Electric conversion of a 1984 Honda CM450

[harrisonpatm]

There's forums out there specifically for electric vehicles, but honestly you folks have been incredibly helpful, more so than other forums, so you get the post!

I was attracted to YouTube videos of electric motorcycle conversions and I kept thinking that it seemed doable, even for my inexperienced talents. So when I came across an old Honda with a dead engine going for $150, I went for it. This was mid-May of this year, about 4 months ago.



Engine was dead, but that didn't matter, and frame was good. I stripped it to pieces, which was my first pleasant surprise, because I was actually able to do it completely in one day by myself with no specialty tools. That's when I got confident and started ordering parts, and taking night classes at YouTube university to brush up on all the skills I would need.



-QS273 8000W-continuous-rated hub motor on an 16-in rim, Kenda Challenger tire
-APT96600 400 amp BLDC motor controller. It has its pros and cons, but mostly I got it because it came with the motor and I had to start somewhere. It's not terrible, but I'll find a different one for the next conversion.

-Battery! I wanted to go LiFePO4, even though it's heavier than Li-Ion, because I'm not an experienced rider and I went for safety, longevity, and temp range (I live in Michigan and might be using this in late fall and early spring) over weight and power. Went with 32650s from Battery Hookup, 12p24s to get that 72v range, just over 5kwh. I picked a JKBMS 200A for it's price point and relatively good reviews online. Daly was another option, but much pricier. We'll see if this one lasts. Battery weight with cables and everything was about 110 pounds, still lighter than the engine was. I don't weld, so I found a welder who was able to make an aluminum case for it that mounted to the frame in the same spot the engine was. If the brackets held the 175-lb engine, they'll hold the battery just fine.


I use a 400-amp contactor, a 200-amp shunt to read amperage off the battery during riding, and a 350-amp breaker for the power train cabling. BMS rated for 200 amps continuous, the controller is rated for 600 amps, the motor is rated for about 130 amps continuous and 250 amps peak... I should be good. Spoiler alert, it does run, but I would love feedback from the forum on how well I selected, rated and built all my components. Who wants to place bets on which component fails first?!?

I cut the bottom of the fuel tank to make room for electronics. Then when the controller arrived, much larger than expected, I cut a hole out of the top as well. Oh well.


For charging, I don't actually have an AC charger yet, but it is planned, either purchase or assemble. For now, I set up 2ea 240w solar panls on the south side of my house. Series connected for now, at about 66Voc, connected to the MPT-7210 boost charge converter to bump it to motorcycle battery voltage. This gives me 4-8 amps at a time, which means I can fully charge the battery in about 10-14 hours of direct sunlight (2-3 days?) This is totally fine for now, I'm not riding miles every day, don't need to charge it every day. This is just going to be my commuter to work 2-3 miles a day, plus joy riding. My future upgrade plan is to upgrade my house's small off-grid battery bank to 48V and use a proper, better, true MPPT controller to collect sunlight at all times, not just when I have the bike plugged in. Then I can treat that as my bike's reservoir, and use a boost converter to tap into the 48V bank to charge my 72v bike; faster, more regularly (better for long-term battery life), and more convenient.

I could go into a lot more details on every little thing, but I'll let pictures speak for themselves. Let me know if you have any comments or feedback. I also have plenty more to do. Namely, need to find a better way to cover up the controller. If anyone is thinking about doing it, I say go for it. This was a surprisingly achievable project.

P.S. The biggest mistake I made was picking a relatively heavy bike. Next time, I'm going smaller and lighter. Smaller battery needed, smaller motor, smaller controller, cheaper, ect...


Checking out the competition...

 

[OffGridInTheCity]

NICE WORK! Do you plan to ride it on the road? Is the original registration/license plate OK still?

 

[harrisonpatm]

NICE WORK! Do you plan to ride it on the road? Is the original registration/license plate OK still?

I certainly plan to road ride it. It's not legal yet, and it never had a title (was scrap) but there's some avenues available. For one, in Michigan, you can ride a sub 49cc moped on roads without a license or registration. So I could do nothing and play dumb if a cop pulls me over and asks why I was going 60.

But I will attempt to apply for a new title. I can register it as an "assembled vehicle", if it passes an inspection, and I at least kept all my receipts for that process.

 

[Redpacket]

You've got to tell us: what does the acceleration feel like especially since it's silent!

