Bafang G311 loud, weak, and overheating

[saucerful]

Hello, I have a Bafang G311 (front hub geared) motor from Grin with Baserunner controller and 52V em3ev battery. The kit is about 4 months old with 400 miles on it. A couple days ago I noticed the following in order:

1. Fluttering sound, not unlike the usual sound, but louder, at times loud enough to make it hard to hear music from a smartphone on handlebars at max volume. The sound is not present when the motor is not getting power (e.g. coasting or just pedaling) and the sound is not much different when the motor is first engaged or in steady state.
2. (After a few miles) Seemed like hitting the throttle was producing less thrust than usual even though the power displayed on the CA is plenty (500 W say).
3. Started going easy on the throttle. But then I noticed motor overheating. I monitored temp in a 5 minute period of an average of 200 W power on pretty flat road, the motor temp went from 75 C to 120 C before getting limited. I don't ever remember it getting temp limited before even going up a steep hill at 500 - 700 W for 5 minutes.
4. Once it got hot the speedometer was jumping all over the place from 15 mph to 25 mph in less than a second, and back down. Even though there was very little change in speed. I stopped using the motor at this point.

I wonder if its one of the issues Grin mentions on their G311 page (stripped gears or loose magnets?) but when I setup the motor I made sure the max torque and phase current were limited as prescribed.

I sent Grin an email though in my experience they can a while to respond especially with the weekend coming. This is my only form of transportation so I'm eager for help diagnosing the issue quickly, hoping you all can help me. I connected it to the Phaserunner suite via USB cable and ran through the autotune and didn't see anything suspicious (except it was loud as it was spinning). Thanks so much!

 

[amberwolf]

Do you see correct hall signals? If not, maybe the PR is driving the motor incorrectly timing-wise, which would cause less torque, more heat, and usually more noise.

Stripped gears in most geared hubs results in no torque, just motor spinning inside hub without being able to drive wheel under load (it will still spin it offground in many cases). Broken or slipping clutch could do this too, and both could cause excessive noises of varying types. But I don't know of any cases where stripped gears or slipping clutch would cause motor to overheat, because they take load *off* the motor. So it's probably not this.

Loose magnets, if htey've all come unglued, could cause lower torque because the magnets are actually slipping aorund the rotor instead of driving it normally. They'll still pull on the rotor, but the drive is now coupled magnetically / frictionally, instead of directly, so it's weaker than usual. This might cause noise also, and could cause heating because the rotor is moving along with the stator instead of "rotating in reverse of it" relative to the stator, so the controller's drive signals may be incorrect at any one instant for it.

 

[saucerful]

Do you see correct hall signals? If not, maybe the PR is driving the motor incorrectly timing-wise, which would cause less torque, more heat, and usually more noise.

Seems so. The software said something like "Bafang hall sensor pattern detected."

Stripped gears in most geared hubs results in no torque, just motor spinning inside hub without being able to drive wheel under load (it will still spin it offground in many cases). Broken or slipping clutch could do this too, and both could cause excessive noises of varying types. But I don't know of any cases where stripped gears or slipping clutch would cause motor to overheat, because they take load *off* the motor. So it's probably not this.

Loose magnets, if htey've all come unglued, could cause lower torque because the magnets are actually slipping aorund the rotor instead of driving it normally. They'll still pull on the rotor, but the drive is now coupled magnetically / frictionally, instead of directly, so it's weaker than usual. This might cause noise also, and could cause heating because the rotor is moving along with the stator instead of "rotating in reverse of it" relative to the stator, so the controller's drive signals may be incorrect at any one instant for it.

The motor is definitely still producing torque, just less so. It's especially noticeable when the throttle is first engaged-- little to no torque. But going up a hill it definitely was still assisting me. So maybe it's loose magnets... I may wait to open it until I hear back from Grin.

What I'd really like to know is am I risking further damage to the motor if I use it in this state? It's tempting to keep using it because it does produce power... But if the magnets are loose could something break? (The controller has thermal limits so it shouldn't get dangerously hot.)

 

[amberwolf]

If the magnets are loose they're sliding around the rotor, and vibrating with the changing stator fields, so frictional wear of the magnets and rotor surface and anything else they are touching will occur. How bad that is depends on the materials everything is made from, and whihc part wears faster depends on what is harder surfaces.

If the gears are damaged, well, they're already damaged so further damage to them is irrelevant...but if pieces of them come off and get ground up in other moving parts it can cause frictional damage to those parts.

If the motor is getting past boiling water temperatures, then it depends on the insulation specifications on the motor wire and other parts in the motor if they will be damaged.

Same for the plastic gears or other plastic parts...

Electronics like the hall sensors in the stator, and the speed sensor on the wiring board may not be spec'd for high temperatures. The spec sheet for Honeywell / Allegro SS41 or SS411 hall sensors shows they're good for 150C, but I don't know if the parts used in your motor are those, or some clone of htem that may have lesser specs--commonly electronics are good for up to around 70C. .

 

[saucerful]

@amberwolf, thanks this is so helpful. I'm suspecting it is a loose magnet... I wonder if this video I took of the Phaserunner autotune with the sound audible would narrow it down? I'll open the motor up soon.

Also I wonder whether the (default) 90 C threshold (when current limiting starts) setting and 130 C max temp settings are too aggressive and should have been lower?

 

[amberwolf]

I'm afraid I can't tell from the sound what it might be.

I dont' know what the temperature limits are for the motor hardware, adhesives, etc.

Some epoxies are very high temperature, and can even be used in internal combustion engine temperatures--I recall a front-engineblock-mounted waterpump that had a stripped screw hole that someone used JB Weld to rebuild the hole with, then tapped the JB Weld with new threads, because there wasn't enough left of the original hole's edges, I think. It lasted the rest of the life of the car.... But some epoxies get quite soft at fairly low temperatures and don't make good magnet glue, especially for SPM type inrunners (which I think this motor is), where centrifugal force trying to fling the magnets off is pretty significant at high motor speeds.

The magnets themselves are probably good for 150C before they begin to permanently change magnetic properties (though they may weaken temporarily while hot they will come back if below whatever their specific problem temperatures is). This random-googlely found page has some info on that Temperature effects on magnets ; I don't know it's precise accuracy but it sounds like the similar other pages I've read before.

WIndings are usually good for a couple hundred degrees C; other wiring insulation varies a lot by type and specific version--it's often printed on the individual wires or the outer sheath, for anything decent.

Hall sensors depends on version and manufacturer, but as previously noted should probably handle even your max...though it depends on where the temperature sensor is whether it is sensing a higher or lower peak or average than the halls (or other parts) see.

If it's under the windings like many, it will heat up really fast compared to anything else but the windings themselves, to help prevent peak temperatures from damaging anything. If it's in the ambient airflow inside, not directly connected to something that heats fairly directly from motor power usage, it could take a lot longer to heat up the air and have the sensor shutdown or rollback power to reduce heat, and everything important could get a lot hotter than the temperatures it reports. Really only the gears/clutch would be about the same or lower temperatures than a sensor in the airflow.

If the planetary gears are plastic, they will get softer the warmer they are (and more brittle the colder they are below their normal range). How much softer vs temperature depends on the specific plastic used, and whether that allows permanent (plastic ) deformation that damages or destroys the gears also depends on that and the torque being applied at the time.