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

[Yanis]

Hi Yanis.
Yes, the new version actually works very well, consumes less and operates very smoothly and regularly up to high cadence.
The main merit is the mspider65 v15 version, but it's not just a copy/paste, I changed something.
When I have more time, I'll explain in detail.

With calibration, the torque sensor range, whatever value it has, is remapped to 160, 'Torque range adj' allows you to change this value.
What is it for?
I'll give you an example, you wrote that you had to reduce all levels of assistance.
If you decreased 'Torque range adj', it is equivalent to decreasing all levels in all modes, changing only one parameter.

As for "Smooth start" you already understood., it is the opposite of "Startup boost".
"Startup boost" applies only to Power mode, while "Smooth start" applies to Hybrid and Torque, and also to Cadence mode.
In Cadence mode, I consider the assistance parameter equivalent to "Pedal torque delta", therefore the same functions can be applied.

Thank you for your explanations.

The smooth start is absolutely great for protecting the blue gear and drivetrain when starting from a dead stop, but I miss the snappy feeling when restarting pedalling after cruising, and I believe there is little chance to damage the blue gear when the bike is in motion already, even at slow speed. Maybe would it be nice to have the option, just like the startup boost, to have the smooth start either from cadence zero or zero speed ?

 

[Coolby]

Dear All,

Currently I have OSF on my tsdz2 with the stock 850c display (I have got it in a kit). Now I ordered a 860c (see exact version on the picture). I have built a bootloader cable, but I have stocked with the update. The display gets power and I can turn it on, I see thr CP210x usb to uart bridge on com port 7, I can open the port in APT Burn tool, but the update firmware shows only waiting despite I push the button of the controller. I see numbers running below at TX, but nothing on RX (just for info, I do not know if it is relevant or not). What can be the problem?

Other questions:
1. Which firmware should be used? I have lost that underground v20.1C.5 beta there is many bin files...
2. After I succeed to change firmware on the display I should change the OSF on the motor too, so which OSF I use with the Stock display shall be changed? For which version?

Thanks for the support.

 

[dgc]

Glad you are asking these questions as I am more or less in the same position as you, Where did you get the bootloader cable info from. My main MTB is running the OSF using the VLCD5 display but on my second MTB would like to try the 860C TSDZ2 combination.

 

[Coolby]

Glad you are asking these questions as I am more or less in the same position as you, Where did you get the bootloader cable info from. My main MTB is running the OSF using the VLCD5 display but on my second MTB would like to try the 860C TSDZ2 combination.

I've built based on this description

Flash-the-firmware-on-860C-850C-using-bootloader | OpenSource-EBike

\\TOC\\

opensource-ebike.org

 

[dgc]

Coolby. Many thanks will have a look at that.

 

[Weissnix]

Hi all,

Yesterday i managed to communicate with my new controller (i'm using ST link v2 clone). I successfully uploaded emmebrusa firmware 20.1C.2-update-3 (settings on image). after that i turned on display 850C (tsdz protocol) and i got 30H error that does not want to go away, i have then connected old controller and error is still there. I tried to upload default 48v settings to the new controller, but no success.

Wires seems to be ok. I haven't checked the continuity with multimeter yet though. i do not know what is wrong with it, Is that possible that bricked the display while flashing new firmware?


another question is regarding my old controller, i managed to flash it few times in the past with the same emmebrusa firmware, and it works, however i was no longer able to communicate with it (that is why i bought new one, as i wanted to install cooling mod but i could not play with setting anymore). There is 0v on GDN and SWIM. is there any way to fix it so i can communicate with it again?
Or at least is it possible to read out current settings somehow?

best regards,

EDIT: i checked the wires with multimeter, all 6 pins are OK


View attachment 353359View attachment 353360

I have here a brandnew controller but it is just a spare part. Last week, I tried to flash. It worked only on the 5th attempt. The 4 wires going to the speed sensor (our programming cables) were not soldered correctly. I have soldered some pins instead of the cables and than I soldered the cables at the pins. Now it works fine.

