TSDZ2 motor continues to run when pedals stopped

DCM1

1 mW
Joined
Oct 2, 2020
Messages
10
Location
Syracuse, Utah
I have a TSDZ2 with an older version of the Open Source Firmware that has been working fine for a couple years. Recently the motor has not been disengaging when I stop pedaling, continuing the same torque as if I was still pedaling. Not repeatable; sometimes it does it, most of the time it works normally. Brake cutoffs are working fine; it doesn't have a shifting cutoff. It does have a hand throttle, but flipping it doesn't seem to have any effect. Next time I'm out I'll unplug it just to see if that has any impact, but considering it's an intermittent problem, that may not help very much.
 
DCM1 said:
... older version of the Open Source Firmware .... Recently the motor has not been disengaging when I stop pedaling, continuing the same torque as if I was still pedaling. ... Next time I'm out I'll unplug it just to see if that has any impact, ....
Unfortunately, this is a persistent problem that has been present with OSF for stock displays from the start.
mbrusa has still not been able to determine the real cause of it, although he has fixed errors that could possibly cause this.
For some, re-flashing the optionbyte only was enough to fix it. It's not clear why this worked for them.

With the old OSF versions it was temporarily solved by turning off the engine or disconnecting the battery.
With the later versions this could sometimes lead to burnt mosfets on the controller.

With the last OSF version v.20.1C.2, mbrusa has put some security checks in the code that will shut down the engine if such a check of sensorvalues isn't good.
Because ghost pedaling isn't reproducible and will only show when you don't expect it once in a time, it isn't completely sure if this cross-checking will prevent ghost pedaling forever.

"E05 – ERROR_MOTOR_CHECK(E03 flashing per XH18)
Inconsistent data in cross-checking of motor parameters. Turn the display off and on again.
"
 
Thanks for the quick reply! I'm on the original (for me) OSF V1.1.0 but will be traveling to the shop where I bought for an update to current version.
 
DCM1 said:
.. I'm on the original (for me) OSF V1.1.0 but will be traveling to the shop where I bought for an update to current version.
v.1.1.0 is not for stock display, but developed for 850C/860C, SW102 by Casainho.
Casainho has stopped developing that version but mbrusa started with developing for these displays too.

I never read about problems with ghost pedaling with the versions for other displays.
That is why I thought you had a stock Tsdz2 display. (XH18, Vlcd5 or Vlcd6)
Imho mbrusa has no cross check included in his latest version v20.1C.4 for other displays.
So in that case there could be another reason for this behaviour.
Maybe some drift of the torquevalues can cause this.
 
Elinx said:
DCM1 said:
.. I'm on the original (for me) OSF V1.1.0 but will be traveling to the shop where I bought for an update to current version.
v.1.1.0 is not for stock display, but developed for 850C/860C, SW102 by Casainho.
Casainho has stopped developing that version but mbrusa started with developing for these displays too.

I never read about problems with ghost pedaling with the versions for other displays.
That is why I thought you had a stock Tsdz2 display. (XH18, Vlcd5 or Vlcd6)
Imho mbrusa has no cross check included in his latest version v20.1C.4 for other displays.
So in that case there could be another reason for this behaviour.
Maybe some drift of the torquevalues can cause this.
Yeah, I didn't mention that I have an 860C display. I was not aware that the display factored into this, as I was suspecting some malfunction in the torque sensors. Then again, I'm a wrench-bender, not a spark-chaser... :roll: :shock: 8)
 
DCM1 said:
..... an 860C display. I was not aware that the display factored into this, as I was suspecting some malfunction in the torque sensors. ...
It isn't the display, but the different developments of the OSF for the different display's.
If it is possible to read the torquevalue while driving, can you read the torque value too if ghostpedaling is there?
If this is a higher value compared with the start value and no feet on the pedals, it could be a reason for ghostpedaling.
 
Elinx said:
DCM1 said:
..... an 860C display. I was not aware that the display factored into this, as I was suspecting some malfunction in the torque sensors. ...
It isn't the display, but the different developments of the OSF for the different display's.
If it is possible to read the torquevalue while driving, can you read the torque value too if ghostpedaling is there?
If this is a higher value compared with the start value and no feet on the pedals, it could be a reason for ghostpedaling.
Well, I went out this morning for about 6 miles. As Murphy would have it, no "ghost-pedaling". Motor stopped within about 1/2 second of pedal stop. I'll get the latest flash the next time I'm down in the city (US$25 in gas, round trip these days :shock: ) so it will be a multi-tasking drive and might be a bit before I get there.
 
Well, have ridden for two months on the new OSF and have had no instance of ghost-pedaling.
I have noticed that the new OSF boots up much quicker than the original; everything seems, from my beginner's view, to be working quite well. :thumb:
Thanks everyone for comments and suggestions, and thanks to the programmers and testers for their hard work! :thumb: :thumb:
 
DCM1 said:
Well, have ridden for two months on the new OSF and have had no instance of ghost-pedaling.
I have noticed that the new OSF boots up much quicker than the original;....
:thumb:
Curious if new OSF is still v.1.1.0, or latest of mbrusa?
 
Elinx said:
DCM1 said:
Well, have ridden for two months on the new OSF and have had no instance of ghost-pedaling.
I have noticed that the new OSF boots up much quicker than the original;....
:thumb:
Curious if new OSF is still v.1.1.0, or latest of mbrusa?
Had to go out to the bike to see, then had to figure out where to find it as the bootup screen blanks out too fast. Figured out that it holds on that screen if I hold the power button down. :eek:

Anyway, it's 860C firmware 0.20.1C-4 and TSDZ2 firmware 0.21.4

Hope that helps.
 
Elinx
Off-topic, but I couldn't find anything in the manual or using the forum search function: Does the display clock not maintain current time when powered down? It's useless if it constantly needs reset to current time. And I'm not going to leave the battery turned on 24/7.
Thanks
 
DCM1 said:
.... it's 860C firmware 0.20.1C-4 and TSDZ2 firmware 0.21.4...
That is the build from mbrusa.
You can find the right topic here

I did search for your "clock" problem too, but could only find such a problem with 850C.

"- Clock field settings is not persistent after restart, when I set Battery Voltage after restart the Time field is preset"
Answer from mbrusa
"850C doesn't have an rtc and I haven't changed anything about it."
 
Elinx said:
DCM1 said:
.... it's 860C firmware 0.20.1C-4 and TSDZ2 firmware 0.21.4...
That is the build from mbrusa.
You can find the right topic here

I did search for your "clock" problem too, but could only find such a problem with 850C.

"- Clock field settings is not persistent after restart, when I set Battery Voltage after restart the Time field is preset"
Answer from mbrusa
"850C doesn't have an rtc and I haven't changed anything about it."
Thanks! I appreciate your time!! As I have a RTC (watch) on my wrist, I've just changed that field to show battery info.
 
Back
Top