TSDZ2 mid drive with 860C, 850C or SW102 displays only -- Flexible OpenSource firmware (Casainho code only)

[casainho]

Any plans to implement an eMTB mode where it automatically steps up through the assist levels depending on torque input?

Are you using the latest firmware version? If not, first try it and then give feedback.

 

[HughF]

Any plans to implement an eMTB mode where it automatically steps up through the assist levels depending on torque input?

Are you using the latest firmware version? If not, first try it and then give feedback.

Not yet, I am going to try and flash up from 0.55 and 0.6.7 (I think that's what I'm running) today.

 

[casainho]

Any plans to implement an eMTB mode where it automatically steps up through the assist levels depending on torque input?

Are you using the latest firmware version? If not, first try it and then give feedback.

Not yet, I am going to try and flash up from 0.55 and 0.6.7 (I think that's what I'm running) today.

You need to install latest version and calibrate your torque sensor.

 

[HughF]

You need to install latest version and calibrate your torque sensor.

Installed, now I will try to find some weights to use for calibration.

 

[sparked]

Olá!!

So after maybe 10 or 20 seconds you keep seeing Wait TSDZ2? Maybe there is a bug but it should timeout and say like error brakes or error TX...

Correct. Even after two minutes it still says 'Wait TSDZ2'. I don't have brakes and didn't change any of the wiring between flashes.

1. Rease motor firmware and test again to see which message you get
2. Flash an old motor firmware version to see if you get an error message

Flashing the motor back to 0.54 and then 0.56 did the trick! 0.54 gave me the right 'firmware error' message. Obrigado casainho!

 

[HughF]

You need to install latest version and calibrate your torque sensor.

Installed, now I will try to find some weights to use for calibration.

Still seeing a few lockups of the display when navigating the menus. And related to my lack of speed when using the 36v high cadence mode (see my post ages ago), I went for a ride back last week and regularly saw pwm values of 100 when doing 31km/h.... On the old code fork for the vlcd6 display I was able to hit 42km/h before I lost assistance.

 

[HughF]

I'm a bit confused by the sensor calibration procedure. How many different weights do I need to find/use? And do I need to do each side, left and right?

 

[skestans]

I'm a bit confused by the sensor calibration procedure. How many different weights do I need to find/use? And do I need to do each side, left and right?

Ideally 7 or 8. You want to go the whole range between 0 kg and however much your sensor tops up at. That will give you more points for the curve rather than guessing. And yes you need to do both sides.

 

[plpetrov]

[/url]
I'm attaching the Motor.c file for those who want to play with it. I did not use the latest version but with the stable release 0.54
I might merge it to the latest one tomorrow if people are interested.

Good timing as I was start working on this. I will look at it. Thanks.

In combination with the negative torque version for the coaster brake hubs or a separate branch?

In combination, because after detecting negative torque, the motor must stop as fast as possible.

I remember that some speed controllers used in RC cars and RC airplanes have dynamic braking implemented in the controller itself. I don’t know if we need that but in case we need to decrease the overrun this is the way to go. There are also some open source implementations. If you think their sources might be useful I can try to find some.

 

[HughF]

I will try and get some weights together and make the calibration. I have just been and done 20kms on the new firmware and there is very low assistance at light pedal pressure, no assistance below 10km/h or so, and when at the highest assistance level the bike tends to accelerate away from you up to about 30km/h before the assistance kind of cuts in and out.

 

[casainho]

I will try and get some weights together and make the calibration. I have just been and done 20kms on the new firmware and there is very low assistance at light pedal pressure, no assistance below 10km/h or so, and when at the highest assistance level the bike tends to accelerate away from you up to about 30km/h before the assistance kind of cuts in and out.

That is a problem with your system. You need to figure out if any sensor is failing or if you did any wrong configuration. You should use the display to see the various sensors values, to help you debug your issue.

 

[r0mko]

I've soldered a temperature sensor to the controller board and now want to check it. How do I see the temparature readout? FW is 0.7.0 (aka 0.6.10)

 

[HughF]

I will try and get some weights together and make the calibration. I have just been and done 20kms on the new firmware and there is very low assistance at light pedal pressure, no assistance below 10km/h or so, and when at the highest assistance level the bike tends to accelerate away from you up to about 30km/h before the assistance kind of cuts in and out.

That is a problem with your system. You need to figure out if any sensor is failing or if you did any wrong configuration. You should use the display to see the various sensors values, to help you debug your issue.

It worked very well with the other code fork, I suspect I need to calibrate the torque sensor.

I have a single speed, no gears so that probably doesn't help. I'm also 65kgs so perhaps that is also a problem?

 

[r0mko]

I've soldered a temperature sensor to the controller board and now want to check it. How do I see the temparature readout? FW is 0.7.0 (aka 0.6.10)

I've made a screen config so it displays the motor temp, but it always shows zero. The LM35 sensor is functioning fine: I've measured voltage between GND and Vs outputs and it shows 0.255V at room temperature. With some heat applied it raises to 0.5V. What I've done wrong?

 

[r0mko]

I've soldered a temperature sensor to the controller board and now want to check it. How do I see the temparature readout? FW is 0.7.0 (aka 0.6.10)

I've made a screen config so it displays the motor temp, but it always shows zero. The LM35 sensor is functioning fine: I've measured voltage between GND and Vs outputs and it shows 0.255V at room temperature. With some heat applied it raises to 0.5V. What I've done wrong?