 

[mbrusa]

@mbrusa, I've been using my VLCD5 with the wheel diameter set to "26" in the hidden settings, which is the actual diameter of my bike wheels. Per your manual, I tried setting it to the largest diameter available for better display resolution, which would be "30." Now, the Soc% is displayed as 9.9 with a full charge instead of 10.1. The battery voltage continues to properly read 40.9.

I just went for a 10-mile ride in Turbo mode and drained a couple of bars from the battery indicator. When I returned, I turned on the lights and checked the Soc%. Now it says 6.5. I believe that is actually 65%, off by a factor of 10. The Watt-hours consumed is displaying 11.8. I believe that should be 118 Watt-hours, also off by a factor of 10. My battery capacity is configured as 374 Wh, and the battery capacity calibration is set to 90%. So if the Soc is 65%, that means 35% was consumed. 374 Wh * 90% calibration * 35% consumed = 118 Wh consumed.

It looks like the Soc% and Wh consumed are being displayed at 1/10th their actual value?

[EDIT]

This is strange. I just reflashed the controller, toggling the "Set parameters on startup" from ON to OFF, and now the Soc% appears to be scaled correctly (not off by a factor of 10). The Wh consumed is still displaying at 1/10th its actual value, but now I realize that must be intentional due to the limitation of the VLCD5 display, which can only go up to 99.9 while Wh consumed will normally be over 100 (so 118 Wh will be displayed as 11.8).

One thing I noticed when I returned to the Configurator is that I was only getting four fields for the Cell Voltage in the Advanced tab; the VLCD5 should have six fields for the six-bar battery indicator. So I toggled the display type to VLCD6 and then back to VLCD5, and then the group of six Cell Voltage fields were activated again. That was the only other setting I changed before flashing.

I hope this information is useful. Thank you.

Sorry, there's one detail I completely forgot about.
The maximum value accepted by the display is 99.9, and is always sent in km/h.
By setting mph, it is the display that converts it, so the maximum displayable value becomes 62.4 (99.9/1.6), to display higher values I am forced to divide them by 10.
Unfortunately it is a limitation of the display.
I also remembered that in the new version that I will release soon, I added the possibility of setting mph also with VLCD6, where the km/miles conversion is not present.
Thinking about it, this new function could also be used with other displays, avoiding the inconvenience of changing the scale at 62.5.
However, there would be another one, the odometer history would remain in km, only those added would be in miles.

 

[larkito27]

Sorry, there's one detail I completely forgot about.
The maximum value accepted by the display is 99.9, and is always sent in km/h.
By setting mph, it is the display that converts it, so the maximum displayable value becomes 62.4 (99.9/1.6), to display higher values I am forced to divide them by 10.
Unfortunately it is a limitation of the display.
I also remembered that in the new version that I will release soon, I added the possibility of setting mph also with VLCD6, where the km/miles conversion is not present.
Thinking about it, this new function could also be used with other displays, avoiding the inconvenience of changing the scale at 62.5.
However, there would be another one, the odometer history would remain in km, only those added would be in miles.

Thank you for that explanation---that makes perfect sense. After my 10-mile ride, the Soc% dropped to 65%, and was displayed as 6.5. I then re-flashed the controller to reset the "Set parameters on startup" setting. When I powered on again, the display must have re-calculated the Soc% based on battery voltage rather than W-h consumed. The new Soc% was 53%, displayed as 53.0. When Ì recharged the battery last night, the Soc% displayed 9.9 again.

I wonder if it might be better to leave the display scale consistent (always divided by 10), or add a setting in the Configurator to allow the user to choose whether to display at full scale below 62.5. One advantage to always dividing by 10 is that it would add less error to the odometer. In any case, it would be helpful to mention this scale change at 62.5 in the manual, as it was confusing to see the numbers change so drastically between rides.

