Tsdz2 firmware open source adapted to vlcd5, vlcd6 and xh18

[hoka]

Last Sunday I was testing my BT, Android app and my patched emmebrusa fw and my unit has went wrong. Therefore I've removed the related links to this solution for safety.

Have you sorted out what went wrong?
I am really interested in solution without display and as few wires as possible so I hope you will share your solution again!

 

[Elinx]

.....
Have you sorted out what went wrong?
....

As you can read here, feketehegyi had installed v20.1C.3-NEW and burned some mosfets on his controller, which was a problem of the latest 20.1C.3 versions for stock display's. Problably nothing was wrong with his patch.
I think his BT solution with v20.1C.1 (for stock display) should be safe enough to use.

 

[feketehegyi]

Last Sunday I was testing my BT, Android app and my patched emmebrusa fw and my unit has went wrong. Therefore I've removed the related links to this solution for safety.

Have you sorted out what went wrong?
I am really interested in solution without display and as few wires as possible so I hope you will share your solution again!

Well, as it was mentioned - that particular emmebrusa fw version has caused it.
Factory default fw can work without display.

 

[hoka]

Last Sunday I was testing my BT, Android app and my patched emmebrusa fw and my unit has went wrong. Therefore I've removed the related links to this solution for safety.

Have you sorted out what went wrong?
I am really interested in solution without display and as few wires as possible so I hope you will share your solution again!

Well, as it was mentioned - that particular emmebrusa fw version has caused it.
Factory default fw can work without display.

OK, I'm a newbie and studying hard...So your implementation works on the factory default firmware AND without any other display or button?

 

[feketehegyi]

Last Sunday I was testing my BT, Android app and my patched emmebrusa fw and my unit has went wrong. Therefore I've removed the related links to this solution for safety.

Have you sorted out what went wrong?
I am really interested in solution without display and as few wires as possible so I hope you will share your solution again!

Well, as it was mentioned - that particular emmebrusa fw version has caused it.
Factory default fw can work without display.

OK, I'm a newbie and studying hard...So your implementation works on the factory default firmware AND without any other display or button?

I've tried lots things with this motor. eg. to go without display using the factory default fw. Actually you need to figure out how to enable the controller. This is actually to short two pins in the display cable. I did it eg. with a jumper like thing to put into the display cable connector.

 

[hoka]

[/quote]
I've tried lots things with this motor. eg. to go without display using the factory default fw. Actually you need to figure out how to enable the controller. This is actually to short two pins in the display cable. I did it eg. with a jumper like thing to put into the display cable connector.
[/quote]

Yeah, I saw that on one of your schematics on github

Will you make your app available again?

 

[Griausmas]

Please help me, I flashed OSF everything was working fine, I was testing all levels of help in Emtb mode, and then suddenly it stopped working at all, tried to unplug the battery but still the same, tried reflashing again with different mode, but still the same.. After many attempts tried to flash back to stock firmware but its not working either.. what should I do? please help.. I only have 100km on it..

 

[Elinx]

Please help me, .... suddenly it stopped working at all, ....

What FW do you have flashed now? stock or OSF
And which version of OSF and what display was/is flashed before?
Which settings you insert in Java configurator?
Does the display still work?

You can try to check already some things
Speed sensor magnet must be far enough (10mm) of the reedcontact.
Torque and PAS sensor. Can you use walkassist or throttle?
If you have vlcd5, you can check the torque sensor with stock FW inside special menu (TE and TE1)
With XH18 and vlcd6 this can be done too, but only with OSF
PAS sensor can be measured with a multimeter on the controller.

If you have installed OSF it is possible to drive the bike with a broken sensor to determine which sensor could be defect.
Other cause could be a burned mosfet on the controller or hall sensor(s) inside the bare motor, but first look at the sensors.

