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

[jeff.page.rides]

Any idea when the beta 20 for 850 C and SW102 could be ready?
Because most of us have the LCD3 displays there should be a way to delete unnecessary things that aren't used by the majority of users so that we can put the torque sensor updates on the LCD3 without losing anything important. If not everyone has to upgrade to a new display. It would be great if we had the opportunity to choose which display we want to use in the future rather be than being forced to use one of the newer displays with more memory.
Thanks, Jeff

Upgrade to a new display is cheap, the SW102 for instance costs only 32 euros.

You guys should not expect that developers will work more again for free so you guys can save 32 euros.

KT-LCD3 is very limited in memory compared to the new ones, the new firmware will need to take a big change and the KT-LCD3 is always have the memory limitation that is a big risk of something that will not the able to be implemented. It is not fair to expect that developers work for free to be always optimizing something that will always be short in memory. Developers have life, their time should be well selected, please do not expect they work for free, use their free time after work, for your specific needs.

The work is big, is getting the users feedback of what is used and not, decide, developing, testing, get users feedback, make new changes after the feedback, etc and do the documentation on the wiki.

I’m SORRY, just thought that leaving off some things that didn’t get used very often would make it easier.

 

[andrea_104kg]

I don't think anyone is going to ask developers to work for free on a project they don't believe in or believe is outdated. But I think it's fair to say if the lcd03 project is permanently closed and even if some developer can if they obviously want to continue.

 

[casainho]

(...) even if some developer can if they obviously want to continue.

I understand what you mean but I think the answer is it depends. The project need developers that are available today to implement with quality but that will be also available to do maintenance in future - no point to get new developers that will just do quick hacks and then reduce the current quality of the project.

I think is better to create a new project (or major firmware version) to prove the project quality and success.

Also, the new features of torque sensor full calibration, custom emtb curve, Bluetooth, should value more than the 32 euros of the new SW102 display.

 

[thineight]

(...) even if some developer can if they obviously want to continue.

I understand what you mean but I think the answer is it depends. The project need developers that are available today to implement with quality but that will be also available to do maintenance in future - no point to get new developers that will just do quick hacks and then reduce the current quality of the project.

I think is better to create a new project (or major firmware version) to prove the project quality and success.

Also, the new features of torque sensor full calibration, custom emtb curve, Bluetooth, should value more than the 32 euros of the new SW102 display.

Of course 32 euros are nothing for a user that spent hundreds for the hardware, some donations, plus time to study and mount it, I think the major hassle (I speak for myself) is to make the initial setup before getting the future updates via Bluetooth working. I mean open up the unit for the first flash.

We are all following the thread with attention and we will individually choose whether we stay with the lcd3 (which support will come to an end sooner to later) that has pretty much all the important features implemented, or to switch to one of the new displays.

Personally I would consider the cheap and compact SW102 but I am a bit afraid of the case opening, therefore I keep reading everything and wait the next steps.. the lcd3 is, so far, more than sufficient :wink:

 

[Rafe]

The compact SW102 is not really a suitable like for like swap with the excellent LCD3 display though. When I first purchased LCD3 for the open source project I was a little underwhelmed by it but soon grew to respect it. It is so easy to read in all light conditions, at wide angles and vibrating away on rough trails even for those who normally wear reading glasses. I also have the Bafang 850 display for the BBSHD the other new option for open source and to be honest even with all its bells and whistles plus glorious technicolour the readability of that off angle or in bright sunlight is not as good as monochrome LCD3

 

[vadda]

I also think that a basic requirement of a bicycle display should be ease of reading in all conditions.
The KT-LCD3 on this I believe is absolutely unbeatable.
I understand that for the memory problem it will not be possible to insert all the new functions, but it would be nice to have the basic features updated with new developments at the same time.
It would be really sad to lose the possibility of using this display, it is certainly not for the expense of € 32

The great gratitude always remains for those who have invested their free time in this fantastic project.

 

[casainho]

I think the major hassle (I speak for myself) is to make the initial setup before getting the future updates via Bluetooth working. I mean open up the unit for the first flash.

This is the current state of easy of flashing our firmware on the first time:

1. Most easy to flash initial firmmware: 850C using the original wires that connect to the motor controller. No need to open the display and solder wires. The display keeps the original water resistance.

2. Most easy to open and solder wires to flash the firmware: KT-LCD3, because the pads are big and easy to open.

3. Most difficult to open and solder wires to flash the firmware: SW102 because it has small pads to solder the wires. However, soon we may be able to install firmware using Bluetooth, no need to open the display and solder wires and then will be the number 1. And the display will keep the original water resistance.

 

[casainho]

I understand that for the memory problem it will not be possible to insert all the new functions, but it would be nice to have the basic features updated with new developments at the same time.

