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

Casainho, we are having real problems with water/mud conditions and the units failing. Is there a possibility that there could be a correlation between the speed sensors failing ( water and mud combined until the speed sensor fails ) and then confusing the processor sufficiently until on power up, the processor won't boot up correctly.Is there anyway of hard rebooting the processor once it gets self into this state ? At the moment once a controller gets into this confused state, it would seem there is no way we can get it back to operate on boot up.

The units before full failure show odd speed readings and the torque sensor readings seem to be at normal levels and operating with pedal input.

Is there anyway we can ignore the speed sensor and just operate on pedal input if we need to ?

Many thanks

Wayne.
 
jbalat said:
Hey guys been a while but just wanted to Share some findings regarding peddle cadence using the 36v motor with 10s and 14s battery in 36v and 36v experimental mode.

Hi jalabat, very nice your test I already use the high cadence mode on the engine powered at 36 and I am very satisfied, I saw on your video that when you go from 42V to 58.8V you only change the 14S battery type, but not change the motor settings (from 36V to 48V) why? when you use a native 36V in the settings do you have to leave it 36V always?
 
Waynemarlow said:
Casainho, we are having real problems with water/mud conditions and the units failing. Is there a possibility that there could be a correlation between the speed sensors failing ( water and mud combined until the speed sensor fails ) and then confusing the processor sufficiently until on power up, the processor won't boot up correctly.Is there anyway of hard rebooting the processor once it gets self into this state ? At the moment once a controller gets into this confused state, it would seem there is no way we can get it back to operate on boot up.
No, and as you explained on the other thread, your controller / TSDZ2 is full of water....
 
casainho said:
Waynemarlow said:
Casainho, we are having real problems with water/mud conditions and the units failing. Is there a possibility that there could be a correlation between the speed sensors failing ( water and mud combined until the speed sensor fails ) and then confusing the processor sufficiently until on power up, the processor won't boot up correctly.Is there anyway of hard rebooting the processor once it gets self into this state ? At the moment once a controller gets into this confused state, it would seem there is no way we can get it back to operate on boot up.
No, and as you explained on the other thread, your controller / TSDZ2 is full of water....
Casainho, we have 3 units all gone US, this particular unit was dry inside the housing.
 
maximusdm said:
Hello,

I need some ideas from people who installed/worked on the torque sensor. SO far I got the service manual from casainho and I have some clues.
I will try to create a tutorial after I have a way of putting it back.

How did you removed the torque sensor from the frame? I mean the actual torque sensor that is pressed inside the frame. What are the options other that bashing it with a big hammer?

Also how did you press it in while not loosing the small springs?

Thanks!

A couple of things that helped for me:

I made a gear puller from the spider by buying some extra long bolts and building up washers on the axle.

Make sure you remove all necessary circlips (there is one hiding behind a rubber bung on the non-drive side). Then I just gently tapped away with a wooden mallet to get it out!

For getting it back in (more important, because the springs need to stay in place), I managed to lash another tool together that pulled it into place using the crank bolts.

For me the two key things were making sure all necessary circlips were out and creating the gear pulling tool for the spider. Actually getting it out was ok because it just needed a gentle tapping.

Hope that helps!
 
mctubster said:
xlxc34 said:
Hi,

I have a very tempermental TSDZ2 which sometimes seems to turn on and run no problem, and other times I get no response unless I REALLY crank down on the pedals, in which case I might get a few 10s of Watts. I thought this might be a dodgy torque sensor, so I bought a new one and it seems to work just fine (checking the raw values on the 9 -> 2 menu it responds really well to changes in torque and has a baseline value of 35). However, I still have this problem where one in maybe every 2 or 3 times I use the bike I get no response, and turning it on and off again doesn't help. Any help with this will be VERY, VERY much appreciated as I am growing tired of starting off my 13 mile ride home with a bike that's 10kg heavier and has a load more resistance on a random basis!

Thanks!

I suspect you are sometimes resting your feet on the pedals while turning on the motor. This stops the torque sensor from calibrating correctly and will give less assistance.

This was a very good call! I just tested it and I can replicate the behaviour I get by turning it on while already pedalling... Totally explains the random nature of the problem! Thanks very much!

