1975 Honda CB 750 Conversion

[BatteryManTakeMeByTh]

I'm finally ready to start posting a real build thread, I think. Though it's definitely going to take several months. I am just CADing up my battery, and I have some questions. I am using Molicel P45A cells at 20s16p. I am designing for max continuous current of 200A at 72v. This is the first battery I have designed like this, so I want to be careful before beginning assembly. I have designed the battery to just be an 20x16 rectangle since it seems I can make that fit into the body of the CB750 frame I have. I am looking at bus bars now, so my main questions are:

1. Should I go for a layout other than square for any particular reason beyond space?
2. How thick should my bus bars be for 200A 72V continuous? I have 46mm by 495mm to work with per positive-negative copper bar and 23 by 495 for the + and - bars at the ends of the battery. I'm looking at WellGo since I've seen other users having good results with their thicker bus bars. I'm hoping I don't have to go for the full 1mm thickness.

Happy to provide more info. I have electrical engineering experience, but batteries are a new sub-field for me.

 

[pwd]

Square or rectangle is good if each cell group is layed out the same, this will give you optimal/easy current sharing.

I can't give you an answer on bus bar thickness but I probably would want an equivalent of 6 AWG copper cable or better, erroring on the side of better. You aren't going to be running 200A all the time on a motorcycle, so I would plan for peak current usage. IE You might run 300A for bursts of up to 10 seconds, which is less demanding overall than 200A continuously.

 

[BatteryManTakeMeByTh]

Square or rectangle is good if each cell group is layed out the same, this will give you optimal/easy current sharing.

That's what I was going for, just to make things simple on myself.

I can't give you an answer on bus bar thickness but I probably would want an equivalent of 6 AWG copper cable or better, erroring on the side of better. You aren't going to be running 200A all the time on a motorcycle, so I would plan for peak current usage. IE You might run 300A for bursts of up to 10 seconds, which is less demanding overall than 200A continuously.

I definitely want to be able to accommodate higher peak, but I'm sort of designing in head room in that 200A continuous number. I don't really expect to need to use the full 14.4kw for very long at all. But, I would rather not have any issues. Six gauge wire is only about 4.1mm, but that'd be super thick for a bus bar. Are we just talking calculate the volume of copper? I guess a better way to phrase that question is: how would I figure out what the equivalent of 6 gauge wire is in bus bar thickness? I can understand just calculating volume, but it seems like it should be more complicated than that to me.

I appreciate the input.

 

[jkrienert]

Pretty sure @pwd is referencing 6AWG in terms of the cables copper core cross-section (13.3 mm^2). So, you'd want to source copper bus bars with the same cross-section or greater, with the shape of that area being somewhat variable. That is, cross-section perpendicular to the flow of current!

 

[BatteryManTakeMeByTh]

Pretty sure @pwd is referencing 6AWG in terms of the cables copper core cross-section (13.3 mm^2). So, you'd want to source copper bus bars with the same cross-section or greater, with the shape of that area being somewhat variable. That is, cross-section perpendicular to the flow of current!

That makes a lot more sense. Thank you for clarifying. Since I have 46ish mm to work with, width-wise, I could probably get away with 0.3 mm thick in that case. So, I would likely size at 0.4 as long as I think I could make that work. Gonna do some more research, but the input is much appreciated.

 

[pwd]

Pretty sure @pwd is referencing 6AWG in terms of the cables copper core cross-section (13.3 mm^2). So, you'd want to source copper bus bars with the same cross-section or greater, with the shape of that area being somewhat variable. That is, cross-section perpendicular to the flow of current!

That is what I meant and you explain it very well.

 

[bananu7]

How thick should my bus bars be for 200A 72V continuous?

I'm aiming for 50mm2 (5x10mm) of solid copper for similar power requirements.

 

[BatteryManTakeMeByTh]

I'm aiming for 50mm2 (5x10mm) of solid copper for similar power requirements.

In my case, that would be over 1mm thick for the bars. Might have to redesign the spacers again :/ Are you using thick bus bars or wire or what?

 

[CRF265]

Just a question, have you purchased the batteries or controller yet? Is there a reason you are set on 72v? Considering the size of your battery and its a road bike, you could easily do a 28s or 30s setup with CALB L148N58A Prismatic cells or similar - they still get 244wh/kg. It would be much easier to assemble. You could then just use copper plate bars between cells 25x3mm and easily achieve 75mm2.