 

[Griausmas]

Thank you for your reply. I flashed 1-20.1C.1-VLCD5-VLCD6-XH18 firmware, first it was ok, but after some ride it stopped working, so I tried changing settings but none helped. My display is VLCD5. display works as before.
Ok, so now I tried flashing back to stock firmware, and made some mistake with option byte, I had to flash original option byte first, then program memory and then data memory . So this time I did everyting normal, and tested it, it started to work again as before... So what is the case here? where I did a mistake? I'm now afraid to flash OSF at all. I don't even know if it will work now on stock as it was before for more than 1km.. Sorry for my English

 

[Griausmas]

And from that site Eco-bike that firmware 48volt is 500w with throttle? because my motor was 500w 48v with throttle..

 

[Elinx]

... it started to work again as before... So what is the case here? where I did a mistake? I'm now afraid to flash OSF at all. I don't even know if it will work now on stock as it was before....)

If you have flashed stock FW back and the engine runs, the sensors are all OK and we only can guess what could be wrong, which is not very effective without OSF.
The only thing we could do for you is that you share your OSF ini file (latest file inside "experimental settings"folder), so we can look if there were some strange values.

About throttle
I don't know if there is a different stock FW for it.
It could be only the controller wiring and 6-pin vs 8-pin connector.

 

[Griausmas]

... it started to work again as before... So what is the case here? where I did a mistake? I'm now afraid to flash OSF at all. I don't even know if it will work now on stock as it was before....)

If you have flashed stock FW back and the engine runs, the sensors are all OK and we only can guess what could be wrong, which is not very effective without OSF.
The only thing we could do for you is that you share your OSF ini file (latest file inside "experimental settings"folder), so we can look if there were some strange values.

About throttle
I don't know if there is a different stock FW for it.
It could be only the controller wiring and 6-pin vs 8-pin connector.

Thank you very much. I hope that every sensor is ok, because now it works, only thing that I don't know is which stock firmware i'm running 500w or 750w. I will upload my OSF ini file if that helps, i hope there is some error here..

 

[Elinx]

... only thing that I don't know is which stock firmware i'm running 500w or 750w. ....
....my OSF ini file if that helps

There is no difference.
500W and 750W are the same.
Tsdz2 has a 36V/4000RPM or 48V/4000RPM engine.

I had a look on your ini.
The first I see was 12S battery (44V nominal) and a cutoff of 44V.
With this cutoff you only can use a full charged battery :wink:
Under cuttoff Voltage the engine shut down.
A li-ion 3.7V cel functioned between 3.1V and 4.2V
So 12S functioned between 12*3.1=37.2V and 12*4.2= 50.4V
So your cutoff should be 37V for 12S

Are your sure you have a 44V battery? 48V is 13S (40V-54.6V)
44V could give loss of performance on stock FW too
Also with just 222Wh battery and a stock cutoff of 40V, the distance you can biking isn't that much.
Other thing I see, you haven't calibrated the torque sensor, which is needed for optimal performance

 

[Griausmas]

... only thing that I don't know is which stock firmware i'm running 500w or 750w. ....
....my OSF ini file if that helps

There is no difference.
500W and 750W are the same.
Tsdz2 has a 36V/4000RPM or 48V/4000RPM engine.

I had a look on your ini.
The first I see was 12S battery (44V nominal) and a cutoff of 44V.
With this cutoff you only can use a full charged battery :wink:
Under cuttoff Voltage the engine shut down.
A li-ion 3.7V cel functioned between 3.1V and 4.2V
So 12S functioned between 12*3.1=37.2V and 12*4.2= 50.4V
So your cutoff should be 37V for 12S

Are your sure you have a 44V battery? 48V is 13S (40V-54.6V)
44V could give loss of performance on stock FW too
Also with just 222Wh battery and a stock cutoff of 40V, the distance you can biking isn't that much.
Other thing I see, you haven't calibrated the torque sensor, which is needed for optimal performance

I think you find a problem I was scared about, thank you so much! I was using 3 lipos from a drone for testing it was 4s each so 12s 50.4 volts, and they were on 3.75 volts per cell, so there was a cutoff, such a stupid mistake.. But I saw my screen was showing 3 bars or something.. With original firmware I think the cutoff was a little deeper so with the same battery it was working.. Thank you so much again for helping me out! I will try flashing OSF again. But how do I calibrate torque sensor after flashing?

