'lightest.bike' 1.7kg 1000w mid drive

[neptronix]

Oh if we want to talk about single geared drives then we should also mention:
Aden Powerkit Sport V2 - 1500W Spoke drive
This has a 500w max version that's a bit cheaper.

 

[glennb]

That kind of torque is better not run through geartrains, but spokes mightn't enjoy it either.

It's hard to get excited about mid-drives ...they're so imperfect due to operating too far upstream - they inflict unnecessary strain and wear on chain, sprockets, freehubs and spokes, So use a hub motor? For high power, sure, but crap for low power and climbing applications.

Tyre friction drive is sensible, except the tire is a poor interface. Rim friction drive is an improvement. Add a "toothed" profile to the rim wall and you're in business. Could be nasty noisy though, and requires grease, and greasy rims would be a hard sell, despite making for simple maintenance.

Anyway, enough rant, thanks for the tips. I had no idea about these new products. I'll check them all out.

 

[neptronix]

Don't want to take this too off topic.. but if you think about it.. everything is a spoke drive in the end..
In gasbike world it's customary to send a few kW through a rag joint and expect it to work.



The lightest.bike drive is programmed in a way where the usually huge low RPM torque is muted ( it has more like an ICE engine's torque curve ). This may mean that it's easier on drivetrains than say, a BBSHD.

Utilizing bike parts for the bike drivetrain means that you could in theory use a stronger drivetrain... with most mid drives, you are limited to whatever choices the manufacturer made.

 

[glennb]

Back off topic ... I think rim tire/rim drives avoid spoke stress. Same way radial lacing is acceptable in 700C wheels with rim brakes, but not with discs.

Still, Lightest is interesting. And if it's gentle with the torque ramp could be interesting with IGH.

I've got an Alfine 11 Gates belted commuter bike that I practically never ride. The belt and oil bath alleviate wear concerns, carbon belt is ridiculously strong, can select any gear instantly with a twist of the wrist ... but hub internals torque limits are an unknown.

 

[limiky]

Hi,
I read all the 42 pages here about the Lightest motor conversion kit.
I’m an owner myself and i put roughly 50 hours of working time on the unit/600km.
I live in Italy, so for me no problem having my motor sent in few days, buying it directly from the Bikee Bike website.
Installation was at first simple, then I figured out i needed some custom work to be done in order to fix small problems.
I bought the maxed out off-road/high-torque/superhero version of the unit, and installed on a capable DH mtb which i couldn’t pedal because of it’s gravity-only oriented geo (but i wanted to use it without paying shuttle services daily).
The only real problem the unit grave me was breaking the internal freewheel when under heavy load.
It happened twice (fast 3 days warranty repairs both times from manifacturer).
Those events happened on some of the rockiest dh tracks in Italy i think, so i thought that it happened because of chain growth under compression of the rear shock (205mm rear travel, almost 2cm chain growth on my banshee legend 650 XL) while the unit was under pedalling load (the internal freewheel is a needle Roller bearing that doesn’t see many Nm from the motor, but if you add the sudden kickback…) so i 3d designed a pair of custom motor brackets and made them cnc’d in heavy thick 6mm stainless steel 304 (overkill, but on a gravity focused bike i didn’t care about weight penalty). Theese brackets have integrated upper and lower pulleys, the lower maximises chainwrap, the upper takes kickback to full zero.
This makes the bike bob with hard pedalling, and no more pedalling platform. You can imagine antisquat and antirise graphs now..
But there is no more chain growth, so no additional stress on the motor internal freewheel.
Also, i wanted a somewhat large 200mm q factor because i’m accustomed to it coming from Brose equipped e-bike, so i cnc’d a custom spindle and fitted crancks i liked more from another ebike conversion seller (with their high POI freewheel, that has no wobble at all).
Drivetrain runs perfect, smooth and quiet despite more drag, but the bike as it was originally was not made to be pedaled too.

Now, for the third time the needle roller one way freewheel bearing broke, and i cannot give the fault to chain growth.
The manifacturer replaced it for the third time, in 2 working days (they were crazy fast and operative as always).
Motor was used full power all the time, stats on the display say max amp used were north of 23, with a 14s 52v battery.
Peak power 1119w, in line with what is expedted.

The manifacturee made a software update regarding the throttle, now i can use it only if I put some force on the pedals, so the motor won’t see instant torque from a standstill situation (both with the bike moving or not), it will only spin the motor if some torque is sensed through the pedal torque sensor.

