TSDZ2 OSF for all displays, VLCD5-VLCD6-XH18, LCD3, 860C-850C-SW102.

[gfmoore]

Is it perhaps the new 48v controller?

 

[jbalat]

https://www.eco-ebike.com/blogs/eco-cycles-instructionals/tsdz2-motor-firmware-programming
I think you can get the stock firmware here..

But before you do…

Just go into the config menu and check your torque sensor is working, check cadence , check your speed sensor, see if there is anything that is out of whack,,

So normally it stop assisting if your speed sensor doesn’t work or your battery voltage is too high or the option bytes are wrong,

Sometimes if the display is not matching the same firmware as the motor….

 

[raylo32]

What about activating "assist with error" as a test? I haven't tried that myself but have read that it could make the motor operate even if it isn't getting all the right input signals.

https://www.eco-ebike.com/blogs/eco-cycles-instructionals/tsdz2-motor-firmware-programming
I think you can get the stock firmware here..

But before you do…

Just go into the config menu and check your torque sensor is working, check cadence , check your speed sensor, see if there is anything that is out of whack,,

So normally it stop assisting if your speed sensor doesn’t work or your battery voltage is too high or the option bytes are wrong,

Sometimes if the display is not matching the same firmware as the motor….

 

[mbrusa]

is c3new version available for download in hex file? on github I see only c3. the display is sw102.

The only version v20.1C.3-NEW released is the one for stock displays.
Version that I have discontinued, although I doubt that the problems reported in the other thread are software-related.
With this version I have done 300 km with two different bikes.
On the other hand, I couldn't do otherwise.
For stock displays I have been testing a new version v20.1C.4 for 10 days, I will continue until the end of the month before letting others try it.

As for 860C there is a v20.1C.3-NEW version tested by some users, but it is not yet released, the news is mainly on the display, for SW102 it will change little.
I think I will release it soon, it is now tested.
With this other bike, with 860C display I did:
500km with v20.1C-860C
300km with v20.1C.3-860C
300km with v20.1C.3-NEW-860C

 

[f1nch87]

is c3new version available for download in hex file? on github I see only c3. the display is sw102.

The only version v20.1C.3-NEW released is the one for stock displays.
Version that I have discontinued, although I doubt that the problems reported in the other thread are software-related.
With this version I have done 300 km with two different bikes.
On the other hand, I couldn't do otherwise.
For stock displays I have been testing a new version v20.1C.4 for 10 days, I will continue until the end of the month before letting others try it.

As for 860C there is a v20.1C.3-NEW version tested by some users, but it is not yet released, the news is mainly on the display, for SW102 it will change little.
I think I will release it soon, it is now tested.
With this other bike, with 860C display I did:
500km with v20.1C-860C
300km with v20.1C.3-860C
300km with v20.1C.3-NEW-860C

Thx for your work, looking forward to the new 102 Version.
So far im happy with the C2. Just a bit of power is missing with two kids in the cruiser behind me and legit 250w in germany. See how i can improve that.

 

[andrea_104kg]

Is it perhaps the new 48v controller?

yes it's 48v controller... why this question?

 

[andrea_104kg]

https://www.eco-ebike.com/blogs/eco-cycles-instructionals/tsdz2-motor-firmware-programming
I think you can get the stock firmware here..

But before you do…

Just go into the config menu and check your torque sensor is working, check cadence , check your speed sensor, see if there is anything that is out of whack,,

So normally it stop assisting if your speed sensor doesn’t work or your battery voltage is too high or the option bytes are wrong,

Sometimes if the display is not matching the same firmware as the motor….

for me it is enough to work with this engine, too big disappointment ... it broke after an hour, changed the control unit and torsometer and still does not go, no guarantee that another engine will go at this point ... Thank you mbrusa who followed me with patience, we tried the torque sensor, (it works even if a little limited, but in any case the control unit receives the data) we tested the power of the electric motor and it is ok, unfortunately the sensor that detects the cadence seems to be not working, but I don't understand how this is possible on two NEW and different controllers. I completely disassembled the engine with a commitment of hours and hours and at this point I don't know what to do anymore. I put the original software back, tried version 19, 20 1b, 20 1c1, flashed hundreds of times. I don't know what to do honestly, certainly not to buy another engine or other parts, since they are all brand new ...
Thanks again to Mbrusa for the patient advice ...

 

[gfmoore]

Is it perhaps the new 48v controller?

yes it's 48v controller... why this question?

Because somewhere Casainho etc are working on how to flash the new controller that has come out. They haven't succeeded yet and since it's summer he said it won't happen till Autumn.

When I find the link to the thread I'll post it or search through my posts where I mentioned it and linked. I'll have to do it tomorrow as its late now.

Gordon

 

[gfmoore]

https://endless-sphere.com/forums/viewtopic.php?f=30&t=111287

 

[andrea_104kg]

it's not a new controller, new controller it's impossible to read write with the old metod, and this it's possible.

 

[Elinx]

..... unfortunately the sensor that detects the cadence seems to be not working, but I don't understand how this is possible on two NEW and different controllers......

