2 years after the summer of peanut butter gear Bafangs???

[John in CR]

I'm startling to wonder if that was just a bad production run of gears installed in the Bafangs we were getting 2 years ago. I haven't heard much mention of broken gears in geared hubs since then. Is it safe to assume that geared hubs are holding up better now, or have I just been missing the failures? I'd like to let go of that preconception if it's not true. Any commuters out there racking up trouble free mileage on geared hubs that aren't babying them with lots of pedal assist?

 

[johnrobholmes]

I ride about 50 miles a week on mine, but it is simply 36v 20a. No worries there. I think the people that were trying to squeeze more power from the bafang have just moved on to bmc/mac sized motors instead.

 

[docnjoj]

I was running 54 volts up to 30 amps with the newer Bafang from John Rob with no problems at all. It is going on a bike for my kid next. The wheel is perfectly true even after about 1000 miles. Way to go John Rob.
otherDoc

 

[spinningmagnets]

Since the Bafang was reasonably affordable, there was a slew of enthusiasts who wanted to find out how hard you could push them, and hopefully find out if there was a reasonably easy/affordable fix for the weak links.

If you don't get the thre plastic planet gears too hot (and soft) they seem to hold up OK. I believe they actually worked well as a cheap and easy "mechanical fuse", as a new set was fairly cheap and easy to replace. If you didn't mind the noise of steel gears, and began over volting/amping the Bafang...you could overheat the windings...much worse and more expensive than three plastic pop-in gears.

If you never shock-load the gears, and you keep the amps down, they should be able to take a lot of abuse. I'd be more comfortable raising the volts rather than raising the amps to get more power. It seems selecting a higher voltage, and then matching the winding to the lowest top speed you want, will give you the best acceleration and hill-climbing torque, resulting in the lowest possible motor heat.

If you'd rather play with higher amps (rather than buy a new winding of motor for higher volts) to save money, I'd recommend getting an infrared thermometer with a remote read-out (mounted on the frame, pointed at the motor). They're not expensive, and you can mount the screen on the handlebars. The outer shell of the motor will not be as hot as the inside, but tackle a few hills in slowly increasing order of steepness.

When you get to a point where you can't keep your hand on the motor (when you stop occasionally to feel it), then the read-out you were getting on the screen will be your warning temp.

I have never used a Bafang, but I read about them due to their affordable price.

 

[zukster]

I've been running a couple of Banfang SWXH for over a couple of years now. One gets
routine use up a steep hill for about 10km M-F. The other several times a week in hilly
up and down conditions.

I smoked a couple of motors when overpowering with a 30A controller and insufficient
battery power. Once I figured out to stick with the 350W 20A Infineon there have been
no problems at all. Even for the motors I burnt out, the gears were okay - the windings
just fried.

The motor delivers considerably better power with 60V as opposed to 40V and nothing
heats up. I ordered metal gears but I have not had to use them.

I get speeds of about 25 mph on the flats on a light bike unassisted. These motors are
fine as long as you don't get carried away over-powering them.

 

[jbond]

I smoked a couple of motors when overpowering with a 30A controller and insufficient
battery power. Once I figured out to stick with the 350W 20A Infineon there have been
no problems at all. Even for the motors I burnt out, the gears were okay - the windings
just fried.

The motor delivers considerably better power with 60V as opposed to 40V and nothing
heats up. I ordered metal gears but I have not had to use them.

I get speeds of about 25 mph on the flats on a light bike unassisted. These motors are
fine as long as you don't get carried away over-powering them.

Could you give some details? Motor spec (eg 26"), sensored/sensorless, rim size (eg 26"), battery (eg 36v15AHr LiFePO), Controller LVC, current, phase (eg 31.5, 20A, 50A)

 

[zukster]

Could you give some details? Motor spec (eg 26"), sensored/sensorless, rim size (eg 26"), battery (eg 36v15AHr LiFePO), Controller LVC, current, phase (eg 31.5, 20A, 50A)

Motor: Bafang SWXH (2007 design, resists heat to appx 180C)
Rocky Speed Mountain Bike 1998 with 135mm dropout in the rear. 26 Inch wheel.

The SWXH fit perfect on my bike with a new 8 cog freewheel (Sunrace). I don't know why it says only for a 6 cog.

Controllers: Infineon 350w 20amp Infineon controllers (36-60Volts). One hall. One sensorless. See
http://endless-sphere.com/forums/viewtopic.php?f=2&t=13433

Either 2 or 3 Canadian Tire 20V Lithium Manganese Yardworks batteries in series for appx 40 or 60 Volts.
Batteries have built-in BMS so they provide the LVC. When one cuts out, the whole circuit cuts off.

