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

[mbrusa]

I don't assume the motor is failing. I've just noticed that when I resume pedaling, there's an additional braking force (more strain on my legs than if I were riding with the motor completely off). This force disappears after a few seconds, when the motor begins to provide assistance again. It feels as if the motor is trying to rotate in the opposite direction, only to reverse direction after a few seconds. I think this is due to the incorrect motor operating parameters I've set. Unfortunately, I don't know which parameters I'm referring to.

The term "motor braking" you used is just an impression, because if the sprag clutch is working properly, the motor doesn't prevent the pedals from turning, even if it's locked. I interpreted it as a delay in reactivating the assistance when you've just stopped pedaling.

I asked you to try v20.1C.5 because in the latest version 5-update-1, I increased the minimum rpm for reactivating the motor, and this should have improved the situation.
I suspected I'd introduced a bug, but I checked carefully and ruled it out.

The delay in reactivating the assistance is a known issue, already reported by others, but not since the latest version, but always. It's a problem that only occurs under certain conditions; it rarely happens to me.
Finding the cause has always been on my to-do list.

I've been analyzing the code carefully these days and think I've figured out the cause of the delay in reactivating the assistance.
However, reading your description, I'm not sure if this is the same problem.
Now enjoy your vacation; I'll send you a new version to try before the end of the month.

Here's the update that fixes the delay in reactivating the motor.
It's not necessary to update the display; if you've installed v20.1C.5-update-1, just update the motor.
https://github.com/emmebrusa/TSDZ2-....5-update-1-860C/TSDZ2-vv20.1C.5-1-1-860C.hex
As soon as you can try it, let me know if it works.

I recommend everyone try this update,
because it's significantly improved even when starting from a standstill.

The stock display version is also updated; anyone who wants to try it can download the master version.
https://github.com/emmebrusa/TSDZ2-Smart-EBike-1/archive/refs/heads/master.zip

 

[piotrby]

Here's the update that fixes the delay in reactivating the motor.
It's not necessary to update the display; if you've installed v20.1C.5-update-1, just update the motor.
https://github.com/emmebrusa/TSDZ2-....5-update-1-860C/TSDZ2-vv20.1C.5-1-1-860C.hex
As soon as you can try it, let me know if it works.

I recommend everyone try this update,
because it's significantly improved even when starting from a standstill.

The stock display version is also updated; anyone who wants to try it can download the master version.
https://github.com/emmebrusa/TSDZ2-Smart-EBike-1/archive/refs/heads/master.zip

Thanks.
I'll try it next week and let you know.

Best regards

 

[arkanum]

Hey,
I have the problem that the engine switches off. It is not possible to reproduce the exact cause. It probably has something to do with slightly higher amperage currents in a higher support level. But I am not sure. I seem to remember that I had a similar problem with the casahino software before. The software ran flawlessly for a few years and then it started to sputter ... a reinstallation fixed the problem. We are now on the road and reinstalling is not an option. Does anyone have any idea whether the memory can be restored with just the vlcd6, a reset of the software or something similar?

Thanks in advance!

 

[arkanum]

The actual software is the embrusa/github version ...

 

[Yanis]

Here's the update that fixes the delay in reactivating the motor.
It's not necessary to update the display; if you've installed v20.1C.5-update-1, just update the motor.
https://github.com/emmebrusa/TSDZ2-....5-update-1-860C/TSDZ2-vv20.1C.5-1-1-860C.hex

I rode about 50 km with this update (with SW102).
I didn't have any reactivation issue, but it seems to have mostly solved another issue I had, which could be related : a "delayed over run" of the motor. In higher assistance settings, when I stopped pedalling the motor would stop immediately, but then less than half a second later it would restart for a very brief moment. This happened just sometimes, but with this update it doesn't seem to happen anymore..
Thanks a lot mbrusa !

 

[jgs1]

Does anyone know if you can enable/disable street mode via the DZ40 display?

