New Bafang Crank-Drives

[teslanv]

It seems to me that these relatively minor challenges suggest that the unit is already a relatively refined design and ought to be appreciated on its many merits (which is what I think teslanv is suggesting!) Not the least being its amazing ease of installation!

Yes, my point exactly. With so many posters on this thread "complaining" that this kit is not yet able to be modified, hot-rodded or otherwise "improved", one would think this kit is less-than desirable. - I would argue that this kit is already optimized to it's fullest and is not capabale, nor does it require "improvement". - It is what it is, and at least for me, has exceeded my expectations. I would buy another in a heartbeat, and do nothing different.

 

[Tom L]

Evening out the difference in PAS levels was the only improvement I thought needed attention.

 

[teslanv]

Evening out the difference in PAS levels was the only improvement I thought needed attention.

yes, I would agree that the PAS could use tome tweaking to provide more options in the lower power levels.

For me, Level 3 of 9 is "just right" and Level 4 is constantly leaving me spinning out as I roll through the gears. - If I am riding on flat ground and not changing gears, I can bump it up to level 5 or 6 and it will feed the controller about 18-22A pretty smoothly.

 

[joshseitz]

As far as I can see, the biggest issues that have emerged here on ES with regard to the BBSxx have been:
- the longevity of the 'white gear' (which may only be an issue with earlier versions anyway)
- potential (with older bike designs) for the slightly recessed back of the main gear case to interfere with some round RH chainstays when installed, requiring use of a 1 or 2mm BB spacer
- with smaller 16"-20" wheels, the need to use an after-market adapter and bigger chainwheels (56-60T) to get maximum 'fun' out of the PAS control.

Have most of you current owners not found the supposed non-linear (on/off) nature of the thumb throttle to be a hindrance? Or throttle delay that I've read about in earlier posts? Coming from a standard geared hubmotor and controller with instant and easily modulated throttle response, I was a bit concerned about rideability.

 

[teslanv]

I ride primarily with the PAS. Trying to shift gears while running on throttle is a no-go. - You have to kill the power before you shift, and then the pause before the throttle kicks back in is really annoying. Kepler's solution is great (Integrating the kill from the ebrake into the shifter trigger). Aside from that kind of solution, I think it's better to use it the with PAS and "learn how to shift" with the dead zone of the PAS.

 

[neptronix]

Have most of you current owners not found the supposed non-linear (on/off) nature of the thumb throttle to be a hindrance? Or throttle delay that I've read about in earlier posts? Coming from a standard geared hubmotor and controller with instant and easily modulated throttle response, I was a bit concerned about rideability.

Aw shit, the Bafang does that? sounds like my MAC motor ( high pole count, spinning at thousands of RPMs, practically an RC motor in itself ) on the old EB2 controllers which couldn't handle it well at all.

If this on-off throttle response bugs anyone ( plus you want more power ), try running an EB3 infineon-type controller from cell_man/em3ev. His controllers are optimized to run on high RPM motors such as the MAC and do great.

 

[mushymelon]

I ride primarily with the PAS. Trying to shift gears while running on throttle is a no-go. - You have to kill the power before you shift, and then the pause before the throttle kicks back in is really annoying. Kepler's solution is great (Integrating the kill from the ebrake into the shifter trigger). Aside from that kind of solution, I think it's better to use it the with PAS and "learn how to shift" with the dead zone of the PAS.

This is why we need the software, I'm actually holding out on this motor until I can remove all the delays and kill the power instantly when off the throttle or the pas. I want full throttle or the same predictable throttle response in all pas modes without paying ridiculous markup for the option. The thing needs to reengage power instantly thats a must, either from ebrakes or throttle.
Looks like theres are all kinds of throttle delays and soft start setting in the software so hopefully it can be done.
Zombies throttle tamer for full smooth throttle response and the ability to reconfigure the lvc cutoff is a must for me as well.
Its not about hotroding with this little motor. Its about setting it up the way I want and the option to change it when I want to.

 

[lazy_mosquito]

Here's the photo I took using BBS 01 blueprints (...) clearly shows we're talking a tight fit there.