I’m here to share thought on this motor unit and, if the freewheel keeps working and won’t break again, i’m trying to understand why the controller is limiting to 23 max amp the motor current (slider goes to 35A, but with Race enabled the controller should let max amps to the unit, and that doesn’t happen).

My actual battery is a BMW american made 14s2p molicel p50b cells, light and small, and the only one that fitted in the triangle of my dh mtb (safe from rocks and central weight over bb, also i had to cnc an enclosure for it because the bike has no bosses to attach the usual battery sled).

Hope to receive some feedback on the amp related/controller question!

I’m attaching some photos of my bike with the motor installed also.

 

[bedesign]

This - "upper pulley"
BTW, impressive work; every household should have a CNC machine in the wardrobe.
In my case, I will implement that only as a non-tensioner, down movement limiter (as I mentioned before).

I also have made 600+ Km, and had an incident where in a sudden situation, I had to by a reflex quickly grab the bike so
the throttle got activated, pushing the front HARD against a wall.
Luckily nothing visibly broke, but it adds up to my and the general wear.

"The manufacturer made a software update regarding the throttle, now i can use it only if I put some force on the pedals,
so the motor won’t see instant torque from a standstill situation (both with the bike moving or not),
it will only spin the motor if some torque is sensed through the pedal torque sensor"

I think we would like to have this generally available for both safety and mechanical and body longevity reasons,
but suggested with the following twist

DEFAULT SETTING:
Pedal use + Throttle = unrestricted

Only Throttle = power and speed limited (6 Km/h ?),
but just enough power to haul a loaded bike uphill
while walking along.

OPTIONAL MODE:
Unrestricted Throttle Use -
for like in an emergency and hurt and unable to
use by pedaling.



Regarding limited current -
For the (14s)2p molicel type with 13mΩ internal res.
35/2P Amp would add 3.9 W of heat to each cell,
while (not specific to molicel) the general IDEAL recommendation is < 5W/cell,
if there is no active (liquid/fan) cooling system.

So there is limited margin for hot weather.

 

[neptronix]

@limiky
Really useful report. Nice bike, too.

When you say internal freewheel, you mean the super huge plastic gear inside the unit, right?
Sounds like it can't handle 23a. I really wonder why they have an option to select 35A now.

Mine is limited to ~19a and only outputs 900w on a 52v battery. I have everything but the high torque upgrade.
The power settings on the screen do absolutely nothing on my unit also.

Extremely disappointing to know that neither of us are getting the power promised, and the drivetrain is the limiting factor.

Let me ask you something about your drive since you have more power than me.
When you hit the throttle from a stall, do you get a jolt of power, like a geared hub motor, or does it take a long time to develop, like a gasoline engine?

On my 1000w unit, my throttle response is more like a gas engine; weak to start; good power at high RPM.

 

[limiky]

For the software mod on the throttle, that i'm trying theese days (to see if my internal freewheel finally stays in one piece), you can ask directly to the manifacturer and they will sens you a qr code, so you can update the firmware by your self. If you want to go back to unrestricted throttle use only and use your bike as a motorcycle, if you ask them they will give you also the qr code of your actual software. Then you can try the new setting, and go back if you want, just by scanning the code and applying with their simple bt procedure.

Motor amp current peak i've seen today is 28.3amps @ 56.6 volts. Clearly under load there was a slight voltage sag. So if the math is correct, 28a x 56v is 1568w.
Battery current said 30amps, slightly more than motor amps.

So in my case, the motor does the power it is said it gives when i bought into it, thats plenty for me.

Medium ascent velocity if i push enough with my legs is more than +2100m/hour. Surely at this rythm the battery dies before doing them, but if you have to hurry for a ride after work, you can be really fast uphill.

For the cnc work, i only 3d sketched and modeled. All the machining work was done in a cnc european factory out of Italy.
It costed a small fortune btw, but performance in suppleness of the rear triangle went up as the chain doesn't interfere with the shock movements.
Now the bike always works as chainless, and that's good on the descents, just don't try to pedal it.
Let alone the weight increment with motor and battery added when battery finishes, you can only point the feont wheel down..

If your frame tubing is round and allows it, the simplest solution to chain wobble up and down is a chain guide clamped to the tube, on my frame this was not possible so i went with what you saw.

