Full Suspension Leaf Bike 1500W Build

[GoFastGus]

Motor & Battery update:
Over the weekend, I bought a 20S4P pack from Bicycle Motorworks. Listing is here. Molicel P50B cells, non-bluetooth BMS with 100A continuous, 160A burst limits. Suffice it to say, I'm gonna watch this thing like a hawk. A waterproof bag will work for now, but eventually I'm going to want a hardshell case with a lock.
To compliment it, I ordered a somewhat custom Leaf Bike motor. It's this motor with a 10.2rpm/V winding at 60V, but it uses this single speed freewheel, this 47mm internal width rim, and an NTC-10K temperature sensor. To be frank, I'm not certain if I got the right motor temperature sensor. I suppose we'll figure it out when I eventually order a CA.

Still need to figure out the right controller. Do I order one rated to my continuous amperage, my burst amperage, or above my burst amperage?

Again, I'll keep y'all posted when this gets here. Let me know what you think!

 

[E-HP]

Motor & Battery update:
Over the weekend, I bought a 20S4P pack from Bicycle Motorworks. Listing is here. Molicel P50B cells, non-bluetooth BMS with 100A continuous, 160A burst limits.

Thanks for the update. I've been looking at that pack, but haven't gotten around to calling them about it. Based on this description, and then looking at the specs of all of the 100A JBD BMS's I could find, that also had temperature protection, I had mistakenly concluded that the BMS supported bluetooth. I guess now my question will be whether they can substitute the bluetooth BMS and what effect that would have on the pack dimensions.

• Temperature-protected JBD 100 amp continuous BMS for reliability under extreme conditions

 

[GoFastGus]

Thanks for the update. I've been looking at that pack, but haven't gotten around to calling them about it. Based on this description, and then looking at the specs of all of the 100A JBD BMS's I could find, that also had temperature protection, I had mistakenly concluded that the BMS supported bluetooth. I guess now my question will be whether they can substitute the bluetooth BMS and what effect that would have on the pack dimensions.

• Temperature-protected JBD 100 amp continuous BMS for reliability under extreme conditions

They told me that it was a JBD ZP20S006 BMS. I would imagine they could replace the BMS with a bluetooth one for a little extra, and it would add about 0.25" to the overall length of the battery (from what they told me).

 

[E-HP]

They told me that it was a JBD ZP20S006 BMS. I would imagine they could replace the BMS with a bluetooth one for a little extra, and it would add about 0.25" to the overall length of the battery (from what they told me).

Great info, thanks!
I didn't know I needed to monitor balancing and the temps while charging, until bluetooth spoiled me lol.

 

[GoFastGus]

I didn't know I needed to monitor balancing and the temps while charging, until bluetooth spoiled me lol.

In a perfect world, I'd love to be able to monitor temps and balancing while charging from my phone, but I didn't want to add 4-5 weeks onto the delivery date when they say it already has balancing and temp protections in place. Adding to my wish list down the line.

 

[GoFastGus]

Controller thoughts:

I'm looking for something that's programmable and that can handle 100A continuous and at least 150A burst. I also need the controller harness to be compatible with a Leaf Bike motor, this sort of headlight, tail light, and turn signals, and these brake levers. As always, I'm open to suggestions.

Option 1: Sabvoton SVMC 72150. Harness looks to have the plugs I need. Programmable via bluetooth. My understanding is this is a pretty tried and true option that would leave me with some room to grow.
Option 2: Siaecosys SIA-CJ72450. About half the price of the Sabvoton. Looks similar to a Fardriver controller, although I can't confirm how similar they perform. Harness looks to have the plugs I need, but I'll double check with the user's manual.
I've looked at Kelly controllers, but they don't seem to have the harness I'd need for the accessories. Maybe if there's some sort of cheap harness adapter available I'd consider them.
Any other options I've missed?