That won't be a tight fit, lots of space though.

And here's how it looks assembled:

I'd personally prefer more space between the cables and the wheel.
But have to say, the thing (36V 500W 25A) is amazing!

Do you have any problems with the low ground clearance? Can you post a photo which shows clearly the clearance with the ground?
thanks

 

[samsavvas]

Have most of you current owners not found the supposed non-linear (on/off) nature of the thumb throttle to be a hindrance? Or throttle delay that I've read about in earlier posts? Coming from a standard geared hubmotor and controller with instant and easily modulated throttle response, I was a bit concerned about rideability.

I have not noticed anything particularly 'non-linear' or poorly modulated about the simple Chinese thumb throttle (the only one I have used). When it first engages there is a sense of 'on/off' but the power applied is clearly very low. Pushed further the switching power levels at the throttle itself seem suitably incremental although I couldn't tell you how many 'levels' there are - I suppose it may depend on how many Hall sensors are used (I may be obvious to some that I am not sure how these throttles work!)

With regard to the levels available from the controller itself, I don't believe that the problem of 'spinning' in PAS mode has much to do with how the PAS power levels are set at the display. I'm not denying that fiddling with the power levels (and finding your own 'optimum' level) is not an effective way of addressing the issue, but in my view it may be better in the long run to address the 'spinning out' issue as more to do with optimising the relationship between the rider's capacity for assisted input and the overall (maximum) power available in PAS. In my experience this is a different from trying to get the PAS levels 'matched' to one's habitual or preferred unassisted cadence.

The assistance brings a new element into the cadence equation and the sum is greater than its parts. I found it necessary to fit a larger chainring and this immediately resolved the 'spinning' issues I had encountered. It may be argued that because the bike has 20" wheels I would have had to do this anyway. However I think it likely that even those riding 26" wheels may find this a useful strategy. I suspect that the BBSxx is optimised for whatever Bafang is focussing on as its largest market - which seems likely to be the millions of 28"x1 3/8" bikes use across Asia. I'm currently using a 56T ring on my 20" wheeled TSR and will likely move to a 60T as soon as I can. Perhaps we should try to gather some experiential data on this regarding what chainwheel sizes best suit what wheel sizes and applications using the smaller BBSxx...

Incidentally, one important aspect of the 36V/250W BBS is that I'm getting a whole week's commuting out of my Ping 10ah battery (24km per day). That's around 125km PAS at least with some occasional throttle as well! I need to get this consumption measured more carefully but the point I'd make is that this seems excellent 'economy' with quite a reasonable level of performance! My experience is limited so I can't compare with hub motor performance but exploring the 'economy/performance' ratio of this lower power unit seems worthwhile.

Savvas.

 

[scapaflo1]

Howdy all, I,ve purchaced and installed a BBSO2 on my quick 3 cannondale. I am awaiting the battery delivery at the moment. Took it for a spin with the feeling that the gearing is to high for me. 48t chain ring with an 18t cog. I want lower gearing. Primarily for the hills not for speed. Cruising at 12 to 15 mph is fine with me. I also have a Nuvince N360 on the bike. Ordered a 22t cog and would like to put either a 36t or38t chain ring on the bike. Can anyone tell me where I can get the chain ring 36t or 38t from? I,m 65 and would like to preserve the knees a little longer versus slogging up hills at 4mph!!! Any advice would be appreciated.........Thanks by the way for a grear forum!

 

[Tom L]

Fallbrook recommend a minimum sprocket ratio of 1.8:1 for the NuVinci N360 (16/29, 17/31, 18/33, 19/35, 20/36, 21/38, 22/40). I'd be careful, you're already putting in more power than it is designed for.

 

[Samd]

The last 350W unit I fitted for a customer did seem to have the throttle well weighted to the low rpm end, much less linear than a general hub controller.
Two more units arriving this week, will be keeping an eye on it.

 

[TroySmith80]

Throttle response and modulation is VERY important to me as well. This unit is so ideal in terms of size, power, packaging, sound levels and price. It seems that the software/controller performance is difficult to evaluate because everyone is getting them in different configurations.

