Waterlogged Bafang SWXH

[shemozzle999]

Hi Mark,

If you are going to remove the rotor from the stator, to expose the Hall devices for removal, you will need to apply a steady pressure to hub shaft to over come the force exerted by the magnets.

While supporting the outer edge of the rotor much like when you removed the cover plate, tapping it with a mallet will not work as the magnets will try to pull it back on.

I went down to my local Kwik-Fit type garage, they usually have an hydraulic press they use for removing wheel bearings - if you are lucky, like I was, the mechanic was fascinated when I showed him the motor and was only to pleased to help out.

 

[shemozzle999]

Thanks d8veh,

No I didn't know that. I was following the instructions that came with it, which say "turn the motor clockwise, along the normal advance direction" for both "winding of the motor" and "Hall of the Motor" tests. May well be an incorrect insruction, I will try this in reverse tomorrow!.

Thanks for the tip
Mark

Those instructions OK but apply to a direct drive motors, yours has an internal free-wheel mechanism so to get the motor inards to rotate, when unpowered, you have to turn it in the reverse direction as d8veh stated.

 

[mgfx]

Thanks again, I thought it was strange that I was getting no reaction at all from the tester. the knowledge on here is fantastic. Will keep you updated tomorrow!

Those instructions OK but apply to a direct drive motors, yours has an internal free-wheel mechanism so to get the motor inards to rotate, when unpowered, you have to turn it in the reverse direction as d8veh stated.

 

[Scimitar]

There's something else you could try - gently dry the motor in the blow from a fan heater or leave it somewhere warm for a few days. This is a common problem in builders' gen sets, where the windings get damp and the output drops to zero. You'd think that the insulation would prevent any effect, but it's only a thin coating on the wires and is permeable.
Drying it out usually cures it and once it's better, it just gets better when it's under load.
Not saying this is for sure what's wrong, but you would be as well to rule it out.

 

[mgfx]

Tried this today. Set the multimeter to 2k (lowest setting). Connecting probes together gave zero. A-B, B-C, C-A all gave reading of 0.001.

Highest setting is 2m. Touching probes together again came to zero. Each phase wire read 1.0 when "earthed" to the motor shell (its back in the wheel).

Connected the tester again, spun the wheel backwards at several different speeds, no leds lit on tester. Connected Hall wires to tester, again no leds.

Tested the tester motor windings section by connecting up to a friends sensorless SWXB motor. All Leds lit up as per tester manufacturers leaflet. So I assume that its working OK.

So, from these results do I assume that the motor has had it?

Testing motor windings with a mutimeter is very easy. Put the multimeter on Ohms. If there are several ohms ranges, select the lowest one. Connect the leads of the meter together, and it will read close to zero ohms. You have 3 phase wires, call them A,B, and C. Put one meter probe on A and one on B. You should get a reading of less than 1 Ohm . Now test between B and C . Finally test between C and A. The 3 readings should be identical or very close. Now put the meter on the highest Ohms range, or just on "Ohms" if there is only one range. Put one probe on any phase wire, and the other probe on the metal "stampings" on which the coils are wound, or if the motor is assembled, on the hub shell. The reading should be exactly the same as if the meter probes are connected to nothing. For this test, you must not touch the metal part of the probes with your fingers. If the stator passes all these tests, the coils are OK. Windings seldom give trouble. Post your test results.

 

[neptune]

Do not be so quick to write off the motor. The results of the first tests, A-B, B-C, and A-C are satisfactory, and show that there are no breaks in the internal coils.
Re the final test. If by 1.0 you are saying 1 Megohm, there is some slight leakage between the coils and "earth", meaning the hub shell, but not enough to be a problem. I t does however show internal dampness, so stick it on a radiator for a couple of days.
Disregarding the hall sensors, a motor is very simple. You have coils wound on a set of iron stampings, and some magnets. That is all. Get , or make a 12 volt test lamp, with a 12 volt flasher bulb. First, test the test lamp on your car battery. Now connect it between A and B wires, and spin the wheel backwards. It should light as the motor acts as a generator . Repeat test, B to C and C to A. It should light in all 3 cases.
Final test. connect one wire of test lamp to hub shell. Whilst spinning wheel backwards, apply other test lamp wire to A, B and C in turn. The lamp should NOT light on any of this group of tests.
If the motor passes all these tests, then it is OK electrically, with the possible exception of the hall sensors. You have proved that all coils are OK, and the motor itself is actually spinning inside the hub.
Please report test results, and ask if anything is unclear.

If your motor passes these test, re-connect the phase wires, and trace the phase wires back to the controller. Make a note of where they go, then disconnect them from the controller. Then repeat the test lamp tests on the disconnected wire ends, thus testing the continuity of the phase wires in the harness.