I read about another user that cannot use more than 900w of power, i think if he contacts the manifacturer, with a qr code they can mod the motor amp draw in two minutes.

 

[limiky]

@limiky
Really useful report. Nice bike, too.

When you say internal freewheel, you mean the super huge plastic gear inside the unit, right?
Sounds like it can't handle 23a. I really wonder why they have an option to select 35A now.

Mine is limited to ~19a and only outputs 900w on a 52v battery. I have everything but the high torque upgrade.
The power settings on the screen do absolutely nothing on my unit also.

Extremely disappointing to know that neither of us are getting the power promised, and the drivetrain is the limiting factor.

Let me ask you something about your drive since you have more power than me.
When you hit the throttle from a stall, do you get a jolt of power, like a geared hub motor, or does it take a long time to develop, like a gasoline engine?

On my 1000w unit, my throttle response is more like a gas engine; weak to start; good power at high RPM.

I'm refereing to the needle bearing press fitted and (maybe) glued on the inside of that big toothed plastic wheel. With 28.3 amps of motor draw and use, i can confirm the plastic wheel is new looking and fresh, so it is plenty strong (at least for continuous 28a). It's the needle roller bearing that went bad.

To your question on throttle response: now the throttle on my bike is able to run power only if i first push the pedals and activate the torque sensor.
When i used it in its original config, it was simply nervous, insanely fast to make the motor spin. No lag as an ice (at least not comparable).

Motor front sprocket is 10t. Bike wheelied with central weight in rear cogs from 28t and bigger, not doing anything to make it wheelie, only small seated movement to the back of the bike, it happened till 15km/h or so.

So no,
It performed as a fast spinning bldc, and it does now if i push on the throttle and then spin the pedals to activate the torque sensor (difficult to perform, but you can). Torque is released immediatly, if i stand an the pedals and i do this, my butt wants to be thrown at the ground.
It's a sudden response in torque i'd say.
Then clearly it is maintained, amd never fades, till max rpm for the given rear cog you are using, but it doeasn't take much time.

If you ask to the company they can sens you the updated qr for full power and use, on my motor i see daily 28 amps.

 

[limiky]

@limiky
Really useful report. Nice bike, too.

When you say internal freewheel, you mean the super huge plastic gear inside the unit, right?
Sounds like it can't handle 23a. I really wonder why they have an option to select 35A now.

Mine is limited to ~19a and only outputs 900w on a 52v battery. I have everything but the high torque upgrade.
The power settings on the screen do absolutely nothing on my unit also.

Extremely disappointing to know that neither of us are getting the power promised, and the drivetrain is the limiting factor.

Let me ask you something about your drive since you have more power than me.
When you hit the throttle from a stall, do you get a jolt of power, like a geared hub motor, or does it take a long time to develop, like a gasoline engine?

On my 1000w unit, my throttle response is more like a gas engine; weak to start; good power at high RPM.

One other thing, the motor is advertised as 1000w.
With my 28 amps motor draw, at 52v nominal, it gives 1452w.

So i am using the power promised, slightly more.
The problem is me, i'd like to try with 35A, just beacause i read on the settings it is possible.
Today i contacted the manifacturer and they said it is possible to use more amps, but they are reclutant doing it now, and for a good reason: i destroyed 3 freewheel bearings till now with the power it has now.
Who knows what will happen making the motor do 1800+ watts.

I already went in the rabbit hole of the designing another freewheel, in the future, if it will fail again, i'm planning to use a commercial csk40pp bearing on the inside of the plastic reduction wheel, safe for 325Nm of load.
Overkill probably, but i could return to throttle use only without braking it.

The display only reads 1119w if i search for this stat on it, but math on the display says this, and feel says something else (more on the 1400w ish as 28x52 gives as result).

At least my display says i make full use of 1000w advertised, so write the company and have your update!

 

[maiz]

Also, i wanted a somewhat large 200mm q factor because i’m accustomed to it coming from Brose equipped e-bike, so i cnc’d a custom spindle and fitted crancks i liked more from another ebike conversion seller (with their high POI freewheel, that has no wobble at all).

I wonder when they will improve this situation

 

[limiky]

I wonder when they will improve this situation

In my unit, the wobble was not much: roughly 2mm from left to right in a full rotation of the crancks.

But i have to be more precise: this was a non problem for me, i changed only beacause i needed a 200mm large Q factor platform.