I have a TSDZ2B that came with the DZ40 and flashing the OSF with the screen setting to VLCD5 works but I can't figure out if it's possible to change the street mode. I've tried following the manual i.e. holding the lights button until 02 appears but it doesn't seem to do anything.

set DZ40 as VLCD5 works fine
I've put the DZ40 on 6 bikes ( bought from Enerprof )
Long press (2sec) on up to emulate lights on/off
To turn off /on street mode you have to long press up twice

 

[dgc]

jgs1, that's interesting, I set my DZ41's as a VLCD5 which as you say works fine for me as well, but I wonder if as the DZ40 these button operations will work for me on the DZ41.
Long press (2sec) on up to emulate lights on/off
To turn off /on street mode you have to long press up twice

 

[pav]

Hey,
I have the problem that the engine switches off. It is not possible to reproduce the exact cause. It probably has something to do with slightly higher amperage currents in a higher support level. But I am not sure. I seem to remember that I had a similar problem with the casahino software before. The software ran flawlessly for a few years and then it started to sputter ... a reinstallation fixed the problem. We are now on the road and reinstalling is not an option. Does anyone have any idea whether the memory can be restored with just the vlcd6, a reset of the software or something similar?

Thanks in advance!

I have recently encountered a similar problem. I use a 48V battery and the motor randomly shuts down when the voltage drops below 47V. No error code appears on the display. Turning the display off and on restores functionality for a while. The problem is not related to motor overheating. The battery is also fine—measured with an external voltmeter under load. I suspect either a bug in the OSF firmware or a faulty electronic component, such as a faulty capacitor in the motor controller.

 

[docw009]

I have recently encountered a similar problem. I use a 48V battery and the motor randomly shuts down when the voltage drops below 47V. No error code appears on the display. Turning the display off and on restores functionality for a while. The problem is not related to motor overheating. The battery is also fine—measured with an external voltmeter under load. I suspect either a bug in the OSF firmware or a faulty electronic component, such as a faulty capacitor in the motor controller.

You might check the cell balance of that battery to be sure it's good. Could have one cell group half charged.

 

[pav]

You might check the cell balance of that battery to be sure it's good. Could have one cell group half charged.

As I wrote, the total battery voltage is stable at around 47V even under load (measured with a voltmeter), yet the motor suddenly shuts down. In the configurator, I have the battery cut-off voltage set to 39V.

 

[arkanum]

I have recently encountered a similar problem. I use a 48V battery and the motor randomly shuts down when the voltage drops below 47V. No error code appears on the display. Turning the display off and on restores functionality for a while. The problem is not related to motor overheating. The battery is also fine—measured with an external voltmeter under load. I suspect either a bug in the OSF firmware or a faulty electronic component, such as a faulty capacitor in the motor controller.

Ok. Could you solve it with a new istallation or any other way?

 

[arkanum]

As I wrote, the total battery voltage is stable at around 47V even under load (measured with a voltmeter), yet the motor suddenly shuts down. In the configurator, I have the battery cut-off voltage set to 39V.

In may case we have two almost new 36v batteries.

 

[Yanis]

You might check the cell balance of that battery to be sure it's good. Could have one cell group half charged.

As I wrote, the total battery voltage is stable at around 47V even under load (measured with a voltmeter), yet the motor suddenly shuts down. In the configurator, I have the battery cut-off voltage set to 39V.

As he wrote, you might check the voltage of separate groups of cells in the battery.
The controller cuts off according to the total voltage, but the BMS monitors cells groups and will cut off if one group drops below around 2.6V. If one group is less charged, the BMS might cut off even if the total voltage of the battery is relatively high.

 

[docw009]

The stock display version is also updated; anyone who wants to try it can download the master version.
https://github.com/emmebrusa/TSDZ2-Smart-EBike-1/archive/refs/heads/master.zip

I tried this new update. It works OK for me in hybrid mode, while the last 1.C.6 release did not seem to give me much assist. I used it with both uncalibrated T/S and then checking off the calibrated boxes, and putting in the zero weight and max weight numbers. Couldn't feel the difference with easy pedalling,

With the VLDC5 though, it does not give me the T/S numbers when I power up, select Eco (or Tour) and press the "V" button.

I feel that the pedal assist feels a little better, using the same assist numbers in the Java configurator with the 1.C.2 release. Maybe smoother or easier. I went back to the older release.

By the way, I went back to the factory firmware for a test ride.It is not as weak as I remembered, as I could discern that eco was switched on, but the assist was very weak. My OSF is current tuned so my eco level is probably between tour and sport on the factory firmware,

 

[sudmanche]

set DZ40 as VLCD5 works fine
I've put the DZ40 on 6 bikes ( bought from Enerprof )
Long press (2sec) on up to emulate lights on/off
To turn off /on street mode you have to long press up twice

Do you get 5 assist modes on the DZ40? As I cannot get level 1 & 2 to be any different to each other.