 

[Deleted member 4366]

It would be exremely unlikely that rain water damaged the windings. The only thing which would damage the windings would be if they were burnt, in which case they'd stink and would be black.

 

[neptune]

@d8veh. That is true. The only possibility is that an internal soldered joint might go open circuit due to corrosion/ chemical action, but only if the soldering was inadequate in the first place, but it could happen. I think the problem is hall sensors, harness or controller.
Actually, a motor with burned out windings could possibly still be repaired by rewinding. In the past, I have rewound all sorts of things including electric motors, car and motorcycle alternators, and even transformers. It just needs lots of patience, and a methodical approach.
As regards test equipment the simpler it is, the less likely it is to mislead you. A test lamp doesn`t lie, provided you test it first.

 

[Deleted member 4366]

True, but if a solder joint had gone, one phase would still be OK. To me, there's water where it shouldn't ought-ta.

 

[Scimitar]

It would be exremely unlikely that rain water damaged the windings. The only thing which would damage the windings would be if they were burnt, in which case they'd stink and would be black.

Dampness on its own causes failure.
If you read my post about alternators, you know what I mean.

 

[mgfx]

Neptune, thanks again for the advice.

I will take the motor out again and leave it to dry out with gentle heat over a few days.

The marking on the multimeter is 2m, so I presume this is Megohm.

I will do the test lamp, takes me back to messing about with cars in the days before ECUs and sealed units!

The only concern is that the tester I have worked fine on a similar motor (Bafang SWXB), connections were made at the same point(, ie onto the pins in the connector closest to the motor), but no leds lit when connected to this one.

I did test the continuity of the phase wires with the mutlimeter on the "beep" setting before starting on this, but will follow your suggestion.

One thought on this, would the penetrating spray, GT85 bike maintainance lubricator, penetrator, water displacer which I had to use to free the rotor from the axle/stator shaft cause problems?

Do not be so quick to write off the motor. The results of the first tests, A-B, B-C, and A-C are satisfactory, and show that there are no breaks in the internal coils.
Re the final test. If by 1.0 you are saying 1 Megohm, there is some slight leakage between the coils and "earth", meaning the hub shell, but not enough to be a problem. I t does however show internal dampness, so stick it on a radiator for a couple of days.
Disregarding the hall sensors, a motor is very simple. You have coils wound on a set of iron stampings, and some magnets. That is all. Get , or make a 12 volt test lamp, with a 12 volt flasher bulb. First, test the test lamp on your car battery. Now connect it between A and B wires, and spin the wheel backwards. It should light as the motor acts as a generator . Repeat test, B to C and C to A. It should light in all 3 cases.
Final test. connect one wire of test lamp to hub shell. Whilst spinning wheel backwards, apply other test lamp wire to A, B and C in turn. The lamp should NOT light on any of this group of tests.
If the motor passes all these tests, then it is OK electrically, with the possible exception of the hall sensors. You have proved that all coils are OK, and the motor itself is actually spinning inside the hub.
Please report test results, and ask if anything is unclear.

If your motor passes these test, re-connect the phase wires, and trace the phase wires back to the controller. Make a note of where they go, then disconnect them from the controller. Then repeat the test lamp tests on the disconnected wire ends, thus testing the continuity of the phase wires in the harness.

 

[mgfx]

Thanks Dave, as you and Neptune have suggested this is something I will try over the next few days

There's something else you could try - gently dry the motor in the blow from a fan heater or leave it somewhere warm for a few days. This is a common problem in builders' gen sets, where the windings get damp and the output drops to zero. You'd think that the insulation would prevent any effect, but it's only a thin coating on the wires and is permeable.
Drying it out usually cures it and once it's better, it just gets better when it's under load.
Not saying this is for sure what's wrong, but you would be as well to rule it out.

 

[mgfx]

Just replied to your earlier post!

I will read your post on alternators, but I have a gut feeling that water/damp is still at the root of this. There was a considerable amount of rust on the stator and rotor magnets and the central shaft, so what the water has done to the joints is anyones guess! Thanks again.

Dampness on its own causes failure.
If you read my post about alternators, you know what I mean.

 

[mgfx]

Hi Shemozzle, I have managed to borrow a small 3 leg gear puller which should do the job. First time I did it, put two wheel nuts on the axle and banged it down hard onto a block of wood whilst holding the sides of the rotor. I still have the bruises from when my hand slipped! Eventually, when the rusty shaft freed up, I was able to force it down enough to release the stator plate. Didn't fancy doing it again though, hence the gear puller!n Thanks again for your unstinting help and advice.

