Bafang G062 / G063 Info

My motors are 6.5T rather than 6T to it helps a little, but ive actually never melted phase wires with the L1019. Even with the vents, the windings themselves typically short before the phase wires do. Then again the venting might help the phase wires too, keeping the internal portion of the phase wires cooler reducing heat buildup
6T has lower phase resistance than 6.5T.
I have had no problems with winding temps (highest peak just 130°C) but wiring inside axle got burned on previous configuration. Also 4mm2 L1019 phases got hotter than I would like.
 
Currently running both water and air cooled modded G062 with 8 awg phases. I'm using both L10 & MT60 for phase connectors.
My bike on this link. DYNO day coming soon.

I wanna see how close my watts and torque guess was! Hurry up on that dyno!

Yeah 6.5t is only 12 strands. I think 6t has better copper fill? Are you sure yours is 15 strands? Possibly why my windings are going before phases. Also not sure if the venting helps the phase wire temps ever so slightly.
 
I wanna see how close my watts and torque guess was! Hurry up on that dyno!

Yeah 6.5t is only 12 strands. I think 6t has better copper fill? Are you sure yours is 15 strands? Possibly why my windings are going before phases. Also not sure if the venting helps the phase wire temps ever so slightly.
I'm waiting for the dyno to be be calibrated during this week. The shop is just as eager to get experience on ebikes 😀
About the strada, Yes. As I did the mods I had to undo the winding connections and brush and clean all individual strands. There are 15 copper strands each 0,55mm.
I'm sure venting does help phase wiring too but becides windings it helps the nylon gears alot! Nylon looses strength with heat...
 
I'm waiting for the dyno to be be calibrated during this week. The shop is just as eager to get experience on ebikes 😀
About the strada, Yes. As I did the mods I had to undo the winding connections and brush and clean all individual strands. There are 15 copper strands each 0,55mm.
I'm sure venting does help phase wiring too but becides windings it helps the nylon gears alot! Nylon looses strength with heat...
Sweet! And good info- thank you.

I'm sure it helps, I left the nylon gears in my brand new motor about 3 weeks ago (after breaking about 4 sets of them) just to give it one last shot at 90 phase amps only- plus they were the nylons with a little bit of steel coming around the sides a bit, the "stronger" version of the nylons. I dropped it into the vented motor case. Shredded the nylons instantly, within 5 miles of HARD riding (full throttle from zero regularly + jumps). Venting didnt seem to make a huge difference, maybe a S&*% load of grease would help idk. I have no damage on the aluminum teeth whatsoever from the steel gears. The major downside is if riding hard it shears the key. Then if you upgrade the key and ride even harder (8ft jumps to flat with throttle on when I land), it can seize the freewheel of the clutch. I broke 3 clutches so far, mostly from big jumps it seems. Most recent clutch is holding.... hoping to not have to throw in the welded clutch. I rotated the snap ring to fit a larger key, instead of 10-11mm its about 13-14mm key, fingers crossed lol.
 
We can't get such insurance in the UK, you have to have type certification, engineer reports and full inspection to make anything above 250w nominal pedal assist.
Or as I do, rarely use the road and make it look legal.
 
Hall sensor board delete- to increase airflow by 20%+ and reduce "hot spots" hidden behind the board
High Temp Insulation Varnish brushed on windings
Oversized Vent holes 2-3x larger than before on both sides
8-10ML ferrofluid on magnets
Custom Hub sinks re installed
22.7" OD tire (low gearing, but not too low, "optimal" gearing for 30mph and below @ 72v)
Stainless/Nylon gears with oversized Stainless Drive Key (5x5x12mm) standard is 5x5x10 low carbon

Stator is ready for new windings- internally coated with high temp insulation, just waiting on the 180c rated copper, I went with higher gauge .81 mm and will attempt the same cross sectional area as the 6t motor (7 strands @ .81), with higher temp coating, higher gauge and 7 turns. Will probably upgrade to 12awg phases as well. - Just waiting on the copper in the mail

Gonna test this out and see what difference it makes to winding temps, then the new wound motor after.
 

Attachments

  • IMG_1307.jpeg
    IMG_1307.jpeg
    294.5 KB · Views: 72
  • IMG_1308.jpeg
    IMG_1308.jpeg
    219.5 KB · Views: 67
  • IMG_1309.jpeg
    IMG_1309.jpeg
    229.8 KB · Views: 71
  • IMG_1310.jpeg
    IMG_1310.jpeg
    472.3 KB · Views: 69
  • IMG_1311.jpeg
    IMG_1311.jpeg
    294.1 KB · Views: 75
I love these mods. Do you feel it helped with the temperature?

