• Howdy! we're looking for donations to finish custom knowledgebase software for this forum. Please see our Funding drive thread

Questions around a 40kW build

ta0s

10 µW
Joined
Oct 26, 2023
Messages
6
Location
Christchurch
I appreciate the stream of inspiration that comes through here.

I'm a lifelong builder of things, and next on my list is an electric motorbike.

I am comfortable with 90% of the build I have in mind, inspired by the photo I've attached. It's a BMX style, but obviously electric. it will be for strictly on road use. I've built many frames for many purposes, and I think this will be a reasonably easy one to build.

I have a reasonable amount of experience with electrical engineering, battery building, and specifically high amperage DC so I don't think I need advice on how to put things together at this stage (unless there is some glaring error in what I'm presenting below).. - more what to put together!

For starters, I want things to peak at around 40kw. I may occasionally want this power to turn my frown upside down, but likely only for a few seconds at a time.

Based on this, I'm looking at the 12kw qs273 17x6 hub motor, at the torquiest winding.

I'm thinking 2 x 12s2p banks of gen2 Nissan Leaf battery modules, 96v total. Each bank running a JBD 300a BMS. For a total of 600 theoretical amps and drawing approximately 3.5C from each battery at this peak if possible. (I've attached a photo of the specs of each module)

To supply a theoretical potential of 57kw, I'm looking at something around a 600amp controller - something like the Kelly Controller KLS96601.

Notwithstanding all the other important details, I'd expect that this set up could deliver my desired 40kw, and probably more. What particularly excites me is the promise of 350+N.m of on tap torque. That sounds very fun.

My questions are:

In general are my numbers achievable/reasonable with the kind of price point of the parts I've listed? I've read a lot about 20-30kw but I haven't been able to find a lot about the 40kw+ space. Is it because this kind of power is demanding too much of the readily accessible Chinese parts? Or is it just that 40kw+ just isn't really desired for the investment required?

With respect to the QS 273 12KW motor (17inch 12KW Hub Motor for super Motorcycle, Bike and Scooter...) - is anyone on here running this sort of wattage successfully long-term? I'm not totally opposed to a mid mount, and I'm aware of the advantages - However for this specific build being a street bike the benefits of building ease and convenience I feel outweighs the benefits of a mid mount. The only thing that could swing me as if someone has built a high-performance bike based on this motor and the handling is really bad due to the high rotating weight. I do want this bike to feel snappy and fun, and I definitely don't want to compromise handling over build-ease

With respect to the BMS's - What are people's preferences in terms of brands (and features). I'm not experienced with such high amperage stand-alone BMSs, but are familiar with lithium chemistry and all of the care requirements of it. I'm also unsure if I need to take into account that the batteries I want to use are LiMn2O4.. The voltage cut-offs are the same - but keen to hear from anyone with experience here.

With respect to the controller - I'm pretty keen on having a regenerative brake function, ideally this would be analog, i.e. variable rate of application - as opposed to a "switch". Does anyone have experienced with high amperage controllers that support this? Other than regenerative brake, I'd looking for something that is as plug and play as possible. I prefer to not have to spend lots of time tuning. My preference would be to hook it up, set the basic configuration up and be able to have the controller and motor produce as close to the configured throughput as possible.

With respect to the concept - I realise this is a "heavy" build. I'm 100kg. Acceleration and handling are both high priorities for this build. Top speed, less so (140kmh would be absolute top, 120kmh satisfactory). If lighter hardware and battery would provide similar "felt" torque, I'd love to understand this better.

Thanks everyone, and looking forward to sharing progress as I go.
 

Attachments

  • 1570219286.jpg
    1570219286.jpg
    64.2 KB · Views: 83
  • BMX-Cub-250cc-1.jpg
    BMX-Cub-250cc-1.jpg
    174.4 KB · Views: 92
Last edited:
Without having tried a big hub like that, I would guess that it would do at least 40kw without problems.
It was probably something like that rovlii had on his drag bike earlier, could be worth checking out in the build thread.

I have run a 205 50h "3000w" with peaks of about 25kw in from the controller, probably significantly less output from the motor..
It was quick with about 60-65kg bike and maybe 90kg rider. It would overheat rather quickly though, but statoraid helped pretty much.
Hubsinks, not so much..
 
