KT motor controllers -- Flexible OpenSource firmware for BMSBattery S/Kunteng KT motor controllers (0.25kW up to 5kW)

[stancecoke]

Did you drive by direct drive motor ?when i started the motor,there were some noise and the motor was shaking . then i changed the "motor special angle " as "160"(the best number),but i could still feel that the motor is shaking.are there any parameters to adjust？

I'm using the BionX IGH3 with our firmware. It's completely quiet except small resonance vibrations in the range of 12-14 km/h.
You can finetune the angles corresponding to the hall events. Xnyle has done this successfully, he implemented a function to show the details in BluOsec.
If there is massive noise, please swap the hall-sensor / phase wires to find the correct scheme.

I'm currently running a Sempu T4 torquesensor using Stancecoke's fw in tqsr mode and this also works extremely well, pretty much faultless in fact.

Nice to hear you finally got it going!

regards
stancecoke

 

[j bjork]

Thank you, I have not noticed any jerkyness in the throttle. But I have not tried very much. What I mean is more the power from standstill. Even if I hold brake and full throttle, and relese the brake I think it is a little slow the first meter or so. I don`t know, maby that is how it always is. It is my first e-bike, so I don´t really have anything to compare to. I have a dd hub and 18fet controller.

 

[geofft]

Thank you, I have not noticed any jerkyness in the throttle. But I have not tried very much. What I mean is more the power from standstill. Even if I hold brake and full throttle, and relese the brake I think it is a little slow the first meter or so. I don`t know, maby that is how it always is. It is my first e-bike, so I don´t really have anything to compare to. I have a dd hub and 18fet controller.

This sounds like you may not be getting a full throttle signal - I think most of these controllers require around 4v on the thr signal line to achieve somewhere near full throttle. I guess you may have already checked this, but if not maybe worth doing a quick meter check on your full throttle voltage?

I'm currently running a Sempu T4 torquesensor using Stancecoke's fw in tqsr mode and this also works extremely well, pretty much faultless in fact.

Nice to hear you finally got it going!

Yeah, I gave up with the T2 type. Although I got it working ok I just didn't like the fact that just resting your feet on the pedals would propel you up the road, so I went for the T4. This gives a much nicer riding experience and they aren't much more expensive than the T2 really...

 

[glebor]

Hi. Which line of code to remove if you want to program the ZWS controller to work properly?

 

[haiyi911]

thanks ，stancecoke.

Hi,Xnyle
i want to aviod the motor startup noise,so i follow "Diagnosing motor startup problems using the app”step,
is my motor angle 105-180=-75??
and how to adapt the number -->Hall Angle 4,Hall Angle6.....Hall Angle5 with different motor?

 

[stancecoke]

Hi. Which line of code to remove if you want to program the ZWS controller to work properly?

Just disable the option "Enable rotor angle correction" in the Java tool.

regards
stancecoke

 

[geofft]

i want to aviod the motor startup noise,so i follow "Diagnosing motor startup problems using the app”step,
is my motor angle 105-180=-75??

I think (with angular measurement) -75deg equates to +285deg? I'm sure someone will correct me if I'm wrong.. :wink:

 

[haiyi911]

i want to aviod the motor startup noise,so i follow "Diagnosing motor startup problems using the app”step,
is my motor angle 105-180=-75??

I think (with angular measurement) -75deg equates to +285deg? I'm sure someone will correct me if I'm wrong.. :wink:

Thanks,you are right!

 

[Xnyle]

Erm, your hall order goes in the wrong direction, as if you are pedaling backwards. Order has to be clockwise. I'm not quite able to spontaneously wrap my head around that, I guess in this case you have to look at the rising edge.

As your Sinus curve also looks like shit I'd say maybe 250°, so motor angle ~175? Id suggest you first fix the order and then look at the diagram.

 

[haiyi911]

Maybe my direct motor hall order is counterclockwise .when the order was clockwise on app,the motor run reverse even though exchange the phase wrie or “motor.c"--hall order.

 

[stancecoke]

Maybe my direct motor hall order is counterclockwise

Don't give up. Try to find the right wire combination, otherwise the firmware can't work properly. (No firmware can work properly, if the wires aren't connected in the right order...)

regards
stancecoke

 

[haiyi911]

Don't give up. Try to find the right wire combination, otherwise the firmware can't work properly. (No firmware can work properly, if the wires aren't connected in the right order...)