I'm running a G062 (Shinko 244 16x3) with a KT40a controller 48v/20ah Samsung battery. Unfortunately I just got a motor position sensor fault as I was about to go out riding. Ever got these? Is replacing Hall sensor doable? I am still baffled how this can even happen. Bike is 2 months old... Thanks

Ps. If not fixable I'm contemplating a dual mode SINE (turn off hall) KT controller with 18 instead of 12 mosfets hoping the controller will run less hot... Any thoughts? I like slow technical trail riding, will motor be smooth enough without hall sensor ?

18 MOSFET dual mode ( looks like only square - that's probably too loud with the Bafang ? )

Or Sine ( gotta get matching display )

Just found this amazing item on AliExpress. Check it out! $63.72 22%OFF | JN 60A Controller 48V 56V 60V 72V Sine wave Dual Mode 18Mosfet Controller with Lock Light Connector for 1000W 1500W 2000W 3000W
 
Last edited:
I love these mods. Do you feel it helped with the temperature?

I'm running a G062 (Shinko 244 16x3) with a KT40a controller 48v/20ah Samsung battery. Unfortunately I just got a motor position sensor fault as I was about to go out riding. Ever got these? Is replacing Hall sensor doable? I am still baffled how this can even happen. Bike is 2 months old... Thanks

Ps. Once the motor works again I'm contemplating a KT50a controller with 18 instead of 12 mosfets hoping the controller will run less hot... Or maybe those Risunmotor ones are better since they work without Hall sensor, any disadvantages? I like the imitation torque sensing PAS of kt...Any thoughts?

Ferrofluid: This made an interesting noise (sounded badass like a supercharger) but as expected within just a couple miles it all flung off. Because its a geared motor spinning at much higher RPM than most DD's it just flung off the magnets and onto the windings and whatever else. Did not work, as expected. I used a cheap ferrofluid, didnt want to spend another 80 bucks on statorade knowing it wouldnt work, but its in line with Grins RPM tables, which also basically say its a no-go.

Hubsinks: These dont do much on a geared motor due to the double shell. Most people would say they are worthless. On a vented motor they do even less, because of how well the venting works the shell does not build up heat like a sealed motor, the shell rarely goes above 60c. I believe they help a little, but probably do almost nothing, they might do a little more on a non vented motor.

Venting: As mentioned in the OP, the venting helps SIGNIFICANTLY. It doesnt make much difference under super high phase amps like 90+ (the motor will still overheat quickly) as the motor can overheat in as little as 30 seconds, so the venting does not have enough time to overcome the heat, but under regular riding conditions it makes a huge impact. It also of course cools down the motor much quicker, rather than needing to wait 5-10 minutes for a hot motor to cool down to "warm" it may only take 2 minutes. With the oversized super large vent holes its the same, doesnt stop the motor from overheating, but the motor cools down insanely quick. I measure 160c temps in the windings sometimes and after as little as 1 minute its back down to 100-125c it seems- so if I do need to stop for a "cool down break" its about 60 seconds or less usually.

You can run the motors sensorless with a sensorless controller, sometimes they cut out a little bit, so I do prefer sensors. Replacing a single hall sensor or the entire hall sensor board isnt a big deal, you can even buy the bafang G060 hall sensor boards I think and just slap a whole new one in with some pretty basic soldering.

I have a KT 30/40/50 amp controllers. From my experience they run hotter than some of the other brand 18 fet controllers. I actually have the 45 amp risun controller. Its not too bad. It can be programmed with the display to reduce battery current which is nice. It puts out around 65-75 phase amps. It seems to be about the same or slightly more phase amps than my 50a kt controller and it runs slightly cooler. Sensorless does work, but not perfectly, it can be a little jerky and cut out a little bit here and there. I currently run one step larger than the risun controller a JN 60a controller, which puts out more amps but more phase amps (90), the sensorless function seems to work slightly better but still not perfect, and it runs much cooler than any other controller, I have it mounted internally on the bomber frame with zero airflow and have never had a problem, it gets maxed out regularly. The only downside is you cannot reduce the power. Its enough power to blow the gears/pin and even the windings if run hard enough. If you do get a risun controller or one of the others be careful of your motor temps, I've melted 5 motors running them hard at those higher power levels. Running 75-90 phase amps will overheat the motor way faster than the 60 or so phase amps of your old controller. Venting is probably a must if you plan to ride it hard, and it has the added benefit of being able to check your winding temps with a laser thermometer. Also worth noting the sensorless controllers do not have a speed sensor, so you will only get a speed readout when on the throttle, when you are coasting it will register 0 mph.
 