A bit off topic, but you might have advice for a rather long- term project with similar power requirements: The small model shown at Login • Instagram
One big enough to carry me requires at least 40 kw for hover, and power-failure safety requires several kw of regenerative capacity driven by autorotating rotors.
There are other desirable electrical features, but do you have any comments on these basics?
 
Last edited:
A bit off topic, but you might have advice for a rather long- term project with similar power requirements: The small model shown at Login • Instagram
One big enough to carry me requires at least 40 kw for hover, and power-failure safety requires several kw of regenerative capacity driven by autorotating rotors.
There are other desirable electrical features, but do you have any comments on these basics?

Taking into account you are going to be running three props.

In my research, this is what I've come up with:
  1. High amperage battery. This is done by using lots of high discharge cells in parallel. Check out molicel p28a for example @ 45 amps.
  2. High voltage. 72v - 108v typical. So, run lots of your high discharge cells in series x lots in parallel. I'm running 2x 20s5p batteries in parallel, each with their own BMS. giving me a total of 20s10p @ 450amp max battery amps. The way the battery is built is critical. For high amperage joins, plain nickel won't do. I'm using a copper nickel solution on my series connections good for 100amps.
  3. High discharge BMSs. I've settled on 2x JK 200a BMSs @ 20s/72v. Only delivering 28kw for this build as 40kw was getting out of my budget at this stage.
  4. High power BLDC controller. I've opted for a Fardriver 721200 which will easily deliver 28kw. You will need 1 per motor.
  5. High power BLDC motor. obviously you have different considerations than I do!
  6. Lots of 0AWG cable to transmit 400a! Rated for 150amp continuous, but will be fine for my 5-10sec pulses of 400a.
Further thoughts,

To run 40kw continuous, in my research you should be considering over-rating everything. Heat is going to be your enemy in batteries, controllers, BMSs, and cabling. For your craft I'd also be looking into fallbacks for your main controllers. It's not uncommon for the cheaper ones to fail, and this isn't an option for you. For the number of batteries you will need, I'd be considering active battery cooling too.

I have more thoughts, but this is the essence of them. I'll keep an eye on your project. Looks awesome!

Cheers.
 
Taking into account you are going to be running three props.

In my research, this is what I've come up with:
  1. High amperage battery. This is done by using lots of high discharge cells in parallel. Check out molicel p28a for example @ 45 amps.
  2. High voltage. 72v - 108v typical. So, run lots of your high discharge cells in series x lots in parallel. I'm running 2x 20s5p batteries in parallel, each with their own BMS. giving me a total of 20s10p @ 450amp max battery amps. The way the battery is built is critical. For high amperage joins, plain nickel won't do. I'm using a copper nickel solution on my series connections good for 100amps.
  3. High discharge BMSs. I've settled on 2x JK 200a BMSs @ 20s/72v. Only delivering 28kw for this build as 40kw was getting out of my budget at this stage.
  4. High power BLDC controller. I've opted for a Fardriver 721200 which will easily deliver 28kw. You will need 1 per motor.
  5. High power BLDC motor. obviously you have different considerations than I do!
  6. Lots of 0AWG cable to transmit 400a! Rated for 150amp continuous, but will be fine for my 5-10sec pulses of 400a.
Further thoughts,

To run 40kw continuous, in my research you should be considering over-rating everything. Heat is going to be your enemy in batteries, controllers, BMSs, and cabling. For your craft I'd also be looking into hub. It's not uncommon for the cheaper ones to fail, and this isn't an option for you. For the number of batteries you will need, I'd be considering active battery cooling too.

I have more thoughts, but this is the essence of them. I'll keep an eye on your project. Looks awesome!

Cheers.
Thanks a lot! My questions are looking far ahead, but I think it wise to know as much as possible about where I’m headed. So I’ll check out all those components you mention.
Can you advise a rough guide as to the efficiency of regenerative braking I might get from motor control systems? And how can “ fallbacks for your main controllers” be constructed? As you noted, failure here would be catastrophic.
 
