New "TSDZ2 Torque Sensor Central Motor"

Anyone tried using Shimano Steps FC-E6000 crank arms? I tried 175mm and when I tighten they are not moving freely. Original crank arms fit well. I have 73mm bottom bracket. Thanks.
 
Demion said:
Anyone tried using Shimano Steps FC-E6000 crank arms? I tried 175mm and when I tighten they are not moving freely. Original crank arms fit well. I have 73mm bottom bracket. Thanks.

Sorry I can't help, but I am looking to purchase some in 175mm. Could you please tell me where you purchased them?
 
famichiki said:
Sorry I can't help, but I am looking to purchase some in 175mm. Could you please tell me where you purchased them?

I ordered them from bike-discount or bike24 a while ago. I am from EU, shipping from Germany.
 
How fast do you have your motor broken? My motor broke down when I had driven 80km. I drove 38 km yesterday and today I had driven 40 km when the rattling noise was started. I turned my nose towards home and after ten kilometers the assist disappeared. The final result you can be seen in the pictures.


BluegearandAxlebroken.JPG

MotorAndAxle.JPG
 
dameri said:
How fast do you have your motor broken? My motor broke down when I had driven 80km. I drove 38 km yesterday and today I had driven 40 km when the rattling noise was started. I turned my nose towards home and after ten kilometers the assist disappeared. The final result you can be seen in the pictures.


BluegearandAxlebroken.JPG

MotorAndAxle.JPG
This seems the very first time someone mentions such issue of axle gear broken.
 
dameri said:
How fast do you have your motor broken? My motor broke down when I had driven 80km. I drove 38 km yesterday and today I had driven 40 km when the rattling noise was started. I turned my nose towards home and after ten kilometers the assist disappeared. The final result you can be seen in the pictures.


BluegearandAxlebroken.JPG

MotorAndAxle.JPG
Typically, this happens when a bearing is not 100% aligned with the axle. This can be caused by a faulty or badly installed bearing, a faulty axle or the casing that is not perfectly straight or well tightened.
In any case, it is a manufacturing fault that should be guaranteed.
I hope for you that you have bought it from a good reseller.

Don't just replace the axle, if the problem comes from the bearing or the housing, it will cause the new axle to break also
 
Finally we own the TSDZ2 torque sensor!!

Up to now we were reading the torque sensor in a wrong way and it could only measure up to 25 kgs. Also there is a big difference from the values read on left or right pedal. This are the output values of my torque sensor (that I bought 3 weeks ago):

image.png


I implemented the linearization of the torque sensor measures and it can now measure up to at least 102 kgs (which is my weight). I also implemented a linearization based on a table with values for left and right pedal and it works very well: before my right pedal was measuring the 102 kgs of my weight but left pedal would measure only 70 kgs, now both pedals measure the same values.

Before my display was showing to me a value of pedal human power of a bit more than 100 watts, no matter the force I did and now, I can go up 400 watts for short peaks. I think strong cyclists will love to see their real pedal human power on the display!!

And by the way, I am pretty sure there is no need to calibrate the sensor as I did before, no need to open the motor and adjust the hall sensor position of the torque sensor.

Calibration process

1. user put values on right pedal, starting at 5 kgs, in steps of 5 kgs up to about 30 kgs. Finally put itself on the pedal with his known weight value. The pair of weight and torque sensor ADC value should be registered and later user will need to decide for 5 points only, 1 at start, other at the last value and 3 values near the knee on the graph, as shown on the graphs.
2. Repeat for left pedal.
3. Put the pair values of weight and torque sensor ADC values, for each pedal, on the specific LCD configuration menu.

One important thing for all this work is that at system power up, the right pedal should be pointing to ground -- as there is the need to identify which pedal we are doing the force and there is no way to identify them, I decided to do like this, having the pedal at a known position at startup and after that everything will work well.

I have the code working on my bicycle and I can see on the LCD the weight value in kgs on the LCD. The tables for linearization are for now hardcoded on the firmware, I can't setup them yet on LCD. When we have the next version of firmware I can add this feature to it.
 
I recently bought a TSDZ2 from an Aliexpress seller, and supposedly it's a 48v 750W motor, and came with an XH18 display. The label on the motor says "48v 4000rpm" but doesn't specify an amperage. Some sources say that all the 48v motors are the same, and differ only in the firmware, but I haven't seen anything definitive. All the images I've seen online of the 48v motor labels either show "15A" or "18A" but my label doesn't specify either (perhaps this is new?).