Does anyone have a part number for the drive side circular sealing ring? I seem to have lost mine whilst replacing a torque sensor for no reason! Although the new one is actually better so I don't feel too dumb... :D
 
Waynemarlow said:
casainho said:
Waynemarlow said:
Casainho, we are having real problems with water/mud conditions and the units failing. Is there a possibility that there could be a correlation between the speed sensors failing ( water and mud combined until the speed sensor fails ) and then confusing the processor sufficiently until on power up, the processor won't boot up correctly.Is there anyway of hard rebooting the processor once it gets self into this state ? At the moment once a controller gets into this confused state, it would seem there is no way we can get it back to operate on boot up.
No, and as you explained on the other thread, your controller / TSDZ2 is full of water....
Casainho, we have 3 units all gone US, this particular unit was dry inside the housing.
Well, I just can think of brake sensors, like if brake sensor signal is active for some reason and then the system will not work, not boot.
 
SW102 and 850C firmware, mobile app

I am very happy with latest 850C firmware and TSDZ2 firmware but still there are always things to improve. I wish next I could some stats like trip time and trip distance in a way they could be reset in a quick way from main menu. But the most needed feature is to configure max motor power in a quick way from main menu.

On SW102, the geeksville developer is being away recently and should not be back until at least the spring. No one is developing this firmware and it needs work, that is why there was no recent releases for it.

My wife wanted to learn developing a mobile app using recent technologies and so we did agree to develop our mobile app for TSDZ2 and SW102. I plan to help getting SW102 firmware working for this, so, I will need to put SW102 firmware working - I think I will work on it somewhere on next weeks.

The idea for the mobile app is to follow the market of this apps. We already looked at the apps suggested by you guys, when I asked about this before.

I think the most basic app will have a main screen like 850C has and will also have the configurations. Another level would be to have map navigation but I am sure this is very complex and I can´t even imagine this to be done soon.

There is one big issue with this app, that is both of us do not have the need to use it. I mean, there is no motivation to do this app other than learning more skills for improving professionally. I think the best motivation is when we use it daily and so we take value out of it daily!! That is why everyone helps and share about ebikes, because we use them daily!!

Ebike-Open-Source-App.jpg
 
jeff.page.rides said:
buba said:
jeff.page.rides said:
jeff.page.rides said:
Hey Buba and Casainho,
I'm very grateful for all the work you guys have done in version 20 it has been working well for me. I am excited about version 21 so that I can set the settings on the torque sensor. Because in version 20 I am still pulling backward on my cranks with a coaster brake version to trick torque sensor into thinking I'm putting more torque on the cranks than I am. I have been doing this ever since I installed my power assist with the stock firmware. It would be great to just put in the setting I like and not have to pull back on my cranks anymore to trick the torque sensor. This will also help those that don't have very strong legs or arms like me and use the standard motor. Because you can't pull back with the standard motor, to trick the torque sensor, With the stock firmware I had to pull back to trick the torque sensor or it was completely unusable. With version 20 pulling back makes a big difference, but the power assist will work without pulling back just not enough for me to enjoy riding. Thanks again for all the time that has been donated to this project, Jeff

I forgot to include that I'm pulling back on the cranks and holding them as I turn on the system. That's how I'm tricking the torque sensor into thinking I'm applying more torque to the cranks then I really am.

Hello Jeff,
Yes, I do remember! Did think about that one for a while. It is difficult to implement on the KT-LCD3 due to memory constraints. I have some ideas that would help but those will only be possible to implement on the new displays. Unless we remove functions on the KT-LCD3 to make some room.

