engineering those brackets, your pack is likely the most expensive part of your build :)

Also, one other little suggestion: don't cut the acrylic straight, try to do some box joints, just like you often see for the wooden box projects made with laser cutters. It will give you more surface for the glue, so it should hold much better, and give a more reliable bond. Also, it will make the alignment a piece of cake, which will help a lot to get everything nice and square.

Can't wait to see that thing, it will be insane!
 
8000W is almost 11hp. Why start with a bike that had 145hp? I know it's a cool bike, but that's like putting a golf cart motor in a Formula One car.
 
Chalo said:
8000W is almost 11hp. Why start with a bike that had 145hp? I know it's a cool bike, but that's like putting a golf cart motor in a Formula One car.

You obviously don't get how power ratings on those motors work. To be fair, no one really does :lol:
They can deliver more than 30KW reliably.

On a friend's bike we got it delivering 42KW with the Mobipus.
Theoretically, Ander's build could go around 55KW peak with that controller.
The only issue is that the motor is likely to saturate earlier than that. so I guess he should expect something like 30-40KW max, if lucky. Which is far from bad, especially considering the instant torque.

Just have a look at the kind of performance this guy got with a similar motor:
https://endless-sphere.com/forums/viewtopic.php?f=12&t=82155
 
Awesome, you can squeeze a "reliable" 40hp from a motor rated for 10.7hp... but you put it in a 145hp (stock) chassis?

"Rad".
 
Chalo said:
Awesome, you can squeeze a "reliable" 40hp from a motor rated for 10.7hp... but you put it in a 145hp (stock) chassis?

"Rad".

I really don't understand why you're spending so much time bitching on every single project.
Isn't the point of this forum to do Electric conversions or did I miss something?
Not mentionning the fact that you could just have a quick look at the list of motorbikes a Zero can take down with a "ridiculous" 60Hp power rating.
Anyway, let's not feed the troll.
 
If you compare torque it makes more sense, I'm putting a motor with ~280nm of torque in a bike that originally had a maximum of 262nm in first gear(102nm on the motor and that's at 8500rpm).

And the choice of bike makes sense regardless of power to me, the frame weighs 12kg so it's plenty light and it being a literbike it's quite roomy for batteries. And perhaps most importantly I think it looks good.

Obviously won't be as fast as it was with the original motor, but if I just wanted to go fast I could have gotten a used R1 for the price of the battery cells alone.
The goal is for it to be as fun as the original at legal speeds.
 
Anders said:
If you compare torque it makes more sense, I'm putting a motor with ~280nm of torque in a bike that originally had a maximum of 262nm in first gear(102nm on the motor and that's at 8500rpm).

That's a decent point.

The all-up weight is probably more like a literbike than something in its HP class, too.

How much did the rolling chassis weigh?
 
Never put it on a scale when I got it so not sure, but the downtubes and subframe are steel so I wouldn't be surpised if it was still close to 100kg with everything except the drivetrain and fuel.
 
lD2Ydif.jpg

Battery box on its own. Changed the design one more time , so now it supports up to three vertically standing battery 10s30p blocks.
jqVKPij.jpg

Double decked lid for the box, controller and disconnect switch (visible) will sit on top as before and the Orion BMS 2 unit and the main contactor inside on the middle layer.
ZGdK0MJ.jpg

Box inside the frame without the side fairing panel. One finished 6s30p pack and one half finished 10s30p block visible in front.
nzhTIIP.jpg

View with the fairing panel on, showing how little of the box will be visible from outside. Since last update the front brakelines have also been fitted.
1yTBbcl.jpg

Turns out the stupidly short original QS phase wires are a few cm too short, so I'll have to extend them.

qXs6sKL.jpg

BMS arrived, with about 12 thousand wires, so have to spend some time figuring this out.
 
Things start getting awesome, kudos on your box build!
If it's not already too late, maybe try to leave some transparent areas so you can check for humidity or other things without having to open the box later.

As for the phase wires being too short, this could be a good thing for you actually. What I suggest you to do is to cut them as short as you can, just so they reach some convenient spot on your swingarm. Then run 3 beefier wires from the controller to this point and attach them together with some crimped/soldered connectors +screws (this connection needs to be the best possible so use pure copper connectors and tighten the screws hard).

This way you can reduce a bit the cables resistance, plus it might look cool to have 3 big cables going to the motor so that's a little bonus :). Another little bonus is that it will make removing the wheel easier later in case you need to (no need to remove the tank and other stuff to get access to the controller, you just open the box, unscrew the terminals and you can free the motor).

On my bike for instance, those phase wires are the original QS ones and while they look beefy enough they still do heat quite a bit after an agressive ride. It's not uncommon for them to reach around 50 degreeC despite the fact that there is a lot of air flow going to them. That's pure loss of efficiency. I often thought about doing this mod and the only thing that prevented me so far is that I'm a bit lazy to do it since my bike works already fine. :roll:
In your case you'll have to work on those wires anyway so that's a good opportunity.

That will imply building a small box to connect the three cables but I assume you can 3D print something nice so that's no big deal. Just make sure it is properly sealed so that no water can come in.

Keep up the good work and keep us posted!
 
Some updates on the progress made last weekend.

iBjXRjz.jpg

BMS,contactor and rubber gasket mounted to the lid. Managed to break off a tab on the vertical sheets, pretty crappy design from my part not having that connect at the end.

0uDMtII.jpg

Getting close to finishing the battery blocks.

NHXQpt7.jpg

Cheap but functional sequential LED blinkers mounted, have yet to fix the broken front fairing.
 
Latest update:
8D24kEZ.jpg

Started wiring the BMS, the Orion 2 has tons of wires....

HWNlTLQ.jpg

ZfF7t8w.jpg

Got the CAN communication working, the CANadapters plug it came with had a missing pin so I had to Mcgyver the Kelly programming lead. Disregard my shitty soldering, not really set up for detailwork at the moment.
 
Back
Top