If my motor is only safe to 720W I'd like to know so that I'll be more careful not to end up with an expensive brick as I continue to play around. I'm still within the Aliexpress dispute window, but I don't imagine that means much in practice.
 

Attachments

  • IMG_5578.jpg
    IMG_5578.jpg
    130.7 KB · Views: 1,537
stumpflumper said:
I recently bought a TSDZ2 from an Aliexpress seller, and supposedly it's a 48v 750W motor, and came with an XH18 display. The label on the motor says "48v 4000rpm" but doesn't specify an amperage.

I received my motor last week and it's the same, no amps on the sticker. I'm not sure whether or not you can view or change it through the XH18 settings as I don't have that. You may need to look at flashing the firmware so you can be sure it's set up how you like.
 
knutselmaaster said:
dameri said:
How fast do you have your motor broken? My motor broke down when I had driven 80km. I drove 38 km yesterday and today I had driven 40 km when the rattling noise was started. I turned my nose towards home and after ten kilometers the assist disappeared. The final result you can be seen in the pictures.


BluegearandAxlebroken.JPG

MotorAndAxle.JPG
Typically, this happens when a bearing is not 100% aligned with the axle. This can be caused by a faulty or badly installed bearing, a faulty axle or the casing that is not perfectly straight or well tightened.
In any case, it is a manufacturing fault that should be guaranteed.
I hope for you that you have bought it from a good reseller.

Don't just replace the axle, if the problem comes from the bearing or the housing, it will cause the new axle to break also

I think the warranty is over because I opened the case. Fortunately, I had a spare motor. I don't know if the cause is in the bearings or where. The case was well attached and the new motor went well into place. All I have to do now is drive and test. If that doesn't work, there's no choice but to buy a new TSDZ2.

I now have one TSDZ2 to spare and I want to have two TSDZ2 engines so one I can drive and I fix other, when the other breaks, driving won't stop.

Edit: I bought TSDZ2 from PSWPOWER
 
it may be an assembly mistake. I had the same problem with a 2017 engine but it was my fault. I wanted to grease the blue gear but the engine did not come out and either tapped lightly with the hammer. Unfortunately the engine shaft is not elastic but fragile and when I reassembled everything after 2 km it broke in the same way. If when they put it together they probably forced it to crack and then broke like in my case.
 

Attachments

  • motore rotto tsdz2.jpg
    motore rotto tsdz2.jpg
    33.7 KB · Views: 1,487
However the dimensions of the motor shaft are ridiculous, you can see in the picture the comparison with a pen. As you can see, I am an engine burn expert, the neighbor is the rotor of a bpm hub :D :D
 

Attachments

  • 201821616354_motor10.jpg
    201821616354_motor10.jpg
    42.8 KB · Views: 1,487
casainho said:
image.png


I implemented the linearization of the torque sensor measures and it can now measure up to at least 102 kgs (which is my weight). I also implemented a linearization based on a table with values for left and right pedal and it works very well: before my right pedal was measuring the 102 kgs of my weight but left pedal would measure only 70 kgs, now both pedals measure the same values.

Before my display was showing to me a value of pedal human power of a bit more than 100 watts, no matter the force I did and now, I can go up 400 watts for short peaks. I think strong cyclists will love to see their real pedal human power on the display!!

Hello Casainho,
I do not quite understand the last message and chart.. perhaps I missed a step somewhere.
We used to see from the last round of discussion that the rest value of the torque sensor is between 25 and 40 (the lower the better) and the upper value for linear variation was about 70, above which we can appreciate the knee of the bi-linear behavior.
Starting from a lower value allows you to have more resolution to read increasing forces before saturate the sensor. That's why, I understood, it's useful to calibrate the sensor with the skrewdriver to have low rest values.

See also my prev. post here with my values:
https://endless-sphere.com/forums/viewtopic.php?f=30&t=93818&p=1480176&hilit=luggage+scale#p1480176

Now I see from your chart that the readings range from 150 to 350.. ?
Could you please clarify what are these readings and why you state that the previous calibration (to bring the rest value down to a low value) is no longer required?

Moreover if you remember, in the past I was reading a rest value of 48 and after few weeks a value of 51.
If this is a general trend observed by many users we may conclude that the torque sensor suffers a sort of "aging", and this maybe can affect some of the algorithms..?

