1975 Honda CB550 Conversion

[mazzola]

For my first build, I've started converting this 1975 Honda CB550 with the intention of turning it into a city commuter that preserves the original aesthetics of the motorcycle as much as possible.



There's a pretty big restoration part of this project (e.g. painting, fixing bearings, leaky forks, front break lines) but this thread will focus on the new electric drivetrain unless folks are interested in my amateur restoration process as well.

I'll be using this bike for cruising around town which means typical speeds of 35mph (56kph) and low range, 15mi. (25km) or so. Sometimes I'll be riding at 35mph with a passenger or cargo on the bike, bringing to total load to 240lbs (110kg). The other need I have is to occasionally commute outside of the city about 25mi. (40km) from one charging station to another with some stretches at 45mph (73kph). This is what I was thinking to use as my "worst case" scenario to design against.

My dream is to keep as much of the aesthetics of the old bike as possible. Since it doesn't have any fairings, this means hiding or blending in a lot of the new components. At the end of the conversion, I want the bike to look as close to stock as possible. Something like this:



I know this is going to mean a lot of tradeoffs and a lot more work when picking my components but I love the looks of these old bikes too much . Since my performance needs are not too extreme, I think this should be possible.

I'll follow up with more detailed posts about how I'm thinking about the motor, batteries, and controller. But for now I'm proposing to use:

* Grin All-Axle front hub motor, fast winding with Statorade, installed in the rear wheel
* 72v, 2.1kwh (20s7p Molicel 21700 P42As) enclosed by the old engine case
* V6 Phaserunner, programmed to 96A phase current, 96A battery current

Let me know what y'all think!

 

[scianiac]

If you like how the bike looks you should just fix up the bike, if you leave all those parts on it just for looks it's going to be so heavy that motor and controller are going to struggle to make it rideable. Keep in mind the dry weight of that bike is 437lbs, will probably be heavier after those parts and that motor may be quite peppy on a 50lb bike but will not be on a 450lb bike. Could it technically be done with an extreme amount of modification, a much larger motor and controller and a small high power density battery, probably?

 

[mazzola]

I'm really only thinking of keeping the engine case and head which is around 20lbs - so that would be the only "extra" weight. Ultimately, I expect the final weight of the bike to be around 220lbs. While the bike was certainly heavy when I first got it, stripped down to just the frame, swingarm, fender, and rear suspension it weighs about 40lbs.



Adding the weight of the other components

Front forks: 20lbs
Front wheel: 15lbs
Gas tank: 5lbs
Seat: 7lbs
2.1kwh Battery: 30lbs
Battery enclosure/engine case: 20lbs
Phaserunner Controller: 1lb
Rear wheel and Grin motor: 24lbs
Various electronics/wiring: 10lbs
Other: 50lbs

I get about 220lbs total. Weight is definitely a tradeoff - it'll never be as light as a ebike or even a lightweight moped like the ONYX. But given that I'll be driving on mostly flat roads and at lower speeds, Grin's trip simulator is saying it's achievable with the components I initially proposed.

 

[Elektrosherpa]

If you want the bike to look like in your second picture, you will need the exhaust pipes, too.
Otherwise it will look totally different.
If you add 4 exhausts to your calculation, you will have at least 20-30 kg (40-60 lbs) more..

 

[scianiac]

It would be possible to make some fake parts, like for instance fake exhaust pipes, the exahusts on those old Hondas are silly heavy.

I also think it's not a great idea to put that much weight on that rear wheel, it will probably take it but I wouldn't feel comfortable with it. You have to remember that rear wheel, meaning the spokes, rim, and tire are all designed for much less weight. I would still consider using a motorcycle hub motor and a matching controller. Or at least a motorcycle rim with matching spokes. Even an in between motor like a leaf or QS and a one size up controller with a moto rim and tire. Then you are much closer to the weight to performance range of common ebikes.

What does the grin sim spit out for acceleration and how it performs going up hills at that weight? I would compare those to similar low speed vehicles like mopeds and scooters.

 

[harrisonpatm]

It'll be interesting to see how you manage to mock it up so it still looks like an engine. I would consider @scianiac suggestion to use fake parts, like fiberglass or 3d printed plastics, instead of steel.

With that bike's weight, 25 mile range off a 2.1kwh battery is a bit too optimistic. Possible, maybe, but it'll be cutting it close. Consider going up to 3-4kwh so you'll have wiggle room.

I believe you'll be okay with 90A battery current, but only okay. If it were my build, I'd want more wiggle room there as well.

I did a 1984 CM450, so check out my build for a real-world comparision on a very similar body.

 

[SlowCo]

For my first build, I've started converting this 1975 Honda CB550 with the intention of turning it into a city commuter that preserves the original aesthetics of the motorcycle as much as possible.