Thanks.
I'v changed the hall order in the code three times，(such as 1-3-2-6-4-5,2-6-4-5-1-3,4-6-2-3-1-5)

switch (hall_sensors) {
case 2://rotor position 180 degree
case 6:
.

and also exchanged the wire combination,so motor was forward(cw）and quiet every situation，but the hall order was still counterclockwise on the APP.It's strange.when the hall order was clockwise on the APP,the motor was backward and noise.maybe my motor hall order is ccw.

when i started the motor,i could hear the noise maybe the commutation voise and feel the motor shake. with the speed increased,the niose disappered.is it normal?
[youtube]https://youtu.be/WtBDmOn0-nE[/youtube]

 

[Valopallo]

Hello. I'm having this annoying feature at the moment so that when I start pedaling, the motor assistance kicks in pretty close to my liking(about 1 sec). But when I stop pedaling and the assistance shuts off, my DD motor slows dramatically and makes the ride a bit like push and pull. What settings should I change at the BluOSEC app to smooth the assistance? The problem wasn't that bad when my throttle was working because I could just smooth the assistance with it.

EDIT: Ramp end and Ramp start are the settings I suppose? So, can I just start to try different values freely or are there any limits or some setting that can damage my system if not correct?

 

[stancecoke]

when i started the motor,i could hear the noise maybe the commutation voise and feel the motor shake. with the speed increased,the niose disappered.is it normal?

yes, that's normal, as the motor starts in 6-Step mode and switches to 60° interpolation at 3erps.

} else if (ui16_motor_speed_erps > 3
						) {
					ui8_motor_state = MOTOR_STATE_RUNNING_INTERPOLATION_60_DEGREES;
				} else {
					ui8_motor_state = MOTOR_STATE_RUNNING_NO_INTERPOLATION_60_DEGREES;
				}

... my DD motor slows dramatically and makes the ride a bit like push and pull....
EDIT: Ramp end and Ramp start are the settings I suppose?

No, this seems to be a matter of the current cal b value, that is set automatically at startup. Please check battery current and current target, in the situation, when the motor should turn free.

regards
stancecoke

 

[haiyi911]

yes, that's normal, as the motor starts in 6-Step mode and switches to 60° interpolation at 3erps.

ok,i see. How to understand the following formula? 256->360°?
ui8_interpolation_angle = (ui16_PWM_cycles_counter_6 << 8 ) / ui16_PWM_cycles_counter_total;

 

[Valopallo]

No, this seems to be a matter of the current cal b value, that is set automatically at startup. Please check battery current and current target, in the situation, when the motor should turn free.

regards
stancecoke

Ok. So BluOSEC screen shows;
Battery current = 0,2A(46/314) when standstill, and when motor is running freely(pedaling tire at air) battery current is 3,6 ~ 4,6A and changing. When rotaing the wheel without pedaling the reading is 0,2-0,4A.

Current target = -0,2A(314) when standstill, and when rotating freely(pedaling tire at air) current target is 32A.

Battery current cal a = 46
Battery current cal b = 314

Obviously this is not correct. My bms current is limited to 15A and single cells are 5A.
What should I do to make this correct?

I also thought to make a video for the upcoming(?) BluOSEC User's manual.
https://youtu.be/ydwLHxxih28

Also my State is 250|1 when not pedaling if it helps.

 

[stancecoke]

Current target = -0,2A(314) when standstill, and when rotating freely(pedaling tire at air) current target is 32A.
Battery current cal a = 46
Battery current cal b = 314

I think your video is in Torque-Simulation mode with highest assist level?!
Cal b seems to be correct. For a S06S Cotroller cal a has to be something about 100.
I've added this information in the Wiki Pages. I'm not sure about the 12 FET, if somebody has reliable information, please let me know, or directly edit the Wiki.

Battery Current cal a: Factor a in the calibration function. 1A = a/10 * ADC value. Required for internal calculation of the current from the 10bit ADC value. For a 6FET and 12FET controller the value has to be something around 100, for the 18FET about 50.

regards
stancecoke

 

[Valopallo]

Cal b seems to be correct. For a S06S Cotroller cal a has to be something about 100.

Ok. Changed the battery cal a value to 100 and now the battery current and target current match very closely.

I think your video is in Torque-Simulation mode with highest assist level?!

Yes. I understood that torgue-simulation mode is automatic because I don't own a Torque sensor. I also understood that S06S is a torque simulation controller. Torque simulation is a mode? What other modes are there? Why did you mention that? It seems I may be missing something very basic here...

 

[stancecoke]

Ok. Changed the battery cal a value to 100 and now the battery current and target current match very closely.

:thumb:
Is the motor still braking that hard, if you stop pedaling? Please try this on the road, as with wheel in the air, it is not representative.

Yes. I understood that torgue-simulation mode is automatic because I don't own a Torque sensor.

Don't worry . If the Controller detects PAS-Pulses and the Torque-Sensor-Mode is not enabled, it works in torquesimulation mode automatically. If you don't mount a PAS, you can go just in throttle-mode.

regards
stancecoke

 

[Valopallo]

:thumb:
Is the motor still braking that hard, if you stop pedaling? Please try this on the road, as with wheel in the air, it is not representative.

