New "TSDZ2 Torque Sensor Central Motor"

[Hillhater]

The normal amount of clearances on the bolts together with the very small radius from the centre of the BB would allow quite a few degrees of rotation, if you're making a stabilizer arm then do allow those tolerances otherwise you will be still reliant on that big nut at the right torque.

For sure the bbracket locknut is the prime security,
That torque link is more - "belt and braces" type of a "security" back up. :wink:

 

[tomtom50]

Guys, I'm not sure what the problem is with the motor moving, if its fully located on the crank side of the BB and the non crank side plate is correctly installed ( yes you may well have to shim it with spacers to get the bolts correctly torqued ) and the large nut is correctly torqued,

Its hard with full suspension bikes because you are only relying on friction to hold it. If you do off road then the locking nut can come loose. Having a torque arm makes sense.

The stock motor attachment reacts torque with the torque arm. The "securing plate"

is not about transmitting torque. It gives a load path for up / down forces on the motor. Without the securing plate a force on the motor housing puts the casting is bending in a way likely to break it like you see in this photo
https://endless-sphere.com/forums/viewtopic.php?f=28&t=79788&start=2775#p1395963

Reacting the torque in friction is not what the design relies on. That is solid engineering practice. Friction is hard to estimate, surface contamination is a factor, torques are hard to ensure and even if a torque wrench is used the preload is not accurately know, etc.

But I can see why people with rear suspensions like the idea of using friction. They don't have a place to install the torque reacting plate. So I ran a few calculations to see how feasible it is. Assumptions:
0.2 coefficient of friction
Torque to react 80 N-m
Bottom bracket effective radius 19mm
No factor of safety
Two friction surfaces (both sides of the bottom bracket)

Pre-load you need from the M33.5 locknut: 10,526N (2366 lb).

To get there you need to apply 66N-m torque (49 ft-lb), but that is not an accurate calculation because differences in thread friction throw it off. I would at least double the torque applied and closely watch to see if it slips. Can the wrench that comes with the unit apply 135N-m (100 ft-lb)? Maybe. It is probably best to extend it. If you extend it and apply 100lbs at 1 foot you are close. Maybe maybe not. That is definitely more preload than an engineer would expect from that wrench. Would it hurt the TDSZ2? Would the preload hurt the TDSZ2? I can't tell for sure. There is an inside radius on the casting that would see more stress.

I like the serrations Bafang puts in their unit. Serrations work, how well they work requires close analysis. If I had rear suspension bike I would try this Nord-Lock serrated washer if you have enough threads showing. It is big enough. It is cheap. Maybe it would work if torqued highly, say 100 ft-lb.

 

[John and Cecil]

I agree that properly tightening the bb nut is going to be the most helpful in eliminating engine rotation on a fs bike. I tightened mine by feel but I suspect it is probably about 75 ft-lbs give or take. I guess it depends on your strength but I like very short wrenches for torquing, this way I don't over torque. My problem has never been under torquing nuts and bolts. I broke so many bolts as a kid that I just know how far I can go before it will break I always had a torque wrench but I only used it for auto and motorcycle cylinder heads, etc. My go-to torque wrench was always a craftsman 3/8" drive tool with a stubby 6" handle. If it was very low torque then I used a 1/4 drive and either one or two fingers.

If you don't have enough room for the serrated washer perhaps a star washer would work too as they are rather thin.

 

[John and Cecil]

duplicate

 

[AWD]

I mentioned this in a post awhile back that the last time I had my motor torn down I drilled a hole through the bottom of the casing and in to the casing that contains the brass gear, in my case at least.

I got a needle type fitting for a grease gun and put a high quality grease cartridge in it. I now apply a few small pumps once a week and everything is staying really quiet and smooth. Here in a bit I'm going to tear it down and see if the extra grease is doing any harm, but as there is no where for it to go really I don't want it to build up too much either and plan on clearing it out and keep doing this procedure over time.



Also going to look into a similar hole in the reduction casing to be able to add a little to that now and then too. The small hole in the bottom of the casing also acts as a drain hole but I have had quite a few of these apart and never seen any evidence of moisture in that are actually.