Due to the 800% less memory on the KT-LCD3 compared to SW102 and 850C, it is not possible to keep the current advanced features while adding new ones.

 

[jimmyfergus]

I've wanted to put a LM35 temperature sensor on my motor since I got it. I should be able to simply apply 5V (to left pin when looking at the flat side) and GND (right) and see something around 0.2V (20 * 10mV) on the middle pin at room temperature, right?

Because first I ordered 2, then a batch of 10, and I've yet to get a plausible looking reading out of them. A lot of my batch of 10 seem to show me about 8mV. My batch of 2 had 2 of their pins shorted, so they were complete garbage.

Yes, obviously I should probably have bought something more reputable, but it still seems weird that all 10 of my second batch are rubbish?

Does anyone recommend a good source?

 

[andrea_104kg]

I will try, but in English it is difficult, to explain myself on what I think about sensor calibration.
I have a levo, the main advantage of the levo is that the more you push the more the engine pushes. But it is not due to pressure. Yesterday with the levo, I faced a long road climb at 6-8% at the most, not a big climb. At full power, I managed to get to 25kmh, forcing relatively little. I tried to pedal with the tip of my toes and the bike was the same!
Today, on the other hand, I made a long climb to 18-20%, very hard, with the tsdz2 (48v 750w). Very strong pedaling (ratio 34/40) the engine developed 650-700w. But it wasn't the best, if I pedaled more slowly the engine went 850w and obviously the speed decreased. So even if you ride hard you can't get the most. It is not a sensor problem, however, not even a problem of cadence, simply the effort on the pedals added to the engine is sufficient and the engine does not go to maximum. If the pedal effort decreases, the engine balances giving more power.
In the turbo levo there is something in the software that prevents this drop, I don't think it's the torque sensor or its calibration. If the sensor is calibrated, in my opinion the only thing you get is that to get the maximum you have to try very hard. And the engine will give less on intermediate effort, not solving the problem. The tsdz2 sensor is very simple and works well. The effort can only be measured if there is resistance, if it is not enough it will not push the engine to the maximum. Even that of the turbo levo should behave like this but it does not and not because it is calibrated better simply the software is "rigged" so that this does not happen. The belief comes to me because even the human effort on the pedals is not maximum at maximum cadence, rather the more the cadence is raised the more the effort decreases, therefore the tsdz2 correctly decreases the power, the levo does not do it. So the sensor calibration has little to do. I don't know if the above is understandable in English with google translator :-( Personal opinion of course ...

 

[elfnino]

Has anyone tried with success to flash 850C display with home made boot-loader (USBtoUART) ?
In my case flashing starts fine data is transmitted but it gets interrupted and I could not get beyond 40%. Really strange that each time it fails with different percentage. I was trying different power sources step-up buck/52V battery/38V power adapter but nothing worked. Any ideas ?

 

[casainho]

Has anyone tried with success to flash 850C display with home made boot-loader (USBtoUART) ?
In my case flashing starts fine data is transmitted but it gets interrupted and I could not get beyond 40%. Really strange that each time it fails with different percentage. I was trying different power sources step-up buck/52V battery/38V power adapter but nothing worked. Any ideas ?

I would say you are the first one.

Note that converter must be 3.3V.

In the case of flashing firmware on the TSDZ2, users report errors and the solution is to use very short cables... Maybe you can try another converter.

Also try remove the charger from your laptop while flashing.

 

[Waynemarlow]

Guys, can anyone put a link up for the latest V20 Beta 1, thanks

 

[maximusdm]

Very strong pedaling (ratio 34/40) the engine developed 650-700w. But it wasn't the best, if I pedaled more slowly the engine went 850w and obviously the speed decreased. So even if you ride hard you can't get the most. It is not a sensor problem, however, not even a problem of cadence, simply the effort on the pedals added to the engine is sufficient and the engine does not go to maximum. If the pedal effort decreases, the engine balances giving more power.

I noticed same issue since version 0.15. With maximum assistance if you pushed very strong the power drops to 300-400w. With slow push and constant I got 850w.
The only thing I did to overcome this was to use higher cadence with High cadence mode enabled. I managed to do difficult slopes with 34-34 ratio.

Maybe Buba or casainho can help with some feedback here.

regards,
Max.

 

[elem]

Yes, in standard mode motor max rpm is 90 but at about 75/80 rpm power begin to decrease .
with a 36v motor it's easy to get power over 100rpm .

As power mode is torque*cadence based, when cadence rich max rpm, resulted power begin to fall, listen, when you'r pedaling at hight rpm you can't do max torque, it's physical, soo increasing your rpm mean decrassing torque .