Ok. I took a spin outside with the bike and it feels a bit more responsive. The original problem didn't seem to change much or at all, though. I'm also guessing this phenomena might purely be because I'm quite a lightweight rider (62kg) and the normal and quite strong magnetic power of the direct drive motor is just what feels almost like the regen braking feeling when not pedaling or using throttle. I also notice the same thing when I'm riding downhill and I shut down the controller and don't pedal. -> The wheel spins freely without the magnetic force of the dd motor, so it's almost like the non-friction-feeling of a geared motor. But sure, If you can help me to try some other settings like ramp end and ramp start with it, please do so. The push and pull feeling is a bit annoying afterall.

Don't worry . If the Controller detects PAS-Pulses and the Torque-Sensor-Mode is not enabled, it works in torquesimulation mode automatically. If you don't mount a PAS, you can go just in throttle-mode.

Ok. Just to make this clear; PAS is not equal to the torque sensor? Right. I have 12magnet PAS and I also have a throttle, but I accidentally broke my throttle's sensor from the handlebar and I'm waiting for a new one arrive from china and I'm using PAS only at the moment because of that. I do notice that when I pedal slower the assistance pulses come at slower pace and when I increase the pedaling speed those pulses also come faster pace. <-This I suppose is simulation and if I had installed an external torque sensor to the rear dropout of my bike or to the chains, I could use torque sensor mode, which would override the need for the simulation. Am I correct?

 

[stancecoke]

The push and pull feeling is a bit annoying afterall.

Please reduce values of Gain I and Gain P, e.g. by factor 10 or 100. You can finetune them later, if you have found a satisfying range.

if I had installed an external torque sensor to the rear dropout of my bike or to the chains, I could use torque sensor mode, which would override the need for the simulation. Am I correct?

There are several bottom bracket torque sensors availabe.

regards
stancecoke

 

[Diva]

Hello, and thanks for the great work

I currently ride with a tsdz2 motor with casahino's firmware, it works really well. But the thing is that I'm pushing the motor too far and I'm sure it won't last long. The cold weather acts as a good aircooling system, but I fear to fry everything in the summer. So I'm thinking of building a mid drive with a sensored outrunner, and power it with a KT controller and torque sensor. I'd like to be in the 700W area.
How many erpm can a KT controller handle? At the beginning of this topic, casahino was talking about 40k erpm as the maximum, is that value still valid or does the "high motor speed" option allow more? The outrunners I'm looking at are near the 100k erpm zone...

Thanks in advance for your help

Didier

 

[casainho]

Hello, and thanks for the great work

I currently ride with a tsdz2 motor with casahino's firmware, it works really well. But the thing is that I'm pushing the motor too far and I'm sure it won't last long. The cold weather acts as a good aircooling system, but I fear to fry everything in the summer. So I'm thinking of building a mid drive with a sensored outrunner, and power it with a KT controller and torque sensor. I'd like to be in the 700W area.
How many erpm can a KT controller handle? At the beginning of this topic, casahino was talking about 40k erpm as the maximum, is that value still valid or does the "high motor speed" option allow more? The outrunners I'm looking at are near the 100k erpm zone...

Thanks in advance for your help

Didier

Why don't you go with the popular and most cheap TSDZ2 mid drive motor that we did develop the OpenSource firmware for it and can handle 500W continuously and 900W peaks?? It has a motor temperature sensor and firmware reduces power when motor is to much hot. Also works with KT-LCD3 or Bafang 850C color LCD.

 

[Diva]

Hi Casahino!

That's what I said: my bike is now powered by an opensource firmwared tsdz2. 36V motor powered by a 48V battery, it rides really nice. But the freewheel of my metal gear is stuck (after maybe 1000km), and I replaced it by a blue gear I had in stock that only lasted 20 km even if I tried not to accelerate too hard (but always at full power).
It rides almost like my former (chipped) Bosch Performance, but I rode more than 8000 km on this one without any problem or overheating (except the battery, 10s4p Samsung 29E that I changed for 30A-able LG's). I'm a bit fed up with all the issues I've had on the tsdz2.
Plus I will take part to the next Sun Trip (in 2020, from France to China with a solar bike), and I would have liked to build my own drive and test it before, so I'd use it on the trip :wink: I'd never do that with a tsdz2...

 

[stancecoke]

How many erpm can a KT controller handle? At the beginning of this topic, casahino was talking about 40k erpm as the maximum, is that value still valid or does the "high motor speed" option allow more? The outrunners I'm looking at are near the 100k erpm zone...

The max speed is limited to PWM frequency / 30 erps, to make sure that we get a proper sinusodial shaped phase current.
In normal mode: PWM frequency = 15625 Hz --> max erps = 521 --> max erpm = 31250
In high speed mode: PWM frequency = 20833 Hz --> max erps = 695 --> max erpm = 41666

You could try if it works satisfying with e.g. 20 points per sine period. I don't own a motor that turns that fast (I'm preferring direct drives ) so I can't try it.

regards
stancecoke