 

[casainho]

I mentioned this in a post awhile back that the last time I had my motor torn down I drilled a hole through the bottom of the casing and in to the casing that contains the brass gear, in my case at least.

I got a needle type fitting for a grease gun and put a high quality grease cartridge in it. I now apply a few small pumps once a week and everything is staying really quiet and smooth. Here in a bit I'm going to tear it down and see if the extra grease is doing any harm, but as there is no where for it to go really I don't want it to build up too much either and plan on clearing it out and keep doing this procedure over time.

unnamed.jpg

Also going to look into a similar hole in the reduction casing to be able to add a little to that now and then too. The small hole in the bottom of the casing also acts as a drain hole but I have had quite a few of these apart and never seen any evidence of moisture in that are actually.

Good idea.

 

[Kisazul]

Hi guys. Somebody tried using a 14s 52v battery (58.8 fully charged) on a standard firmware. I originally purchased a 48v 15a motor. Its parameters using st link is: 4E 77 00. I changed them according to the parameters specified here on the forum: 57 82 11. But the motor does not work correctly when I start it from the throttle trigger. The problem is like on this video https://youtu.be/x5Ijvyysy5A . Can anyone tell me what the problem is and how to solve it?

P.s. I also tried to set via st link the motor to 36 volts and use 10s 42v fully charged battery , it works fine, but has a small cadence.

 

[feketehegyi]

Hi guys. Somebody tried using a 14s 52v battery (58.8 fully charged) on a standard firmware. I originally purchased a 48v 15a motor. Its parameters using st link is: 4E 77 00. I changed them according to the parameters specified here on the forum: 57 82 11. But the motor does not work correctly when I start it from the throttle trigger. The problem is like on this video https://youtu.be/x5Ijvyysy5A . Can anyone tell me what the problem is and how to solve it?

P.s. I also tried to set via st link the motor to 36 volts and use 10s 42v fully charged battery , it works fine, but has a small cadence.

I've not tried to modify these values yet, but according to eyebyesickle's PDF here
https://endless-sphere.com/forums/viewtopic.php?f=28&t=79788&start=2650#p1393994
is not the "11" as amperage value too small?

 

[Kisazul]

I've not tried to modify these values yet, but according to eyebyesickle's PDF here
https://endless-sphere.com/forums/viewtopic.php?f=28&t=79788&start=2650#p1393994
is not the "11" as amperage value too small?

it seems that 11 hex = 17 dec. but in any case I tried to 11, 14, 17 amp - no changes. it seems as if voltage still high.

 

[KD5ZXG]

0x82h = 0.4V*((8*16)+2) = 52V

Tried setting your median a little higher +1.2V, 0x85h maybe??
No sense trying much higher than 58.8V, which would be 0x93h.
Your original number 0x77h probably equated to only 47.6V.

I dunno what those mean in terms of real high and low cutoff.
Especially since a separate byte for low cutoff exists, but isn't
clear from anecdote wether that other ever had any effect.

I've got a 36V motor (wound 36V, maybe not programmed such)
that Eyebycicle setup per his preferences I don't know exactly.
My motor ain't mounted, BMS unfinished, cells not yet welded.
But there's an 80V 19A Agilent supply here at my bench, I've no
excuse not to have tested the soft Voltage cutoffs yet.

At work, I've every other programming dongle known to man,
but no STLink. Swear I've seen one here before, but small
desk drawer items may come and go as people turnover.

 

[jbalat]

I implemented a new configuration menu where now user can:
1. define on LCD the battery numbers of cells (used only to calculate battery state of charge to indicate on LCD battery bars)
2. define on LCD the battery low cut-off voltage

Are we up to version 6 now
Cant see it on github ?

 

[Hillhater]

Hi guys. Somebody tried using a 14s 52v battery (58.8 fully charged) on a standard firmware. I originally purchased a 48v 15a motor. Its parameters using st link is: 4E 77 00. I changed them according to the parameters specified here on the forum: 57 82 11. But the motor does not work correctly when I start it from the throttle trigger. The problem is like on this video https://youtu.be/x5Ijvyysy5A . Can anyone tell me what the problem is and how to solve it?

The motor in the video, is responding to a battery of too high voltage...IE , above HVC setting.
Try with a 48v pack to prove the problem.
Then you will have to set the HVC point to suit your pack.. 60v maybe ?

