KT controller setup help please

[SteveDillon]

Hi,
I'm putting together an ebike system using a KT controller (22A sine wave), LCD08 and a Shengyi DGWX2 motor with 36v battery.
At the moment, it's all connnected up 'on the bench' to check it all works.
I've set up the controller using what I hope are sensible values for my (at the moment) throttle only system as follows
P1=72, P2=6, P3=1, P4=0, P5=15, C1=2, C2=0, C3=0, C4=0, C5=10, C6,7,8,9=0, C11=0,C12=4,C13=0,C14=2,L1,L2=0,L3=1,L4=5

What happens is:
As I apply throttle, the motor starts fine (I'm hoping the hall wiring is OK), but as I apply more throttle and the motor speeds up, at some point the motor suddenly decides to be very noisy and doesn't run at all smoothly. The point at which the motor enters this noisy mode can be changed by altering the value of C5. For example, if C5 is 3 the motor enters noisy mode at very low speed and throttle input. This makes me think there is some sort of current limiting going on.

Other odd things as follows:
The battery voltage is reading 38.5v, but only one bar on the battery graphic is 'lit' ??
The speed reading is very very low.
The power reading seems very low as well; with C5=10, the power only seems to get to just 18W before the motor enters noisy mode, and with C5=3, only 9W is achievable.

I've connected the motor to my wife's ebike which has a Cycle Analyst and Grinfineon controller. It all works fine, and the CA displays a much more sensible power usage (around 50W at high speed and no load).

It seems that the KT controller is getting itself a bit confused on voltage and current, but I can't find any settings to resolve this. Is it still possible that the hall sensors are wired wrongly ? My limited understanding of this sort of thing leads me to think not; why would it run smoothly up to a certain speed (power, current) then snap into noisy rough mode ?

Any suggestions gratefully received.

 

[E-HP]

I'd try setting P5=0 and monitor the voltage to get more data when testing the behavior

 

[SteveDillon]

Thanks for the reply - I've done that and noticed virtually no change in the voltage as the throttle / speed increases. Maybe dropped 0.2v but nothing significant.
Used a DVM on the battery input to the controller which agrees with LCD08 display.
Even when the motor is in it's rough noisy mode, the battery voltage doesn't drop significantly.
Normally, I release the throttle as soon as the motor gets noisy (it's pretty bad), but if I keep the throttle on, the controller seems to shut down completely after a few seconds.
The controller behaviour seems to point to current limiting, but the display only shows 18W which doesn't tally.

I've tried various values for p5 - no difference in the fault.

 

[E-HP]

Try swapping the phase wires:

https://endless-sphere.com/forums/download/file.php?id=249668&mode=view

 

[SteveDillon]

Tried all 6 combos of phase wires - the original wiring is best, all others are either stuck or very rough from the word go. The original runs smoothly at first, then suddenly goes rough.
The fact that changing C5 (current limit adjust) can change the speed at which the motor starts to run badly, leads me to suspect some sort of current limit going on. However, the display power reading is very low throughout - with C5=10, the motor becomes noisy at an indicated power of 18W or just 500mA at my 36V.
I still reckon there is something going on with the current limit kicking in too early but I can't think of how to confirm/fix this .

Could the hall wiring cause this sort of effect ? I can't see how but I'm clutching at straws now.

 

[E-HP]

What are the wheel size and max speed set to in the general settings?

 

[SteveDillon]

max speed to 72kph, wheel size to 700c.
Have tried changing pretty much all settings to no avail - motor quiet at first until 18W power indicated, then noisy.
I have also noticed that applying a large throttle input relatively quickly will cause the motor to be noisy from low speed, whereas a very gradual increase in throttle will allow the motor to spin nicely up until the 18W power point.

I'm beginning to wonder whether it's something to do with spinning the motor prettyquick without any load, can't understand why that should be, but am struggling to understand what I'm seeing.

 

[SteveDillon]

Oh well.....
Was homing in on the current limiting circuitry of the controller - shorting out the current limit resistor got rid of the problem but is obvs. not a real solution. The problem could be made worse simply by putting my DVM probe on the current limit amplifier output which shouldn't happen. My mate who knows analogue design told me that the particular amp. design used in the controller can often oscillate at high frquency, so suggested a couple of mods.

Unfortunately I forgot that the on board caps hold charge for a long time, and so shorted something to something else with a good spark - resulting in a dead controller and unhealthy smell when powered up.