Regardless of cell type, higher voltage with p45 cells is still worth considering. Advantage of higher voltage is that you will draw less current all the time which will generate significant less heat - heat increases by I^2 so it is exponentially better. Less heat results in less losses, you also get better efficiency at higher motor rpm with high voltage where you are not needing to generate extra heat through field weakening. Only drawback is its more dangerous.

 

[BatteryManTakeMeByTh]

Just a question, have you purchased the batteries or controller yet? Is there a reason you are set on 72v? Considering the size of your battery and its a road bike, you could easily do a 28s or 30s setup with CALB L148N58A Prismatic cells or similar - they still get 244wh/kg. It would be much easier to assemble. You could then just use copper plate bars between cells 25x3mm and easily achieve 75mm2.

Regardless of cell type, higher voltage with p45 cells is still worth considering. Advantage of higher voltage is that you will draw less current all the time which will generate significant less heat - heat increases by I^2 so it is exponentially better. Less heat results in less losses, you also get better efficiency at higher motor rpm with high voltage where you are not needing to generate extra heat through field weakening. Only drawback is its more dangerous.

I've already purchased the molicel P45A cells, and 320 of them no less.

I can definitely consider higher voltage with the heat considerations. I've done a lot of theory, but real power concerns aren't my forte, so I appreciate that insight. Better efficiency would be very nice. In terms of danger, I understand that higher voltages are more dangerous for high power applications. However, I had understood that "current is more dangerous" for lack of a more precise phrase. Is that just misleading, or am I missing another part of the puzzle here? Otherwise, it would seem to me that higher voltage and lower current would be less dangerous. Is it just that the peak current of the battery isn't going to be reached going through a person? I guess just practically, what is the dangerous part here? Just looking to learn

 

[CRF265]

The danger is current flowing through your body but you need voltage to overcome the resistance in your body for current to flow. 72v and 100v are both dangerous but 100v has more potential for lethal harm.

With 320 cells, pick your voltage and based on the maximium input voltage of your controller.

 

[BatteryManTakeMeByTh]

The danger is current flowing through your body but you need voltage to overcome the resistance in your body for current to flow. 72v and 100v are both dangerous but 100v has more potential for lethal harm.

With 320 cells, pick your voltage and based on the maximium input voltage of your controller.

Ok, that's more or less what I thought. Just wanted to make sure I was on the right track.

Any controller suggestions? I was looking at ANT stuff, and they seem to have some good options. I just want to know of the balance current is going to be enough.

At 28s I can run 11p without buying any more cells, and this gives me about 44Ah, which would make maybe 20 min at a peak current of 120 amps down from 200. That's not too bad, but I may want more cells at that point.

 

[adsc]

I mean, U = R*I, which means that I = U/R, and since R of young, healthy, dry skin is around 100k ohm, the expected current flow is below 1 mA, which is absolutely inconsequential. You can see video of guys who touch the end terminals of a 96V pack with their bare hands and nothing happens.

BUT: If some conductor pierces your skin, or your skin is already damaged where it touches, or your skin is wet, that R value is much lower, and it might just be low enough to become a problem. It doesn't take much current to cause your muscles to contract, which could already lead to a fatal accident by itself during riding.

I remember well from working on 48V telephone lines in the military that with wet leather gloves you sometimes got an uncomfortable shock that kind of paralyzed your fingers when you slipped with the screwdriver.

 

[BatteryManTakeMeByTh]

I mean, U = R*I, which means that I = U/R, and since R of young, healthy, dry skin is around 100k ohm, the expected current flow is below 1 mA, which is absolutely inconsequential. You can see video of guys who touch the end terminals of a 96V pack with their bare hands and nothing happens.

BUT: If some conductor pierces your skin, or your skin is already damaged where it touches, or your skin is wet, that R value is much lower, and it might just be low enough to become a problem. It doesn't take much current to cause your muscles to contract, which could already lead to a fatal accident by itself during riding.

I remember well from working on 48V telephone lines in the military that with wet leather gloves you sometimes got an uncomfortable shock that kind of paralyzed your fingers when you slipped with the screwdriver.

