Leaf / leafmotor / leafbike high efficiency 1500w motor

[ebike4healthandfitness]

Not really. The choices are so random that there is not clear use case, and just a random wish list. You’re welcome to go the twenty questions route. (I didn't suggest any ebike parts. I only expressed my preference to see an M635 build).

The poster is asking for help and is unsure about many things. The choices are not random. There is a clear use case. It is not a random wish list.

 

[neptronix]

For my next bike I’m looking at either a torque sensing Bafang M635 torque sensing BBSHD or a Leaf 1500w.

Can anyone advise how expensive and time consuming, complicated to set up the Leaf will end up with

- regen braking
- a controller capable of 4-5kw bursts
- dual battery setup potentially (eg 2 x 48v or 52v batteries run in parallel for double amps of nominal rating l, plus regen on dual batteries if they are paired correctly)

I think there were some other factors I was going to consider but can’t remember now. I’m leaning towards the Leaf for ease and simplicity with drivetrains, but am not sure if efficiency will be all right with a fair bit of serious hill climbing.

The other option to wait for is the Proton mid drive from CYC capable of 3kw+, But it will be expensive and a long wait

If you're using a 6T you're not going fast, so it's okay to use a lesser battery. You'll be abusing the max output for much shorter periods of time versus using a 4T which will suck ~2x the amps continuously at top speed.

Buy the motor wheel, two torque arms from ebikes.ca, a VESC from spintend, and the largest battery ( 20AH+ ) you can buy and after a little wiring and programming the controller, enjoy.

Hub motor bikes can be built in a day and rarely require any maintenance. You might like that versus a mid drive.
You might not like it if you have protected climbs above a 7% grade. The leafbike can handle some hills, but it is still a mid sized hub motor at the end of the day.

 

[E-HP]

The poster is asking for help and is unsure about many things. The choices are not random. There is a clear use case. It is not a random wish list.

Well maybe it's pie in the sky when it comes to the motor choices relative to the criteria.

"Bafang M635 torque sensing BBSHD or a Leaf 1500w."

- regen braking
- a controller capable of 4-5kw bursts
- dual battery setup potentially (eg 2 x 48v or 52v batteries run in parallel for double amps of nominal rating l, plus regen on dual batteries if they are paired correctly)
- a fair bit of serious hill climbing.

There's a lot of folks trying to chip in on supporting the many newbie questions. I'm going initially with reviewing the post at face value, assuming the posters have stated what they want, either in the current thread, or their other threads on the same subject, rather than go through the long Q&A. It's easier for me to read than type, so looking at the info sprinkled around in their posts is my approach. For example, if the poster want 5kW for whatever reason, I'm not questioning it.

I think the Leaf could probably fill all of the requirements, where the other two can only meet one (being able to run parallel batteries). But not enough info to tell, so I stated my preference. I bet you'd like to see an M635 build too.

The Q&A method may be preferable, but we're all just trying to help out, so I think there's room for different approaches.

"In my testing the P45B was a better performer in every way over the Molicel P42A:

The P45B had almost 9% higher capacity.
The P45B had about 33% lower DC IR (at 100% charge).
The P45B’s charge rates are 4.5A standard/13.5A max (70°C cutoff) and the P42A’s are 4.2A standard/8.4A max (60°C cutoff).
The P45B had longer run times at all discharge current levels."

I've been looking at these spec sheets so many times after reading through Eastwood's battery build thread. He's using the P42A. Any of these cells, except the 50E, could meet my needs for capacity and current capability, but I'm allergic to voltage sag. Regardless of how capable a cell/pack is, I think voltage sag is the biggest robber of performance. Is there a way to determine voltage drop of these cells using the discharge graphs provided in most spec sheets? Is it as simple as taking the state of charge and comparing the discharge curves of a low discharge scenario and a high discharge scenario, and doing the math?
I see around 0.6V between the 0.2C and the 45A curves (0.4V in the voltage range/state of charge I'm usually running my pack at), so does that mean for 20S5P, the approximate sag for the pack would be 20 x 0.6 = 12V when discharging at a 225A?
I'm going to do another review of the curves. I'm thinking the cells with the greater capability will sag less, but for me it's worth checking.

 

[harrisonpatm]

I see around 0.6V between the 0.2C and the 45A curves, so does that mean for 20S5P, the approximate sag for the pack would be 20 x 0.6 = 12V when discharging at a 225A?