I say this was a non-problem because the wobble could be seen only rotating the freewheel spider backwards (CCW) by hand, and dragging with it the crancks.
With the cranks stationary, and rotating the freewheel spider clockwise as when you operate the motor only with throttle, there was zero wobble.
The chainring probably works an a plane inclined by some degree from full perpendicular to the spindle axle, but no problem in chain derailment. The chain always stayed up in my case, even with the original freewheel and cranks.

Also, backpedal is non existent, the freewheel relases the spider from the cranck if you try. So i could only see the problem when no chain was on, and when i could spin the spider-crancks assembly ccw pushing on the chainring.

Eventually, in my opinion nothing to be fixed by the company regarding axle, axle bearings, isis spindle, crancks, freewheel and spider assembly for freewheel.

P.s: by manifacturing new cnc'd brackets that support the motor, i made the central bb "hole" large enough to accomodate bigger bb cups (external ones, diameter 52mm, stainless steel bearing, from hope). The seat for the isis bb cups are only like 43-44 mm in diameter so external bb cups won't fit (they will, but not enought thread engaged on frame shell, and won't go fully in contact with the bracket).

 

[neptronix]

Okay, i think i had some misunderstandings based on your first post.
So you think the screen is lying, hm? i never gave a second measure on mine.

Thanks for letting me know about this code. I didn't get help from the company after asking 3 times so i gave up asking.

So you are blowing this freewheel at 28A or so, it sounds like you don't have the gear preserving tune which reduces the low RPM torque... but since the power controls in the app don't do anything, i don't think you can do that.

What would be the correct thing to do is dramatically reduce the phase amps to battery amps ratio. But i am not sure if you can tune this. Give it a try in the app.

Example of what is stock battery to phase amp ratio for a geared hub:
20A batt, 50A phase ( tons of torque from a stall, reducing as the speed increases, opposite of a gas motor )

If we change this to 20A batt, 30A phase, we flip this where the motor is more like a gas motor. Maybe wimpier than that.
( this is a bad choice for a hub motor but good choice for a mid drive, if we want to preserve gears and chains )

 

[limiky]

Okay, i think i had some misunderstandings based on your first post.
So you think the screen is lying, hm? i never gave a second measure on mine.

Thanks for letting me know about this code. I didn't get help from the company after asking 3 times so i gave up asking.

So you are blowing this freewheel at 28A or so, it sounds like you don't have the gear preserving tune which reduces the low RPM torque... but since the power controls in the app don't do anything, i don't think you can do that.

What would be the correct thing to do is dramatically reduce the phase amps to battery amps ratio. But i am not sure if you can tune this. Give it a try in the app.

Example of what is stock battery to phase amp ratio for a geared hub:
20A batt, 50A phase ( tons of torque from a stall, reducing as the speed increases, opposite of a gas motor )

If we change this to 20A batt, 30A phase, we flip this where the motor is more like a gas motor. Maybe wimpier than that.
( this is a bad choice for a hub motor but good choice for a mid drive, if we want to preserve gears and chains )

I think i cannot change this setting, because in Race mode sliders are useless: if you touch the slider of a parameter in the app, it does nothing. It simply is unrestricted with a hard stop at 28 something amps as producer confirmed.

I can only read the stats on the display, hoping they are right.

As far as amps are concerned on the stats, there are 2 different fields on the display, one is called Motor Amps and the other is Battery Amps. I'm attaching a photo of today's ride with the two readings.
This was a start on a hard 20% uphill road from stand still in hardest gear, so low motor rpms and (i think) maximum load.

I dunno if i can and if it will work (maybe it is motor specific for serial number) @neptronix but if you want to try i can send you the qr code i had before throttle mod.
All unlocked for me, throttle too, but i fried the freewheel, use it and try to upload it at your responsibility!

The field that reads motor torque said 67Nm max, i suppose at the 10t motor sprocket is where this was measured. Multiply it, considering rear biggest cog is a 40t, compensate for some losses, and you can understand why i fried the freewheel. Max sensed torque field on display had a huge incredible number well over 300Nm, than freewheel broke.
Numbers stayed the same, so freewheel broke again, three times.
This was the power reading out of the motor with theese settings, and it is the same now, only throttle differs and you have to push pedals to activate it.
I have qrs of the motor without throttle mod.

 

[neptronix]

Unfortunate that you have to chose between too high of power and not enough. There is no way to tune it to the middle.
These numbers on your screen and the amps don't line up. I question the screen's accuracy.

