Motor going both forward and reverse with KT controller

[brhooma]

Hi,

I bought a KT36ZWSRTD to replace a controller on a foldable Chinese bike (Fiido D3).
The motor spins forward on low speeds (no display so I can't tell what speed) and then spins in reverse when I twist the throttle a bit more. Do I have a faulty controller or am I doing something wrong?. Any help would be appreciated.

 

[amberwolf]

that sounds as if the throttle is perhaps hooked up to the wrong connector.

perhaps the one it's hooked up to is actually a forward/reverse connector that responds to voltages above about 2.5v by reversing the rotation?

 

[brhooma]

there are only two connectors with three wires, throttle and pas.

 

[amberwolf]

then i don't know; iv'e never seen a controller that could behave that way using either the throttle or the pas connector.

 

[amberwolf]

the only other thing i can imagine is if the phase / hall wires are not in the correct order. i have never actually seen rotation direction change from this, and don't see how it could happen, but there's nothing else that could be caused solely by amount of throttle and change rotation direction, either.

hall/phase wires don't always (usually don't) match color for color, so you may have to experiment with the order to get both the right direction and correct operation (including low no-load / offground power usage).

 

[AHicks]

I don't mean to seem smug here, but I think a call to the people that supplied the controller might be in order, or possibly, to KT themselves? Ask them for their opinion. Suddenly reversing mid throttle like that is pretty crazy.....

 

[brhooma]

hall/phase wires don't always (usually don't) match color for color, so you may have to experiment with the order to get both the right direction and correct operation (including low no-load / offground power usage).

Thanks for the advice, I did that and manage to get the wheel to spin in the right direction but now it makes loud noise when accelerating, I guess I need to experiment with the order more.

I don't mean to seem smug here, but I think a call to the people that supplied the controller might be in order, or possibly, to KT themselves? Ask them for their opinion. Suddenly reversing mid throttle like that is pretty crazy.....

I did get in touch with seller and they kept insisting that my motor is faulty even after I showed them the motor working perfectly fine with a different controller and their solution was to send the controller back to China to get a partial refund.

 

[E-HP]

Thanks for the advice, I did that and manage to get the wheel to spin in the right direction but now it makes loud noise when accelerating, I guess I need to experiment with the order more.

Possibly this diagram posted by TommyCat would be of assistance:
https://endless-sphere.com/forums/viewtopic.php?f=2&t=103373&p=1512036#p1512036

 

[docw009]

Controller is advertised by one ebay seller with an optional dual mode that allows it to work with sensorless or sensored motors. Switches to sensorless if sensors aren't working. Sounds scary to me. That should be a option set by user, not the controller.

The above is the only scenario I could see for a unit to change direction.

By the way, it must be a direct drive motor if you can see it change direction.

 

[amberwolf]

Controller is advertised by one ebay seller with an optional dual mode that allows it to work with sensorless or sensored motors. Switches to sensorless if sensors aren't working.
<snip>
The above is the only scenario I could see for a unit to change direction.

this does not happen on the grinfineons that do this (autoswitch to sensorless when sensors fail). they continue to operate in the correct direction as before.

 

[brhooma]

Tried all the combinations. Motor either stuck, spisn in the right direction but makes noise or silent but goes both forward and reverse. I guess I'm just going to run it without connecting the hall sensor for now while I find a solution.

 

[walrus]

Old thread, I know, but I just had the exact same problem with a KT controller and this was the only place where I found anyone mentioning this issue. So for anyone who has the same issue, here's how I fixed it.

In the controller parameters, there is option L3. This option only exists for dual mode controllers that can work with sensored and sensorless brushless motors. If this option is set to 0, the controller will only activate non-hall (sensorless) mode when there is a hall sensor failure. If this option is set to 1, the controller "will choose the proper model to use according to the controller system optimization" as stated in the documentation. This value defaults to 1 and setting it to 0 fixed this problem for me.

Be aware that this still means that in case of a hall sensor failure, for example if the sensor cable becomes unplugged, the motor will reverse direction immediately and with force! Depending on the vehicle you are riding, this may have catastrophic results. I recommend using the motor in sensorless mode if there is a significant risk of getting hurt if this occurs!

A bit more technical detail if you are interested. The motor I have is direct drive, has 30 magnets and a hall sensor angle of 36° (360°/30*3, the hall sensors are spaced three magnets apart). I suppose that this unconventional hall sensor angle is causing the problems. With the hall sensors connected, the direction of rotation is somehow opposite to the direction that results without hall sensors connected. Therefore, it is safe to assume that the "controller system optimization" at some higher speed decides that it prefers to ignore the hall sensor signal. But this then results in an immediate reversal of direction of rotation.

On the plus side, I now have a hacky reverse switch by simply disconnecting the hall sensors. My controller doesn't have a reverse mode by default.

I haven't yet tested while actually riding. So it may turn out that the controller doesn't actually work better with this hall sensor arrangement being used. So I might still prefer to run this motor sensorless, if there are any issues.