You might even need an "engine noise system" so other drivers notice you!

 

[harrisonpatm]

You've got to tell us: what does the acceleration feel like especially since it's silent!

I can only speak for my experience, because I have almost no experience with conventional motorcycles, just a few hours for the motorcycle endorsement on my driver's license. All I can say is that it's incredibly smooth and silent and quiet, and the instant torque is something else. I still need to get used to moving fast on two wheels, but the acceleration is incredible. I don't have to worry about shifting gears, and can pay all my attention to amperage and voltage.

On that note, I'm seeing a voltage sag of max 8% when I really give it throttle, and the battery is giving all it can. The internet seems to think that a 5-10% voltage sag when pulling really high current is normal. Especially if the voltage bounces right back, which is does (yay new cells). Does that seem within range?

You might even need an "engine noise system" so other drivers notice you!

Loud pipes save lives. But I actually get more head turns when people see me, because I can see them thinking, "why can't I hear you?!?"

 

[OffGridInTheCity]

On that note, I'm seeing a voltage sag of max 8% when I really give it throttle, and the battery is giving all it can. The internet seems to think that a 5-10% voltage sag when pulling really high current is normal. Especially if the voltage bounces right back, which is does (yay new cells). Does that seem within range?

5-10% good or bad? Just out of interest, how are you calculating the percent? - For example, do you mean that 58v (no load) is dropping 10% to 52.2v and then bouncing back to 58v. On a 14s lithium-ion that would be 4.14v -> 3.73v / pack. As you go lower down the voltage curve you'll hit the 3.4v-3.5v discharge knee and at that point go right on down.

The example above is a pretty hefty drop, however....

Is the load (amps / cell) with the specs of the cells and does it work for you?
- If you're exceeding the cells pecs, I presume it will drastically shorten their life. This may be OK for you.
- Does the overall experience work for you? For example, a 10% drop will shorten operational range as you hit low voltage cut-off significantly earlier than if it was only a 2% drop. But if the range/performance is OK for you, then I'd say it's OK

For context when my powerwall was only 80kwh, a 150a @ 50v load (7,500w) would drag the voltage down by ~0.5v when it was at 50v. This means it would hit the 49.5v inverter cut-off and then bounce back to 50v. 0.5v/50v = 1%.

 

[harrisonpatm]

This is all good info, thanks, and in the long run everything is relative.

5-10% good or bad? Just out of interest, how are you calculating the percent? - For example, do you mean that 58v (no load) is dropping 10% to 52.2v and then bouncing back to 58v.

78v no load dropping to 74v when I go full throttle 0-35mph, pulling close to 150amps. This is a 5.2% drop.

- If you're exceeding the cells pecs, I presume it will drastically shorten their life.

Hopefully not. Each cell was rated for 18A max discharge, at 12p thats 216A, which I hopefully never hit, and if I do, only for a second.

This may be OK for you.

You know, it kinda is. You're helping me put things into perspective. While I do want this to last awhile, it is a learning experience, and maybe what I'll learn is how to better spec a battery build.

- Does the overall experience work for you? For example, a 10% drop will shorten operational range as you hit low voltage cut-off significantly earlier than if it was only a 2% drop. But if the range/performance is OK for you, then I'd say it's OK

I have yet to do a proper top-to-bottom range test, it's on my list this week. But you're right: as I got to the bottom of my capacity, a 5%voltage drop caused the BMS to cutoff due to a few cells in series being under voltage. Realisticly though, as LiFePO4 has such a sharp dropoff after 3.0V, I'm sure I didn't have much real capacity left even after voltage sag. So I could be picky, but again, first build, learning curve and all. Plus from what I read, both the BMS SOC readings and the SOC of the relatively cheap panel voltage meter on the handlebars, need a couple full charge-discharge-charge cycles before their percentage is accurate. And I also need to get better at knowing the battery life and range. Asking about voltage sag is basically something I didn't fully wrap my head around during the planning stages, so I'm wanting to learn about it. Thanks.

 

[harrisonpatm]

I have yet to do a proper top-to-bottom range test, it's on my list this week

My 5.5kwh battery on a 300-350 pound bike with a 200-pound person on it went 42 miles on a single charge. Even mix of suburban stop-and-go traffic, as well as faster cruising. In case anyone wants a reference for their own projects.