I did get a chance to ride one briefly. It was a 48v kit, 750W i believe. The thumb throttle was tied to the selected PAS level, so that in level one, full throttle was very little power, but in level 4 it was much more powerful. On level 0 the throttle was disabled. As in most every e-bike i've ever ridden, PAS was very poorly implemented with excessive lead and lag times when i start or stop pedaling, and no torque sensor, only RPM sensing.

How difficult and costly would it be to use a different controller with this kit, one that would give really great throttle response and modulation? How about a torque sensing PAS? Do those work well?

I'd like to take the demo bike for a longer ride and i will do so when i have time and appropriate weather. Hopefully this weekend.

 

[dirkdiggler]

The prices are slowly dropping. Saw the 750w on aliexpress for $483 shipped to the US.
I took the kit off my busted frame today. The crank puller killed a few thread on the crank arm due to the cheaper quality metal. Be careful if you have to remove your cranks.
Hoping to get a donor bike soon. All the bikes I like and that will fit this kit seems to sell faster than I can respond.

 

[Ken Taylor]

I want to order one of these but every time I try come away confused. According to the comprehensive article by Spinningmagnets at http://www.electricbike.com/bafang-bbso2-750w-mid-drive/ the only mechanical differences between bbs01 and bbs02 is a 5mm wider stator in bbs02 which leads to 1 lb more weight. He says there are seven stock controller/stator combination options:
250W = 24V X 18A
250W = 36V X 15A (greenbikekit.com)
350W = 36V X 18A
500W = 36V X 25A (Ali Express)
350W = 48V X 15A (BMS-Battery)
500W = 48V X 20A
750W = 48V X 25A (trusted vendor of the year em3ev.com), (lectriccycles.com) or also (greenbikekit.com)

He says the BBS01 and BBS02 have a different wind but it's not clear if all of each type have the same wind. He says the BBS01 can handle 17 amps but not 24 amps. Splitting the list mechanically we get.
BBS01
250W = 24V X 18A
250W = 36V X 15A (greenbikekit.com)
350W = 36V X 18A (I guess this is the one greenbikekit.com are now selling as 36V 350W)
350W = 48V X 15A (BMS-Battery)

BBS02
500W = 36V X 25A (Ali Express)
500W = 48V X 20A
750W = 48V X 25A (trusted vendor of the year em3ev.com), (lectriccycles.com) or also (greenbikekit.com)

So what is the remaining differences? Samd got an over voltage error message when connecting 42V to a 36V BBS01 see http://endless-sphere.com/forums/viewtopic.php?f=28&t=50104&p=867139#p867139 but there is a 48 volt version. I guess the cut-off voltage and maximum current is only a function of controller programming and can be adjusted but do they all have the same winds. The ideal cadence is the same regardless of voltage so I would guess the different voltages would require different winds. Are these guesses reasonable?

After that there seems to be programmable maximum speed cut-offs and programmable variations on PAS and throttle function which I also find confusing and which sellers don't usually specify.

 

[neptronix]

You guys are thinking about the hall effect sensor in the throttle, but actually the high eRPM of the motor plus the controller's inability to accurately modulate it is the more likely culprit.

Unfortunately many production systems come with this flaw. MAC/BMC motor setups had this problem until the EB3 infineon-like controllers came about. On the EB2 controllers, a direct drive motor was smooth and easy to modulate - as soon as you hooked up the MAC motor, it performed terribly. Optibike's high end model most definitely had ( and may still have ) the abrupt on-off throttle modulation problem too, and an inability to ease the power on gradually.

You can try swapping throttles, but i doubt you will get very far.

 

[mwkeefer]

You guys are thinking about the hall effect sensor in the throttle, but actually the high eRPM of the motor plus the controller's inability to accurately modulate it is the more likely culprit.

Unfortunately many production systems come with this flaw. MAC/BMC motor setups had this problem until the EB3 infineon-like controllers came about. On the EB2 controllers, a direct drive motor was smooth and easy to modulate - as soon as you hooked up the MAC motor, it performed terribly. Optibike's high end model most definitely had ( and may still have ) the abrupt on-off throttle modulation problem too, and an inability to ease the power on gradually.

