New "TSDZ2 Torque Sensor Central Motor"

m3lonbr3ad said:
casainho said:
Well, at least the cranks I got, 4 of them, were all market with L on backside. And I had to open a second pair of pedals because the right on me was not threading and so I had to put another left pedal on the right side. Or I did something wrong??

You cannot put a R pedal in an L crank arm. As bigwheel and others mentioned the threading is reversed on one side so that pedaling does not cause the pedal to loosen itself. So you need a R crank arm with an R pedal and L pedal with L crank arm.
c
I think I understand the need to have reverse thread on R crank as also a R pedal. Happens that the crancks from TSDZ2 are both market as being "L"... and I also think I had to use 2 left pedals and that is why a pedal got loose... I put the glue and let's see what happens.

m3lonbr3ad, can you please look at yours TSDZ2 crancks on the backside to see which leters are marked??
 
Casainho, it is unfortunate and also odd that you ended up with all L crank arms. It is clearly stamped on both TS and Bafang cranks on the backside by the pedal hole either L or R.

Hopefully the glue method works for you until you can source the correct R crank.

I might add that conventional bikes, as they have the chainring(s) integral to them, is certainly less confusing. I use Schlumph drives on My two hubmotor bikes which require the same cranks as we are dealing with here. Which is a good thing as they are expensive and these are not and I was able to use one to replace a right cranks that had wallowed in me.
 
Seems the motor I got doesn’t quite fit this frame. Any suggestions? Original B.B. is a 68. Anywhere sell shims that would fit on the gear side maybe?

25CD5778-F311-4546-B3DF-F2B35F4648A0.jpegE39EAB68-D830-4799-B5E2-E91D3616E2AD.jpeg

Ok I see it came with a couple longer screws. But still scared to tighten B.B. but with plastic hitting one side.
 
Switched to the long screws with double spacers did a little twisting and it looks less than 1mm off now.

3AFCE666-8E6C-414E-B2A2-351840FD3BDA.jpegC55FCA40-725E-4BB6-BBCF-583C12928D78.jpeg
 
wpenner said:
Switched to the long screws with double spacers did a little twisting and it looks less than 1mm off now.

You should try to twist some more I think. That drive side should be firmly up against the bottom bracket shell. If you can twist more, see if that protrusion can swing away from the drive side chainstay.
 
casainho said:
You will need 2 left pedals, while usually you buy a pair of pedal L + R but you will need 2 L pedals. Also, I think that glue for using on the pedal threads is a must, at least for the right side where you will using a left one and will tend to get loose.

That is funny, I used the 2 pedals that came with my bike. I guess Suntour SR cranks use 2 left pedals too?
 
wpenner said:
Seems the motor I got doesn’t quite fit this frame. Any suggestions? Original B.B. is a 68. Anywhere sell shims that would fit on the gear side maybe?

25CD5778-F311-4546-B3DF-F2B35F4648A0.jpegE39EAB68-D830-4799-B5E2-E91D3616E2AD.jpeg

Ok I see it came with a couple longer screws. But still scared to tighten B.B. but with plastic hitting one side.

Is that steel frame, and is it an expensive frame? If it was steel I might consider using a hammer or a very dull chisel or perhaps even a vise to put a small dent in just the right place on the frame. It may be risky though, depending on the frame and how much weight you will be carrying, and I certainly would not try it with anything but a steel frame. Perhaps someone with experience will chime in, I have never done anything like that to a bike frame. Good luck.



Otherwise put a large washer in there, but that is going to offset the pedals even more than normal. I checked my pedals today on my 73mm bb frame, i can put 3 fingers between the crank arm and frame on 1 side and 4 fingers on the other side. I don't really notice when pedaling though, I guess I got used to it.
 
I had to remove the torque arm so I could rotate the motor another couple mm to skip the protruding part. I guess I’ll throw some zip ties through the hole for now.

