TSDZ2 Suddenly not giving any pedal assistance

Joined
May 1, 2019
Messages
16
I purchased a TSDZ2 48v 500w with a VLCD5 in June, during which time it worked flawlessly. Suddenly on my way home from work the bike stopped giving assistance, on the rest of the way it would give assistance seemingly at random. Now the bike doesnt seem to give any assistance at all for the whole duration of any trip. I've tried troubleshooting a bit, and these are my findings:

  • Walk assist seems to work as normal
  • Motor makes some sound when pedaling, but does not give any assistance (sound intensity seems to be the same on all assistance levels)
  • Battery has been fully charged, and VLCD5 display shows the correct charge status
  • Ive tried to get the torque sensor readings from the display, but the "hold i + power for 10 seconds" method does not work on my display
  • [EDIT] Speed sensor is reading correct speed

I welcome any suggestion for further troubleshooting or knowledge of this problem with open arms :idea:
 
I don't have any power assist (at least not that I can notice), although I hear the motor making some noise. I dont have any throttle to test the motor output without pedaling, but the walk-assist seems to work as normal.

Torque sensors being iffy makes sense... Just gotta find a way to read the sensors :lol:
 
Try the alternate instructions:

P.S. Every post mentioning the hidden menu says hold info+power 10 seconds, but that does not work on mine. Mine requires first to press info once, then hold info+power 3 seconds, then info 3 times. I figured this out by dumb luck after being very frustrated using the wrong instructions. Just wanted to point that out in case it helps anyone else in the future.
 
Schmuel The Cruel said:
I don't have any power assist (at least not that I can notice), although I hear the motor making some noise. I dont have any throttle to test the motor output without pedaling, but the walk-assist seems to work as normal.

Torque sensors being iffy makes sense... Just gotta find a way to read the sensors :lol:

If the motor is making noise, but the system isn't driving the chain, it almost certainly means some part of the system is detecting your control inputs and spinning it (or trying to), but isn't able to drive the output sprocket and chain. That probably means the clutch or gears has failed.

If the walk mode does drive the output sprocket and chain correctly, and the system moves forward (with you on the bike) just like it should, then it can't be the gears or clutch.

If walk mode works without you on the bike but not with you on it, it could still be the clutch slipping once enough torque is applied to it.
 
Schmuel The Cruel said:
.....
I welcome any suggestion for further troubleshooting or knowledge of this problem with open arms :idea:
De Tsdz2 controller is very sensitive for a bouncing reedcontact of your speedsensor.
This happends if the magnet is too close.
Try about 10mm distance.
 
Elinx said:
Schmuel The Cruel said:
.....
I welcome any suggestion for further troubleshooting or knowledge of this problem with open arms :idea:
De Tsdz2 controller is very sensitive for a bouncing reedcontact of your speedsensor.
This happends if the magnet is too close.
Try about 10mm distance.

I forgot to mention that the speed sensor seems to read the correct speed. Would it still be possible for the speed sensor to be the culprit, in that case? I'll try to readjust as you suggest to see if that solves the issue :)
 
amberwolf said:
Schmuel The Cruel said:
I don't have any power assist (at least not that I can notice), although I hear the motor making some noise. I dont have any throttle to test the motor output without pedaling, but the walk-assist seems to work as normal.

Torque sensors being iffy makes sense... Just gotta find a way to read the sensors :lol:

If the motor is making noise, but the system isn't driving the chain, it almost certainly means some part of the system is detecting your control inputs and spinning it (or trying to), but isn't able to drive the output sprocket and chain. That probably means the clutch or gears has failed.

If the walk mode does drive the output sprocket and chain correctly, and the system moves forward (with you on the bike) just like it should, then it can't be the gears or clutch.

If walk mode works without you on the bike but not with you on it, it could still be the clutch slipping once enough torque is applied to it.

Thanks for the suggestions! Are you referring to the blue nylon gear, or another gear in the motor?

If the clutch was slipping, would I still be able to pedal normally without assitance?

Walk mode works with me on the bike. I'll see if I have time to open up the motor this weekend to see if there is any visible and obvious damage anywhere. Will update the thread once I find (or dont find) the issue.
 
Schmuel The Cruel said:
.....
I forgot to mention that the speed sensor seems to read the correct speed. Would it still be possible for the speed sensor to be the culprit, ...
Yes, bouncing meaning a very lot of pulses in a very short time. It depends how the calculation for the speed is done.
Rotate the magnet 10mm away from sensor and see what happends.
 
pxl666 said:
after flash torque sensor can be read on all displays
You have a vlcd5. With that display and original stock firmware it is also possible to check the torquesensor.
You must acces the hidden menu and check TE and TE1 value.
Without load on the pedal these values should be equal.
With your weight on the pedal you should see a higher value with TE1

Acces to hidden menu could be a different with vlcd5:
See this Vlcd5 manual
Or that Vlcd5 manual
 
Try the alternate instructions:
Sorry for not following this up sooner. I'm back after being away from my bike for a very long time.

I have spent the day troubleshooting the motor. The alternate instructions for accessing the torque sensor readings worked.

There is something strange going on, because sometimes the T and T1 shows 00000. Sometimes T and T1 shows 94, and applying pressure to the pedals steadily increase T1 up to about 130, as expected. However, when testing I noticed that the T and T1 would show normal readings and when applying pressure it would also provide assistance for a few seconds before both T and T1 showed 00000 again.