Thanks
 
andrea_104kg said:
However the dimensions of the motor shaft are ridiculous, you can see in the picture the comparison with a pen. As you can see, I am an engine burn expert, the neighbor is the rotor of a bpm hub :D :D
I am being pulling 1000 watts in bursts from my TSDZ2 and I had no problems with it.

That motor shaft may be small but on the other side, users like me prefer a small and light motor, everything adding weight for having a more powerful motor is unnecessary.
 
thineight said:
casainho said:
image.png


I implemented the linearization of the torque sensor measures and it can now measure up to at least 102 kgs (which is my weight). I also implemented a linearization based on a table with values for left and right pedal and it works very well: before my right pedal was measuring the 102 kgs of my weight but left pedal would measure only 70 kgs, now both pedals measure the same values.

Before my display was showing to me a value of pedal human power of a bit more than 100 watts, no matter the force I did and now, I can go up 400 watts for short peaks. I think strong cyclists will love to see their real pedal human power on the display!!

Hello Casainho,
I do not quite understand the last message and chart.. perhaps I missed a step somewhere.
We used to see from the last round of discussion that the rest value of the torque sensor is between 25 and 40 (the lower the better) and the upper value for linear variation was about 70, above which we can appreciate the knee of the bi-linear behavior.
Starting from a lower value allows you to have more resolution to read increasing forces before saturate the sensor. That's why, I understood, it's useful to calibrate the sensor with the skrewdriver to have low rest values.

See also my prev. post here with my values:
https://endless-sphere.com/forums/viewtopic.php?f=30&t=93818&p=1480176&hilit=luggage+scale#p1480176

Now I see from your chart that the readings range from 150 to 350.. ?
Could you please clarify what are these readings and why you state that the previous calibration (to bring the rest value down to a low value) is no longer required?

Moreover if you remember, in the past I was reading a rest value of 48 and after few weeks a value of 51.
If this is a general trend observed by many users we may conclude that the torque sensor suffers a sort of "aging", and this maybe can affect some of the algorithms..?

Thanks
I am really happy to discuss what I did learn about all this, sharing with others so we can improve forward.

Looking at that graphs, I think that even If I adjust the hall sensor position and the rest value lowers to something like 30 instead of the 46 (184/4 = 46), the other points will be just the same. So lowering the rest value means increasing the high sensitivity range but see that even if I didn't that, the torque sensor can still measure my weight which I think will be mostly the max weight I will put on the pedals. So, is it really need to do the other process that is hard and risky to damage the torque sensor? -- even if was needed, this second part I did would anyway be needed to do, so, if possible I think we should skip the first part because the reasons I did mention.

The values of ADC are now 4x times higher because I decided to use the max ADC resolution of 10 bits instead of previous 8 bits. This is a must because of the low sensitivity range. I think we could even try to go with like 12 or 14 bits, by doing oversampling which seems to be a popular technic but for now I think it works very well.

The change on the resting values, sometimes if you rotate the pedals backwards it will go back like 2 or 3 units. I think it is not an issue.

The math I did and implemented on the code considers the value starting at the rest value and counts the steps up to the next point, so, if it changes a bit every startup, there is no issue (at startup the rest value is recorded). Also, this units at rest value represents kind of little change on the final value, like each unit is 0.7 kgs, so 3 units of variation is like 2.1 kgs, that is a very low value/step that we can not control/feel the difference with our legs.
 
andrea_104kg said:
it may be an assembly mistake. I had the same problem with a 2017 engine but it was my fault. I wanted to grease the blue gear but the engine did not come out and either tapped lightly with the hammer. Unfortunately the engine shaft is not elastic but fragile and when I reassembled everything after 2 km it broke in the same way. If when they put it together they probably forced it to crack and then broke like in my case.

So if it is assembly mistake, then it happened in the factory. TSDZ2 was brand new.

I installed new motor inside and drove today over 50 kms and TSDZ2 was working very good.
 
andrea_104kg said:
However the dimensions of the motor shaft are ridiculous, you can see in the picture the comparison with a pen. As you can see, I am an engine burn expert, the neighbor is the rotor of a bpm hub :D :D

I agree, little bigger more durability. I don't care very much for weight, because it's electric bike. Exept when I carry my bike to my carage which is on the third floor (home).
 
wpenner said:
andrea_104kg said:
it's not a short circuit it's just too much heat. I have repeatedly seen the temperature sensor reach 105 degrees. I think it's a miracle if my engine is still able to turn. a short circuit does not burn the windings, the motor stops first. This is the reason that led me to radical changes shown in another post. I noticed studying the sensor that the worst situation is not in off-road but in long road climbs faced at maximum power. just 1km at 8% at max power to overheat the engine.