SWXH Info Stamped on Motor

8FUN ->
Voltage: 36VDC (clear sticker)
Output: 250W (same clear sticker)
26 (green sticker)
ZL20003 2 0123400.9
Permanent Magnet brushless high speed hub motors used for electric bicycle
M5X8L
Then, one motor had
BFY806H11541
and the other had
BFY902H11328

Approximate weight:

Yardworks 20V - 3.5 lb. - 1.6kg, Bafang SWXH - 2.9 kg - 6.5 lb, Infineon Controller 36V/350W - 240g - 0.5 lb.
Bike - Full Suspension Cross Country MTB - 29.5 lb. Currently not using rack. Batteries Pipe clamped to frame triangle.

Bike + (3 x bateries) + motor + cntrlr ~= 47 lbs.

 

[John in CR]

Great info Zukster. Thanks for sharing it. May your little Bafangs last as long as I know my DD hubbies will.

Also, +1 on max 20A controller. I melted my one and only bafang trying to run it with a 30A controller. Never even got to ride it. Gears were virgin, but windings toast. I think maybe Bafang got a bad batch of gears or something was too much out of tolerance, and they have since corrected it. They're still pumping the motors out. Too bad that BPM is sensorless. I'd scoop one right up, but I'm afraid it will be voltage limited due to electrical rpm limitations of sensorless.

 

[jbond]

Zukster, thanks for all that. Your 40v pack with a fully charged voltage of 42v in LiMn is what everyone else calls a 36v LiOn. It's 10s. The "20A" controller, I'd guess is programmed to 20A-50A for battery-phase limit[1]. The motor is a stock size running at it's stock settings and will run out of assist at round about 20mph. 3 battery packs is 15s which is 63v fully charged, nominal 54v. This should give a no-load rpm of about 30mph. I guess that's where your 25mph came from.

The only curious bit in this is what happens to motor temperature if you run for a long time at 54v-20A. That's a KW of input power to the controller, so maybe 900w to the motor. At max efficiency that's 150W of waste heat in the motor. If you can get away with it, that suggests that a 48v battery with the same current limit should be ok with some headroom.

John, Don't be afraid of sensorless or the BPM. I'm running a BPM with a LZDSZ controller (it came like that) on 36v LiMn. I don't know what the controller is actually set for but it says 9A rated, Max 18A on the sticker. It's a 26" code 11 in a 26" rim. At very low speeds, you can feel an occasional brief stutter as it starts. Above walking pace you'd never notice. No-Load max is 24mph. Real world speed on fresh batteries is 20mph. I've yet to find a hill it wont climb, even off road. I've got every reason to think that the BPM will take considerably more voltage and current. 48v-25A feels like it should work fine, maybe higher. YMMV! I've heard of one person using 48v of A123, BPM code 10 (18") in a 26" rim, Infineon sensorless controller set to 30A-80A. All from cellman. Claimed top speed is 27mph. http://www.pedelecs.co.uk/forum/electric-bicycles/7528-bpm-kit.html

[1]These settings seem to be pretty imprecise. It needs a calibrated CA to really know how accurate they are.

 

[voicecoils]

I ride about 50 miles a week on mine, but it is simply 36v 20a. No worries there. I think the people that were trying to squeeze more power from the bafang have just moved on to bmc/mac sized motors instead.

Yeah, that's what I did...moved on to ebike-sf BMC V2S

88v30A turned out to be totally unreasonable for the tiny bafangs. In hindsight it should have been obvious... live & learn.

36v20A sounds just about right. 30A with a smooth starting controller would likely be fine too.

 

[jayuk2011]

Hi

Ive only just joined the forum but have been viewing for about a year. I thought i would join as i am the on with the bpm kit that jbond has mentioned above. I am absolutly loving the kit, its performing exactly as i wanted. I have the following:

Bpm 500w 36v code 10 with sensors (18") running in a 26" rim also has the black paint job too
A123 9.2ah 48v
Infineon 36v 350w controller 30a 80a phase
twist throttle
28 to 31mph
Although i stick to around 20 - 25
I certainly would not want anything faster.

I will add some photos soon..

Jay

 

[John in CR]

John, Don't be afraid of sensorless or the BPM. I'm running a BPM with a LZDSZ controller (it came like that) on 36v LiMn. I don't know what the controller is actually set for but it says 9A rated, Max 18A on the sticker. It's a 26" code 11 in a 26" rim. At very low speeds, you can feel an occasional brief stutter as it starts. Above walking pace you'd never notice. No-Load max is 24mph. Real world speed on fresh batteries is 20mph. I've yet to find a hill it wont climb, even off road. I've got every reason to think that the BPM will take considerably more voltage and current. 48v-25A feels like it should work fine, maybe higher. YMMV! I've heard of one person using 48v of A123, BPM code 10 (18") in a 26" rim, Infineon sensorless controller set to 30A-80A. All from cellman. Claimed top speed is 27mph. http://www.pedelecs.co.uk/forum/electric-bicycles/7528-bpm-kit.html

[1]These settings seem to be pretty imprecise. It needs a calibrated CA to really know how accurate they are.