 

[piotrby]

Here's the update that fixes the delay in reactivating the motor.
It's not necessary to update the display; if you've installed v20.1C.5-update-1, just update the motor.
https://github.com/emmebrusa/TSDZ2-....5-update-1-860C/TSDZ2-vv20.1C.5-1-1-860C.hex
As soon as you can try it, let me know if it works.

I recommend everyone try this update,
because it's significantly improved even when starting from a standstill.

The stock display version is also updated; anyone who wants to try it can download the master version.
https://github.com/emmebrusa/TSDZ2-Smart-EBike-1/archive/refs/heads/master.zip

Good morning.
Initial tests done:
Tuesday = 60 km
Wednesday = 55 km
Each ride – started with a 100% charged battery (and after the ride, the display showed approximately 70% charge). I have a 15Ah 48V 72Wh battery, and the motor consumes approximately 0.5%/km.

I used the eMTB assist and the assist power ranged from 170/254 to 230/254.

After installing the new software on the motor, I returned to all previous settings (exactly as described in the post from July 25, 2025).
Driving observations – Tuesday:
Pros – The engine engaged the power steering significantly faster (the "perceptible engine braking" effect was reduced to an acceptable level compared to the previous version of the engine software).
Cons – During the entire ride, I turned the engine off and on eight times to clear the E7 error.

Driving observations – Wednesday:
I will discuss this ride by dividing the route into the first stage = 35 km and the second stage = 20 km.
Stage One
The motor responds very well to the power assist. It doesn't jerk when changing gears up or down, but provides quick and smooth assistance. Assistance also occurs smoothly and quickly after a break in pedaling. The motor didn't display any errors.
Stage Two
The motor displayed error code E7 (turn off/on to clear and continue riding) four times. The motor engaged the power assist differently after pedaling again – sometimes correctly (like in the first stage), and sometimes it took longer. The motor temperature didn't exceed 40-45 degrees Celsius.
I tried to find a correlation between pedaling style and power assist time, and my observations are very subjective. If I apply more pressure to the pedals (press harder), the motor response delay increases (I wait longer for the power assist to engage). Gentle pedaling causes the power assist to engage more quickly. Sometimes, changing gears (I have a 1x11 drivetrain) causes the power assist to engage more quickly. I also noticed (also a subjective observation) that when climbing a small hill like a river embankment (3-4 m high and steep), too small a gear reduction and sudden, strong pressure on the pedals causes error E7
Questions:
What else should I check (test)?
How can I improve my current engine settings? (To eliminate E7 and further improve engine performance)

 

[mbrusa]

I've already answered your question about E07.
Try increasing the "Overcurrent error delay" value; the default is 2. Try increasing it gradually until you don't get false errors.
The maximum value is 5; you can also disable the function by setting the value to zero.
If you're looking for better responsiveness when starting or restarting, don't use eMTB mode.
Use Power mode with "Startup boost" enabled and "Startup boost at zero" set to cadence.
If you'd like a more aggressive start, try increasing the "Startup boost torque factor" value to 350/400, but no higher.

 

[Bday988]

Hey all, I added the lm35 temperature sensor to my build but it is not working right. I have OSF 4.2 on the display and motor. The sensor is reading high at 55C at room temperature so something is already weird. I tried multiple sensors. But what is really strange is when the sensor is enabled, I get no assist. I am at a loss of what is wrong. For now, I just disabled the sensor in the settings and I am able to ride normally. I would really like to get this working if possible.

Any ideas? Thanks.

 

[piotrby]

I've already answered your question about E07.
Try increasing the "Overcurrent error delay" value; the default is 2. Try increasing it gradually until you don't get false errors.
The maximum value is 5; you can also disable the function by setting the value to zero.
If you're looking for better responsiveness when starting or restarting, don't use eMTB mode.
Use Power mode with "Startup boost" enabled and "Startup boost at zero" set to cadence.
If you'd like a more aggressive start, try increasing the "Startup boost torque factor" value to 350/400, but no higher.