Good to think about for sure. I'm not necessarily as worried about dying as starting a fire or causing myself permanent harm. I am going to try to do all I can in terms of precautions, and thankfully I'll be doing some of this work at a maker space with plenty of more experienced folks to ask

 

[BatteryManTakeMeByTh]

Just a question, have you purchased the batteries or controller yet? Is there a reason you are set on 72v? Considering the size of your battery and its a road bike, you could easily do a 28s or 30s setup with CALB L148N58A Prismatic cells or similar - they still get 244wh/kg. It would be much easier to assemble. You could then just use copper plate bars between cells 25x3mm and easily achieve 75mm2.

Regardless of cell type, higher voltage with p45 cells is still worth considering. Advantage of higher voltage is that you will draw less current all the time which will generate significant less heat - heat increases by I^2 so it is exponentially better. Less heat results in less losses, you also get better efficiency at higher motor rpm with high voltage where you are not needing to generate extra heat through field weakening. Only drawback is its more dangerous.

Looking for parts, I am now remembering why I went with 72V. Controllers and motors for higher voltages seem kind of hard to come by. Seems like I'd probably want to stick to 96V for consumer electronics. Thoughts? I'd love to just be missing something.

Also, what motor should I go with? Still just whatever QS motor fits my specs? I'm shooting for 10kW, but I don't see many of those in the right size. Has anyone bought from SOTION before, like this motor?

Thanks again

 

[CRF265]

I dont have any experience with hub motors so I cannot comment on what type to get, but generally a 72v motor will take 96v fine, it will just have a higher max rpm. The controller duty cycles/phases the DC battery voltage to feed the motor 3 phase AC voltage and varies the voltage with rpm.

In simple terms a hub motor rated to 800rpm at 72v will only see 18v from the controller at 200rpm, 36v at 400rpm etc, until it reaches max rpm where motor voltage is equal to battery voltage. Raising the maximum voltage to 96v theoretically increases max rpm to 1066. The voltage the motor sees at lower rpms will be the same for either a 72v or 96v battery.

There are still alot of options at 96v if thats what you want to go with. My thoughts and experience is higher voltage is always better, for same power less current, less sag, less heat.

You mentioned ANT earlier, ANT is a BMS, I have used a couple and never had an issue.

 

[thebes]

Looking for parts, I am now remembering why I went with 72V. Controllers and motors for higher voltages seem kind of hard to come by. Seems like I'd probably want to stick to 96V for consumer electronics. Thoughts? I'd love to just be missing something.

There are many options for controllers, 3shul, fardriver, votol, vesc labs, etc. Most if not all 72v rated motors would be fine at 28s and higher, they would just run at a slightly higher rpm.

Also, what motor should I go with? Still just whatever QS motor fits my specs? I'm shooting for 10kW, but I don't see many of those in the right size. Has anyone bought from SOTION before, like this motor?

A qs273 """"8000w"""" or one of the qs268s is plenty for ~10kw. Sotions hubs seem to be rebranded qs motors as the internals are extremely similar and their claims, at least for their inrunners, are a bit exaggerated so

 

[thebes]

At 28s I can run 11p without buying any more cells, and this gives me about 44Ah, which would make maybe 20 min at a peak current of 120 amps down from 200. That's not too bad, but I may want more cells at that point.

Dote note that the power out and the capacity would be the same as you still have 320 cells, its just that at a higher voltage you have to run less current for an equivalent power output.

 

[BatteryManTakeMeByTh]

I dont have any experience with hub motors so I cannot comment on what type to get, but generally a 72v motor will take 96v fine, it will just have a higher max rpm. The controller duty cycles/phases the DC battery voltage to feed the motor 3 phase AC voltage and varies the voltage with rpm.

In simple terms a hub motor rated to 800rpm at 72v will only see 18v from the controller at 200rpm, 36v at 400rpm etc, until it reaches max rpm where motor voltage is equal to battery voltage. Raising the maximum voltage to 96v theoretically increases max rpm to 1066. The voltage the motor sees at lower rpms will be the same for either a 72v or 96v battery.

There are still alot of options at 96v if thats what you want to go with. My thoughts and experience is higher voltage is always better, for same power less current, less sag, less heat.

You mentioned ANT earlier, ANT is a BMS, I have used a couple and never had an issue.