 

[Kisazul]

Hi guys. Somebody tried using a 14s 52v battery (58.8 fully charged) on a standard firmware. I originally purchased a 48v 15a motor. Its parameters using st link is: 4E 77 00. I changed them according to the parameters specified here on the forum: 57 82 11. But the motor does not work correctly when I start it from the throttle trigger. The problem is like on this video https://youtu.be/x5Ijvyysy5A . Can anyone tell me what the problem is and how to solve it?

The motor in the video, is responding to a battery of too high voltage...IE , above HVC setting.
Try with a 48v pack to prove the problem.
Then you will have to set the HVC point to suit your pack.. 60v maybe ?

today I will try to discharge the battery below 54.6 volts (for the HVC 48 volt battery 13 * 4.2 fully charged). If it works, it means that the setting on the DATA MEMORY tab does not work correctly as indicated in the instruction for the 52v battery (0x82h = 130hex => 130 / 2.5 = 52)

 

[julienh]

hi,

I've mount a tsdz2 36V 350W on my old mountain bike. At this day i've ride about 600km (road and track in Alpes near Annecy, FR)
It's my first ebike experience and i'am really bluffed.

I'am tring to test my battery limits : it's an home made 600Wh (10s5p) with Sanyo 18650GA and bluetooth smart BMS.
Usally when the battery voltage decrease the power tend to decrease too --> OK why not.

But yesterday at the end of my ride my motor surprised me : when the cell voltage was about 3.2V (32V for the pack) the assistance seemed to be nul : in fact the batterry supllied just less than 2A.

At this low voltage my power pack should be able to deliver more power/amps until it reach the cutoff voltage (2.5V for the datasheet but i set it to 2.8V on the bms)

Is it a classical behavior ? Do you have a such experience when you battery reach his end capacity ?

 

[casainho]

hi,

I've mount a tsdz2 36V 350W on my old mountain bike. At this day i've ride about 600km (road and track in Alpes near Annecy, FR)
It's my first ebike experience and i'am really bluffed.

I'am tring to test my battery limits : it's an home made 600Wh (10s5p) with Sanyo 18650GA and bluetooth smart BMS.
Usally when the battery voltage decrease the power tend to decrease too --> OK why not.

But yesterday at the end of my ride my motor surprised me : when the cell voltage was about 3.2V (32V for the pack) the assistance seemed to be nul : in fact the batterry supllied just less than 2A.

At this low voltage my power pack should be able to deliver more power/amps until it reach the cutoff voltage (2.5V for the datasheet but i set it to 2.8V on the bms)

Is it a classical behavior ? Do you have a such experience when you battery reach his end capacity ?

Install our OpenSource firmwares for TSDZ2 + LCD3 to unlock custom battery low voltage cut-off as also automatic operation of motor with constant power while battery discharges. You also unlock max TSDZ2 battery voltage (52V) and current 18 amps.

 

[Kisazul]

As I expected, after the battery goes down to ~ 56V, the motor starts to work. I used 0x82h. Even raising to 0x93h the situation does not change. In addition, I tried to change the current (11, 12,14, 17 and other values), but on my G.T. Power Rc I see constantly the maximum current strength of 14-15 amps. And the power of the motor is about 750 watts. It can be concluded that the current and operating voltage settings (LVC and HVC) are located elsewhere, or these settings are programmed for the throttle. Perhaps the maximum values ​​when using the pedals will change?

 

[Kisazul]

Guys, maybe someone who has a 52 volt version motor will be able to share original firmware for testing?

 

[casainho]

I implemented a new configuration menu where now user can:
1. define on LCD the battery numbers of cells (used only to calculate battery state of charge to indicate on LCD battery bars)
2. define on LCD the battery low cut-off voltage

Are we up to version 6 now
Cant see it on github ?

