AnDiabhal said:
Hi all just signed up today and have just received my ebike battery for my project, but after a bench test I have an issue, first of all....
Spec -
Vevor 2000w 48v DC brushless 3 cables and hall sensor plug
Vevor 48v 2000w 34A controller (with shunt mod)
48v 20AH 50A BMS ebike battery
Note that modifying the shunt means the controller can no longer determine actual currents, so it can't operate as designed. Depending on the system and load, it may be unable to run the motor correctly in some circumstances. (or it may damage the controller if the load is high enough, as it won't know the load is that high and it won't be able to protect itself against that).
Probably doesn't have anything to do with the problem you're seeing, but something you might want to know about.
Bench testing setup and when speed selector
switch is "off/middle" voltage at motor is 20V
When selector is "low" voltage is at 8V
When selector is "high" voltage is 60V
FWIW, voltage at the motor is PWM'd DC into an inductor (the motor) and so is not easily/simply measureable with just a voltmeter. With the meter, depending on how it samples and at what rate, you may even see voltages higher than battery voltage (and higher than actual average voltage on the motor) because of inductive spikes or BEMF from the spinning motor.
Issue I am experiencing is when selector is in high the motor audibly and visually is running slower than "middle" speed. Would there be something stopping the motor reaching "high" speed?
There's a couple of possibilities.
The most likely thing is the wiring combination between motor and controller for the phase and hall wires is incorrect. Color matching is not necessarily correct, even for parts that come in a "kit". Some controllers have a self-learn, or auto-identify, or similar function. If it does, use that, and see if it corrects the issue.
If it does not correct it, or if it doesn't have autolearn, then if you have an amp meter (or wattmeter) that can measure the no-load current from battery to controller, it should be a couple of amps or so, generally, for full throttle at the max motor speed. If it's significantly higher, it is probably a wrong combination. This can be fixed, by procedurally switching pairs of phase wires until the current drops and motor speed is normal. If rotation is in the wrong direction, you can then swap hall wires until rotation is correct and current is low, and motor speed is normal.
Another possibility is that the controller is unable to correctly read the hall signals when it's spinning faster, so it slows down until it can. This isn't likely, unless this only occurs when you are using full throttle on the high speed selection, and it operates normally once you drop below some amount of throttle on that same selection. If this is the case, it may be noise in the hall lines (fixable with shielding around them, or it might be the controller not being able to read that high an RPM (not likely but possible), or the hall sensors placed in the motor where they simply don't generate good clean signals.
If it weren't for the motor voltage readings you get, I'd've guessed that the switch was miswired or built wrong, so that the "open" connection was on the "high" marking and the high connection was on the "medium" marking.
I have an RPM reader at work I'll be able to see exactly what I'm getting, will try tomorrow and update! Hoping my controller ain't goosed! 
