NB Power direct drive motor with KT controller unresponsive

[CargoMechanic]

Hey guys,

I have just fixed damaged motor cable for a friend of a friend on an NB Power direct drive motor. The cable was damaged where it exits the motor axle. I have shortened the cable and re-soldered the hall effect sensor and phase wires. This is a job I have done numerous times without any issues. I have triple checked my connections inside the motor and out and everything appears good.
Unfortunately when I connect the motor up to a KT controller and 36v battery to test, the motor is completely unresponsive, there isn't any motor shuddering or noise, nothing. I have tried adjusting the P1 motor characteristics setting on the KT display but it doesn't make any difference.
I have just tested a 500w planetary geared motor with the same controller and it runs perfectly.
Unfortunately I don't have the original ebike kit with me to test the motor. I believe this is the kit here: http://www.nbpowers.com/pd.jsp?id=33

Any advice would be much appreciated.
Thanks, Jasper

 

[fechter]

Did you try measuring the voltage on the hall signal wires while turning the motor by hand? Each signal wire should toggle between 0-5v as the motor is turned.

 

[E-HP]

If the motor is “stuck”, you could try swapping the blue and green phase conductorsnto see if you get movement, and go from there.

Testing BLDC motor's Phase Wiring - Hall Sensors and Wiring. - Electricbike.com Ebike Forum

Testing electrical components of a typical hub motor with a digital multi meter. First take a few minuets and look over the wiring going into the motor for cuts, scrapes, missing insulation and other wiring issues or poor connections that should be resolved first. Smell for...

electricbike.com

 

[amberwolf]

If the motor cable was damaged while the controller was operating, or the contorller was operated after the cable was damaged, it's likely there was contact between at least one phase wire and at least one hall wire had their conductors touch, shorting battery-level voltages to the 5v-level parts, which can damage or destroy the hall sensors.

Additionally, the controller that was on the bike the motor is from is itself probably damaged; FETs and/or MCU inputs for the halls.

If the phase wires shorted to the hall 5v power line, it could even have damaged or destroyed any or all 5v devices powered from the controller (throttle, PAS, etc). Some controllers have a diode in that supply line to the motor halls to help prevent this, but not all fo them do, and it doesn't always have a high enough voltage rating to prevent it.



If the KT has a self-learn and/or is a dual-mode (sensored/sensorless) controller, you can try running it with either or both of those, the latter of which will get around the hall sensor issue.

 

[CargoMechanic]

If the motor cable was damaged while the controller was operating, or the contorller was operated after the cable was damaged, it's likely there was contact between at least one phase wire and at least one hall wire had their conductors touch, shorting battery-level voltages to the 5v-level parts, which can damage or destroy the hall sensors.

Additionally, the controller that was on the bike the motor is from is itself probably damaged; FETs and/or MCU inputs for the halls.

If the phase wires shorted to the hall 5v power line, it could even have damaged or destroyed any or all 5v devices powered from the controller (throttle, PAS, etc). Some controllers have a diode in that supply line to the motor halls to help prevent this, but not all fo them do, and it doesn't always have a high enough voltage rating to prevent it.



If the KT has a self-learn and/or is a dual-mode (sensored/sensorless) controller, you can try running it with either or both of those, the latter of which will get around the hall sensor issue.

This is great thank you. I'll check out the controller when I get a hold of it. I've tested the hall sensor wires with a multimeter. The 5v supply is pulled down to 2v as soon as the motor is plugged in to the controller. Blue and green sensor wires sit at 2v without voltage fluctuation when the wheel is turned. Yellow seems to be working fine. Would you recommend replacing the PCB with the hall sensors or just the sensors?

 

[amberwolf]

I've tested the hall sensor wires with a multimeter. The 5v supply is pulled down to 2v as soon as the motor is plugged in to the controller.

That means the 5v supply in the controller is damaged and can't supply enough current anymore (in which case the controller won't operate at all while the motor is plugged in, as the MCU won't be powered/working), and/or the 5v wire in the motor cable is shorted to something else (probably a hall signal). That may be a direct wire-to-wire short, or it may be an internal short in a hall sensor.

Documenting and then disconnecting the wires from the PCB inside the motor and testing for shorts between each one and every other wire in the motor cable will find an inter-wire short.

While those are disconnected, testing between ground, signal, and supply pins of each hall will find a shorted hall.

You can also test the controller 5v. While the halls are plugged in, check the votlage at the throttle 5v line. If it is also dropped down, the controller is probably also damaged from the motor cable damage.

Blue and green sensor wires sit at 2v without voltage fluctuation when the wheel is turned. Yellow seems to be working fine. Would you recommend replacing the PCB with the hall sensors or just the sensors?

If you can find a PCB identical to the one you have, with the same sensors all facing the same way as yours in the same positions as yours, you can replace the whole PCB if that's easier.

Otherwise you simply replace all three hall sensors with honeywell / allegro / etc SS41F (or G) / SS411F (or G) or equivalent. The typical characteristics they need are open collector, latching, 3 to 4v minimum / max 20 to 30v (or higher) supply range, etc. (see the spec sheets for the 41/411 for comparison). 411 usually operates at a lower voltage than 41, G has a higher temp rating. There are other variations on the spec sheets with other differences.

Note that a whole PCB with sensors is likely to have unknown clone sensors on it, so you won't really know what properties / specs they have. Using genuine sensors (sourced from Mouser, etc vs ebay and the like) may be more reliable.

None of the options will prevent a failure from a motor cable short or other outside-specs damage; that requires redesign of the system to prevent such things.