You can try swapping throttles, but i doubt you will get very far.

Depending on the unit and whom you ordereded it from and how they programmed it using controllerst.exe, there is in fact not only a throttle delay but an actual ramp with values multiplied by 10 ms to 100 ms depending upon the setting, a primary throttle delay (presumably to eliminate shock if it's been applied) and then 9 separate levels of progressive power each with a delay sequence between.

PAS works the same except the 9 levels increment the % of PAS provided.

Under the hood the controller software is quite versatile or seems to access many settings which will be useful for "fine tuning".

I am concerned about the White Plastic Gear shattering on someone, was that a BBS02 or BBS01? I've had my unit in stock form running between 48v and 55.5v nominal packs and playing around a little bit on a 20" folder for test purposes - but I have that lag time others have spoken of, 1/2 throttle isn't 1/2 throttle it's a ramp from 1 to 4 or 5 setting of throttle over X time, it can be disabled in the EEPROM / SETTINGS but there really should be something to prevent someone from flipping themselves off or overloading the motor? Eh when the Open Source software to read and write BBS01/02 units stock controllers we will see what the ingenious ES crowd comes up with for shared settings.

Regards,
Mike

 

[samsavvas]

Something from the Lighting Rods Mid Drive thread earlier today:

MWKeefer (Mike) said:
"I also have an open source version of the BBS-01/BBS-02 software I've been working on for some time which I will be releasing in about the same time so you will be able to custom program the onboard controller for those who wish to retain the stock controller over an external."

Savvas

 

[beazee]

Do you have any problems with the low ground clearance?

There's 6 inches clearance on 26" wheels.
No issues so far, except getting the bike in/out of the elevated shed.
Should have more pix over the weekend.

 

[kampua]

Hi just a question for those who are in Australia, if I buy BBS02 from EM3ev (China) I have the options to ship via EMS (Express Post) or DHL (Courier), my memory is that if I ship by courier I will need to pay custom brokerage fee but with EMS I don't. Is this correct?

 

[scapaflo1]

Fallbrook recommend a minimum sprocket ratio of 1.8:1 for the NuVinci N360 (16/29, 17/31, 18/33, 19/35, 20/36, 21/38, 22/40). I'd be careful, you're already putting in more power than it is designed for.

Thanks for the info. Looks.like a 42t ring with a 22t cog will work and get me close to where I want to be. Will give an update when all is tested. Will be using the stock ring for a while with 22t cog until Fallbrook gets 42t rings in May. Thanks again, Al

 

[beazee]

I'm currently on 46T:21T (2.19:1 ratio) on NuVinci N360. Bike itself is 45kgs, driver+passenger+lunch boxes make another 130kgs, with 36V 500W climbing comfortably any inclines around.

And the manual says:

The minimum sprocket ratio (front chain ring tooth count/rear sprocket
tooth count) is 2.1:1 for pedal-assist bicycles with combined power from the
250W (rated) motor and rider. For bicycles with only rider power transferred
through the NuVinci N360 CVP (non eBikes and front hub motor eBikes), the
minimum sprocket ratio is 1.8:1.

http://www.fallbrooktech.com/sites/default/files/videos/press-releases/20100910_N360%20for%20250KW%20e-Bikes.pdf

 

[TroySmith80]

You guys are thinking about the hall effect sensor in the throttle, but actually the high eRPM of the motor plus the controller's inability to accurately modulate it is the more likely culprit.

Unfortunately many production systems come with this flaw. MAC/BMC motor setups had this problem until the EB3 infineon-like controllers came about. On the EB2 controllers, a direct drive motor was smooth and easy to modulate - as soon as you hooked up the MAC motor, it performed terribly. Optibike's high end model most definitely had ( and may still have ) the abrupt on-off throttle modulation problem too, and an inability to ease the power on gradually.

You can try swapping throttles, but i doubt you will get very far.