Yes, but are you actually going to be discharging that much, on a regular basis? If so, and if you really want to avoid voltage sag, your best bet is just going to be using more cells in parallel. Can you fit 7p, 8p? At 8p, 225A would be shared to 28amps per cell

 

[E-HP]

Yes, but are you actually going to be discharging that much, on a regular basis? If so, and if you really want to avoid voltage sag, your best bet is just going to be using more cells in parallel. Can you fit 7p, 8p? At 8p, 225A would be shared to 28amps per cell

I can barely fit 20S5P, without a BMS, in my triangle, and won't be discharging anything close to that. I'd like less than 5V-6V sag at 100A max discharge. I'm eyeballing about 0.3V between the 20A and 0.2C curves at a state of charge of 3000mAh on the graph, so for a 20S5P pack at 100A, that would be 6V. Maybe you could check my math, because that sounds really good.

 

[harrisonpatm]

I can barely fit 20S5P, without a BMS, in my triangle, and won't be discharging anything close to that. I'd like less than 5V-6V sag at 100A max discharge. I'm eyeballing about 0.3V between the 20A and 0.2C curves at a state of charge of 3000mAh on the graph, so for a 20S5P pack at 100A, that would be 6V. Maybe you could check my math, because that sounds really good.

That seems about right. Like you say, it's an eyeball measurement for now, because sag will also depend on how thick your cables are

 

[ebike4healthandfitness]

There's a lot of folks trying to chip in on supporting the many newbie questions. I'm going initially with reviewing the post at face value, assuming the posters have stated what they want, either in the current thread, or their other threads on the same subject, rather than go through the long Q&A.

He never said he wants the Leaf or the Bafang M635. He merely said he is looking at the Leaf and the Bafang M635. He also has questions about the Leaf. He even mentions waiting for CYC Proton as an option at the end of the post (despite him saying it being expensive and a long wait) because he is not sure about either Bafang M635 or Leaf 1500w (due to concerns the set-up he wants is time consuming, expensive, complicated and as well as having lack of efficiency during a "fair bit of serious hill climbing").

 

[E-HP]

He never said he wants the Leaf or the Bafang M635. He merely said he is looking at the Leaf and the Bafang M635. He also has questions about the Leaf. He even mentions waiting for CYC Proton as an option at the end of the post (despite him saying it being expensive and a long wait) because he is not sure about either Bafang M635 or Leaf 1500w (due to concerns the set-up he wants is time consuming, expensive, complicated and as well as having lack of efficiency during a "fair bit of serious hill climbing").

So you have a problem with me stating that I’d like to see an M635 build, but not offering a suggestion to the poster? Is there a rule I’m missing or do you have an issue?

 

[vladrov]

I have a HUGE problem - they can't find a longer axle.

Do you have any ideas on what can be done? How do you fit two torque arms on both sides? Are your axles also this short?


The math again:

30 mm on the cassette/freewheel side, 35 mm on the disc side.

1.Minimum 1 mm washer between the frame and the motor
+
2. 11 mm frame (or to be exact, even more ~ 11.5 mm)
+
3. 11 mm Grin Torque Arm V7

This means - 7 mm for the nut on the cassette side and 12 mm for the disc side.

The thread pitch of M14 is 2 mm. This gives a maximum of 3 turns of the nut on the stronger side and around 5 on the weaker cable side.

And this is assuming everything fits perfectly with no tolerance or buffer. I don't see how the torque arm on the cassette side will fit and that's the stronger side.

Do the Chinese not think at all when they build something, there must be engineers out there? And why is the disc side the longer side when a torque arm is needed on the solid side? Very frustrating.

 

[Chalo]

If you use a clamping torque arm on the cassette side, it can be extra thick because you won't need a nut on that side.

If you have some threads to work with, you can use the nut to compress the stack before tightening the TA. But you don't have to get it tight enough to risk stripping threads, only enough to squeeze out any gaps.

 

[scianiac]

This is similar to how my Leaf is setup, both sides have the nut barely threaded on, one only about 3mm and the other a bit more but they are really only to compress things while I tighten the massive stainless torque arms to a crazy clamping force. The torque arms are in turn also clamped around the frame. Each arm is 6mm thick with 2 on each side so 24mm total of clamping area with two M8 class 10 bolts torqued to 22 ft-lbs I'm not worried about the axle going anywhere.