Thanks for your valuable feedback.
Regarding error E7 – (I'll implement these settings). For the test, I wanted the bike settings to be the same as with the previous firmware.
I'll experiment with other support settings, such as eMTB, and with "Startup Boost."
What puzzles me and I'll have to observe further is that with the same motor settings, its responsiveness varies (sometimes it responds faster, sometimes slower – subjective impressions).
This is not related to the motor temperature indicated by the LM35 sensor and is not caused by a drop in battery voltage (after riding, V=approximately 50V).
Best regards

 

[piotrby]

Hey all, I added the lm35 temperature sensor to my build but it is not working right. I have OSF 4.2 on the display and motor. The sensor is reading high at 55C at room temperature so something is already weird. I tried multiple sensors. But what is really strange is when the sensor is enabled, I get no assist. I am at a loss of what is wrong. For now, I just disabled the sensor in the settings and I am able to ride normally. I would really like to get this working if possible.

Any ideas? Thanks.

Good morning.
I had a situation where I bought an LM35 temperature sensor (visually and the markings matched), but it was some kind of transistor, not a temperature sensor.
I suggest checking the sensor.
Connect:
1 – ground
2 – signal
3 – V=+5V
and at room temperature it should be: signal=250mV, and at 100°C = 1V.
When heated from room temperature, the signal increases from 250mV → 1V.
Check the sensor.

 

[Bday988]

Good morning.
I had a situation where I bought an LM35 temperature sensor (visually and the markings matched), but it was some kind of transistor, not a temperature sensor.
I suggest checking the sensor.
Connect:
1 – ground
2 – signal
3 – V=+5V
and at room temperature it should be: signal=250mV, and at 100°C = 1V.
When heated from room temperature, the signal increases from 250mV → 1V.
Check the sensor.

Thanks for the reply. Let me clarify what I saw. Before I connected the lm35 to the controller, I was reading 4.6v across the 5v and GND. So that looks good. But when I read the voltage across the throttle wire and ground, I was reading 0.16v. This was without the sensor even hooked up. And the display was reading 32C. When I hooked the sensor up, the reading on the throttle line was .52v and the display said 50C which is way above room temperature reading that should be .22v. It seems to me that the throttle input line is being fed power by the controller when it should be reading at 0v if it is purely an input line. Is there a resistor somewhere? My controller is not a coaster brake controller to my knowledge. I looked and I did not see a resistor next to the throttle input. I hope all of that made sense. This is pretty perplexing.

 

[arkanum]

I was flashing the motor 2x to play with the configuration. But now i cant flash it anymore. During the flashing i get no error or warning. Flashing successful. But no soc at start up, no speed detected and no power from the motor.
Any ideas what i can change in the flashing procedure?
I use the latest software and configurator. Have a vlcd6...

 

[Yanis]

I was flashing the motor 2x to play with the configuration. But now i cant flash it anymore. During the flashing i get no error or warning. Flashing successful. But no soc at start up, no speed detected and no power from the motor.
Any ideas what i can change in the flashing procedure?
I use the latest software and configurator. Have a vlcd6...

Try flashing the stock option bytes file (available somewhere, just google it) with STVP software.
It did the trick for me when I had the same symptoms once.

 

[arkanum]

Try flashing the stock option bytes file (available somewhere, just google it) with STVP software.
It did the trick for me when I had the same symptoms once.

Mega! Thank you, seems to work.

 

[piotrby]

Thanks for the reply. Let me clarify what I saw. Before I connected the lm35 to the controller, I was reading 4.6v across the 5v and GND. So that looks good. But when I read the voltage across the throttle wire and ground, I was reading 0.16v. This was without the sensor even hooked up. And the display was reading 32C. When I hooked the sensor up, the reading on the throttle line was .52v and the display said 50C which is way above room temperature reading that should be .22v. It seems to me that the throttle input line is being fed power by the controller when it should be reading at 0v if it is purely an input line. Is there a resistor somewhere? My controller is not a coaster brake controller to my knowledge. I looked and I did not see a resistor next to the throttle input. I hope all of that made sense. This is pretty perplexing.

I understand that the LM35 was connected in place of the throttle valve? (Meaning the throttle valve was disconnected and disassembled, and the LM35 was installed in the controller?)
Then the sensor legs should have:
1 – ground
2 – signal (at room temperature = approx. 250mV)
3 – power supply V=+5V