 

[E-HP]

Controller thoughts:

I'm looking for something that's programmable and that can handle 100A continuous and at least 150A burst. I also need the controller harness to be compatible with a Leaf Bike motor, this sort of headlight, tail light, and turn signals, and these brake levers. As always, I'm open to suggestions.

Option 1: Sabvoton SVMC 72150. Harness looks to have the plugs I need. Programmable via bluetooth. My understanding is this is a pretty tried and true option that would leave me with some room to grow.
Option 2: Siaecosys SIA-CJ72450. About half the price of the Sabvoton. Looks similar to a Fardriver controller, although I can't confirm how similar they perform. Harness looks to have the plugs I need, but I'll double check with the user's manual.
I've looked at Kelly controllers, but they don't seem to have the harness I'd need for the accessories. Maybe if there's some sort of cheap harness adapter available I'd consider them.
Any other options I've missed?

If a CA is in your plans, then most things in your original post can be accomplished using the CA (temp monitoring/thermal rollback, etc.) without a programmable controller. Might be worth considering since a cheap generic $70 controller may do the trick and save you money. The cheap ones with a 3 speed switch will provide field weakening when set to high.

 

[GoFastGus]

If a CA is in your plans, then most things in your original post can be accomplished using the CA (temp monitoring/thermal rollback, etc.) without a programmable controller.

Sounds like CA is a good alternative to getting a display in the first place.

Might be worth considering since a cheap generic $70 controller may do the trick and save you money.

Intriguing! I'll take a look on Ebay/Amazon/AliExpress for 72V 150A controllers.

The cheap ones with a 3 speed switch will provide field weakening when set to high.

Good point, field weakening is probably worth looking for at this power level.

 

[DanGT86]

I don't know exactly what your riding style/environment is like but I can tell you its hard to find a situation where you need 100amps continuous on a 72v bike. It takes a very steep hill climb or a really soft terrain to need 7000watts continuous. Your motor will likely not be able to do that for long.

Just pay attention, when controller shopping, to phase vs battery current ratings. With a high powered hub motor you may be pulling burst phase/motor currents of 2.5-3x your battery current.

I love the CA3 platform. I learned more about ebikes and electronics from using that device than any other endeavor. The detailed users manual is almost like an ebike 101 class. It can make any controller reasonable to use as well as adding thermal protection and various forms of pedal assist. I keep one in my shop for monitoring battery charging and sometimes even just as a volt meter. If you are tinkering and swapping parts all the time its a great thing to have despite looking a little dated.

I have some old cheap controllers that I can no longer program due to XPD being dead. The CA3 is what is keeping them out of the trash can since they would be violent without CA throttle settings and power limits.

 

[E-HP]

I love the CA3 platform. I learned more about ebikes and electronics from using that device than any other endeavor. The detailed users manual is almost like an ebike 101 class. It can make any controller reasonable to use as well as adding thermal protection and various forms of pedal assist. I keep one in my shop for monitoring battery charging and sometimes even just as a volt meter. If you are tinkering and swapping parts all the time its a great thing to have despite looking a little dated.

I have some old cheap controllers that I can no longer program due to XPD being dead. The CA3 is what is keeping them out of the trash can since they would be violent without CA throttle settings and power limits.

The CA is like the preamp of a stereo system. It has the brains, and you just add as much of the brawn as you need with a big dumb controller. People say it’s pricey, but it costs far less to buy a CA and dumb controller than a controller with all its brains built in, and once you blow that expensive controller, you’re stuck buying another one, instead of replacing the cheap dumb one.

 

[GoFastGus]

The CA is like the preamp of a stereo system. It has the brains, and you just add as much of the brawn as you need with a big dumb controller. People say it’s pricey, but it costs far less to buy a CA and dumb controller than a controller with all its brains built in, and once you blow that expensive controller, you’re stuck buying another one, instead of replacing the cheap dumb one.