Hi Mark,

If you are going to remove the rotor from the stator, to expose the Hall devices for removal, you will need to apply a steady pressure to hub shaft to over come the force exerted by the magnets.

While supporting the outer edge of the rotor much like when you removed the cover plate, tapping it with a mallet will not work as the magnets will try to pull it back on.

I went down to my local Kwik-Fit type garage, they usually have an hydraulic press they use for removing wheel bearings - if you are lucky, like I was, the mechanic was fascinated when I showed him the motor and was only to pleased to help out.

 

[shemozzle999]

Hi Mark,

I would not worry about the test unit for a while.

You first need to establish if the cable and its connections to the motor are sound.

I dont know if you are proficient with a solding iron but as the coil wires are commoned together through the motor coils the only way I know to be certain that you have continuity from end to end of the cable wiring and no shorting between wires (i.e. cable damage) is to removed any sleeving where the 3 cables connect to the copper coil wires and disconnect them.
Check with the buzzer on the meter for end to end continuity and also from wire to wire to ensure that there is no shorting between wires. If they check out ok reconnect them and re-sleeve using fresh heat shink tubing.

I would then repeat this process with the Hall wiring.

Photos would help!

If it all checks out then you can move on to using the tester knowing the cable is intact.

 

[mgfx]

Hi Shemozzle,

Good idea, I will separate the motor again, undo the Phase wires and test continuity as you suggest, whilst the motor is "drying out". I have a feeling that all will be well with the wiring, as there in no external damage and apparantly all was working fine until it was "abandoned" in the rain for two or three months (just the front wheel really, rest was covered in a shed, and the battery was indoors) then it wouldnt run at all.
But you are right to suggest eliminating the wiring as a potential cause.

I am OK with a soldering iron, its the eyes that need the help!

Will put some photos up soon,will wait until dismantled again so I can show the wiring.

 

[mgfx]

Update.
Rigged up a test lamp, got a glow from A-B B-C A-C.
Difficult to connect to hub shell whilst spinning it , but holding a bare wire on to it, or on to the axle brought no light!

Motor is now stripped again, and sitting on a radiator, will leave it there for a few days.

Any thoughts on the GT85 causing a problem?

Do not be so quick to write off the motor. The results of the first tests, A-B, B-C, and A-C are satisfactory, and show that there are no breaks in the internal coils.
Re the final test. If by 1.0 you are saying 1 Megohm, there is some slight leakage between the coils and "earth", meaning the hub shell, but not enough to be a problem. I t does however show internal dampness, so stick it on a radiator for a couple of days.
Disregarding the hall sensors, a motor is very simple. You have coils wound on a set of iron stampings, and some magnets. That is all. Get , or make a 12 volt test lamp, with a 12 volt flasher bulb. First, test the test lamp on your car battery. Now connect it between A and B wires, and spin the wheel backwards. It should light as the motor acts as a generator . Repeat test, B to C and C to A. It should light in all 3 cases.
Final test. connect one wire of test lamp to hub shell. Whilst spinning wheel backwards, apply other test lamp wire to A, B and C in turn. The lamp should NOT light on any of this group of tests.
If the motor passes all these tests, then it is OK electrically, with the possible exception of the hall sensors. You have proved that all coils are OK, and the motor itself is actually spinning inside the hub.
Please report test results, and ask if anything is unclear.

If your motor passes these test, re-connect the phase wires, and trace the phase wires back to the controller. Make a note of where they go, then disconnect them from the controller. Then repeat the test lamp tests on the disconnected wire ends, thus testing the continuity of the phase wires in the harness.

 

[Deleted member 4366]

Where are we now? I can't understand the test you've done and it's too much to read everything again. Try this simple test: get a torch bulb and join two wires to it; connect the ends of the wires to any pair of phase wires; hold the axle still and spin the wheel backwards. You should see the bulb pulsing light. The lower the voltage of the bulb, the brighter it should be. A 1.5v one would be best, but don't spin too fast or it might blow. Repeat for the other two pairs of phase wires. If the phases generate, then they'll work as a motor.

Next check the hall wires 5v supply and each phase should pulse when you turn the wheel backwards. If that's ok as well, then you only have to get the phase wires and halls in the correct sequence. I guess you checked the colours accross the connectors.

 

[neptune]

I have never heard of GT85, but I assume it is similar to WD40. The laquer or varnish used on winding wire is very tough, and there are very few solvents or chemicals that effect it. I would say damage from GT85 is very unlikely, or the motor would not pass the tests described in your last post.

 

[mgfx]

GT85 is a bicycle maintainance spray, "lubricator, pernetrator, and water displacer with added PTFE" according to the tin!