I didn't know much about the inner workings of electric motors. I was always more into PCB design. That all makes a lot of sense, and I appreciate the info quite a lot.

I definitely would like to go higher voltage, but all the tutorials and stuff I've seen online was around 72. Then, coupled with my lack of knowledge about exactly what the specs meant for the motor, I wanted to play it safe. With some better info from someone who's done something like this, I think I'll look around for 96V controller and BMS then.

A qs273 """"8000w"""" or one of the qs268s is plenty for ~10kw. Sotions hubs seem to be rebranded qs motors as the internals are extremely similar and their claims, at least for their inrunners, are a bit exaggerated so

That's sort of what I figured since I have heard from many on this forum that they run their motors well above "rated" and have no issues. I just want to triple check before I do something stupid and have to immediately swap parts. I'd be hoping for 10kw continuous without issue.

The sotion stuff definitely looked too close to the QS motors to be completely unique. It would make sense if it was just rebranded. By inrunners are you just talking about mid-drive motors? Not surprising things would be exaggerated, but that's all the more reason to look for something spec'd a bit higher imo.

Dote note that the power out and the capacity would be the same as you still have 320 cells, its just that at a higher voltage you have to run less current for an equivalent power output.

I understand that, and I will definitely be taking that into account when trying to implement acceleration levels, but I do still want to design around minutes at peak current, even if that is now a higher power with a higher voltage pack. I just want to have in my head about how long I can go if I really floor it. Otherwise, I think any real numbers for how long I can ride will be contingent on the weight of the bike and the voltage sag under load and so on. Just trying to get a rough range estimate is all. I do appreciate the reminder, though.

 

[BatteryManTakeMeByTh]

Hi again, all. I am planning to go for 28 or 30s, and I was looking at Fardriver controllers. I found this one for up to 108V (ND1081200). Would that work for 30s? I get nominal of 111V, and while that's not far off, I'd like to be sure before buying. I'd rather go with 30s if possible.

 

[harrisonpatm]

ND1081200 | far-driver

ND1081200 10000-12000W 84V 96V 108V Sine Wave Programmable BLDC Motor Controller With Regen For Electric Scooter Motorcycle

www.far-driver.com

Highest voltage for the ND1081200 is 128v.

30s x 4.2v max charge per cell = 126v.

On paper this works fine. For more wiggle room, you could either go with 28 or 29s, or you could set your max charge voltage to be 4.15v per cell, which is healthier for their lifespan anyway. 4.15 x 30s = 124.5v

 

[j bjork]

Are they really available now? A few years ago you couldn't get any of the 108v fardrivers.
And it seems like the ones sold before that had a habit of braking.
It can for sure have changed since then.

 

[BatteryManTakeMeByTh]

ND1081200 | far-driver

ND1081200 10000-12000W 84V 96V 108V Sine Wave Programmable BLDC Motor Controller With Regen For Electric Scooter Motorcycle

www.far-driver.com

Highest voltage for the ND1081200 is 128v.

30s x 4.2v max charge per cell = 126v.

On paper this works fine. For more wiggle room, you could either go with 28 or 29s, or you could set your max charge voltage to be 4.15v per cell, which is healthier for their lifespan anyway. 4.15 x 30s = 124.5v

Thank you, I will definitely check that out. I will want to limit the charge for sure.

Are they really available now? A few years ago you couldn't get any of the 108v fardrivers.
And it seems like the ones sold before that had a habit of braking.
It can for sure have changed since then.

This is the next task lol, not sure if they are or not. Gonna have to ask directly I think

 

[BatteryManTakeMeByTh]

Alright, yall. I need some help grouping my cells, I think. Since I've moved from 20s to 28 or 30s, this design no longer works out into nice neat rows. How would y'all think about grouping these? I don't mean to just crowd-source the design, but any input would be helpful. Here's how I have the full pack laid out in the frame so far. Should I try for triangles or just sorta go with the flow? Just trying to figure out how to organize things so the bus bars aren't a nightmare.

 

[adsc]

It doesn't look like you have enough space for 28s16p. If you have only single tubes on top where the tank used to sit, you could maybe separate the pack into two halves that extend upwards past those tubes on the left and right of it. Or if the depth (using the above picture, depth would be the blue axis going into the picture) within the frame is large enough, you could do two 14s16p next to each other.