Bafang 750w Geared Hub Motor Maintenance

[BrianMH13]

Havn't posted here in a while (no $ to do an ebike build) but I've got over 1,500mi on my Aipas A2 at this point. A couple handfuls of rides ago I noted that the hub motor got louder than usual fairly abruptly. Nothing crazy but it got louder and started getting a little harmonic vibration that I could feel. I figured it could be the planetary gears getting worn or needing grease.

I opened up the hub today and found what I imagined. The gears had very little grease on them still.

Aipas mentioned nothing about servicing the hub motor in the user guide maintenence section which is not exactly cool but whatever I suppose (I knew better)

I dont think there is any significant wear on the gears so I figure it should be ok after I service it.

I dont have any e bike gear motor specific viscosity grease so I'm thinking my two short term service options are:

1. Collect the grease thats been driven off the gears and re-apply it or...

2. Use the grease I do have (Park Tool Polylube 1000) PPL-1.

Short term service interval is ok. In other words because its easy to get into this motor I don't mind obtaining the correct spec grease and re-lubing it with that again very soon.

The Park Tool stuff is lower viscosity or less sticky. I'd use less of it so that it wouldnt go all over inside the hub.

I'm leaning towards relocating the grease thats in the hub back onto the gears and running it.

I also may be able to get ebike motor gear grease from a local shop in the next few days.

Opinions or advice?

 

[gromike]

Number 1 would be better than number 2, in my opinion. After your inspection of the gears and finding no obvious issue, the problem might be a bearing (or something else).

Now, if you want to have some fun, ask what grease to use for regreasing.

 

[amberwolf]

It's unlikely to be a gear problem (or grease) if there is no gear damage.

The grease on the gears will virtually all be squished off of the teeth on their very first meshing if they are correctly designed, so that they properly mesh.

The only time you'll see grease thickly (or even visibly) coating tooth surfaces is if they aren't meshing correctly, so that whereever they never touch, you'll see grease.

(same thing for heatsinks and hot parts--if you can see the heatsink paste between the parts, there's too much of it and the parts aren't making contact, either the parts are deformed and not lapped to fit properly or they are not fixed together properly).


It's much more likely that you have worn bearings in the motor covers, or bushings /bearings on the parts inside the motor (like the rotor/sungear, which spin the fastest).


It can also be loose spokes, especially if the noise is cyclic / periodic with wheel rotation.

It can also be loose axle-mounting hardware (nuts, washers, etc) so that the motor is vibrating the frame differently than before.

It can also be a cracked frame (usualy near the top of seat stays) changing the frame vibrations.

It can also be a hall sensor problem, so that the controller is driving the motor differently (if it's a dual mode contorller, many of those run sensorless if a hall fails, but the sensorless is often not sinewave and so it's louder).

 

[BrianMH13]

I performed option 1 and went for a quick ride. It seems the sound/vibration is still present. I checked the case bearings and they seemed good. One was pretty dry and had quite a bit of resistance. I popped the shield and greased it. It seemed to run smooth but yeah there seems to be a problem still. It's not a axel hardware thing I'm pretty certain nor spokes. I'll check for frame cracks but I doubt thats it either. I imagine its bushings/bearings in the drive system or the sensor. I remember when it started. It was all of a sudden and after about 7 miles ride. I had experienced previously three times where the motor stopped responding once at the same 7mi into a ride. When I cycled the power it worked again afterwards. I wonder if that was indicative of the sensor going out.

I don't know what type of controller it has. I'll look that up.

The description on the Aipas website doesn't give any specs only description that may be suggesting its dual mode saying (low motor running noise)

Could I determine if it has a hall sensor by how many pins are in the connector?

 

[amberwolf]

I performed option 1 and went for a quick ride. It seems the sound/vibration is still present. I checked the case bearings and they seemed good. One was pretty dry and had quite a bit of resistance. I popped the shield and greased it. It seemed to run smooth but yeah there seems to be a problem still.

If a bearing was hard to turn, it usually means at a minimum there was corrosion or debris intrusion. That usually means damage / pitting / etc. which acclerates further wear (even with relubrication),. even if you can't see it with the naked eye.

If it ever ran rough, it probably has an even more damaged race and/or balls; tha'ts usually what causes the roughness.

I remember when it started. It was all of a sudden and after about 7 miles ride. I had experienced previously three times where the motor stopped responding once at the same 7mi into a ride. When I cycled the power it worked again afterwards. I wonder if that was indicative of the sensor going out.

That can be overheating, causing the sensor to not operate correctly. Damage can be cumulative.

There are posts and threads here with how to test hall sensors in various ways. Also on ebikes.ca in the Learn section.

The description on the Aipas website doesn't give any specs only description that may be suggesting its dual mode saying (low motor running noise)

That makes it that much more likely that it's now running sensorless and so it's probalby driving the mtoor with squarewave (noisy) instead of sinewave (quiet).

Could I determine if it has a hall sensor by how many pins are in the connector?

Almost all of them have hall sensors, except for certain tiny geared hubmotors.

If there are more than three pins in the connector, then there is probably at least a speed sensor (three more pins) and probably full UVW / ABC halls (five pins, two for 5v/ground, three for signals). PIns beyond those can be a speedometer-only sensor, thermistor, etc.