Acceleration stuck after throttle release

lfbferraz commented 1 year ago

Variant

ADC

Control type

Commutation

Control mode

Voltage

What is the bug and how to reproduce it ?

Hey guys,

I apologize for writing in the bugs section, but I couldn't find an appropriate section to share my issues with you experts.

I modified my daughters Ride on Car, removed the cheap 25w motors with plastic gearbox to the hoverboard chassis with the motors fitted to the original plastic wheels and a few more modifications.

I'm using a RC tx/rx connected to the right PWM cable, using only the speed input, and I've changed the steering coefficient to 0 to ignore any disturbance from the open steering channel.

The car runs fine when the wheels are off the ground or unloaded. The problem is that when my daughter is in the car, I press the accelerator lightly and the car accelerates hard, and after I release the accelerator it continues accelerating for a few seconds. I almost crashed the car with my daughter in it and it's very dangerous.

Thing I tested: Change voltage to torque - does the same Change RATE, FILTERS, COEFFICIENTS, I_*MAX and N_MOT_MAX, also no improvements.

I have to say it was running fine and issue started after:

I changed 8'' wheels from a Hoverboard XL to 6'' wheels from a common hoverboard
I connected the RC receiver at 5v through a DC-DC (15 to 5v) on the same cable as the right board cable, to use a steering servo, which is very weak at 3v. (the right cable in theory supports 5v, and in my mind I wonder what difference it makes if the no-load operation is fine?)
I put a kill switch and fuse between the battery and the mainboard (I'm glad I did this before I let my daughter ride)

I don't actually have a lot of time to dedicate to this project, so I wanted to ask if anyone has any clues as to what might be wrong?

The next thing I plan to test is remove the steering servo and power the RC receiver back to 3v. Also build a new code with a 50 rpm cap and a very low rate for safety).

Thanks in advance Luis

Candas1 commented 1 year ago

Hi,

Have you checked this?

The pwm implementation is prone to noise, it needs rework. But there are advices on the wiki page to reduce the noise.

lfbferraz commented 1 year ago

Hi Candas1

Yes, I did red, but to be honest didn't care enought as it was previouly working. I've just twisted the PWM and GND cables together and kept it short, around 5cm, but still connected to the board connector. I will try to solder a capacitor and make it even shorter, as I can't find a shielded cable or ferrite around.

Thanks

Candas1 commented 1 year ago

Maybe you can also try another variant to prove if the problem is the pwm signal.

Candas1 commented 1 year ago

Any news ?

lfbferraz commented 1 year ago

Yes and no. I manage to change the transmitter to board cable to a shorter one, twisted with the gnd and soldered a capacitor on the board input. But didn't had time to test yet.

Hope to test it next week.

On Thu, May 4, 2023, 3:04 PM Candas1 @.***> wrote:

Any news ?

— Reply to this email directly, view it on GitHub https://github.com/EFeru/hoverboard-firmware-hack-FOC/issues/394#issuecomment-1534745884, or unsubscribe https://github.com/notifications/unsubscribe-auth/AZUHI3VOC735WDQCXS75H3LXEOSNTANCNFSM6AAAAAAWYZQN7M . You are receiving this because you authored the thread.Message ID: @.***>

lfbferraz commented 1 year ago

Sorry for the long delay to update it.

The capacitor + short twisted wiring + disconnecting some LEDs and the audio system from the main board have solved the issue. Another issue was that I set the filter so high that after releasing the accelerator, it was tamping down the current so slow that it was taking more than 1 second to cut the acceleration.

Closing the topic and sorry for raising it.

EFeru / hoverboard-firmware-hack-FOC