 

[Elinx]

... With original firmware I think the cutoff was a little deeper
.... how do I calibrate torque sensor after flashing?

Correct, I think about 40V instead of your 44V for OSF
With a cutoff of 37V it should be solved.

The calibration must be done after flashing OSF. The "how to" you can find in the manual page 4

In short:
Set assist level 1
Push light button 6 times (3x2), when E04 flashing, push again 2x
Read 3 digits for zero offset and with full weight max. range.
After this calibration of the torque sensor, insert the measured values in the javaconfigurator and enable
"torque adv. on startup" and "torque sensor calibrated"
Flash controller again.

 

[Griausmas]

... With original firmware I think the cutoff was a little deeper
.... how do I calibrate torque sensor after flashing?

Correct, I think about 40V instead of your 44V for OSF
With a cutoff of 37V it should be solved.

The calibration must be done after flashing OSF. The "how to" you can find in the manual page 4

In short:
Set assist level 1
Push light button 6 times (3x2), when E04 flashing, push again 2x
Read 3 digits for zero offset and with full weight max. range.
After this calibration of the torque sensor, insert the measured values in the javaconfigurator and enable
"torque adv. on startup" and "torque sensor calibrated"
Flash controller again.

Thank you so much for your help! Now everything is good, and I'm calm because there is no problems with the motor
i'll try to calibrate my torque sensor and flash it again , because without torque sensor calibrated it was little strange feeling to ride OSF.

 

[Elinx]

.....Now everything is good, ....
i'll try to calibrate my torque sensor and flash it again , because without torque sensor calibrated it was little strange feeling to ride OSF.

I'm glad it's solved relative easy for you now

Realize there is a lot of difference how stock FW and OSF handles the engine. This isn't only torque sensor related.
OSF is more responsive for the complete power range and because of fieldweakening more effective for lower battery Voltage, as in your case with 12S-44V battery.
The torquesensors have all a different sensitivity. That is why calibration is needed for the best result.

 

[abrahad]

I have now flashed with OSF 20.1c and am impressed with the improvement over the standard firmware.

I tried today to use the VLCD5 display to change some parameters but cannot figure out how to do this. For example, LEVEL 0/E02 should enable/disable SET PARAMETER. How do I change this setting to enable SET PARAMETER? Is it with the +/- button whilst the display is flashing? Can't seem to get this to work.

Do I need to use the configurator and tick "Set parameter on startup" then re-flash?

 

[Elinx]

.....VLCD5 display ....
For example, LEVEL 0/E02 should enable/disable SET PARAMETER. How do I change this setting ....

I have explained here how to change parameters with VLCD5.

You can choose to enable "set parameters" inside Java configurator to change other parameters directly.
Some disable it, if they are afraid that someone change some parameters by accidence.
But if you are the only one that uses the ebike you can make "set parameters" default.
You don't need to reflash the controller.

Level 0-E02: L0 - 2 pushes on light button to toggle this on (or off).
Level 0-E04: L0 - 3x2 pushes on light button to save all currrent settings.

 

[Badmotor]

Hello.

It's been a while since I posted here. So I've been using open source for about 2 years and it's been great. However lately I began experiencing some issues. When the battery is about half way discharged and I have to go uphill, the motor would suddenly shut down. If I try to push pedals harder it would make a very weak struggling noise, as if it wasn't getting enough power. I shifted the gear lower and it was suddenly ok again.

my motor is 36v and battery is the same. I checked the battery connections and some of them were oxidized. I took it apart and cleaned every connector as well as I could but that didn't solve the issue.