It sounds like external controller is the way to go for now.

How are the temps on this motor at such high power?
On my 1000w unit, i could never get the motor temp much past 50C even if i abused it.

 

[limiky]

I've never seen more than 50°c too, even after hauling with a cable a friend on am enduro bike in max power and throttle only for 650mt of altitude gain, full throttle, paved road.

No matter how much i stress it, it stays relatively cool.

That's why i'm hoping the freewheel needle bearing stays intact, if so manifacturer is ok with feeding more power to the unit by modifying controller phase amps, but first i have to test it in the worst conditions (practically doing trials full gas up the rockiest dh tracks).

Under load phase amps : battery amps is roughly 3 : 1 ratio. Then, mid to hig speed and less load, ratio wants to go 1 : 1.

I have to say the motor unit is, for its size, powerful, and i'm happy about how it performs (when it is not under warranty repair).

Bike is pretty much the same as manoeuvrability because all the added weight is over the bb, and as low as it can be. Motor and battery weight as a CYC photon motor only in the end, that's not much weight.

Not the fastest, sure. But the limit offroad, with this kind of power, is try not to roll over your back or make the tyre slide.
With the big cogs used in the cassette (40t or more) you can crawl up a vertical wall if you master aequilibrium at circus level.

If you don't want to drill the motor to make possible to hook an external controller, and are ok with some waiting, i use the bike every day.
A torture test every day for 2 weeks should give good measure of how it performs without the freewheel engaging like a hammer on the shaft (after throttle mod), and by now it seems that this last throttle mod works and preserves the free wheel.

If it will be alive in two weeks, i'll contact the seller and ask for an update on motor power and max amps.

 

[limiky]

Unfortunate that you have to chose between too high of power and not enough. There is no way to tune it to the middle.
These numbers on your screen and the amps don't line up. I question the screen's accuracy.

It sounds like external controller is the way to go for now.

How are the temps on this motor at such high power?
On my 1000w unit, i could never get the motor temp much past 50C even if i abused it.

"These numbers on your screen and the amps don't line up. I question the screen's accuracy."

That's what i wanted to point out when saying "

The display only reads 1119w if i search for this stat on it, but math on the display says this, and feel says something else (more on the 1400w ish as 28x52 gives as result)."

I don't know if redings are correct. Surely, if i reset the stats, they are consistent, and after a ride, max values are substantially the same.

 

[neptronix]

I'm surprised you still can't get over 50c at 20-33% more power.

Yeah, let me know how this modification goes.
It would be much better if they reduce the phase amp than this fix.

I would find it annoying, let me know how you see it.

ps with such a powerful pack, if you're actually drawing 28A battery, 57v x 28 = 1596w peak at full charge considering you have very low IR battery pack ( molicel is badass )

 

[neptronix]

I finally got the nerve to drill the case; luckily i got the angle right. But i did beat up the connectors by letting the drill hit them. That was hard to avoid, but not a dealbreaker.

The job isn't for the squeamish.

The holes should be big enough for 14 ga wires. I'll run about 1 inch of 14GA through the case then the remaining wire going to the controller will be 12 ga.



I think this would be more than suitable wiring for 35A batt / 60A phase. Voltage calculator tells me that the 1 inch 14ga section is responsible for 0.033v of drop @ 60A, which is acceptable.

If on limiky's bike, the phase amp to battery amp ratio is 2.5:1 ( my pessimistic guess ), that would make the max phase amp 70A. -10A might be enough safety margin to not blow freewheels.
So my initial tune will be 30A batt / 60A phase with a 1 second ease into the power. Basically, stock BBSHD power, minus some initial torque.

14 ga wire on the way for the experiment.

 

[limiky]

I'm surprised you still can't get over 50c at 20-33% more power.

Yeah, let me know how this modification goes.
It would be much better if they reduce the phase amp than this fix.

I would find it annoying, let me know how you see it.

ps with such a powerful pack, if you're actually drawing 28A battery, 57v x 28 = 1596w peak at full charge considering you have very low IR battery pack ( molicel is badass )

Today's ride, 800mt ascended on road to reach the start of a run.
Went up in full power mode at 34km/h, constant slope lenght 12km for 800mt elevation gain.
Big work for my legs, i helped the motor as much as i could to go fast, to see if i could stay under 25mins.

It went well for the motor, no problems at all, real fast battery drain (expected).