I'll follow up with more detailed posts about how I'm thinking about the motor, batteries, and controller. But for now I'm proposing to use:

* Grin All-Axle front hub motor, fast winding with Statorade, installed in the rear wheel
* 72v, 2.1kwh (20s7p Molicel 21700 P42As) enclosed by the old engine case
* V6 Phaserunner, programmed to 96A phase current, 96A battery current

Let me know what y'all think!

I would suggest to choose a bigger, stronger motor for a heavy build like this. The Grin motor is more intended for bicycles and light mopeds. A QS205 with a compatible controller is much more suited for your project: QS205 Hub Motor
They are here on this forum and very helpful. Ask them what you would need to order to make everything work "plug&play": QS Motor thread on E-S

 

[harrisonpatm]

A QS205 with a compatible controller is much more suited for your project: QS205 Hub Motor

Maybe, but that swingarm likely has 230-235mm dropout, the 205 might be too small

 

[SlowCo]

Maybe, but that swingarm likely has 230-235mm dropout, the 205 might be too small

It's a steel swing arm and possible to bend it to make the QS205 fit.

 

[mazzola]

you will need the exhaust pipes, too

I'll probably leave the exhaust pipes off. While it adds a lot of chrome to the look of the bike, I think this is just one of those parts that I'll have to forego because of weight and they won't serve another purpose like the engine will as a battery enclosure.

Even an in between motor like a leaf or QS and a one size up controller with a moto rim and tire

Yeah I think I'll probably have to beef up the controller. But I like the idea of having a plug-n-play between Grin's Phaserunner and hub motor. I'm definitely going to use a motor rim and tire. There's a 32 spoke aluminum moto rim that should save on weight too. But all in all I'm expecting the rear wheel to be about the same weight as the original since the Grin All-Axle is crazy light and the old wheel has heavy drum brakes (which I'd replace with disc or use regen) and s heavy sprocket.

What does the grin sim spit out for acceleration and how it performs going up hills at that weight?

I mapped out my "worst case" scenario on the Grin trip simulator and broke it down into 5 segments with 35 and 45mph speed limits. It all looks good! Even in the higher speed segments, the bike is able to maintain the speed limit for 99% of the time and dropping <5mph or so on steep hills for a couple seconds which I'm okay with. It also doesn't overheat by the end of the trip, even with a high, Texas ambient heat of 86F.

Segment 1
Segment 2
Segment 3 (45mph)
Segment 4
Segment 5 (45mph)

The only scenario that is a little worrisome is when there's a 4mph N/S headwind. Then it looks like there's a chance of overheating

With that bike's weight, 25 mile range off a 2.1kwh battery is a bit too optimistic.

According to the sim, my worst case scenario will use only use 1.66kwh total and that's assuming no regen and constant high speed /high drag instead of the actual route which will require lots of stops. But I think I'll end up having some more room in the battery enclosure so I might bump up the kwh.

I did a 1984 CM450

Oh, awesome! I'll definitely take a closer look. Would love to hear more about your rear disc brake installation - but I'll post it over in your build thread.

It's a steel swing arm and possible to bend it to make the QS205 fit.

I was very close to getting the QS205 but my biggest concern was the swing arm. Yes it's steel, but it also has a 243mm dropout, so bending it to 200mm seems like I'd be bending it too much but I'm unsure if it's actually acceptable.

 

[scianiac]

I should have been specific, I meant the other grin simulator and looking at the acceleration numbers which are ok if a bit slow at first but to get to those higher crusing speeds is going to take some time. So is the idea the grin motor allows a much longer axle to be installed so you don't have to modify the swing arm? Another option would be to just special order a different motor with a longer axle, I've heard many manufacturers will do this for not much additional cost. I see you're point of wanting a plug and play solution and I think it will work but you're going to have to be mindful of weight when building it and acceleration will be a bit leisurely. It wouldn't hurt to just play around with both grin sims and some wider stator motors and higher current controllers and compare the rough performance to other things. Perhaps lookup another slow bike/scooter/moped you've ridden acceleration wise to get an idea of how it will ride. If you're happy with the comparison go for it, if not consider the other options. Even a 35H leaf is a fair bit more stator than the 27H? grin.

 

[SlowCo]

I was very close to getting the QS205 but my biggest concern was the swing arm.

Your biggest concern should be that the Grin all axle motor is way too weak for your application...

Maybe consider the mid drive motor with build in gear reduction so you can keep the original rear wheel: Mid drive motor 1/2.35

 

[mazzola]

looking at the acceleration numbers which are ok if a bit slow at first but to get to those higher crusing speeds is going to take some time

How are you figuring out the acceleration with the motor simulator? I've never done it before but what made sense to me was to try and sample it once per second by taking the initial speed and acceleration (e.g. 0mph and 3.26 mph/s) and then getting the next acceleration measurement at the new speed (e.g. 3.26mph), which should be roughly one second later. And then I averaged them altogether to get an equivalent constant acceleration.