 

[casainho]

I noticed same issue since version 0.15. With maximum assistance if you pushed very strong the power drops to 300-400w. With slow push and constant I got 850w.
The only thing I did to overcome this was to use higher cadence with High cadence mode enabled. I managed to do difficult slopes with 34-34 ratio.

Maybe Buba or casainho can help with some feedback here.

On 850C display we plan to add visual guide for when the motor hits his own limits and then power may drop: max battery current; max motor RPM; max PWM duty-cycle; max motor temperature, etc.
On the graph, there will be yellow and red zones depending when the variables are near the limit. You will be able to look at last 1h/30m/15m of each variable and see if the max value was hit.

I hope the same will be implemented also on SW102 although with different visual warnings.

 

[elfnino]

Note that converter must be 3.3V.

With bootloader box the 3.3V pin is not used. The UART chip is powered via the main source of power 27V or more.

In the case of flashing firmware on the TSDZ2, users report errors and the solution is to use very short cables... Maybe you can try another converter.

Also try remove the charger from your laptop while flashing.

Thanks for the hint when I removed the laptop charger and used battery to power the LCD it flashed till 100%.
I think the data transfer gets corrupted easily by some EM fields generated by power supply. What is strange I have never had such issue before with the same USBtoUART adapter when I was flashing smart sockets like Sonoff..

Unfortunately even when flashing was fully finished the display doesn't work it shows just white screen
I tried re-flash it again but same result only white screen.

 

[casainho]

Note that converter must be 3.3V.

With bootloader box the 3.3V pin is not used. The UART chip is powered via the main source of power 27V or more.

In the case of flashing firmware on the TSDZ2, users report errors and the solution is to use very short cables... Maybe you can try another converter.

Also try remove the charger from your laptop while flashing.

Thanks for the hint when I removed the laptop charger and used battery to power the LCD it flashed till 100%.
I think the data transfer gets corrupted easily by some EM fields generated by power supply. What is strange I have never had such issue before with the same USBtoUART adapter when I was flashing smart sockets like Sonoff..

Unfortunately even when flashing was fully finished the display doesn't work it shows just white screen
I tried re-flash it again but same result only white screen.

Good to know that you were able to flash.

The white screen, please verify if you ride the bicycle, the motor pulls you as usual. Try to increase the assistance with up/down buttons while trying.

Also you should try to flash original firmware, which I don't have but I think you may find on google and download. The original firmware can be flashed the same way and you should be able to recover the display.

If it works as expected with original firmware, then please share with me the original firmware and I also need to one of that displays and test to see if I can change/ try different init commands sequence as other users did report to see the image inverted and flashing the firmware without the bootoader and where able to stop the execution of firmware at startup.

 

[elfnino]

Note that converter must be 3.3V.

With bootloader box the 3.3V pin is not used. The UART chip is powered via the main source of power 27V or more.

In the case of flashing firmware on the TSDZ2, users report errors and the solution is to use very short cables... Maybe you can try another converter.

Also try remove the charger from your laptop while flashing.

Thanks for the hint when I removed the laptop charger and used battery to power the LCD it flashed till 100%.
I think the data transfer gets corrupted easily by some EM fields generated by power supply. What is strange I have never had such issue before with the same USBtoUART adapter when I was flashing smart sockets like Sonoff..

Unfortunately even when flashing was fully finished the display doesn't work it shows just white screen
I tried re-flash it again but same result only white screen.

Good to know that you were able to flash.

The white screen, please verify if you ride the bicycle, the motor pulls you as usual. Try to increase the assistance with up/down buttons while trying.

Also you should try to flash original firmware, which I don't have but I think you may find on google and download. The original firmware can be flashed the same way and you should be able to recover the display.

If it works as expected with original firmware, then please share with me the original firmware and I also need to one of that displays and test to see if I can change/ try different init commands sequence as other users did report to see the image inverted and flashing the firmware without the bootoader and where able to stop the execution of firmware at startup.

absolutely I could see some packets are being sent from 850C with the white screen on because the Rx led flashes at high rate on the USBtoUART adapter which indicates the incoming UART traffic.
To my biggest surprise the motor control as well works just fine and I am able to change the assist level !

I was trying to search for the original Bafang 850C firmware but so far no luck .. May be someone else could give me a hint where to get it..

 

[casainho]

absolutely I could see some packets are being sent from 850C with the white screen on because the Rx led flashes at high rate on the USBtoUART adapter which indicates the incoming UART traffic.
To my biggest surprise the motor control as well works just fine and I am able to change the assist level !

I was trying to search for the original Bafang 850C firmware but so far no luck .. May be someone else could give me a hint where to get it..

Please the PM I sent. Please tell me what happens with that firmware, will be display be also all white and motor works??