There is a small rubbing sound once per revolution of the pedals looks like the chainring hitting the plastic a tiny bit. Anyone else have this? Not sure if I should bother adding the 2.5mm shim. Might fix the sound as motor still barely touching chainstay and maybe bending it .5mm

[youtube]https://youtu.be/l1Uxt96qclE[/youtube]
 
Maybe it is hitting one of the bolt heads on the motor too. It is probably best to hold the tire off the ground and figure out exactly where it is hitting. You might just need to replace a bolt head with something lower profile. Or maybe the chainring is warped?

Screw the torque arm, I just used a heavy duty zip tie put on in an "x" pattern and the motor has not budged at all so far in 350 miles :)
 
wpenner said:
I had to remove the torque arm so I could rotate the motor another couple mm to skip the protruding part. I guess I’ll throw some zip ties through the hole for now.

There is a small rubbing sound once per revolution of the pedals looks like the chainring hitting the plastic a tiny bit. Anyone else have this? Not sure if I should bother adding the 2.5mm shim. Might fix the sound as motor still barely touching chainstay and maybe bending it .5mm

[youtube]https://youtu.be/l1Uxt96qclE[/youtube]

I havent watched it too closely but I am noticing a slight wobble in the chainring. Try snugging up all the bolts and nuts. If it still rubs, you might need to give it a tweak to get the clearance you need. By tweak I mean correct the warp with a little force. Worst case you might need to replace the chainring.
 
When I did the conversion to the Bafang offset ring I need some spacers and the only ones I could find were made of nylon that were close. A little too close and hard to set evenly. The chainring has a slight wobble and rubbed the plastic cover intermittently. U

Upon removing the chain ring to space it out I saw it had also slightly worn down the forward cap head bolt that holds it on. I got two thin washers and spaced it out with those and put a button head bolt in and no more rub.
 
wpenner said:
Not sure if I should bother adding the 2.5mm shim. Might fix the sound as motor still barely touching chainstay and maybe bending it .5mm

Add the shim. Bad things can happen if you are tightening the gear reduction housing against the chainstay and not the bb as it was designed. Aluminum bb spacers are available in many widths. Tongsheng has a video showing them denting the chainstay to remedy this. That is also a bad idea.
 
Coaster Brake version, Broken Torque sensor.

I recently received a TSDZ2 motor and a spare torque sensor assembly, for me to strip the motor and replace the sensor. It was assumed rather than known that the sensor was broken, so I fitted the motor to a spare mountain bike and tested it.
The display lit up, everything looked normal but on riding it there was no assistance at all.
On stripping the motor and removing the original torque sensor I found it was cracked in the tube that surrounds the pedal spindle.

Short story: Replacing the torque sensor fixed the problem and it worked well, however:

Long story: When I compared the broken sensor and the one I had been given to fit, they were not identical. The old one was shorter in the pedal spindle and in the tube that surrounds it, and it had two bearings on the left hand end of the pedal spindle. The new one had a longer spindle, longer tube and only room for one bearing.
The length from the mounting for the large gear to the circlip on the pedal spindle that retained the inner race of the lefthand bearing was the same on both.

If I had fitted the new assembly with only one bearing, it would have been able to slide sideways by the width of the bearing (7mm), so I machined up a spacer bush the same OD as the bearing and the same width. I slid this bush in before the bearing and tried to assemble. Weirdly, it seemed the bush was 1mm too wide, so I made another 6mm wide and everything assembled fine with virtually no side play on the spindle.
Looking at the photos of torque sensors for sale from various suppliers, they all looked the same as the newer (single bearing) one, so I assume the motor I had was an early one and the design has changed in later motors.

Another thing I noticed was that the big gear was solid, with no sprag clutch (one-way bearing) in the middle. This meant that the pedal spindle rotated with the chainring at all times, so if the motor was driving the chainring it was also driving the pedals.
I realized that this was almost certainly a "Coaster Brake" version of the motor, as pedalling backwards DID drive the chain backwards and would be able to activate a coaster (back-pedal) brake.
It also means that it would be very unwise to use a throttle with a coaster brake version of this motor!!
Test Riding the bike, I noticed that if I was pedalling hard, then stopped suddenly, I could feel the pedals "kick" forward briefly. Obviously it took a fraction of a second for the torque sensor / controller to sense that the pedal torque had dropped to zero and stop powering the motor. Not a major issue, but enough that I would prefer not to use the coaster brake version unless I actually needed that function.
 