Buda, thanks for the reply. If It's too much trouble to do on the LCD3 I can buy a new display, Handcycles with coaster brake motors don't use torque assist, cadence assist, walk assist, Cruise, City mode or a throttle and Street mode. If it's not too much trouble to remove all those and there could be room to put in the torque sensor settings that would be great, but if not will just upgrade all the current LCD3 on the units we've sold with the 850C
Thanks, Jeff

I'm just curious about the new version 850C_v0.6.2 and v0.51.0.
Does it have all the usability and safety features that were put into version v20 that was a huge improvement over v19 for me, disabled, and other riders?
Also, does this newest version allow you to preset the torque value like when I pull back on the cranks as I turn on the system?
Thanks, Jeff
 
jeff.page.rides said:
I'm just curious about the new version 850C_v0.6.2 and v0.51.0.
Does it have all the usability and safety features that were put into version v20 that was a huge improvement over v19 for me, disabled, and other riders?
Also, does this newest version allow you to preset the torque value like when I pull back on the cranks as I turn on the system?
Thanks, Jeff

Yes, I am a little confused as well. We have been waiting a long time for 850C on v20. v20 is the main development branch with lots of amazing features and capabilities. So now we have v50 based on v19? Did I read some time back that this is just temporary to add the torque calibration? When will we get back to v20 that we spent the summer testing and getting ready for release? Was v20 ever released? Will 850C ever work on it or will that only be v21? Sorry, but I am confused. Perhaps I missed some updates on this topic.
 
zappan said:
Hi jalabat, very nice your test I already use the high cadence mode on the engine powered at 36 and I am very satisfied, I saw on your video that when you go from 42V to 58.8V you only change the 14S battery type, but not change the motor settings (from 36V to 48V) why? when you use a native 36V in the settings do you have to leave it 36V always?

Yes the 36v motor is different from the 48v motor regardless of what voltage you run through it.
 
famichiki said:
Very interesting, thanks. What were your settings for maximum battery current and motor power limit? I'd be interested to know if the torque is reduced in high cadence mode and if so, is it across the whole range or only when you go above what would be considered normal maximum cadence.
I use 18a max and no limit on power. Saying that I did notice when you are pulling close to 18a you get quite a lot of battery sag, almost 4v in my case which could explain why I'm not getting max power.
I haven't noticed any reduction in torque using experimental mode although there is a natural drop off as cadence gets high. In normal mode the drop off is more severe since ithere s a software limit. Whether we can optimise torque vs speed using an advance curve is something i would really like to do but i don't have the equipment unfortunately.
 
zappan said:
jbalat said:
....... using the 36v motor with 10s and 14s battery ,,,,,,
........ .... in the settings do you have to leave it 36V always?
jbalat said:
....... 36v motor is different from the 48v motor regardless of what voltage you run through it.
@Zappan, because it is a 36V motor you don't need to change motortype to 48V.
Normally in the settings you change the battery type to 14S, so you can see the discharging of it.
But if you change the battery only to 52V than you have a very full "36V battery" and no discharge indication on display anymore.
 
Rydon said:
Sorry, but I am confused. Perhaps I missed some updates on this topic.
Same here. Some idea of a roadmap to V 0.20 with KT and both bafang displays would be very appreciated.
Also confusing to hear KT LCD3 can't be developed further due to lack of memory and to see the Bafang displays stuck on 0.19 and LCD 3 working on a beta 0.20.
 
jeff.page.rides said:
I'm just curious about the new version 850C_v0.6.2 and v0.51.0.
Does it have all the usability and safety features that were put into version v20 that was a huge improvement over v19 for me, disabled, and other riders?
No and I never tried v20 so it is even hard for me to say, I just saw and tried to understand the changes Buba did to v19.

Currently I am the only developer of this firmware.

For what I could understand, Buba did a great effort to make the firmware to work on the coast brake version of TSDZ2 you use. I really respect the work and I think has great value for users like you, I think it is really important and I wish to have that on current firmware. Unlike Buba, I don´t have resources nor motivation to go through the development, testing and maintenance of this feature -- I don´t use this feature for myself so why I would buy a specific motor, make a specific testing bench and keep it in my very little space home so I can develop, test and maintain this feature??

My motivation is to ride a lot, do new adventures, in this new Spring and Summer, with my family. My motivation is not to solve others issues but I am really happy to share with others what I developed for me.