Thanks a lot! My questions are looking far ahead, but I think it wise to know as much as possible about where I’m headed. So I’ll check out all those components you mention.
Can you advise a rough guide as to the efficiency of regenerative braking I might get from motor control systems? And how can “ fallbacks for your main controllers” be constructed? As you noted, failure here would be catastrophic.
No sorry, I don't know the regenerative efficiency. Someone here would have tested at least some of the more popular controllers I'm sure.... There would be a few ways you could approach the challenge of a fallback, but if it was my project on second thought I'd probably be looking at how other electric aircraft have done it.... Perhaps there is a preferred controller that already is provisioned with a failsafe system...? Getting above my pay grade!
 
I'm perhaps a tad bit late to this thread but I'd just like to clarify a couple things
To supply a theoretical potential of 57kw, I'm looking at something around a 600amp controller - something like the Kelly Controller KLS96601.
The 96601 will only do about 2-300 battery amps, the "600 amps" is a phase amp rating, it is not very well explained on their site but their manual does elaborate more. I can confirm that it will actually do the 300ish battery amperage rating though, mine was doing ~270a@104v nominal pretty comfortably with some zero batteries and a qs205, and I've had it up a little higher.
With respect to the controller - I'm pretty keen on having a regenerative brake function, ideally this would be analog, i.e. variable rate of application - as opposed to a "switch". Does anyone have experienced with high amperage controllers that support this?
As for variable regen, afaik Kelly, 3shul, and tronic are the only consumer oriented controllers that support it at such a high voltage, of course you have curtis and sevcon but I consider them far far out of the consumer sphere due to their price (for example a 96v gen4 size 6 or 8 sevcon is a couple thousand).
 
I appreciate the stream of inspiration that comes through here.

I'm a lifelong builder of things, and next on my list is an electric motorbike.

I am comfortable with 90% of the build I have in mind, inspired by the photo I've attached. It's a BMX style, but obviously electric. it will be for strictly on road use. I've built many frames for many purposes, and I think this will be a reasonably easy one to build.

I have a reasonable amount of experience with electrical engineering, battery building, and specifically high amperage DC so I don't think I need advice on how to put things together at this stage (unless there is some glaring error in what I'm presenting below).. - more what to put together!

For starters, I want things to peak at around 40kw. I may occasionally want this power to turn my frown upside down, but likely only for a few seconds at a time.

Based on this, I'm looking at the 12kw qs273 17x6 hub motor, at the torquiest winding.

I'm thinking 2 x 12s2p banks of gen2 Nissan Leaf battery modules, 96v total. Each bank running a JBD 300a BMS. For a total of 600 theoretical amps and drawing approximately 3.5C from each battery at this peak if possible. (I've attached a photo of the specs of each module)

To supply a theoretical potential of 57kw, I'm looking at something around a 600amp controller - something like the Kelly Controller KLS96601.

Notwithstanding all the other important details, I'd expect that this set up could deliver my desired 40kw, and probably more. What particularly excites me is the promise of 350+N.m of on tap torque. That sounds very fun.

My questions are:

In general are my numbers achievable/reasonable with the kind of price point of the parts I've listed? I've read a lot about 20-30kw but I haven't been able to find a lot about the 40kw+ space. Is it because this kind of power is demanding too much of the readily accessible Chinese parts? Or is it just that 40kw+ just isn't really desired for the investment required?

With respect to the QS 273 12KW motor (17inch 12KW Hub Motor for super Motorcycle, Bike and Scooter...) - is anyone on here running this sort of wattage successfully long-term? I'm not totally opposed to a mid mount, and I'm aware of the advantages - However for this specific build being a street bike the benefits of building ease and convenience I feel outweighs the benefits of a mid mount. The only thing that could swing me as if someone has built a high-performance bike based on this motor and the handling is really bad due to the high rotating weight. I do want this bike to feel snappy and fun, and I definitely don't want to compromise handling over build-ease

With respect to the BMS's - What are people's preferences in terms of brands (and features). I'm not experienced with such high amperage stand-alone BMSs, but are familiar with lithium chemistry and all of the care requirements of it. I'm also unsure if I need to take into account that the batteries I want to use are LiMn2O4.. The voltage cut-offs are the same - but keen to hear from anyone with experience here.

With respect to the controller - I'm pretty keen on having a regenerative brake function, ideally this would be analog, i.e. variable rate of application - as opposed to a "switch". Does anyone have experienced with high amperage controllers that support this? Other than regenerative brake, I'd looking for something that is as plug and play as possible. I prefer to not have to spend lots of time tuning. My preference would be to hook it up, set the basic configuration up and be able to have the controller and motor produce as close to the configured throughput as possible.