Attachments

  • TS Tq Sensor 2 (2).jpg
    TS Tq Sensor 2 (2).jpg
    60.6 KB · Views: 2,428
  • TS Tq sensor 3 (2).jpg
    TS Tq sensor 3 (2).jpg
    55.2 KB · Views: 2,428
  • TS Tq Sensor 4 (2).jpg
    TS Tq Sensor 4 (2).jpg
    54.2 KB · Views: 2,428
Tools and techniques for changing the Torque Sensor

Sorry, for some reason I didn't take any photos while doing this torque sensor replacement apart from the pictures of the torque sensors themselves.. see my previous post.

The torque sensor assembly is in two parts: the rotating assembly and the stationary coil.

If you only need to replace the rotating assembly, you do not need to take the cover off the controller, but if replacing the stationary coil as well, you will need to remove the controller cover so you can separate the connector from the coil to the controller.

Remove both pedal cranks, remove the cover over the large gear, remove the circlip that retains the big gear, then pull the big gear off. (For details of how to do these steps, find the very helpful video by forum member JBALAT) https://www.youtube.com/watch?v=u-SY0rGMEMo
Once the big gear is off, remove and put aside any shim washer between the gear and the rotating coil of the sensor, then move to the left-hand side of the motor.
Remove the oil seal from the pedal spindle by sliding a bent O-ring pick or similar between the spindle and the inner lip of the seal and hooking it out. in my case the seal came out easily and without damage.
A long-nose, straight pair of external circlip pliers with small tips is needed to remove the circlip on the pedal spindle as it is buried fairly deeply in the housing behind the seal.
Remove the shim washers behind the circlip and keep for later. This is tricky , but if you turn the motor on its side they might drop out.

Now turn the motor on its side so the pedal spindle is vertical with the left-hand side down, and tap upwards on the end of the spindle to drive the torque sensor out upwards. This is because the stationary coil sits between the rotating coil and the housing, and is pushed outwards into contact with the rotating coil by three small springs. If you tap out the torque sensor with the motor horizontal, the stationary coil falls outward, the springs jump out and hide behind any available object and are devilishly hard to find.. (don't ask how I know that..). If the motor is on its side, the stationary coil is held down by gravity and the springs will remain in place until you remove them.
If you want to replace the stationary coil, separate the connector at the controller and carefully lift the coil out. Retain the springs in place with a little dab of tacky grease and lower the new coil into place, aligning the recesses with the springs. The wires are stiff enough that threading them back through the opening in the housing is fairly easy. Rejoin the connector halves at the controller.

If you want to replace the bearing(s) at the left of the pedal spindle, tap them out with a drift from the right-hand side.

Reassembly is simply the reverse of the disassembly. Use anti-seize or grease an anything you want to come apart easily next time, and clean the motor gear and big gear and replace the grease with a good quality grease (Mobil 28 and Lucas Red ' n ' tacky have been recommended on this forum).

The bearing(s) at the left-hand side of the spindle are 28mm OD, 15 ID 7 wide, type 6902RS, and the oil seal is the same size as the bearings.

Sorry for the long post, but hopefully it will help somebody one day (even if only to keep possession of those little springs!
Dave
 
After 2 days, I got it working and I finally can swap the TSDZ2 original LCD to KT LCD3. Many thanks to hurzhurz for improving the TSDZ2 motor controller original firmware.