Last edited:
Ferrofluid: This made an interesting noise (sounded badass like a supercharger) but as expected within just a couple miles it all flung off. Because its a geared motor spinning at much higher RPM than most DD's it just flung off the magnets and onto the windings and whatever else. Did not work, as expected. I used a cheap ferrofluid, didnt want to spend another 80 bucks on statorade knowing it wouldnt work, but its in line with Grins RPM tables, which also basically say its a no-go.

Hubsinks: These dont do much on a geared motor due to the double shell. Most people would say they are worthless. On a vented motor they do even less, because of how well the venting works the shell does not build up heat like a sealed motor, the shell rarely goes above 60c. I believe they help a little, but probably do almost nothing, they might do a little more on a non vented motor.

Venting: As mentioned in the OP, the venting helps SIGNIFICANTLY. It doesnt make much difference under super high phase amps like 90+ (the motor will still overheat quickly) as the motor can overheat in as little as 30 seconds, so the venting does not have enough time to overcome the heat, but under regular riding conditions it makes a huge impact. It also of course cools down the motor much quicker, rather than needing to wait 5-10 minutes for a hot motor to cool down to "warm" it may only take 2 minutes. With the oversized super large vent holes its the same, doesnt stop the motor from overheating, but the motor cools down insanely quick. I measure 160c temps in the windings sometimes and after as little as 1 minute its back down to 100-125c it seems- so if I do need to stop for a "cool down break" its about 60 seconds or less usually.

You can run the motors sensorless with a sensorless controller, sometimes they cut out a little bit, so I do prefer sensors. Replacing a single hall sensor or the entire hall sensor board isnt a big deal, you can even buy the bafang G060 hall sensor boards I think and just slap a whole new one in with some pretty basic soldering.

I have a KT 30/40/50 amp controllers. From my experience they run hotter than some of the other brand 18 fet controllers. I actually have the 45 amp risun controller. Its not too bad. It can be programmed with the display to reduce battery current which is nice. It puts out around 65-75 phase amps. It seems to be about the same or slightly more phase amps than my 50a kt controller and it runs slightly cooler. Sensorless does work, but not perfectly, it can be a little jerky and cut out a little bit here and there. I currently run one step larger than the risun controller a JN 60a controller, which puts out more amps but more phase amps (90), the sensorless function seems to work slightly better but still not perfect, and it runs much cooler than any other controller, I have it mounted internally on the bomber frame with zero airflow and have never had a problem, it gets maxed out regularly. The only downside is you cannot reduce the power. Its enough power to blow the gears/pin and even the windings if run hard enough. If you do get a risun controller or one of the others be careful of your motor temps, I've melted 5 motors running them hard at those higher power levels. Running 75-90 phase amps will overheat the motor way faster than the 60 or so phase amps of your old controller. Venting is probably a must if you plan to ride it hard, and it has the added benefit of being able to check your winding temps with a laser thermometer. Also worth noting the sensorless controllers do not have a speed sensor, so you will only get a speed readout when on the throttle, when you are coasting it will register 0 mph.
Thank you so much for the extensive reply. So with the JN Controller, if a hall sensor goes bad during ride, it automatically shift to senseless? But do you get a message on your display that you are running Sensorless? just so you know that you eventually want to repair it... why is the hall sensor blowing in the first place though 🤔

JN also makes 15 MOSFET controller that would probably have around 60 phase amps...it's future proof for higher voltage battery 48-72v or I just get the 48/52

But would it be cooler than the KT45 with 18 mosfets...


What brothers me is that the JN clearly stated KT48SVPR which stands for kunteng sine wave however they sell it with ukc1 display, I already got the lcd8h from my kt controller and I would prefer not to have to invest in another display... What are you running? Are these compatible?

Do you have a picture of the insides of your Risunmotor controller? Is it true they have overheating protection shutdown compared to kunteng?

How do you like the JN pas vs KT?
 

Attachments

  • Screenshot_20231122-122313~2.png
    Screenshot_20231122-122313~2.png
    1.5 MB · Views: 33
Last edited:
Thank you so much for the extensive reply. So with the JN Controller, if a hall sensor goes bad during ride, it automatically shift to senseless? But do you get a message on your display that you are running Sensorless? just so you know that you eventually want to repair it... why is the hall sensor blowing in the first place though 🤔

JN also makes 15 MOSFET controller that would probably have around 60 phase amps...it's future proof for higher voltage battery 48-72v or I just get the 48/52

But would it be cooler than the KT45 with 18 mosfets...