Thank you for the fine work you're doing on this project. It has added a tremendous amount of value to my e-bike and enjoyment to my bicycle riding.

 

[endlessolli]

..... in the new version that I will release soon, I added ....

@mbrusa : Can't wait or - as the italians say (I think):
Non vedo l'ora
(I have just learned this)
If you need an early tester - I am available
(As I am currently in vacation in northern Italy where the weather is as bad as we never have experienced bevor in this region....)

 

[Coolby]

Dear All,

Currently I have OSF on my tsdz2 with the stock 850c display (I have got it in a kit). Now I ordered a 860c (see exact version on the picture). I have built a bootloader cable, but I have stocked with the update. The display gets power and I can turn it on, I see thr CP210x usb to uart bridge on com port 7, I can open the port in APT Burn tool, but the update firmware shows only waiting despite I push the button of the controller. I see numbers running below at TX, but nothing on RX (just for info, I do not know if it is relevant or not). What can be the problem?

Other questions:
1. Which firmware should be used? I have lost that underground v20.1C.5 beta there is many bin files...
2. After I succeed to change firmware on the display I should change the OSF on the motor too, so which OSF I use with the Stock display shall be changed? For which version?

Thanks for the support.

Any advice on these topics?

 

[Coolby]

Any advice on these topics?

I have checked the wiring of my bootloader again. Rx goes to Tx, Tx to Rx based on the labeling of the usb-uart modul and the attached scheme of the display (with mirroring due to the female part of the cable). The display turns in with the usb-uart plugged in with 29v voltage, so that part seems okay as well. Also I can open the port and with that a red led turns in on the usb-uart, but after that when I push upload firmware, there is just waiting waiting and nothing happens when I push buttons on the controller.

What can be wrong?

 

[dgc]

Correct me if I am wrong but the marked up photo it looks like you have display TX female going to the controller TX male and the display RX female going to the controller RX male. The display TX should transmit signals to the controller RX and the display RX receives signals from the controller TX. I have had this problem on industrial control systems a few times.
The GND, PL and P+ seem to be marked up correctly. This maybe why you are getting a "waiting" on the computer.

 

[mbrusa]

...
I wonder if it might be better to leave the display scale consistent (always divided by 10), or add a setting in the Configurator to allow the user to choose whether to display at full scale below 62.5. One advantage to always dividing by 10 is that it would add less error to the odometer. In any case, it would be helpful to mention this scale change at 62.5 in the manual, as it was confusing to see the numbers change so drastically between rides.
...

I agree, changing scale to 62.4 wasn't a good idea, it confused me too!
Data that can exceed this value is best always divided by 10.
Alternatively, for those who want a larger scale, there will be the new function for VLCD6, also applicable to other displays.

 

[mbrusa]

I have checked the wiring of my bootloader again. Rx goes to Tx, Tx to Rx based on the labeling of the usb-uart modul and the attached scheme of the display (with mirroring due to the female part of the cable). The display turns in with the usb-uart plugged in with 29v voltage, so that part seems okay as well. Also I can open the port and with that a red led turns in on the usb-uart, but after that when I push upload firmware, there is just waiting waiting and nothing happens when I push buttons on the controller.

What can be wrong?

Trivial question, have you tried to reverse Rx with Tx?

 

[mbrusa]

@mbrusa : Can't wait or - as the italians say (I think):
Non vedo l'ora
(I have just learned this)
If you need an early tester - I am available
(As I am currently in vacation in northern Italy where the weather is as bad as we never have experienced bevor in this region....)

Yes, unfortunately this is a very rainy period, too bad for your holidays.
I'm late with the road tests.
When I have time to rehearse, it rains.
When it's sunny I have other commitments.
The new version is ready, I still have to do some checks and one final road test.
If you want to join the Burnt Controller Club, welcome!
You may have wondered why the standard display version is stuck at .2.
Because when I made someone try versions 3, 4, 5, it was a massacre of controllers, yet it never happened on my bikes.
Now I think I have identified the cause, so "incrociamo le dita". In German perhaps they say “Ich drücke dir die Daumen”?