 

[ebike4healthandfitness]

So you have a problem with me stating that I’d like to see an M635 build, but not offering a suggestion to the poster? Is there a rule I’m missing or do you have an issue?

There should be a rule because you are not helping the poster and instead you are stating what you want from him when he never asked you for that. Honestly he would be better off going to another forum.

 

[neptronix]

Well technically they are not out yet. It was leaked in a annual report you can read here published back in March. Estimated completion of R&D in 2024. Those figures should be taken with a grain of salt but I would bet if they are off its not by far considering where I found that photo was published before the 50B figures were out. But the 60 in the same suggests 6ah and the "P" nomination is for their high discharge cells. Probably will be another year or two before mass market release though. P50B's has had samples out since the beginning of the year (at least) and still hasn't hit mass market, most guesses seem to be EOY. Also worth noting is the crazy XA3 cells they are working on. 4ah and 90c discharge and a 40c charge (seems to be a 10s peak though). https://media.taiwancement.com/web_tcc/en/report/annual/2023 Annual.pdf

Thanks for the info, that's why i haven't heard of them. Still pretty amazing.
P45B is still a good cell and if i was building a 21700 pack for a leafbike 4T & the cost wasn't insane, it would be my cell of choice i think - impressive piece of engineering
https://media.taiwancement.com/web_tcc/en/report/annual/2023 Annual.pdf

Also >5ah 21700's have been a thing for a bit now. First one that comes to mind is the LG m58, 5.8ah which has been out 2 years now. And then what ever cell Vapcell rewraps for their F60 6ah. Although those cells are not rated for high amps, and to really get more capacity out of them vs your typical 5ah cell they need to be ran at really low amps (<5). What could really be interesting though is from my understanding if you build a high discharge cell like the P60B you are sacrificing capacity in the design. So the question is if Molicel works on a high capacity cell (which they designate with "M") what would the capacity be? Possibly 7ah, maybe even 8ah?!?

Yeah i'm aware of those and not too excited about them because we're losing power density to energy density.. only useful in a medium to high speed scenario if you're willing to build a honker of a battery.

 

[E-HP]

There should be a rule because you are not helping the poster and instead you are stating what you want from him when he never asked you for that. Honestly he would be better off going to another forum.

Fair enough, you're right. I acquiesce. I was being flippant, maybe a bit punchy from not sleeping. I do want to see an M635 build though.

Can anyone advise how expensive and time consuming, complicated to set up the Leaf will end up with

- regen braking
- a controller capable of 4-5kw bursts
- dual battery setup potentially (eg 2 x 48v or 52v batteries run in parallel for double amps of nominal rating l, plus regen on dual batteries if they are paired correctly)

Op, my apologies. Here is the cost breakdown of my bike as an example. The current controller doesn't support regen, but the previous two did. I'm learning to live without it.

• In blue are the ebike specific conversion costs.
• In white are the bike costs, including tires. The frame and fork were Craigslist finds (I recall when I was mountain biking, the frame was $1.5-$2k when it came out, so I called as soon as I saw it).
• In yellow represents my last minute decision to relace my hub to 24". I guess the front rim ($82.00) should have been highlighted as well.

The lipos in parallel with my triangle pack has changed to two 7S 5Ah packs, but otherwise costs are complete, but I just noticed my rear rack isn't represented.

In your case, you'll need to substitute your battery costs and figure out how and where you'll be mounting them. You're in the best position to figure that out when you have the bike in front of you. You might want to mock up something out of cardboard or posterboard to see where/how they may fit. You don't say how many Ah you are planning to have. My guess is that you'll spend more on batteries than my setup if you go with anything decent, and a better controller.

Currently my bike peaks right around 7kW. I hit 9kW off the line once using the last controller.

As far as effort goes, you could assemble everything in a day or two*; another day if you're lacing your wheel.
*You can install everything in less than a day, but you may need/want to do/fabricate something custom for things like your torque arms, etc.

As far as hill climbing goes, for any motor type, the key is to keep is spinning fast enough to keep it cool. Airflow or additional cooling can help mitigate heat, but the slower your bike needs to go, the more you need gearing to keep the motor spinning fast enough. With Statorade, the Leaf can climb the steepest paved streets in my area without overheating, if I keep my speed to 20mph (~1 mile climbs). It will climb steeper-shorter hills on dirt at around 10mph-12mph, but can't do steep slow switchbacks since you're basically stalling the motor at that point. That's where you need a mid drive and gearing to keep the motor spinning fast enough.

