Promovec Carrier 2/6 error code 2

[karlpedal]

This is a danish 36V cargo bike Christiania Light Electric with build year 2019. It's got a Promovec (also a danish company) Carrier 2/6 controller labeled 51101-D03-225-S-0. I believe the controller will go by the description Carrier 2, Carrier 6 or D03. A reddit user states that Promovec just rebrands various parts, so maybe this controller is well known by the e-bike community. I'm new to e-bike tinkering but know my way around electronics and microcontrollers, etc.

Here a product image with pinouts from the epoxy(?) potted controller: https://static.cykelshoppen.dk/ai/1600/storage/images/products/51101-S04-195-E-0.jpg

The display is showing a error code 2 (just the digit 2 where speed is displayed) along with a wrench symbol, indicating a controller error (the documentation does not elaborate further on the particular issue). I have measured resistance between battery plus and minus to the three phase pinouts in the 9-pin connector. I've got 36 kohm on all three phases from plus and 65 kohm on all three phases from minus (six mosfets in total) which I believe indicates there's nothing wrong with the mosfets. Should I continue to examine the motor for a possible short? Does anyone know if it's possible to hook up the controller UART for better troubleshooting?

 

[fill]

Often you can find 2 x controller-display manuals 1x a shorter easy to read document, and also a longer more in depth detailed manual with all configuration details and which will include all the troubleshooting decodes for errors displayed.
including the keywords ".pdf" in a search will list the page count of relevant results in the list so you can discount those with less than 20? pages for a more efficient search.

#1 whats the fully charged battery voltage? and is it displayed correctly on the bikes display screen. ?

#2 Is every plug pushed fully home and inserted fully?

A controller error? - are lights powered through the controller and have you added any after market additions??
light switching mosfets? are frequently limited to ma capacity and can fail with excessive draw with unknown side effects.

The other 'common failure' is the hall sensors in sensored hub drives, which can fail with motor overheating, while not directly indicated by the display its worth ruling out with a simple test imho, just rotate the wheel backwards to illicit a response from active sensors

 

[E-HP]

The display is showing a error code 2 (just the digit 2 where speed is displayed) along with a wrench symbol, indicating a controller error (the documentation does not elaborate further on the particular issue).

Error code 2, while nor universal among various displays, often refers to a brake sensor issue. You could try disconnecting the brake cutoffs switches to see if the error clears, then troubleshoot from there.

 

[karlpedal]

Sorry for slow update. My appreciation for your troubleshooting suggestions!

I did some more research on the particular hardware (still don't know which model the motor is). According to Promovec documentation error code 2 "indicates errors in power consumption" and suggests "measure battery voltage and check motor wire for damage".

However, I charged the battery and measured 38.84 VDC (not connected to controller). Went on to do some measurements on the motor pinout (the wheel is still mounted on the bike). I found the attached pinout after a Google search and believe it's similiar to mine. Measuring between phases and GND I only see a very high resistance that's slowly dropping. Doing a diode measurement on the "phase sensor" (hall sensor) pinouts gives the same voltage (0.7 V if I recall that correctly) on all three. I got a slightly higher readout doing a diode test on the thermistor/motor-speed-pinout. Note that the battery wasn't connected during measurements.

Maybe I should continue measuring by feeding the +5 V pinout from an external power source and spin the wheel to see how signal from the hall sensors turn out. I do have access to an oscilloscope.

 

[E-HP]

Good update. Sounds like you’re making good progress.

Maybe try fully charging the battery to see if it will reset the error.

 

[amberwolf]

Went on to do some measurements on the motor pinout (the wheel is still mounted on the bike). I found the attached pinout after a Google search and believe it's similiar to mine. Measuring between phases and GND I only see a very high resistance that's slowly dropping. Doing a diode measurement on the "phase sensor" (hall sensor) pinouts gives the same voltage (0.7 V if I recall that correctly) on all three. I got a slightly higher readout doing a diode test on the thermistor/motor-speed-pinout. Note that the battery wasn't connected during measurements.

Maybe I should continue measuring by feeding the +5 V pinout from an external power source and spin the wheel to see how signal from the hall sensors turn out. I do have access to an oscilloscope.

Hall sensor testing in motors is best done as a voltage test wiht them connected to the controller and the controller turned on but not operating the motor. It's unlikely you'll get much useful from a passive test (like a diode test) unless the output transistor failed "hard" open or shorted (to ground, normally). You'll see the voltage on each signal pin toggle between around 0v and around 5v as you manually slowly spin the wheel backwards.

If that's not possible you'll need to provide your own 5v power to the sensors *and* a pullup on each hall signal via a resistor (1kohm-10kohm) to the 5v.

That's because motor hall sensors generally are open-collector types, so they don['t supply any signal on the output, they only ground whatever is supplied to them on the signal line from the controller, whenever they are turned on by a magnet passing***.


***exactly how they are turned on depends on the specific sensor, most are bipolar latching so they only change state when the polarity of the magnet changes, whcih on motors generally means every other magnet, or pair, so "pole pairs" is a common term for the magnet count.

 

[fill]

If Fully charged ?? <39v is NOT Healthy for a 36v battery pack., afaik you really want a voltage over 41v and the closer to 42v the better.. Check/confirm fully charged battery voltage,

Grin host on ebikes.ca a site with a good troubleshooting knowledgebase with step by step test directions for things like hall sensors etc.. (cant seem to post links atm?)

 

[karlpedal]

Hall sensor testing in motors is best done as a voltage test wiht them connected to the controller and the controller turned on but not operating the motor. It's unlikely you'll get much useful from a passive test (like a diode test) unless the output transistor failed "hard" open or shorted (to ground, normally). You'll see the voltage on each signal pin toggle between around 0v and around 5v as you manually slowly spin the wheel backwards.

If that's not possible you'll need to provide your own 5v power to the sensors *and* a pullup on each hall signal via a resistor (1kohm-10kohm) to the 5v.

That's because motor hall sensors generally are open-collector types, so they don['t supply any signal on the output, they only ground whatever is supplied to them on the signal line from the controller, whenever they are turned on by a magnet passing***.


***exactly how they are turned on depends on the specific sensor, most are bipolar latching so they only change state when the polarity of the magnet changes, whcih on motors generally means every other magnet, or pair, so "pole pairs" is a common term for the magnet count.

Wheel unmounted and fitted in a truing stand. I tried getting some measurements by feeding the hall sensors an external +5V supplied from an Arduino Uno. Female jumper wire connectors fit neatly over the pins in the connector coming from the wheel. After reading your comment I see why that didn't work out as I thought.

This is the setup I need to get some readouts, right?

Another tricky part is the potted controller circuitry. It prevents doing any kind of measurements on the sensor pins when connecting the wheel to the controller + battery. I have to work something out, like a breakout connector, that allows measurements.

 

[amberwolf]

This is the setup I need to get some readouts, right?

Yes. The resistor can be anything from around 1kohm to 10kohm for results that would probably be reliable; the hall can probably sink around 5-10mA to maybe 30-50mA (some can do more), so safer not to go any lower than say, 500ohm. Higher is usually ok, but the higher the resistance the more susceptible to induced noise the signal will be.

Another tricky part is the potted controller circuitry. It prevents doing any kind of measurements on the sensor pins when connecting the wheel to the controller + battery. I have to work something out, like a breakout connector, that allows measurements.

I've done that for testing the hall signals on a GMAC connected to a Phaserunner, using an extension cable cut open to access the wires inside.