My time is very small, for instance, I had high expectations for SW102 display but once the main developer stopped, the firmware is not updated anymore. I simple can´t do more development and I really want to enjoy riding with my family this new Spring and Summer of 2020 -- I will never trade that for solving the others issues -- on next 8 months I will have more days like this one instead of passing hours on the workshop developing and maintaining the firmware:

2020-01-12-09-36-25-1.jpg



Rydon said:
Yes, I am a little confused as well. We have been waiting a long time for 850C on v20. v20 is the main development branch with lots of amazing features and capabilities. So now we have v50 based on v19? Did I read some time back that this is just temporary to add the torque calibration? When will we get back to v20 that we spent the summer testing and getting ready for release? Was v20 ever released? Will 850C ever work on it or will that only be v21? Sorry, but I am confused. Perhaps I missed some updates on this topic.
I hope you understand what is my motivations by reading the previous message -- it is mainly to solve my problems (not the others problems) but I am really happy to share with others what I developed for me.

While V20 was being developed by Buba I was developing and testing the torque sensor full calibration. In the end, I did understand that V20 emtb mode masks the torque sensor lack of linearity so both features could not be on the same firmware or at least I think all this changes need to be done in a slow, small step, way so we can well understand them. I think no one understand what the emtb values do really mean as also what previous torque sensor values did mean but now, it is very clear, it is kgs -- it is very easy for user to understand and validate that the torque sensor is working well, in full range. We now know (and we can validate and show to user, and improve) that there are TSDZ2 units on the market with good and less than good torque sensors, that can give bad user experience on ebikes -- I mean riding an ebike without torque sensor.

Also V20 had the new power modes that I saw confusion on users when that asked here for help to understand them - again, I think we need to go in small steps.

So, I diverged to a new firmware version starting from V19 and I was looking at V20 sources and changes Buba did from V19. I wish over time to bring the improvements from V20 that I value most. For instance, the lights being able to flash and the flash time being configurable -- I do not have lights controlled by TSDZ2 on my ebikes so I have no motivation to bring that feature, test it and maintain it. I really prefer to focus on a good motor control, like I did on V51 of motor firmware, by improving the motor current/power control resolution by 4x.

Uturn said:
Also confusing to hear KT LCD3 can't be developed further due to lack of memory and to see the Bafang displays stuck on 0.19 and LCD 3 working on a beta 0.20.
Bafang display has instead advanced to currently V0.51 (read the wiki to be up to date), with torque sensor full calibration/linearization and 4x more sensitivity and, motor control with 4x more resolution.
 
Will SW102 development continue by other developers? (atleast to add the same functions as the 850C display)

Not sure if I should order one now considering the state of development, but want a minimalist display.
 
btslo said:
Will SW102 development continue by other developers? (atleast to add the same functions as the 850C display)

Not sure if I should order one now considering the state of development, but want a minimalist display.
There are no developers currently. I would like myself to make it at pair with 850C but I am not sure I will be able to do it.
 
Waynemarlow said:
Casainho, we are having real problems with water/mud conditions and the units failing. snip
Many thanks

Wayne.
Wayne
2 weeks ago a 48v 750watt 6 pin motor on a 29er just stopped. No noise, nothing. Stopped for a coffee, wouldn’t power up.

I narrowed the fault to the controller, so replaced it. The motor has done 2,000 miles in just over a year and it’s been very wet here. The old controller showed signs of a little dribble in one corner but this was an old leak. I found the 6 pin higo from the controller to be strangely easy to remove, then checked the continuity and all 6 pins were buggered. Water had partially shorted the lot.
 
Thanks re the pins, I'll check as I'm starting to think that it may be something like this.

I've been doing a bit of reading on how to " waterproof" these connectors better and it really just maybe as simple as making sure we have put a few dabs of dielectric or silicone grease on the pins before we first put the connector together, if there are no voids then no water can penetrate.

Equally I think on the later engines that the weak point is really just the cover over the entry of the wires, which doesn't sit particularly flat on the exit point of the wires + has another exit point and I consider the dab of silicone at the motor casing looks like it could be prone to failure . I'm going to full that void with a good waterproof grease, if the water can't get in there in the first place, then it isn't going to get through where the wires enter the case.

One of our lot did a Dawn French, Vicar of Dibley puddle moment https://www.youtube.com/watch?v=gWKUeV5B4mI which ended up with his bike right up to the handlebars in a small stream, I was pleasantly surprised at how little water at such pressure, entered the motor. The seals must be pretty good to handle that sort of dunking.
 