Motor temperature went max 62° (first time in the sixties ever, but also first time giving it all as power is concerned).

I think it won't go any higher than that, maybe something more if only throttling, but on such an ascent it would be slower than using leg power to help, and i like to pedal my bike.

 

[limiky]

I finally got the nerve to drill the case; luckily i got the angle right. But i did beat up the connectors by letting the drill hit them. That was hard to avoid, but not a dealbreaker.

The job isn't for the squeamish.

The holes should be big enough for 14 ga wires. I'll run about 1 inch of 14GA through the case then the remaining wire going to the controller will be 12 ga.

View attachment 368503

I think this would be more than suitable wiring for 35A batt / 60A phase. Voltage calculator tells me that the 1 inch 14ga section is responsible for 0.033v of drop @ 60A, which is acceptable.

If on limiky's bike, the phase amp to battery amp ratio is 2.5:1 ( my pessimistic guess ), that would make the max phase amp 70A. -10A might be enough safety margin to not blow freewheels.
So my initial tune will be 30A batt / 60A phase with a 1 second ease into the power. Basically, stock BBSHD power, minus some initial torque.

14 ga wire on the way for the experiment.

I realistically think you can go higher in battery amps.
I'm saying this because, if you read what chief engineer of this motor Matteo said in the first pages of this topic, motor was tested and didn't fail at 1700w.

My thought is: the motor will run no problem till 2kw, but the problems regarding the mechanical transmission of power could fail.

If you'll find the limits of electrical behaviour of the unit exceed what the producer is selling it for, and by a margin that is somewhat a reason to upgrade to an external controller in a safe way, maybe i'll put some work in sketching, designing and machining some transmission mod to existing elements in order to use the power without sacrificing longevity of mech components (i.e. the needle roller one way bearing freewheel that will eventually brake).

Hyper curious to see how the motor performs with external controller and your settings.
Let me know if i can give help by giving you infos on my motor behaviour, both by answering questions and taking photos/videos!

 

[limiky]

Don't know if it helps, but i observed the so called on display "motor amps" and "battery amps", and tried to understand how values changed in time when accelerating hard.

Maybe this helps in order to understand battery to phase amp ratio.

Test done from standstill climbing on a gentle slope, almast flat; small input with micro-rotation of the cranks to activate thumb throttle, than full thumb throttle to max speed without help of rider (46kph max with my gearing, 10t motor cog, 11t cassette smallest cog, biggest burliest heaviest slowest rolling tyres ever made michelin dh22's 27.5inches and heavy by itself dh alu bike).

Start: from 0 to 10kph:

Battery amps quickly went to around 20, max was 23.something; motor amps went 0 to 28.4 in a flick of an eye.

Mid acceleration: 10 to 30kph

Motor spins faster and faster and begins to make noise (not because something is braking, high pitch noise from faster spinning).
Battery amps went progressivly down from high 20s to 10-12amps; motor amps stayed the same 28 or so, going never lower than 24 to 30kph.

Mid to highest speed: 30 to 46kph:

Motor whistleing, but other than high pitch sound no vibrations at all and smooth operation.
Battery amps jump up again, going around 20.
Motor amps progressivly lower from highest value to 20 too.

I understand, looking at this, that battery to motor amp ratio is not fixed in time as motor spins, and changes roughly going not in perfect linear progressions from 30:10 to 20:20 amps.

 

[neptronix]

Thanks so much for your reports as i wait on wire.
This is great testing and confirms what i thought in the beginning, that the motor is significantly underrated, any limits would be mechanical ones since this is a lightweight design.

62C after a short but hardcore torture test means we have a lot of thermal headroom. With an external controller, this increases.
That means the drive is probably safe during summer
I agree that there is a lot of potential in the copper and magnets.

It does sound like your curve is not flat but it's not particularly powerful, maybe 2:1 ratio & in the case of the amps reducing as it gets closer to the speed, this means we're not using artificial speed limiting, this sounds like BEMF kicking in and naturally limiting the speed of the motor ( good )

I wonder if motor amps means phase amps, or another value i have seen before on the ebikes.ca motor simulator.
If it means phase amps then i should be a lot more conservative with my tuning and start with 28a batt / 28a phase, lol.

I agree that we could crank up the battery current a lot, maybe the voltage too.

52v produces a good cadence ( fast but not too fast ), i think in order to use 60v you need to use the dual chainring version of this drive and utilize some gearing difference to bring the pedal RPM down. This is doable.