KT LCD3 showing the following data from TSDZ2 motor:
- battery voltage
- motor power in watts
- bicycle wheel speed
- motor assist level
- brakes state
- battery state of charge

[youtube]75Na5RDLZAk[/youtube]

The motor power is clearly going to zero when it should not, this is clearly an issue that I need to improve. For the calcs, I am using the same filtered and stable voltage we can can on LCD. I guess the only source of issues is the current value sent by the controller... maybe I am doing something wrong somewhere...

After, I want to start calc the motor used Wh for the current trip.
 
Drum said:
Coaster Brake version, Broken Torque sensor.

Another thing I noticed was that the big gear was solid, with no sprag clutch (one-way bearing) in the middle. This meant that the pedal spindle rotated with the chainring at all times, so if the motor was driving the chainring it was also driving the pedals.
I realized that this was almost certainly a "Coaster Brake" version of the motor, as pedalling backwards DID drive the chain backwards and would be able to activate a coaster (back-pedal) brake.
It also means that it would be very unwise to use a throttle with a coaster brake version of this motor!!
Test Riding the bike, I noticed that if I was pedalling hard, than stopped suddenly, I could feel the pedals "kick" forward briefly. Obviously it took a fraction of a second for the torque sensor / controller to sense that the pedal torque had dropped to zero and stop powering the motor. Not a major issue, but enough that I would prefer not to use the coaster brake version unless I actually needed that function.

Thanks, Dave (Drum), for taking the time to write up your torque sensor replacement details. It was useful to me and I am sure will be useful to many more in the future. I have something to add to what you discovered about "coaster brake" versions of the motor. I am fairly sure that the 'kick' you feel after stopping pedaling is simply the delay in how quickly the motor stops after you stop pedaling. The coaster brake version of the motor does not have this kick. I have had them apart and the controller on the coaster brake version is marked with a 'bk' in the part number sticker. The 'bk' is also in the model number on the motor casing. So in addition to the one-way bearing difference you discovered, there is a different version of the controller that is probably programmed with a different time-of-stop delay. You can eliminate this problem by replacing the controller. Although, if you are using a throttle or the walk-assist feature you will lose those as the coaster brake version of the motor has them disabled at the factory for the reasons of safety you have already stated.

In the future, this time-of-stop delay should be a parameter in open source firmware to support coaster brake versions of the motor. Ideally it would have a coaster brake support mode that in addition to eliminating the delay, would also turn off throttle and walk-assist for safety reasons.
 
Thanks Rydon, great to have that extra info.

This forum is amazing.. a lot of knowledge spread among many people all round the world gets brought together in one place for the benefit of all of us.
 
Having performed that operation the first time and having the same issue as Drum it is good to forewarn everyone for sure. Also we have found the best way to put it back together is with the drive side facing up. Best if the motor is out but lay the bike on its side if doing it while on the bike.The springs and resulting tension while compressing can cause the inner ring to not stay centered and seat properly.
 
casainho said:
The motor power is clearly going to zero when it should not, this is clearly an issue that I need to improve. For the calcs, I am using the same filtered and stable voltage we can can on LCD. I guess the only source of issues is the current value sent by the controller... maybe I am doing something wrong somewhere...
After, I want to start calc the motor used Wh for the current trip.

Good work again !
I noticed something similar with a normal watt meter in line. I think you just need to smooth/average the amps over 1 or 2 seconds maybe??

Can the VLCD5 display all this info or not ?
 
casainho said:
After 2 days, I got it working and I finally can swap the TSDZ2 original LCD to KT LCD3. Many thanks to hurzhurz for improving the TSDZ2 motor controller original firmware.

KT LCD3 showing the following data from TSDZ2 motor:
- battery voltage
- motor power in watts
- bicycle wheel speed
- motor assist level
- brakes state
- battery state of charge

[youtube]75Na5RDLZAk[/youtube]

The motor power is clearly going to zero when it should not, this is clearly an issue that I need to improve. For the calcs, I am using the same filtered and stable voltage we can can on LCD. I guess the only source of issues is the current value sent by the controller... maybe I am doing something wrong somewhere...

After, I want to start calc the motor used Wh for the current trip.