 

[mctubster]

Yes, obviously I should probably have bought something more reputable, but it still seems weird that all 10 of my second batch are rubbish?

Does anyone recommend a good source?

Sparkfun(US)
Element14(UK,AU)
CoreElectronics(AU)

 

[mctubster]

I also think that a basic requirement of a bicycle display should be ease of reading in all conditions.
The KT-LCD3 on this I believe is absolutely unbeatable.
I understand that for the memory problem it will not be possible to insert all the new functions, but it would be nice to have the basic features updated with new developments at the same time.
It would be really sad to lose the possibility of using this display, it is certainly not for the expense of € 32

The great gratitude always remains for those who have invested their free time in this fantastic project.

I 100% agree with Rafe and Vadda. Display visibility is king - even though I have purchased a 850c aware of the display limitations because I like data

The 850c is poor in bright sunlight, not even as good as a cheap cellphone. I used one for a year on my Bafang BBS01b. I don't know much about the SW102 but assuming it is cheap OLED I suspect it will be even worse in bright sunlight. I will likely buy a SW102 to try out also but again I have to agree, if you are on a mountain bike on a trail nothing beats a LCD segment display, both size and contrast are far superior on the LCD3

I'm secretly hoping buba will be able to optimise a few things, and maybe move some static variables into the motor controller. I wonder if two manual calibration points say 15kg and 50kg (looks a bit bi-linear) would be close enough for the torque sensor tbh. Prob need a larger sample set to say for sure.

Many thanks either way to all of the contributors - the TSDZ2 and its many options are far superior to Bafang and even Bosch systems (what I have used previously). I have donated to casainho in the past, I will donate to buba now

Cheers

 

[microchip7000]

Hi, I upgraded to 0.20 alpha 10. I did 1.000 km with 0.17 and 1.000 km with 0.19. My bike's battery is 1kw and the engine is 36v 350w. With the old versions the watt meter was very precise, 1 kw of consumption and the battery was completely discharged at 29v, while with the 0.20 it marks about 20% more consumption, reached 0% I still had a lot of battery, I increased from the menu the battery capacity up to 1.200w to arrive correctly.
Even instant consumption obviously marks 20% more. There is some calibration to do to adjust the shot.
for the rest everything is fine, excellent starting from a standstill and much more linear than the 0.19.
Hello and congratulations for having reached these levels :thumb:

 

[buba]

First of all, I want to apologize to the community for being absent. Have become time constrained again but I still try to follow the thread as much as possible. A lot of things have been discussed and that is great!

I have personally nothing to add to the discussion for now. But I do want to clarify that I look forward to the stable release of the 0.20.0. That is a priority for me because there are a LOT of changes and I want to get them out there. The community has provided several great suggestions and ideas that have made the firmware even better! What an awesome collaboration!

It will be interesting to see how the development moves forward and what can be done to further improve the experience. I am very excited to see the changes the new developers will introduce and also the ideas Casainho will implement! It will be some great work and I truly look forward to see and try it all out!

For the time being, I hope everyone finds the 0.20.0 Alpha version a great upgrade from the 0.19.0! Because it is going to get even better with the 0.20.0 Beta :wink: More about this in my reply to Microchip7000 below:

Hi, I upgraded to 0.20 alpha 10. I did 1.000 km with 0.17 and 1.000 km with 0.19. My bike's battery is 1kw and the engine is 36v 350w. With the old versions the watt meter was very precise, 1 kw of consumption and the battery was completely discharged at 29v, while with the 0.20 it marks about 20% more consumption, reached 0% I still had a lot of battery, I increased from the menu the battery capacity up to 1.200w to arrive correctly.
Even instant consumption obviously marks 20% more. There is some calibration to do to adjust the shot.
for the rest everything is fine, excellent starting from a standstill and much more linear than the 0.19.
Hello and congratulations for having reached these levels :thumb:

Thank you for being so detailed in your feedback! You have done an amazing job explaining the problem and giving me a very clear picture of what I need to do! I appreciate that a lot so the least I can do is to solve the problem :wink:

As soon as I got home today I did some work on the bike that enabled me to connect some instruments and do some tests. Can confirm that the ADC conversion values are not perfect. Error range: 1 - 20 %. I have already sorted this out and everything should be more accurate.

Oh... and one more thing! How about more torque and power? :wink: Yes, Solving this bug has enabled more power and torque! The changes are already submitted in the pull request and will be available in the 0.20.0 Beta 1 as soon as Casainho has time to take a look and release the beta version. I now recommend everyone to update to 0.20.0 Beta 1 as soon as it is available.

A big thanks to the community!

 

[buba]

Guys, can anyone put a link up for the latest V20 Beta 1, thanks

Not yet released but hopefully soon!