I just did some runs with my CSC Monterey (60v, 24ah, 1.5 kw nominal, 181lbs) and it was doing 0-20mph in 5.42s which works out to be an average of 3.68 mph/s which honestly is a good amount of pep for me. For the Grin, at full throttle with my proposed setup, I get an average acceleration of 3.06 mph/s which makes a 0-20mph time of 6.54s. That seems like it will be perceptibly slower, which would be a dealbreaker for me with the Grin All Axle.

So is the idea the grin motor allows a much longer axle to be installed so you don't have to modify the swing arm?

Exactly - the stock axle is 20mm which fits perfectly into the grin motor and I wouldn't have to modify the swing arm.

Another option would be to just special order a different motor with a longer axle

I've reached out to QS and all they could offer was a 200mm dropout. I've been looking all over for something with a longer axle but QS and Leaf have the longest I can find at 200mm and 190mm respectively. Replacing an hub motor axle with a custom-length one seems like I'd need a specialty part made and too much trouble. I won't be able to fabricate a new swing arm and I've already tried to buy a compatible swing arm from a smaller Honda CB, but didn't have any luck. So now I'm thinking my only option is to get a shop press to narrow the swing arm down to ~200mm so I can use more powerful hub motors.

the Grin all axle motor is way too weak for your application

Yeah, it's starting to sound like it.

 

[mazzola]

Ah, but I'm just realizing that if the Grin doesn't have enough acceleration, then I just need to increase the phase amps from the controller. So if I run the simulator using a 300A controller like the makerX GO-FOC G300 then I get an average of 4.92 mph/s for a 0-20mph in just over 4s which sounds pretty good to me

 

[scianiac]

Well not so fast, the grin simulator does not take into account saturation so actual power output will be quite a bit less once you go well into saturation which you probably are considering that is a lot of current for such a small motor. That and of course overheating, melting phase wires, etc. Grin might actually know what the saturation current is which you could stick into the sim and see where that gets, doing that while also shaving some weight.

There is a reason most moto conversions either just replace the engine with a motor and use the same chain drive or use a motorcycle specific hub motor. You probably could fit a quite powerful motor inside the gearbox if you removed enough from the inside but it would not be a simple project.

Honestly, and this would have to be done carefully, you could probably just fabricate the clamping torque arms / drop extensions so they are just some real solid big chunks of steel that get welded to the inside of the dropouts and extend far enough inward to clamp a 200mm axle. They would have to of course be strong enough and welded well. Could also combine this with a little bit of swingarm compression.

 

[mazzola]

that is a lot of current for such a small motor. That and of course overheating, melting phase wires, etc.

I reached out to Grin but have yet to hear back from their engineering team. But I figured out that the phase wires on the all axle are most likely 12ga based on this motor cable that they sell. So 300amps (even the 200A that are actually needed to get 4.92mph/s) seems like it might be pushing the limits given Grin's Marine motor running 70 phase amps continuous in still air melted the L10 connector and Dr. Bass's test in still air which nearly melted a 12ga wire at 100A for 1 minute.

I tried compressing a spare swingarm using a threaded rod and was able to bend it a little, however it bent only at the axle slots causing them to no longer be parallel. Also taking a look at the construction of the swingarm, it seems unlikely I'll be able to compress it because of the thick reinforcement (highlighted in yellow) that is welded on to the steel tubes.



So with the Grin All-Axle being too small/underpowered and my swingarm being difficult to modify, I'm looking into other thru-axle motors. The best I could find is the Crystalyte H40 UFO series which has a 12mm thru axle option. I've reached out to Glide Cruisers who sells Crystalyte motors and am waiting to hear back from them. Let me know if there are other distributors that can ship to the US or if I should reach out to Crystalyte directly.

The plan with an H40 motor would be to use the Honda Grom rear axle which is 12mm x 305mm (only 5mm longer than the stock CB axle) and use 12mm/20mm bushings in order to fit it into the axle slots. Using the TC40100 as a stand-in, I'm leaning towards the H40100 since I assume they have similar performance.

You probably could fit a quite powerful motor inside the gearbox if you removed enough from the inside

I've taken the engine fully apart but there's not a lot of space inside for a motor. The space where the final drive shaft was is only 111mm wide, 120mm tall and 153mm long.



I've looked at the QS 90 70h, ECP4 Brushless motor, ETM Brushless motor, QS 138 4000W V3, and QS 120 and they are either dimensionally too big or too underpowered and liable to overheat, especially since there won't be much airflow inside the engine case.

 

[Tony01]

I’ve got a bike like that in worse condition that I’m going to restomod. It’s a waste of a good motorcycle to convert one to electric. I plan to put a modern front end, mag wheels, fab up a swingarm, modify the frame, do some motor work porting and increasing compression. It’s going to take a lot to have an EV setup that is anywhere near the stock 550 performance and not even talking about range.