When I want to free the motor from it's shell and run it at 100V50A I need sensored, because sensorless hasn't proven to handle high enough rpms or RC motors would be all the craze for high power. Sensorless controller solutions get too expensive for durability with bike loads at low rpm.

 

[FMB42]

Thanks for all the great info everyone.

I've been holding off on my first FS MTB 36v rear hub e-build due to my concerns with the high unsprung (~5Kg) weight of the average 500w DD hub motor. I felt that the lower unsprung weight, and supposedly improved hill climbing power, of a geared hub would be the answer. However, I've been worried about geared hub motor gear failures under more than "moderate" off road use conditions.

I'm now thinking that the BMP motor might be the way to go (even though it's only 1Kg lighter than the average 500w DD unit).

 

[zukster]

May your little Bafangs last as long as I know my DD hubbies will. Too bad that BPM is sensorless.

I hope they last as long as DD too but I doubt it. In the long run, the DD should be the winner due to less moving parts. If I get
another year or two more out of the Bafangs, I'll be happy. They get used a lot. The only partial complaint I have is that when it gets
below freezing or very wet outside, I've had to treat them with ACF-50 or Fluid Film to keep the engine internals from shorting.

http://endless-sphere.com/forums/viewtopic.php?f=2&t=12984&start=60#p218898

It mainly seems to be in freezing weather when the bikes are stored outside. When the motor cools, condensation gets into
the motor. The treatment does keep them going. credit for this trick goes to Affliction - if he's still around.

I you're not running the BPM much over 20A 60V, I wouldn't worry to much about it being sensorless if you can get a controller
from ecrazyman/kewin. He hasn't got back to me lateley though, so I don't know if he's still around either. The 350W 20A
sensorless controller works great. I can't really tell it appart from the hall controller anmore, now that my batteries don't come
off the charger as high.

Just for more info on the Yardworks batteries and the Yardworks chargers. The 3S (batteries) 15S (cells) pack has never
measured more than 61.5V off the chargers. You can charge with them all hooked up and the controller on even! If anyone
is going to use one of these 20A 350W Infineon's with Lipo, I would recommend undercharging a bit. Lipo recommended
max being 4.2V x 15 cells = 63V is too borderline for the 63V capacitors most of these controllers will have. I would not
charge the pack over 62V or over 4.13V per cell. Might be better for the batteries in the long run too. I will probably be
trying this over the next few months myself. Will let you all know how it works out.

The only curious bit in this is what happens to motor temperature if you run for a long time at 54v-20A.
That's a KW of input power to the controller, so maybe 900w to the motor. At max efficiency that's 150W of waste heat in the motor.
If you can get away with it, that suggests that a 48v battery with the same current limit should be ok with some headroom.

Not sure about this motor math. Are you saying that you're wasting energy dumping 60V (61.5V off the charger) into
the motor. From experience riding with my son and a 40V pack, we get about the same distance going full speed...?
Or close enough not to worry about it. Oh, the motor does not heat up at full throttle up 5-7% grade with either 40 or 60V packs.
With the 30A controller, both packs caused overheating problems in the motor and the controller. I wanted to reprogram the
controller with a 20A current limit, but no one would get back to me with the original Parameter Designer programming
parameters to modify and re-flash the controller with. Anyway, the two small 20A controllers work great.

 

[BenMoore]

Too bad that BPM is sensorless. I'd scoop one right up, but I'm afraid it will be voltage limited due to electrical rpm limitations of sensorless.

A custom order of sensored Bafang BPM motors in a range of speeds was made by me last year :wink:
Most the motors are in Australia but some are with cellman in China for cheaper global distribution.
I'll put details up in the For Sale section in the next few days.

- Ben

 

[Gregory]

Just to relate my experience today.

Running an older 2008, but little used Bafang in 26" rim at 55V with peak amps at 29A.
Destroyed the gears. No pedaling. Some small hills.
So my 1700W with a 100kg rider was too much.

It was OK running 22A (1200W) previously on an infineon controller. I will either try to source some replacement gears and unsolder the shunt or hit cell_man up for a new MAC.

Probably the latter.

 

[zukster]

Finally after 3 years of use, my son stripped the nylon gears in his rear drive Bafang SWXH 26
Inch wheel model being run by Ecrazyman/keywin 350w/20a Infineon controller and 60V to it.

I had already ordered some metal gears from keywin a couple of years ago (I don't know if keywin
is around anymore), so I just had to install them. On of the nylon gears had actually cracked and
jammed the motor. Not wanting to take anymore chances, I used 3 metal gears.

Its not too loud when the motor has been treated with ACF 50 or fluid film inside to protect it.
See http://www.endless-sphere.com/forums/viewtopic.php?f=2&t=12984&start=60#p218898

Also a note. The insides of the motor are in good shape, so the above treatment worked and
did its job of protecting the motor though a couple of winters.

 

[John in CR]

Thx Zukster, it's always good to hear good results. 20a and 60v sounds like a good performer too.