But most of the time the sensors just showed 00000....
 
You have a vlcd5. With that display and original stock firmware it is also possible to check the torquesensor.
You must acces the hidden menu and check TE and TE1 value.
Without load on the pedal these values should be equal.
With your weight on the pedal you should see a higher value with TE1

Acces to hidden menu could be a different with vlcd5:
See this Vlcd5 manual
Or that Vlcd5 manual
Thanks for the instructions, I managed to get access to the torque sensor readings using some alternate method. Please see my post above for the results.

Sorry for the late reply, I have been away for some time.
 
Last edited:
...I managed to get access to the torque sensor readings ....
94-130 are good values, but 0000 means there is no torquesignal.
The reason could be a bad connection, no Voltage or broken torquesensor.
But imho that last two options aren't the reason, because it will never work
It could be the one or two wires of the torquesensor from which the isolation is broken at the wirefeed.
 

Attachments

  • torquewires.jpg
    torquewires.jpg
    129.9 KB · Views: 3
94-130 are good values, but 0000 means there is no torquesignal.
The reason could be a bad connection, no Voltage or broken torquesensor.
But imho that last two options aren't the reason, because it will never work
It could be the one or two wires of the torquesensor from which the isolation is broken at the wirefeed.
Right... So basically I should open up the motor and check the torque sensor for damage. Do you think soldering the wires is a feasible solution if their connection is broken?
 
..Do you think soldering the wires is a feasible solution if their connection is broken?
If that is the reason, it is also important how and why they are damaged, because if you leave the real cause untouched, the problem comes back after soldering.
 
If that is the reason, it is also important how and why they are damaged, because if you leave the real cause untouched, the problem comes back after soldering.
Thank you for your help, I really appreciate it. I will try to have a look at the torque sensor later this week. Will keep this thread updated.
 
Last edited:
If that is the reason, it is also important how and why they are damaged, because if you leave the real cause untouched, the problem comes back after soldering.
Alright, so I've finally found time to disassemble the motor. I retrieved the torque sensor and can't really see any visible damage it or bad wiring.

However, what I did find was that the "ring" on the main shaft (?) was warped. Come to think of it I have noticed some creaking when pedaling, but didn't think much of it at the time. See pictures attached. Do you have a suggestion for what this indicates?
 

Attachments

  • 20230605_221804.jpg
    20230605_221804.jpg
    1.1 MB · Views: 4
  • 20230605_201258.jpg
    20230605_201258.jpg
    991.6 KB · Views: 3
  • 20230605_201057.jpg
    20230605_201057.jpg
    1.1 MB · Views: 5
  • 20230605_201107.jpg
    20230605_201107.jpg
    1 MB · Views: 4
  • 20230605_201048.jpg
    20230605_201048.jpg
    1 MB · Views: 5
  • 20230605_194237.jpg
    20230605_194237.jpg
    1.3 MB · Views: 6
As long as you have the motor open, Look inside the blue gear. There are struts inside (needle ratchet gear) which are delicate and wear out. On my motor, I wore out 2. Also, the screw holding the white wire was stripped so I had intermittent then no power. A larger screw solved that.
I gave up and bought a newer model motor that doesn't have the blue gear. Upon finishing installation, the screen was black. No battery. The black wire from the motor to the battery connection was slipped out of the connector. Slipped it back in, e6000 glue and 200 miles so far so good.
 
.... can't really see any visible damage it or bad wiring.

However, what I did find was that the "ring" on the main shaft (?) was warped. ....
It looks that you don't have a circlip in that spindlegroove, but only a "warped"ring.

If possible, have you measured the complete torquesensor wiring?
Meaning the hallsensor wiring (red/black) and both coils for wireless transmission of the torque signal to the controller.
 
It looks that you don't have a circlip in that spindlegroove, but only a "warped"ring.

If possible, have you measured the complete torquesensor wiring?
Meaning the hallsensor wiring (red/black) and both coils for wireless transmission of the torque signal to the controller.
I do have the circlip, I just removed it when inspecting the warped ring.

Sadly I dont have easy access to a multimeter, so I dont think I'm able to measure the torque sensor wiring. But when using the bike it sometimes gives assistance and torque sensor readings at the beginning of the ride before abruptly disconnecting, so there is some signal going through.

Do you think it could be an OS issue and that switching to OSF could help?

EDIT: I also saw your old reply referencing to this thread, Elinx. Maybe the issue is caused by a faulty stationary coil? If the signal is unreliable maybe it can be attributed to the coils not transmitting the signal properly.
 
Last edited:
.....

Sadly I dont have easy access to a multimeter, ...

Do you think it could be an OS issue and that switching to OSF could help?
..
Because your readings dissapear, but not always, I think it would be something with the wiring/coils/connector.
If that is the cause, OSF will not help to solve that.
It is a pity you don't have a multimeter, so you must trust visible obeservations.
 
Because your readings dissapear, but not always, I think it would be something with the wiring/coils/connector.
If that is the cause, OSF will not help to solve that.
It is a pity you don't have a multimeter, so you must trust visible obeservations.
I'll check with my neighbor if he has one I can borrow. Which points of the sensor to your suggest I try to measure?

I assume I should also try doing it with the torque sensor attached to the body so that the coils can transfer signal as well?
 
Back
Top