With respect to the concept - I realise this is a "heavy" build. I'm 100kg. Acceleration and handling are both high priorities for this build. Top speed, less so (140kmh would be absolute top, 120kmh satisfactory). If lighter hardware and battery would provide similar "felt" torque, I'd love to understand this better.

Thanks everyone, and looking forward to sharing progress as I go.
Just a couple thoughts...

1. If you wish 40kW you will not go bellow 120V, otherwise you risk to melt the copper windings due to heat. At 120V this is up to 300A peak, which is quite a lot. Beyond 300A, even pure copper heats. Bear in mind that at such amp draw, your battery connectors risk melting. Hence you need a smart battery settup.

2. Instead of building one large battery pack, try to make at least two lower amperage batteries. If one connector or BMS blows, you still have the other. It provides redundancy and will not let you stranded on a deserted road with a 200+ pounds ebike. Charging is also faster with twin batteries, you cut the time in half.

3. Ditch the regen. Your motor will make a lot of heat when swallowing those amps. You DO NOT want to add more heat to your stator windings. The amount of electric power you would save is marginal to the risk your motor would be subjected to. Brake pads are way cheaper than electric motors.

4. See if there is already an existing ebike frame which you could improve or alter to have more power. It might save you some time when you do not have to reinvent the wheel. Litterally.

5. When it comes to Chinese specs, accept all that with a grain of salt. The quality varies. Some stuff is really ok, some other components and parts can be pure crap. If they say it is a 12kW motor, try it, test it, you might find out that it provides 12kW only when some ideal conditions are met. Good luck if you need to return a component.

6. The QS273 12kW motor is heavy. Freaking heavy. The motor is around 24kg (52 pounds), the aluminum rim and covers with bolts are more than 8kg (18 pounds). Bear in mind that this is around 32kg or 70 Lbs at your rear end. Forget the small diameter disc and brake if you really want to stop that motor from 40kW power. You need a larger disc and better caliper.

7. Build light, but make it strong. If you need acceleration, then forget the low count windings on electric bicycles (3 to 6). You will probably need 7 to 10 turns (T) in your stator coils. You can reach speed with a higher voltage (120V) up to 130km/h (80 mph) and leave the speed to over 130km/h to the controller if it has that option.

8. Count the weight. The motor, the batteries, the frame, try to get the bike lighter than your own weight without the batteries. Try to design the parts so that you can easily take the batteries out, dissasemble the rear or front wheel for tire change, or to pack your bike in your car trunk. See that you can dissassemble and reassemble the bike in under 15 minutes.

9. Use proven cells, I always use the LG and Samsung 21700 cells which are installed in proven electric cars. deisgn your packs so that they are light and easy to carry, with light handles or straps (I use straps to save weight). Get a decent smart BMS with bluetooth so you can change settings or monitor your battery without having to connect manually.

10. See bikes which are similar to your idea on Ali Baba. If you go through the effort to make a bike, maybe someone else would like this design, too. If they wish to buy such a thing, you'd be the right person to help in making a decision or making an alteration. Maybe you could even make a buck by helping others make their dreams come true.

Just my 2c worth.
 
Just a couple thoughts...

1. If you wish 40kW you will not go bellow 120V, otherwise you risk to melt the copper windings due to heat. At 120V this is up to 300A peak, which is quite a lot. Beyond 300A, even pure copper heats. Bear in mind that at such amp draw, your battery connectors risk melting. Hence you need a smart battery settup.

2. Instead of building one large battery pack, try to make at least two lower amperage batteries. If one connector or BMS blows, you still have the other. It provides redundancy and will not let you stranded on a deserted road with a 200+ pounds ebike. Charging is also faster with twin batteries, you cut the time in half.

3. Ditch the regen. Your motor will make a lot of heat when swallowing those amps. You DO NOT want to add more heat to your stator windings. The amount of electric power you would save is marginal to the risk your motor would be subjected to. Brake pads are way cheaper than electric motors.

4. See if there is already an existing ebike frame which you could improve or alter to have more power. It might save you some time when you do not have to reinvent the wheel. Litterally.