My (paved) test hill. Starting temp 37C, temp at the top 98.7C. Statorade for cooling.

https://endless-sphere.com/sphere/attachments/current-cost-092522-jpg.323886/

Offroad temperature testing. Motor is at 48.7C and heats up quickly since replenishing the Statorade was overdue. It really should have made it up both of the hills otherwise.

 

[Mongo]

I do want to see an M635 build though.

Hi @E-HP, I don't want to distract from this thread's main topic, but related to the M635 motor, High Voltage reported in a recent video that several sources state that Bafang has stopped production on the M635 motor - for now, forever, related to production resource priorities, who knows.

 

[E-HP]

Hi @E-HP, I don't want to distract from this thread's main topic, but related to the M635 motor, High Voltage reported in a recent video that several sources state that Bafang has stopped production on the M635 motor - for now, forever, related to production resource priorities, who knows.

Thanks. Kind of a double bummer, since in his videos it doesn't seem like it worked as well with the stock controller as when he converted to the ASI controller. He didn't have the sound turned up, but I'm guessing it would be similar to the BBSHD for noise, which is about the amount of noise I'd be willing to put up with.

 

[E-HP]

Hi @E-HP, I don't want to distract from this thread's main topic, but related to the M635 motor, High Voltage reported in a recent video that several sources state that Bafang has stopped production on the M635 motor - for now, forever, related to production resource priorities, who knows.

He posted another video yesterday explaining what he had heard, which is basically that there’s a lot of excess inventory since ebike demand has gone way down after the pandemic. Makes sense, since my neighbor just bought another factory ebike and mentioned how cheap it was and how there are a lot of deals now. It’s a Brose powered mid drive for not much more than a no name ebike.

 

[yantra yatra]

For my next bike I’m looking at either a torque sensing Bafang M635 torque sensing BBSHD or a Leaf 1500w.

Can anyone advise how expensive and time consuming, complicated to set up the Leaf will end up with

- regen braking
- a controller capable of 4-5kw bursts
- dual battery setup potentially (eg 2 x 48v or 52v batteries run in parallel for double amps of nominal rating l, plus regen on dual batteries if they are paired correctly)

I think there were some other factors I was going to consider but can’t remember now. I’m leaning towards the Leaf for ease and simplicity with drivetrains, but am not sure if efficiency will be all right with a fair bit of serious hill climbing.

The other option to wait for is the Proton mid drive from CYC capable of 3kw+, But it will be expensive and a long wait

Thanks for the feedback and apologies for the lack of nitty gritty details and understanding. I’m used to mid drives and have been hesitant to go to the leaf hub motor.

I gain about 300m over 2km on one of the semi-regular rides. But usually the hills on my rides though steep are only for about 500m length. So I think the Leaf should handle it.

The other features I am looking for are
- torque sensing
- temp sensor

It sounds like maybe dual battery with regen is a bit too complicated at the moment but I’d like that capability for regen (eg when just using one battery), and also capability to draw from dual batteries - eg with current 48v 17.5ah, to get double amps.

It seems like the cost is similar to or slightly more expensive than the M635. The complication seems to be more than the M635 for installation, finding the right components, and less for regular maintenance.

I’m really interested particularly in the relative price and difficulty of setup of the Leaf compared with the M635. I don’t want to dedicate a huge amount of my life to it and want something that goes without huge amounts of learning and maintenance required. Sounds like the M635 is being cancelled though so maybe the Leaf is the way to go.

Have a BBS01, BBS02 (underpowered, no torque sensing, drive train issues) and a T07 DM01 (failed motor, motor swung up and damaged the frame, and would like more power ideally).

(Already have bikes to put a motor on, a couple of batteries that will likely get replaced sometime soon. I’m 70kg. Would go for the 4t Leaf on 26” wheels)

 

[ebike4healthandfitness]

Thanks for the feedback and apologies for the lack of nitty gritty details and understanding. I’m used to mid drives and have been hesitant to go to the leaf hub motor.

I gain about 300m over 2km on one of the semi-regular rides. But usually the hills on my rides though steep are only for about 500m length. So I think the Leaf should handle it.

The other features I am looking for are
- torque sensing
- temp sensor

It sounds like maybe dual battery with regen is a bit too complicated at the moment but I’d like that capability for regen (eg when just using one battery), and also capability to draw from dual batteries - eg with current 48v 17.5ah, to get double amps.