Would this EB3 controller work well with the BBS0X? Or is there another controller that would? Presumably there is something that would give quick and precise throttle input we're looking for. Maybe even an RC car or airplane controller? I don't know much about this stuff.

Is there already a thread about that, or should we start one? I actually kind of hate the enormous threads like this one that cover a lot of different topics around a given product, because specific information in this thread becomes very difficult to find. It becomes a needle in a haystack situation. I think it's much better to have many different threads that are each smaller in scope, so that people searching for something can more easily find it.

 

[samsavvas]

Hi just a question for those who are in Australia, if I buy BBS02 from EM3ev (China) I have the options to ship via EMS (Express Post) or DHL (Courier), my memory is that if I ship by courier I will need to pay custom brokerage fee but with EMS I don't. Is this correct?

I've rec'd stuff from OS via both EMS and DHL. I've only ever paid duties (if that's what you are referring to) on items over $1,000. I have never paid 'brokerage' fees.

Savvas.

 

[neptronix]

I can't give you a guarantee that it will work well, but i can say that it works well on a MAC motor, which is 16 poles and spins at 1,600rpm at the speed it was designed. I've spun it to ~3,000rpm easily ( 72v - 4kW ) and it was still controllable and partial throttle levels still worked fairly well on an EB3, but not perfect. There is a world of difference between lesser controllers than the EB3 and the EB3 when controlling such a high rpm & high pole count motor.

The symptom with lesser controllers on my MAC was always these: ( and i tried quite a few controllers to compare because it was really pissing me off )

1) throttle has an off-on response and nothing in between. This causes a spike of power when the throttle is engaged at all from 0%, which is both unpleasant and wears on the gears or any drivetrain components as well.
2) If you hold the throttle at a partial level, then the power will kick on/off randomly and will just be a jittery affair..
3) No way to ease the power on, even from a stall.
4) The higher the voltage the controller is given, the worse this problem is.
5) The faster the speed the motor is spinning, the worse this problem is.

If this motor has a 30:1 reduction ratio as GEBattery says, then it spins at ~2,600 rpm nominally. I believe that the motor is a 16 pole motor like the MAC has.
If the above specs are correct, then the bafang motor is easily one of the most difficult common ebike motors to control.

My theory: the motor is spinning so fast and there are so many poles per rotation that the controller can't keep up and instead of losing sync and doing all kinds of goofy things while trying to modulate the power to the motor via PWM, it gives up and fails to a pure on/off mode.
This may not happen with the 36v BBS01 units because the voltage is lower, and maybe the RPM is as well??

You'd be surprised at how many production systems do this. I was extremely disappointed to see the $8,000 optibike do the same thing my MAC motor was doing with a controller that was not designed for the high electrical RPM of the motor. just tapping the throttle made the clutch/freewheel in the expensive rohloff hub thunk and clunk loudly. Same thing with my MAC motor, which is why a lot of early MAC/BMC users had serious problems with chewing up gears and freewheels, yet you rarely hear of those kinds of problems today.

You guys are thinking about the hall effect sensor in the throttle, but actually the high eRPM of the motor plus the controller's inability to accurately modulate it is the more likely culprit.

Unfortunately many production systems come with this flaw. MAC/BMC motor setups had this problem until the EB3 infineon-like controllers came about. On the EB2 controllers, a direct drive motor was smooth and easy to modulate - as soon as you hooked up the MAC motor, it performed terribly. Optibike's high end model most definitely had ( and may still have ) the abrupt on-off throttle modulation problem too, and an inability to ease the power on gradually.

You can try swapping throttles, but i doubt you will get very far.

Would this EB3 controller work well with the BBS0X? Or is there another controller that would? Presumably there is something that would give quick and precise throttle input we're looking for. Maybe even an RC car or airplane controller? I don't know much about this stuff.

Is there already a thread about that, or should we start one? I actually kind of hate the enormous threads like this one that cover a lot of different topics around a given product, because specific information in this thread becomes very difficult to find. It becomes a needle in a haystack situation. I think it's much better to have many different threads that are each smaller in scope, so that people searching for something can more easily find it.