So I decided to flash the latest version of open source which is v20.1C.1 and reduced battery max current. I set it to 13 instead of 15 which was there before. Well, as I went out to test it while the battery still was at half charge, the motor would shut down even faster and this time it would throw out error E08. This time shifting gears wouldn't help and even restarting the motor would still produce the same error within a few seconds. I checked the release notes and found this

7 - Added error E08 due to speed sensor failure or missing magnet.

So is it all because of speed sensor problem? Is it supposed to shut down just because speed sensor isn't working? I have replugged and and shifted it a few times, but that didn't help. I wonder if there's maybe some wire that has been bent too much inside the motor? I was avoiding opening the motor because of lack of time. Or could it also be that my battery is simply dying and can't provide enough current ? I've used it for about 3 years now.

Any advice is appreciated. Thanks.

 

[kaziq]

When the battery is about half way discharged and I have to go uphill, the motor would suddenly shut down. If I try to push pedals harder it would make a very weak struggling noise, as if it wasn't getting enough power. I shifted the gear lower and it was suddenly ok again.

I have a very similar issue with a battery that I made myself. For me the motor shuts down at the 1/3 of capacity (2 or 3 bars on the VLCD5 display). I tested it with an artificial load, and found out that the BMS was shutting down the battery at 3.4V per cell. I guess I didn't match the cells enough, and some of them get to lower voltage quicker than others, and BMS disconnects everything to avoid damaging the weakest cells.

I think I will just adjust the battery level bar voltages to make the display show real capacity left. For you it may be also other things, like oxidized connections, that cause large voltage drop when the motor wants more power. Perhaps connections inside the battery are degraded too (it shouldn't happen for professional batteries, but is an option).

You also may be experiencing battery cell degradation over time, and BMS protecting the weakest cells.

 

[mbrusa]

...

7 - Added error E08 due to speed sensor failure or missing magnet.

So is it all because of speed sensor problem? Is it supposed to shut down just because speed sensor isn't working? I have replugged and and shifted it a few times, but that didn't help. I wonder if there's maybe some wire that has been bent too much inside the motor? I was avoiding opening the motor because of lack of time. Or could it also be that my battery is simply dying and can't provide enough current ? I've used it for about 3 years now.

Any advice is appreciated. Thanks.

Yes, you have a speed sensor problem.
You can try to set "Assistance with sensors error".
Copy and paste from the manual.

Attention, the presence of an error disables assistance in all modes. It is however possible to force assistance even with an error if this is caused by a problem with a sensor, torque, cadence or speed.
You will have to choose the assistance mode that does not involve the use of the faulty sensor.
In the configurator "Lights mode 3" must be set to 10 - "Assistance with sensors error".
Enable at level 4-TURBO, E04 - ASSIST WITH SENSORS ERROR, by pressing the lights button 2 + 2 + 2 times until E04 flashes. Attention, this function cannot be set in the configurator, at power on it is always disabled, it is however possible to save the setting in eeprom on the display.
Use only in case of need, with this function enabled there are limitations in assistance.

Regarding the battery, it may be that as the battery ages the capacity decreases and the drop increases.
What capacity does the battery have?

 

[Elinx]

... ... Or could it also be that my battery is simply dying and can't provide enough current ? I've used it for about 3 years now.

A battery is as good as it weakest cell. The BMS must keep all cells on the same Voltage. If just one cell has a too low Voltage the BMS cut the power. After some time and degredation of the cells a BMS isn't fast enough anymore to keep all Voltages equal. There a two types of a BMS.
The cheapest type keeps the Voltages of the highest cells down by a resistor till other cells are the same too. This cost a lot of time and can only compensate just a small difference in Voltage and does only work when charging.
The other type is more active and compensate faster between the cells. The higher cells charges the lower ones too. Also with discharging. So the difference between cells keep small as possible in time.