I uploaded now the new version of firmware, it is version 0.2 and this are the files:
- TSDZ2_flexible_opensource_firmware-v0.2.hex
- KT_LCD3_flexible_opensource_firmware-v0.2.hex

Get them here: https://github.com/OpenSource-EBike-firmware/TongSheng_TSDZ2_motor_controller_firmware/tree/master/TSDZ2_flexible_opensource_and_KT_LCD3_firmwares

 

[tomtom50]

Also, I need to confirm, but I think my doc needs editing.... That was an old TS value chart... I looked at my saved profiles, and I use '83' for 52v, not 82. I think you need 83 or it won't run at full 58.8... waiting to confirm and I'll edit it

Thanks, eyebyesickle. I followed your instructions and it saved. I changed 78 & 14 to 83 & 17.

Great, let me know if there are any problems, some versions need the program memory flashed to a different file also :/

Which I can provide as needed and if it comes up alot I'll edit the doc. 82 works fine for 52v though it seems, it's was the program memory problem I describe above in this case...

I was so confident it would work. I connected and saved the values, battery is at 58.5V, and no assist. I looked at
https://opensourceebikefirmware.bitbucket.io/kunteng_lcd3/TSDZ2_configure_battery_voltage_and_motor_current.html tried 82 & 83, tried setting the min V as well

No assist. How to flash the program memory to a different file?
Thanks

 

[feketehegyi]

I'm controlling my TSDZ2 over Bluetooth from my Android phone.
[youtube]X1xmpxRbYmc[/youtube]

 

[jbalat]

Re 52v motor eeprom this should get you to 58.8 or maybe 60 HVC
See link here
https://github.com/OpenSource-EBike-firmware/TongSheng_TSDZ2_motor_controller_firmware/tree/master/TSDZ2_original_improved_and_KT_LCD3_firmwares

Also there is a bin to hex tool here but not sure how to use it
https://github.com/OpenSource-EBike-firmware/TongSheng_TSDZ2_motor_controller_firmware/tree/master/TSDZ2_original_improved_and_KT_LCD3_firmwares/tools

Once in hex format you’ll u can use the windows tool stlink visual programmer, browse to the Memory tab and flash that to the current tab.

Also just noticed the memory address from above is 00004000 however from Casainho the address is zero as per pic below, I assume this gets flashed at the 4000 memory address ?

 

[eyebyesickle]

Also, I need to confirm, but I think my doc needs editing.... That was an old TS value chart... I looked at my saved profiles, and I use '83' for 52v, not 82. I think you need 83 or it won't run at full 58.8... waiting to confirm and I'll edit it

Thanks, eyebyesickle. I followed your instructions and it saved. I changed 78 & 14 to 83 & 17.

Great, let me know if there are any problems, some versions need the program memory flashed to a different file also :/

Which I can provide as needed and if it comes up alot I'll edit the doc. 82 works fine for 52v though it seems, it's was the program memory problem I describe above in this case...

I was so confident it would work. I connected and saved the values, battery is at 58.5V, and no assist. I looked at
https://opensourceebikefirmware.bitbucket.io/kunteng_lcd3/TSDZ2_configure_battery_voltage_and_motor_current.html tried 82 & 83, tried setting the min V as well
52V Settings.png
No assist. How to flash the program memory to a different file?
Thanks

I'll post the link tomorrow... It seems now the program memory has to be changed too, so I think the guide is going to change to just editing motor amperage, but having to upload the voltage profile....

 

[tomtom50]

Looking forward to it. I look at the posts and I don't know where to start. I found a file called TSDZ2_52V_eeprom.bin, but I have no idea how to use it. Since I don't know how to play with microprocessors I'm afraid I'll make it worse.

 

[tomtom50]

Guys, maybe someone who has a 52 volt version motor will be able to share original firmware for testing?

I think we are having the same problem. Editing the nominal voltage doesn't seem to affect the max voltage setting. The battery voltage has to drop to 55V before the assists works.

https://github.com/OpenSource-EBike-firmware/TongSheng_TSDZ2_motor_controller_firmware/tree/master/TSDZ2_original_improved_and_KT_LCD3_firmwares

lists TSDZ2_52V_eeprom.bin, which may be the firmware you are looking for?

 

[tomtom50]

If the controller is off I can use ST Visual Programmer and read and edit the DATA MEMORY tab values. But if the controller is on I am told I don't have the right drivers and I cannot connect (Windows 7)

Is this normal? I have read that the firmware is flashed with the controller on, in this case I would need to somehow tweak the drivers.