It seems like the cost is similar to or slightly more expensive than the M635. The complication seems to be more than the M635 for installation, finding the right components, and less for regular maintenance.

I’m really interested particularly in the relative price and difficulty of setup of the Leaf compared with the M635. I don’t want to dedicate a huge amount of my life to it and want something that goes without huge amounts of learning and maintenance required. Sounds like the M635 is being cancelled though so maybe the Leaf is the way to go.

Have a BBS01, BBS02 (underpowered, no torque sensing, drive train issues) and a T07 DM01 (failed motor, motor swung up and damaged the frame, and would like more power ideally).

(Already have bikes to put a motor on, a couple of batteries that will likely get replaced sometime soon. I’m 70kg. Would go for the 4t Leaf on 26” wheels)

If T07 DM01 (a torque sensing 1000w mid drive) failed (and you also say you would like more power than T07 DM01) I think replacing it with another 1000w motor (the M635) is not a good idea.

P.S. Do you still have your 1000w rear hub motor?

 

[E-HP]

The other features I am looking for are
- torque sensing
- temp sensor

I have a torque sensor. Just too lazy to install it. There are a few options for adding a torque sensor, outside of the mid drive options. A Cycle Analyst plus any controller is one option. Lishui and KT controllers have options including open source firmware.

A temp sensor and voltmeter are required data, in my opinion. Depending on your topography, regen can really extend your range, and influence what routes you take. But you need the temp sensor if you're using regen because it can heat up your motor pretty quickly.

Here's a few options for torque sensing collected from various threads:

Displays and/or Controllers supported by Open Source Firmware (custom/enhanced firmware developed by the user community):
Controllers -
KT controllers (certain models 0.25kW up to 5kW)
Lishui controllers (certain models)
Displays -
KT LCD3 used in conjunction with TSDZ2 mid drive
VLCD5-VLCD6-XH18, LCD3, 860C-850C-SW102 used in conjunction with TSDZ2 mid drive

 

[ebike4healthandfitness]

Yantra yatra have you thought about using your BBS02 with your 1000w DD hub motor?

BBS02 with shorter than stock crank arms combined with your 1000w DD hub motor + statorade + higher power than stock controller is probably your best bang for the buck.

 

[Chalo]

Yantra yatra have you thought about using your BBS02 with your 1000w DD hub motor?

BBS02 with shorter than stock crank arms combined with your 1000w DD hub motor + statorade + higher power than stock controller is probably your best bang for the buck.

Hub motor side covers often break just from vigorous pedaling, even without superhuman pedal power. I wouldn't risk it.

 

[E-HP]

Hub motor side covers often break just from vigorous pedaling, even without superhuman pedal power. I wouldn't risk it.

I would think either the eight tiny screws, or the aluminum they are screwed into, would fail first.

 

[ebike4healthandfitness]

I would think either the eight tiny screws, or the aluminum they are screwed into, would fail first.

This is what he is talking about:

Hub motor face plate cracking

Hi all I am having a problem with my rear wheel hub motor face plate cracking at the bace of the Freewheel. I am on my second faceplate. Is any one else had this problem. I have read that it's the extra Talk put on by me peddling but my bike is made commercially it's a Fat bike with 500W rear...

endless-sphere.com

Its mainly from using the outermost gears (i.e the tall gears) and putting a lot of torque on the gear through the chain. This because the outermost gears are furthest away from the supporting bearing on the axle and when a lot of torque is used the axle bends. On a non hub motor with 10mm axle this is a very common occurrence (resulting in permanent bending or breakage of axle) and even the ones made of Chromoly still do this (switch to freehub fixes this problem).

 

[E-HP]

This is what he is talking about:

Hub motor face plate cracking

Hi all I am having a problem with my rear wheel hub motor face plate cracking at the bace of the Freewheel. I am on my second faceplate. Is any one else had this problem. I have read that it's the extra Talk put on by me peddling but my bike is made commercially it's a Fat bike with 500W rear...

endless-sphere.com

Its mainly from using the outermost gears (i.e the tall gears) and putting a lot of torque on the gear through the chain. This because the outermost gears are furthest away from the supporting bearing on the axle and when a lot of torque is used the axle bends. On a non hub motor with 10mm axle this is a very common occurrence (resulting in permanent bending or breakage of axle) and even the ones made of Chromoly still do this (switch to freehub fixes this problem).

Well that's a good excuse for pedaling less...for safety reasons of course lol.