If you had flashed v20.1b before and now the latest v20.1C.1, fieldweakening is enabled.
That works at the cost of more current with lower Voltage. That is why your motor shuts down faster.

An advice is to open the battery, measure the Voltage of each 10 cellblocks and charge the weakest one seperately or discharge the highest ones with a resistor till all cellblocks are equal.
After this you can use your battery some time depended how bad that weakest cell(s) is.
If there is some space in the batterycase you can add an active balancer parrallel to the BMS, which keeps the Voltage between cells the active way. No guarantee this works for you, because the unknown state of the weakest cell(s).
Eventually replacing the weakest cells with the same celltype is also an option, but in that case I advice always to use an active BMS/balancer.

 

[Badmotor]

When the battery is about half way discharged and I have to go uphill, the motor would suddenly shut down. If I try to push pedals harder it would make a very weak struggling noise, as if it wasn't getting enough power. I shifted the gear lower and it was suddenly ok again.

I have a very similar issue with a battery that I made myself. For me the motor shuts down at the 1/3 of capacity (2 or 3 bars on the VLCD5 display). I tested it with an artificial load, and found out that the BMS was shutting down the battery at 3.4V per cell. I guess I didn't match the cells enough, and some of them get to lower voltage quicker than others, and BMS disconnects everything to avoid damaging the weakest cells.

I think I will just adjust the battery level bar voltages to make the display show real capacity left. For you it may be also other things, like oxidized connections, that cause large voltage drop when the motor wants more power. Perhaps connections inside the battery are degraded too (it shouldn't happen for professional batteries, but is an option).

You also may be experiencing battery cell degradation over time, and BMS protecting the weakest cells.

I see. I was hoping it wouldn't be because of the battery.. It just started happening so suddenly.

Yes, you have a speed sensor problem.
You can try to set "Assistance with sensors error".
Copy and paste from the manual.


Regarding the battery, it may be that as the battery ages the capacity decreases and the drop increases.
What capacity does the battery have?

Alright. Thanks. I solved the sensor problem now, but not the power drop. I didn't know the battery issue can manifest itself in such a way. The capacity was quite low from beginning , 36v, 12 ah. But it never shut down before. Only when 1 bar was left.

 

[Badmotor]

... ... Or could it also be that my battery is simply dying and can't provide enough current ? I've used it for about 3 years now.

A battery is as good as it weakest cell. The BMS must keep all cells on the same Voltage. If just one cell has a too low Voltage the BMS cut the power. After some time and degredation of the cells a BMS isn't fast enough anymore to keep all Voltages equal. There a two types of a BMS.
The cheapest type keeps the Voltages of the highest cells down by a resistor till other cells are the same too. This cost a lot of time and can only compensate just a small difference in Voltage and does only work when charging.
The other type is more active and compensate faster between the cells. The higher cells charges the lower ones too. Also with discharging. So the difference between cells keep small as possible in time.

If you had flashed v20.1b before and now the latest v20.1C.1, fieldweakening is enabled.
That works at the cost of more current with lower Voltage. That is why your motor shuts down faster.

An advice is to open the battery, measure the Voltage of each 10 cellblocks and charge the weakest one seperately or discharge the highest ones with a resistor till all cellblocks are equal.
After this you can use your battery some time depended how bad that weakest cell(s) is.
If there is some space in the batterycase you can add an active balancer parrallel to the BMS, which keeps the Voltage between cells the active way. No guarantee this works for you, because the unknown state of the weakest cell(s).
Eventually replacing the weakest cells with the same celltype is also an option, but in that case I advice always to use an active BMS/balancer.

Thanks for the suggestions. I'll look into it. I'm still thinking of checking inside the motor first , just in case there's some hardened grease or any other unexpected troubles. I have a "backup" battery, which is from another bike and might be possible to repurpose for mine. The issue is I don't know how much it has been used. Although your idea sounds interesting as well.