The gap that is between the cadence-sensor on the controller and the magnets on the torquesensor is imho unknown.
With disassamble & reassemble the controller and torque-sensor have you observed the alignment?
If the torque-sensor has axial play, I don't know what happens with the cadence signal.
See this post too, for what I mean.

 

[jbalat]

You can try using torque mode and set firmware to ignore the errors
See wiki
https://github.com/emmebrusa/TSDZ2-Smart-EBike-LCD3/wiki

 

[andrea_104kg]

..... unfortunately the sensor that detects the cadence seems to be not working, but I don't understand how this is possible on two NEW and different controllers......

The gap that is between the cadence-sensor on the controller and the magnets on the torquesensor is imho unknown.
With disassamble & reassemble the controller and torque-sensor have you observed the alignment?
If the torque-sensor has axial play, I don't know what happens with the cadence signal.
See [b][u]this post[/u][/b] too, for what I mean.

the control unit is fixed with 2 screws and the sensor comes out of a hole, how can you get the alignment wrong? there is no other way to fix the controller..

 

[andrea_104kg]

You can try using torque mode and set firmware to ignore the errors
See wiki
https://github.com/emmebrusa/TSDZ2-Smart-EBike-LCD3/wiki

the only way to make it work, with the help of mbrusa was "start without pedaling" and emtb mode at strartup. But it runs very poorly, the engine goes by itself with the weight of the feet on the pedals and does not stop, impossible to drive.

 

[andrea_104kg]

..... unfortunately the sensor that detects the cadence seems to be not working, but I don't understand how this is possible on two NEW and different controllers......

The gap that is between the cadence-sensor on the controller and the magnets on the torquesensor is imho unknown.
With disassamble & reassemble the controller and torque-sensor have you observed the alignment?
If the torque-sensor has axial play, I don't know what happens with the cadence signal.
See [b][u]this post[/u][/b] too, for what I mean.

This Is the gap https://youtube.com/shorts/wAMScSoZEuE?feature=share but it's Impossible to reduce.

 

[Elinx]

... But it runs very poorly, ....

In your case a software calibration with a poor sensitive sensor has result in almost an "on/off switch".
The sensitivty of torque can be hardware calibrated for a much better response, by moving the hall sensor a bit, after loosing the screws. See also the wiki
In that case you adjust the zero adc value to a lower value, which means a larger working range of the torque sensor.


The main problem is the PAS/Cadence sensor, which doesn't work with both controllers and both torque sensors. Why?
Imho your old controller and torque sensor aren't devective
You have to find the common reason, why the hallsensor on both controllers can't sense the magnets on the PAS/cadence ring.
The axial play of the torque sensor could have influence, but I don't know how much.

 

[andrea_104kg]

I put a copper washer to limit the movement of the torsiometer but no result

 

[gfmoore]

it's not a new controller, new controller it's impossible to read write with the old metod, and this it's possible.

Ahh okay, my bad.

 

[andrea_104kg]

I opened the engine again for the umpteenth time.
the oscillation of the torsometer is due to the fact that it slides on the roller bearing. I made a custom copper washer, now the torsometer turns perfect less than one mm from the pas sensor but nothing works anyway. therefore it is not a problem of axial play of the torque meter. I'm desperate I don't know what to do anymore ...

 

[Elinx]

... I'm desperate I don't know what to do anymore ...

Maybe a strange thought, but did you also replaced the magnet ring with the torquesensor?

 

[andrea_104kg]

The magnete ring it's in the torsiometer and it's all NEW.

 

[andrea_104kg]

the new pieces arrived Thursday, just assembled, with the original firmware it worked, after the first flash it immediately stopped working.

 

[andrea_104kg]

unfortunately I did not make the backup, I tried to put back the original firmware taken from the eco-ebike site, including option byte but it no longer worked. it may be that we have changed something in the control units in the reading of the pas sensor? and the current code no longer fits?
what other cause can prevent cadence reading? the speed sensor shows the speed correctly, can it be broken anyway? (even disconnecting it, however, nothing works) a short in the display cables can prevent cadence reading? (but they are not tightened by cable ties and the display works without errors)
I try to think of them all, even absurd ...

 

[Elinx]

The magnete ring it's in the torsiometer and it's all NEW.

I know, it was a strange thought, but I'm just trying to figure out why the same error can occur on two different sensor/ecu versions. In that case there must be a common reason.
That the engine didn't work after flashing seems logical if one sensor doesn't work.

One reason I can think of is that the PAS sensor emits a different signal which OSF doesn't recognize anymore, but the supplied stock FW did. Mainly because flashing downloaded stock FW didn't work either.
But in that case there should be more people with such a problem and a new controller.
I can't believe that the PAS/cadence sensor is broken on both controllers.

Another reason could be that the PAS/cadence sensor on both controllers is blowed up after flashing and starting the controller.
Don't ask me how, I only think about possible common reasons.
Question: Was your battery connected with flashing? That could be a reason

FYI Vlvuk has (maybe) replaced the hall sensors after frying it. I haven't found an update of the result.

 

[andrea_104kg]

No my battery It's not connected when flashing...
I think in the end I will have to give up and throw the motor in the garbage can ... and not even buy another one ... I don't see any solutions