5. When it comes to Chinese specs, accept all that with a grain of salt. The quality varies. Some stuff is really ok, some other components and parts can be pure crap. If they say it is a 12kW motor, try it, test it, you might find out that it provides 12kW only when some ideal conditions are met. Good luck if you need to return a component.

6. The QS273 12kW motor is heavy. Freaking heavy. The motor is around 24kg (52 pounds), the aluminum rim and covers with bolts are more than 8kg (18 pounds). Bear in mind that this is around 32kg or 70 Lbs at your rear end. Forget the small diameter disc and brake if you really want to stop that motor from 40kW power. You need a larger disc and better caliper.

7. Build light, but make it strong. If you need acceleration, then forget the low count windings on electric bicycles (3 to 6). You will probably need 7 to 10 turns (T) in your stator coils. You can reach speed with a higher voltage (120V) up to 130km/h (80 mph) and leave the speed to over 130km/h to the controller if it has that option.

8. Count the weight. The motor, the batteries, the frame, try to get the bike lighter than your own weight without the batteries. Try to design the parts so that you can easily take the batteries out, dissasemble the rear or front wheel for tire change, or to pack your bike in your car trunk. See that you can dissassemble and reassemble the bike in under 15 minutes.

9. Use proven cells, I always use the LG and Samsung 21700 cells which are installed in proven electric cars. deisgn your packs so that they are light and easy to carry, with light handles or straps (I use straps to save weight). Get a decent smart BMS with bluetooth so you can change settings or monitor your battery without having to connect manually.

10. See bikes which are similar to your idea on Ali Baba. If you go through the effort to make a bike, maybe someone else would like this design, too. If they wish to buy such a thing, you'd be the right person to help in making a decision or making an alteration. Maybe you could even make a buck by helping others make their dreams come true.

Just my 2c worth.
That's some pretty good advice there, I'd say the same.

I never had the opportunity to try the 12KW QS273 motor, but I tried the 3KW, 6Kw and 8KW versions.
I was able to push about 22KW through the 3KW motor and I seem to recall around 35KW with the 8KW motor. I still don't know how much the 6KW can take as I just installed it on my bike a few days ago, but should be something in between.
So I'm fairly confident the 12KW can take 40KW easily.

It is a very heavy motor indeed, in my opinion it might be a problem for a BMX frame build, the bike will be very unbalanced, that can be quite dangerous.
In my opinion the OP possibly underestimates what 40KW of electric power might feel like. A BMX would already proper scary with just 20KW.
 
Taking into account you are going to be running three props.

In my research, this is what I've come up with:
  1. High amperage battery. This is done by using lots of high discharge cells in parallel. Check out molicel p28a for example @ 45 amps.
  2. High voltage. 72v - 108v typical. So, run lots of your high discharge cells in series x lots in parallel. I'm running 2x 20s5p batteries in parallel, each with their own BMS. giving me a total of 20s10p @ 450amp max battery amps. The way the battery is built is critical. For high amperage joins, plain nickel won't do. I'm using a copper nickel solution on my series connections good for 100amps.
  3. High discharge BMSs. I've settled on 2x JK 200a BMSs @ 20s/72v. Only delivering 28kw for this build as 40kw was getting out of my budget at this stage.
  4. High power BLDC controller. I've opted for a Fardriver 721200 which will easily deliver 28kw. You will need 1 per motor.
  5. High power BLDC motor. obviously you have different considerations than I do!
  6. Lots of 0AWG cable to transmit 400a! Rated for 150amp continuous, but will be fine for my 5-10sec pulses of 400a.
Further thoughts,

To run 40kw continuous, in my research you should be considering over-rating everything. Heat is going to be your enemy in batteries, controllers, BMSs, and cabling. For your craft I'd also be looking into fallbacks for your main controllers. It's not uncommon for the cheaper ones to fail, and this isn't an option for you. For the number of batteries you will need, I'd be considering active battery cooling too.

I have more thoughts, but this is the essence of them. I'll keep an eye on your project. Looks awesome!

Cheers.
What model JK BMS did you get? Test it out yet? Pros/cons? Thanks for the details.
 
With these hub motors running at such high power I would have thought there would be issues with the axle breaking seeing as all of the torque is running through it. Has anyone ever heard of these failing?
Another good reason not to use regen braking too as the constant twisting back and forth would make that axle even more prone to fail.
 
Back
Top