Where are you finding non-programmable controllers that can take over 100A for less than $200? I haven't seen anything on Alibaba, Amazon, or Ebay. The closest I can get is a Fardriver ND72360 for about $200.

 

[GoFastGus]

Torque arm update: before I start cadding anything, I'm thinking about one of two types of torque arm designs.
This design seems fairly easy to rough out of mild steel. The fastener would clamp on the flat parts of the axle. What improvements could I make to this design?


This other design looks more compact, but I don't think it would work as well, since it's clamping on the curved parts of the axle. Is this just for the bin, or can this design be saved?

 

[DanGT86]

I like the split clamp design and think its the best way to keep it secure over time.

The top pic is the better option since the clamp bolt will squeeze the motor axle flats like a wrench. Also keeping it in line with the dropout allows wheel removal.

Personally Id prefer the top one but use the IS brake tab bolts to attach to. its a simple design and only requires slightly longer screws.

 

[AdR]

Torque arm always clamp on the flat. How you plan to secure it to the frame.

 

[GoFastGus]

Torque arm always clamp on the flat. How you plan to secure it to the frame.

I've done stainless steel U bolts before, and they're my first inclination. Second choice would be attaching to the brake brackets on the brake side. There's nothing to bolt to on the cassette side.

 

[GoFastGus]

use the IS brake tab bolts to attach to. its a simple design and only requires slightly longer screws.

That's true. I could cut a torque arm out of mild steel, drill and tap the fastener holes, and use grade 10.9 or 12.9 hardware. I have some worries about this idea though.
First, the small distance between the brake fasteners and the axle means a smaller moment of inertia, which means more force gets transferred into the frame. I don't have enough engineering experience to quantify the strength of the frame and decide from there how far from the axle I should clamp my torque arm, so I can only guess at how big of an issue this would be.
Second, the fasteners would be locating the torque arm (is this the right term?) and preventing it from twisting. My intuition says fasteners are better at clamping things together than at stopping things from twisting. The brake bracket is 8mm thick. Is that enough material for the fasteners to grab onto to absorb the majority of 10kW?
Once again, I'm not an expert engineer, so if any of my concerns are flawed, please let me know.

 

[DanGT86]

The brake bracket is not an uncommon location to mount a torque arm. Braking forces are violent enough to lock up the wheel so unless your motor is capable of doing standing burnouts I believe braking force is always more than motor force.

I would make the torque arm such that it goes under the bolt heads like a washer so the bolts are still clamping like you said. I only mentioned longer because they would need to be longer by the thickness of the torque arm material. I like the brake bracket because the torque arm forms an extra gusset that locks the whole rear of the frame together.

I think you would be shocked if you realized how little lever arm it takes to stabilize the motor. The torque arm can be very short it just needs to clamp solidly on the axle. Its the abrupt shock load and rocking back and forth that destroys the drop outs not the total amount of torque. A clamping torque arm made of 1/8" thick steel will outlast a sloppy one made of 1/2" steel in my experience. Your main priority is zero slop at the axle.

Here is an example of some I made recently for a friend. The stronger brake side one is clamping. The drive side one I just made very close tolerance to the axle and gave it a little torque to take the slack out before the final tightening.

 

[99t4]

I don't think it would work as well, since it's clamping on the curved parts of the axle. Is this just for the bin, or can this design be saved?

You are correct there, as clamping the rounded axle portion is not effective; do not even consider it.

Go with something similar to what DanGT86 shared with you above:

You don't need to reproduce the triangular cutouts. You can make it longer so it attaches via the two brake caliper IS tabs.

 

[E-HP]

Torque arm update: before I start cadding anything, I'm thinking about one of two types of torque arm designs.
This design seems fairly easy to rough out of mild steel. The fastener would clamp on the flat parts of the axle. What improvements could I make to this design?
View attachment 385466

This other design looks more compact, but I don't think it would work as well, since it's clamping on the curved parts of the axle. Is this just for the bin, or can this design be saved?
View attachment 385467

Are you going to be using axle nuts? I was surprised that my Leaf's axle was so short compared to the old MXUS motor it was replacing. With the washers and a 1/4" thick Grin torque arms in place, there's barely enough exposed thread for the axle nut. It depends on the frame dropout width, too, but the MXUS had almost 1/2" of exposed thread with the nut in place.