 

[limiky]

Thanks so much for your reports as i wait on wire.
This is great testing and confirms what i thought in the beginning, that the motor is significantly underrated, any limits would be mechanical ones since this is a lightweight design.

62C after a short but hardcore torture test means we have a lot of thermal headroom. With an external controller, this increases.
That means the drive is probably safe during summer
I agree that there is a lot of potential in the copper and magnets.

I wonder if motor amps means phase amps, or another value i have seen before on the ebikes.ca motor simulator.
If it means phase amps then i should be a lot more conservative with my tuning and start with 25a batt / 25a phase, lol.

Reading the manual and interpreting from the messagges with the company producing it, i’m thinking motor amps on the display are phase amps.

I also tried, with race button enabled, to slide the cursor that sets “ target current”. It works. If I slide it to lets say 7 amps, no matter how hard i push the pedals or the thumb, maximum battery amps go to 7 on display, motor works accordingly.

This works to ca20 amps, then the you can max out the slider all the way to 35A but display will never see that value.
5amps at 52v btw felt exactly like my old Brose or bosch cx motor in acceleration and pedalling dynamics, so i think the slider overall doesn’t lie about cutting amps to a certain value (250w continuous felt like right).

With phase amps, as they go to 28.5 in my motor with ease, i think you should be good with 30.
Maybe you can start at 1500w, my motor roughly does this as you calculated, so it should be safe 30a batt/30a phase.
If it doesn’t brake, and it won’t (the motor, not the drivetrain internals) you can try progressively increase phase amp for a stronger output torque.

Excited to know how it goes and what you’ll do! Really hoping to acheive power density of 1w/1gr

 

[limiky]

I tried the ebike.ca simulator, great tool!
But: lightest motor is not listed, so i went with using bbshd and bbs02 to try what happened touching the current parameters.
I noticed you can input custom parameters for a given motor, but i cannot find them anywhere for the bikee bike lightest.

In the end i learned:

1) battery amps flatten the torque curve from mid speed till max speed for a given phase amp value, so if batt amps = phase amps, torque curve is practically flat from zero to almost max motor rpm.

2) If batt amps > phase amps, a flat curve till max motor rpm is produced. The lower the phase amps, the lower the flat curve on Y axis (torque).

3) if batt amps < phase amps, the first half of the torque curve changes. The more phase amps, the higher low speed torque starts. From half to max speed the curve flattens again and sets on a given torque value dictated by batt amps.

If all of this is right and i understood what i saw, pushing 28.4 amps of battery current instead of the 20 max i’ve seen on display should be safe both thermally and mechanically.
The motor actually sees 28.4 phase amps every acceleration under 25kph and is ok.
if max battery amps are bumped up, the last part of the torque curve will flatten, max speed will increase by 3-4kph, but mechanical forces on drivetrain components will stay exactly the same.
Graphs for torque, speed, acceleration and load won’t change when altering battery current alone: curve becomes flat and torque stays where it is even if you set battery current to 100amps, if you don’t touch phase current.

So my guess is: battery current should be set to whatever you have as battery/bms/wiring. For me and my config, 50 amps battery is safe by a good margin (Short wires 12ga everywhere, 60amp bms continuous 120burst, good fresh cells).

Then: if you leave phase amps at 28, motor will have a constant acceleration till max speed. If you increase them you increase beginning torque.

I don’t know where the limit for the needle roller bearing in the reduction plastic gear is, but i know that it is a hfl2026 model that has 28,5Nm of permissible torque before breaking (Not multiplied torque, but proper motor shaft torque).
This is a big value there! I really don’t know why it went bad three times. Anyways the simulator cannot simulate torque on motor shaft so we have to guess…

Bafangs stay under 100°C on small inclines with 35 phase amps (the simulator says). The light magnesium case of the motor has worst thermal conductivity (and so worst dissipation) than aluminium as raw material, on the other hand its walls are very thin and light and as someone said thermals seems good.

After all this thinking i’m already at work for a bearing that won’t fail under more than 28.5Nm at the motor shaft.
It will be heavier by a bunch (100gr more than what there is now so more inertia to make it spin and also slight more drag as it is bigger) but it wont fail, the plastic gear will strip before.
Who knows if i can steal info and make a brass gear with a stronger freewheel bearing..i’ll ask the producer to know the gear specs..