My fault. I set the temperature sensor mode to throttle. Must have set to temperature. Now it works fine

 

[casainho]

Flashing the motor back to 0.54 and then 0.56 did the trick! 0.54 gave me the right 'firmware error' message. Obrigado casainho!

So your issue is solved, everything working as you expect?

 

[sparked]

Flashing the motor back to 0.54 and then 0.56 did the trick! 0.54 gave me the right 'firmware error' message. Obrigado casainho!

So your issue is solved, everything working as you expect?

Yes, because you asked I took it for a very short spin and it rode wonderfully. I did notice the boot-up time (from the initial button press to the information screen) takes a bit longer than before: I timed it at 7.24 seconds.

 

[casainho]

Here is a testing release for TSDZ2 motor firmware that solves the overrun and improve the TSDZ2 coast braking version:

https://github.com/OpenSource-EBike-firmware/TSDZ2-Smart-EBike/tree/coast_overrun/releases/0.56.2

Please give me feedback so I can release it.

 

[r0mko]

I would deeply appreciate a special TSDZ2 firmware version torque-only PAS mode. Using the pedal power for motor current calculation with a high-cadence mod (36V motor with 48V battery) leads to very uneven assistance: you got no torque at low cadence and startup, but at higher RPM the motor gets mad and tries to propel me to the outer space with almost no effort on pedals.
I can also try to build it by myself, but I need some hints on the code from casainho.

 

[casainho]

I would deeply appreciate a special TSDZ2 firmware version torque-only PAS mode. Using the pedal power for motor current calculation with a high-cadence mod (36V motor with 48V battery) leads to very uneven assistance: you got no torque at low cadence and startup, but at higher RPM the motor gets mad and tries to propel me to the outer space with almost no effort on pedals.
I can also try to build it by myself, but I need some hints on the code from casainho.

Did you calibrate your torque sensor?

What is your average human power at low cadence and at high cadence?

 

[r0mko]

I would deeply appreciate a special TSDZ2 firmware version torque-only PAS mode. Using the pedal power for motor current calculation with a high-cadence mod (36V motor with 48V battery) leads to very uneven assistance: you got no torque at low cadence and startup, but at higher RPM the motor gets mad and tries to propel me to the outer space with almost no effort on pedals.
I can also try to build it by myself, but I need some hints on the code from casainho.

Did you calibrate your torque sensor?

No, mine turns out to be quite linear. Torque sensor is not a problem, it works fine with the original FW, the problem in such a setup is the pedal power. At lower RPM the pedal power is low, so the motor does not help me to gain speed, and at higher cadence it is already high to maintain speed, so I don't need so much assist anymore. Now the motor acts in the exactly opposite way. I want to change it by removing the pedal RPM from the current calculation formula and use only pedal torque. In my opinion this would make the assistance more responsive and uniform across the RPM range.

 

[HughF]

I would deeply appreciate a special TSDZ2 firmware version torque-only PAS mode. Using the pedal power for motor current calculation with a high-cadence mod (36V motor with 48V battery) leads to very uneven assistance: you got no torque at low cadence and startup, but at higher RPM the motor gets mad and tries to propel me to the outer space with almost no effort on pedals.
I can also try to build it by myself, but I need some hints on the code from casainho.

This is the exact same thing I experience - you are better at describing it than me. I had great performance with the code fork that supports the original displays, but the latest code for the SW102 display is not very nice.

 

[r0mko]

I would deeply appreciate a special TSDZ2 firmware version torque-only PAS mode. Using the pedal power for motor current calculation with a high-cadence mod (36V motor with 48V battery) leads to very uneven assistance: you got no torque at low cadence and startup, but at higher RPM the motor gets mad and tries to propel me to the outer space with almost no effort on pedals.
I can also try to build it by myself, but I need some hints on the code from casainho.

This is the exact same thing I experience - you are better at describing it than me. I had great performance with the code fork that supports the original displays, but the latest code for the SW102 display is not very nice.

I'm not aware what was in the mentioned fork, but that what we are dealing with, uses pedal power to calculate the motor current. I've posted my thoughts earlier: https://endless-sphere.com/forums/viewtopic.php?f=30&t=93818&p=1528169#p1528169

 

[HughF]

I would deeply appreciate a special TSDZ2 firmware version torque-only PAS mode. Using the pedal power for motor current calculation with a high-cadence mod (36V motor with 48V battery) leads to very uneven assistance: you got no torque at low cadence and startup, but at higher RPM the motor gets mad and tries to propel me to the outer space with almost no effort on pedals.
I can also try to build it by myself, but I need some hints on the code from casainho.

This is the exact same thing I experience - you are better at describing it than me. I had great performance with the code fork that supports the original displays, but the latest code for the SW102 display is not very nice.

41721060002331

I'm not aware what was in the mentioned fork, but that what we are dealing with, uses pedal power to calculate the motor current. I've posted my thoughts earlier: https://endless-sphere.com/forums/viewtopic.php?f=30&t=93818&p=1528169#p1528169

Interesting, thanks for the link to your previous discussion. I think I will fit my old display back and re-flash with the Marcoq fork and take it for another run.

 

[casainho]

At lower RPM the pedal power is low, so the motor does not help me to gain speed, and at higher cadence it is already high to maintain speed

I may not be understanding correctly but this does not make sense as the average pedal power should be similar over the full cadence range (10 RPM to 90 RPM).

You should calibrate your torque sensor and see how your pedal power value changes over the full cadence range, that is why I asked you for your values otherwise there is not much point discussing without valid data.