 

[GoFastGus]

Here is an example of some I made recently for a friend. The stronger brake side one is clamping. The drive side one I just made very close tolerance to the axle and gave it a little torque to take the slack out before the final tightening.
View attachment 385475View attachment 385473View attachment 385474

This is a great idea for the brake side! Unfortunately, the cassette side doesn't have any holes to bolt to, so I'll be making something to clamp onto the frame for that side. Thanks!

 

[GoFastGus]

Are you going to be using axle nuts? I was surprised that my Leaf's axle was so short compared to the old MXUS motor it was replacing. With the washers and a 1/4" thick Grin torque arms in place, there's barely enough exposed thread for the axle nut. It depends on the frame dropout width, too, but the MXUS had almost 1/2" of exposed thread with the nut in place.

Probably not on the disc brake side, maybe on the cassette side. I agree, the Leaf Bike axles are shorter than one would prefer.

 

[E-HP]

Probably not on the disc brake side, maybe on the cassette side. I agree, the Leaf Bike axles are shorter than one would prefer.

Maybe you could just use a Grin torque arm on the brake side so you can get a nut on it?

 

[GoFastGus]

Maybe you could just use a Grin torque arm on the brake side so you can get a nut on it?

Probably, but I'd prefer to make something out of mild steel so I can get more experience CADing and machining. Plus, it'll be cheaper.

 

[GoFastGus]

Update:
Battery's here, and it fits in the bike frame. 20S4P Molicel P50B cells, JBD 100A continuous 160A pulse BMS. XT90 discharge connector, XT60 charge connector.



Next step is ordering a frame bag and a charger. For a bag, I'm thinking about something from Amazon, like this from Ibera or Gorix. For a charger, I'm probably not going to spring for a Satiator unless everyone says it's absolutely necessary. I found some options from Amazon, like this one from QKDIM (5A), or this one from YZPower (4A). I could also get one from Bicycle Motorworks (3A). At this lower price point, I'm pretty sure I'm giving up the ability to alter the amperage based on the level of charge. My question is more so is that feature worth $250-300 extra?

Side note about connectors: The XT90 that came with this battery is gonna get replaced by two ring crimps. I'm not sure how the waterproofing is gonna work out, but I'll figure that out down the line.

Just got my controller today as well: Sabvoton SVMC72150 V2. $210 USD, including tax and shipping off of Ebay.

Looks like the seller ordered this from Siaecosys: the box, bagging, and controller all have Siaecosys branding on them. I don't know if the barcode is the serial number or a UPC or SKU or something, but it's better than nothing in making sure it's legitimate.

Came with the USB cord and bluetooth module.

The connectors, as well as the hall sensor adapter. This pic clearly shows how the battery and phase wires are connected via screws, so I'll need to use ring crimps unless there's a better way.

Framed handlebar stem, 2005 Marzocchi Bomber 888 VF fork, Revolution 32 wheel, 26x2.5" Maxxis Hookworm tire. Pressing in the headset this weekend so the battery won't need to be the chock for long.


In the mail:
Amazon throttle w/ key, 203x2.3mm rotor, Sram Code caliper
Leaf Bike motor
Fox RP23 seal kit
Aliexpress ebike DOT brake lever w/ 7/8" MC
CA3 HC, 0.5mOhm shunt, 10ml Satorade

 

[n00b]

When working with your cardboard battery templates, don't forget to figure in at least minimal dense foam along the edges to protect the battery from vibrations and shock. Highly recommended.

From now on, I factor in some dense foam and plywood.