What brothers me is that the JN clearly stated KT48SVPR which stands for kunteng sine wave however they sell it with ukc1 display, I already got the lcd8h from my kt controller and I would prefer not to have to invest in another display... What are you running? Are these compatible?

Do you have a picture of the insides of your Risunmotor controller? Is it true they have overheating protection shutdown compared to kunteng?

How do you like the JN pas vs KT?

I'm still experimenting with sensorless controllers. Some seem to work slightly better than others. But for hardcore offroad riding (rocks/bumps/jumps) they dont seem to work great. From my experience they can cut out at times, sometimes quite frequently. It can even be a safety issue, there have been times where I tried to apply power to wheelie over an obstacle and it cuts out, and then I nose dive into a G out or something. I would recommend running sensored. I have actually never blown a hall sensor on any of my motors yet and replacing a sensor, or replacing the whole board isnt too big of a job.

Yes I think it automatically goes to sensorless on the JN but I'm not 100% sure. Worst case you can just unplug the hall sensor. Both JN and RISUN have self learning wires for sensorless, the risunmotor controller seems to need to use the self learning wires, the JN controller seems to work fine without needing to run the self learning system. The benefit of the risunmotor controller is you can reduce battery current with an SW900 or UKC1 display if you want, I dont believe the JN controller is programmable whatsoever so you are stuck at the full 60 amps / 90 phase amps no matter what. It works with SW900 display or UKC1 display but only the basic functions work such as PAS levels, speed, battery voltage and lights, you cannot program the power on the JN controller. Not sure if it works with KT displays or not. The listing says Kuenteng but im not sure if thats actually true. Also on either of those controllers they do not have a speed sensor so you will get a speed reading when on the throttle but while coasting it will show 0 mph, meaning the trip miles will be inaccurate. If you do the JN controller and run hard (offroad) as I mentioned in the OP, you will probably break the nylon gears, break the drive pin, and possibly overheat/melt the motor. Its a lot of power for a G062 motor, venting is almost a MUST if you run this controller, and even with venting you would want to keep an eye on temps. Both work with PAS and both work with Lights, but im not sure you can program the PAS settings quite as much as with a KT controller if that matters.
 
Thank you for sharing your experience. I see there is a lot to evaluate... Right now I will focus on getting it up and running again...

Update i just opened motor and measured all three hall sensors and they all work, wire from controller to inside of motor also working, yet I'm getting Hall error even with original controller that came with bike... did a display reset. Now error is gone but when I'm using the throttle or walk mode I see it on the display but motor makes a single click sound not moving.. I get resistance when spinning wheel either direction unless I unplug controller, then it's fine..

KT Error message now

"Motor or controller short circuit fault"

Controller dead? You had several so you might know.. I measured blue phase diode to battery ground short hmm also 8 of my 12 mosfets are positive... I ordered some new mosfets

The original one still says " Hall error." It does not have a reset option, it pretty basic

After unplugging motor battery controller and putting together again now I'm getting throttle fault.
By now I probably had every code possible 😅 something is on the fritz
 
Last edited:
Does the JN protect itself from overheating? Ever had a blown MOSFET ?
Idk man sounds like it may be your controller. I have multiple controllers/batteries/motors so its easy for me to diagnose, I just plug into a different controller for diagnostics.

Idk if the JN has temperature protection. I run it internally with no airflow, maxed out, and have never had a problem. Worst case is $60 bucks for a new one. It will probably blow your motor one way or another however, if you are riding hard.
 
This is the 45 amp controller, the dunes are just the right size that I don't risk overheating anything and can use full power even though the sand is quite a drag on the bike, running sensorless on a cheap Chinese controller, it works fine, only loses track once every few rides and I am always pushing it to the limit.

 
It was a couple of mosfets broken on my blue line, soldering this stuff is too tiny tried and impossible to get these out without wrecking the board. Ordered the 50amp (max 2700w) 18 MOSFET kt sine killer BF deal. Will limit C5 to 1500w or maybe 2000w these should not overheat as quick plus I will frame cool the thing. Let's see how it goes. Now I wait for Ali 😎
 
Last edited:
This is the 45 amp controller, the dunes are just the right size that I don't risk overheating anything and can use full power even though the sand is quite a drag on the bike, running sensorless on a cheap Chinese controller, it works fine, only loses track once every few rides and I am always pushing it to the limit.

Do you run it on full power (C5 setting )? Is controller inside frame ?
 
Back