 

[endlessolli]

Yes, unfortunately this is a very rainy period, too bad for your holidays.
I'm late with the road tests.
When I have time to rehearse, it rains.
When it's sunny I have other commitments.
The new version is ready, I still have to do some checks and one final road test.
If you want to join the Burnt Controller Club, welcome!
You may have wondered why the standard display version is stuck at .2.
Because when I made someone try versions 3, 4, 5, it was a massacre of controllers, yet it never happened on my bikes.
Now I think I have identified the cause, so "incrociamo le dita". In German perhaps they say “Ich drücke dir die Daumen”?

Ja ich drücke die Daumen
-> A burnt controller would not be so nice… but I would like to help with testing. Is there anything I could watch out for during testing to avoid a burnt controller? ( I use 36V version, xh18 Display, power assist mode, limit current at 16A, but ride mostly at 5 to 8 A with speedlimit set to 25km/h and Temp sensor enabled. I use no startup assist / boost)

 

[mbrusa]

It wasn't a problem with the settings or the way of use, I think the cause was the undefined Irq uart priority level.
I'm 100% sure, but confirmation will only come when many people try...

 

[dgc]

Yes, unfortunately this is a very rainy period, too bad for your holidays.
I'm late with the road tests.
When I have time to rehearse, it rains.
When it's sunny I have other commitments.
The new version is ready, I still have to do some checks and one final road test.
If you want to join the Burnt Controller Club, welcome!
You may have wondered why the standard display version is stuck at .2.
Because when I made someone try versions 3, 4, 5, it was a massacre of controllers, yet it never happened on my bikes.
Now I think I have identified the cause, so "incrociamo le dita". In German perhaps they say “Ich drücke dir die Daumen”?

I think we have all got a tea-shirt / sweatshirt for that.

 

[Coolby]

E

Correct me if I am wrong but the marked up photo it looks like you have display TX female going to the controller TX male and the display RX female going to the controller RX male. The display TX should transmit signals to the controller RX and the display RX receives signals from the controller TX. I have had this problem on industrial control systems a few times.
The GND, PL and P+ seem to be marked up correctly. This maybe why you are getting a "waiting" on the computer.tzanks

Trivial question, have you tried to reverse Rx with Tx?

Thanks all, I switched tx-rx and after that I was able to program. Now I hope the display is okay, but it shows the attached screen and only if I keep pushing the on-off button. Is this normal, it will turn in if the motor will be connected? (I am not sure everything is fine, because in some videos the screen fully turns in).

Thanks

 

[Coolby]

E



Thanks all, I switched tx-rx and after that I was able to program. Now I hope the display is okay, but it shows the attached screen and only if I keep pushing the on-off button. Is this normal, it will turn in if the motor will be connected? (I am not sure everything is fine, because in some videos the screen fully turns in).

Thanks

Picture

 

[mbrusa]

Display and motor must be connected

 

[Coolby]

Display and motor must be connected

Thanks, now I need my bike back from service and try.

I should change the motor firmware as well (now it is osf with stock display), which osf version I need?

 

[dgc]

That is great, just waiting for an adapter so I can program the 860C and I will give it a go on my other bike.

 

[mbrusa]

Thanks, now I need my bike back from service and try.

I should change the motor firmware as well (now it is osf with stock display), which osf version I need?

You need the same version you used for the display.
Where you downloaded the.bin file for display there is also the.hex file for the motor.
TSDZ2-v20.1C.5-beta-1-860C.hex
To flash you need ST Visual Programmer

 

[Weissnix]

I need a new Display, my 850C has broken So i can buy a 860c Bafang uart version, right? But they have other connectors. My old 850C had a 6 pin connector (male), an my flasher has the counterpart. So how can I connect the bafang uart display to my motor and flasher?