I would like to ask Casahino a question, I think it is better to make it visible to everyone.
When the original display version of Marcoq was created, I opened another post dedicated to it. Gradually users have dedicated themselves to studying it and are improving it. It seems that soon we will have version 20 for lcd03 also on the original display. Can I do the same for the lcd03 version? If there is someone who intends to develop it, he can post his results there. By now this tread is dedicated to the 850 display software.
 
Waynemarlow said:
Thanks re the pins, I'll check as I'm starting to think that it may be something like this.

I've been doing a bit of reading on how to " waterproof" these connectors better and it really just maybe as simple as making sure we have put a few dabs of dielectric or silicone grease on the pins before we first put the connector together, if there are no voids then no water can penetrate.

Equally I think on the later engines that the weak point is really just the cover over the entry of the wires, which doesn't sit particularly flat on the exit point of the wires + has another exit point and I consider the dab of silicone at the motor casing looks like it could be prone to failure . I'm going to full that void with a good waterproof grease, if the water can't get in there in the first place, then it isn't going to get through where the wires enter the case.

One of our lot did a Dawn French, Vicar of Dibley puddle moment https://www.youtube.com/watch?v=gWKUeV5B4mI which ended up with his bike right up to the handlebars in a small stream, I was pleasantly surprised at how little water at such pressure, entered the motor. The seals must be pretty good to handle that sort of dunking.
I think that the engine is not waterproof in any case.
Water can penetrate from the bottom bracket, especially from the torque sensor side. The plastic cover is certainly not waterproof and neither is the bearing
 
andrea_104kg said:
I would like to ask Casahino a question, I think it is better to make it visible to everyone.
When the original display version of Marcoq was created, I opened another post dedicated to it. Gradually users have dedicated themselves to studying it and are improving it. It seems that soon we will have version 20 for lcd03 also on the original display. Can I do the same for the lcd03 version? If there is someone who intends to develop it, he can post his results there. By now this tread is dedicated to the 850 display software.
I created this thread for my project, that has a specific philosophy. Advanced features and a good display to support that features as also give information to user so he can understand what is happening and to educate it about ebike motors, battery, etc. I am not working only on firmware, I also put a good effort on the wiki, with the TSDZ2 FAQ with known issues, possible causes and repair.

KT-LCD3 can't hold the new recent added features and new future ones, so the firmware for it should be freezed, this way we can move forward with the project with the much more capable displays 850C and SW102.

You are free to create another thread but the wiki will focus only on this project.
 
casainho said:
jeff.page.rides said:
I'm just curious about the new version 850C_v0.6.2 and v0.51.0.
Does it have all the usability and safety features that were put into version v20 that was a huge improvement over v19 for me, disabled, and other riders?
No and I never tried v20 so it is even hard for me to say, I just saw and tried to understand the changes Buba did to v19.

Currently I am the only developer of this firmware.

So, I diverged to a new firmware version starting from V19 and I was looking at V20 sources and changes Buba did from V19. I wish over time to bring the improvements from V20 that I value most. For instance, the lights being able to flash and the flash time being configurable -- I do not have lights controlled by TSDZ2 on my ebikes so I have no motivation to bring that feature, test it and maintain it. I really prefer to focus on a good motor control, like I did on V51 of motor firmware, by improving the motor current/power control resolution by 4x.

Please include the link to the wiki and will V0.51 allow you to set the sensitivity of the torque sensor?
Thanks
 
jeff.page.rides said:
Please include the link to the wiki and will V0.51 allow you to set the sensitivity of the torque sensor?
Thanks
https://github.com/OpenSource-EBike-firmware/TSDZ2_wiki/wiki/How-to-install-the-Flexible-OpenSource-firmware

Once you calibrate the torque sensor, you use the assist level to scale up the motor power compared to pedal human power.

An example: I do comfortable 100 watts of pedal human power with my legs. If doing 100W and want motor to assist 500W, then I set assist level to factor 5.0.

You will need to test how much pedal human power you usually do and then decide for motor max power you want and then select an assist level factor.
 
Back
Top