Well done with hard work.

Can we have Cadence in the same field of Voltage? Can be selected in the Display? Thanks
 
jbalat said:
Good work again !
I noticed something similar with a normal watt meter in line. I think you just need to smooth/average the amps over 1 or 2 seconds maybe??

Can the VLCD5 display all this info or not ?
Good to know that real current is actually like that. Thanks.

No, VLCD5 just can show what is shows. It can't for instance, read the motor current that we are now sending from the the motor controler.
 
AZUR said:
Can we have Cadence in the same field of Voltage? Can be selected in the Display? Thanks
We can't right now because the motor controler is not sending that information. We wish to have that and I are looking at the original firmware to understand how can we improve it to send also pedal cadence.

I have a digital hook scale that I think I can use to measure force on the pedals, so I can calc how much torque is done on the pedals. After I need to use cadence, to get human ppwer on pedals: torqie * human power.
 
casainho said:
After 2 days, I got it working and I finally can swap the TSDZ2 original LCD to KT LCD3. Many thanks to hurzhurz for improving the TSDZ2 motor controller original firmware.

KT LCD3 showing the following data from TSDZ2 motor:
- battery voltage
- motor power in watts
- bicycle wheel speed
- motor assist level
- brakes state
- battery state of charge

[youtube]75Na5RDLZAk[/youtube]

The motor power is clearly going to zero when it should not, this is clearly an issue that I need to improve. For the calcs, I am using the same filtered and stable voltage we can can on LCD. I guess the only source of issues is the current value sent by the controller... maybe I am doing something wrong somewhere...

After, I want to start calc the motor used Wh for the current trip.

I have felt something with the TSDZ2 motor that may be related to the reason why the current intensity is zero for a few moments.

I have a TSDZ2 with 350W 36V.

I felt that when I was pedaling continuously, on a road with a slight slope of 1%, the bicycle was working with sobs. It was as if the TSDZ2 engine briefly stopped assisting the pedal.

In another ebike I tested, a Haibike SDuro Fullnine 5.0, with Yamaha PW-SE, 250W engine, I did not feel anything like it.

The pedaling assistance was permanent.

In TSDZ2 it is as if the engine is making, sometimes, a counter force.

I have a friend who has a Specialized turboLevo with a Brose engine and no brake sensors on the brake levers.

He says the engine has a double clutch system that when you stop pedaling, the engine is in "free wheel" as in bicycles hub.

Because of this, he says, the BROSE engine does not have a braking effect.

The impression I had was that when, at a certain stage of the pedaling, it was less strong (less torque), the TSDZ2 stopped working.

Perhaps the engine will consider a reduction in pedaling force (less torque) to be the same as a moment of braking, so stop working.

Perhaps, for this reason, the current is zero for a few moments.

This may be a way for the TSDZ2 to work because it does not use brake sensors in the brake levers.

The owner of a bicycle shop that has been selling several brands of ebikes for several years has been experimenting with my bike and felt the same. It was he who made me realize I had a problem.

It may also be a matter of torque sensor sensitivity.

If we put brake sensors on the brake levers, this may help to solve this problem and program the TSDZ2 in another way.

What I do not know is how they are programmed the other engines.

This is my first ebike and only few km of experience.

My written English is not very good. Sorry.
 
I got motor power working and I am very happy!! Anyone interested installing KT LCD3 for the TSDZ2? If so, I can write a simple guide for the installation.

The motor pulls 950W (18 amps) with my 48V charged!! With a 52V battery, would pull over 1000W.

It is clear on the video when I am breaking (symbol active on LCD) and stopped pedal, as battery voltage decreases when motor power increases and wheel speed also increases :)

[youtube]bNXXfhM9N10[/youtube]


casainho said:
I did some quick tests on an accentuated climb.
Assist level 1: max around 6 amps
Assist level 2: max around 12 amps
Assist level 3: max around 18.4 amps
Assist level 4: max around 18.4 amps
 
Back
Top