It is strange because I’ve done this ride over 100 times. Yes it is steep and I did not have my temp sensor configured yet. Guessing motor is around 100c+ as I do 2Ah 48v in 10 minutes, 4km.

Isn't 80c a little conservative? Is 80c the highest known safe value, worried my motor might cut out daily on my commute at that low of a value. BUT I don't want it to burn my new one out also, so willing to reduce power output to save it. Those with temperature sensors, what do you have yours set at?
in my experience (I burned 2 bpm hubs and a q128h hub) the burn is never instantaneous but the result of subsequent heating that first weakens the enamel and then burns it completely. Usually the enameled wire used for motors is in class h, it should withstand up to 200 degrees. So 80 degrees are just fine. The speech of the weakening of magnets is different, but it is not so easy. I rewrapped a burned bpm and found no magnets weakened. The burnt inlet wire is because the plastic sheath is less resistant to temperature. Usually only one of the 3 phases is burned because the heating is not uniform. I don't think (but I'm not sure) that the water can make the engine burn, if a short circuit is made, it happens before the winding and stops everything. But it can be a combination of factors, water on enamelled wire already weakened by temperature. In this case, an instantaneous short circuit is realized.
 
I have a digital thermometer connected to the motor, no connections to the control unit, I monitor the temperature manually. I noticed that the greatest increase in temperature occurs at high speeds uphill, even not too steep climbs, but faced at speed with long ratios. It also depends on whether the engine is already warm when we tackle the climb. With my 48v if the engine temperature is 70 degrees and I face a climb of 8% at maximum power, it takes less than one km to see on the display of the thermometer 104 degrees. Obviously it is a very dangerous temperature, also because it is detected outside the engine, inside it is certainly higher. If you then insist on facing climbs because you do not know the engine temperature, burning is safe. I recently made a rather radical change by putting copper fins on the engine and I noticed an incredible improvement. It is not only the presence of the fins but now the whole left part of the engine heats up, while before it remained barely warm and this removes so much heat from the engine. The heating phase is very long, on quiet routes the engine never manages to reach over 70 degrees. only with extreme paths it reaches about 85 degrees and goes down very quickly while with the original engine the descent of the tempreatura was very slow.
In the picture you can't see but I closed the holes with heat-resistant silicone
 

Attachments

  • alet5.jpg
    alet5.jpg
    52.5 KB · Views: 1,415
dameri said:
andrea_104kg said:
However the dimensions of the motor shaft are ridiculous, you can see in the picture the comparison with a pen. As you can see, I am an engine burn expert, the neighbor is the rotor of a bpm hub :D :D

I agree, little bigger more durability. I don't care very much for weight, because it's electric bike. Exept when I carry my bike to my carage which is on the third floor (home).

I also think that the engine is really undersized, the minimum possible. The bafang bpm have more solid engines.
Unfortunately we have to be content with this, which makes miracles so small it is
 
casainho said:
Finally we own the TSDZ2 torque sensor!!

Will doing the calibration process mean we don't need to keep our feet off the pedals at startup any more? Does it need to be re-calibrated before a different person rides the bike?
 
famichiki said:
casainho said:
Finally we own the TSDZ2 torque sensor!!

Will doing the calibration process mean we don't need to keep our feet off the pedals at startup any more? Does it need to be re-calibrated before a different person rides the bike?
No, it should work for everyone and you need to keep your feet out of pedals and the right pedal down in direction to the ground!!

You need to download this firmware version from here (based on 0.19.0, you must have 0.19.0 on LCD):
https://github.com/OpenSource-EBike-firmware/TSDZ2-Smart-EBike/archive/torque_sensor_calibration.zip

You need to put you sensor own values on file ebike_app.c and then build the firmware:

Code:
uint16_t ui16_torque_sensor_linearize_right[TORQUE_SENSOR_LINEARIZE_NR_POINTS * 2] =
{
  // ADC 10 bits step, steps_per_kg_x100
  0, 16,
  260, 16,
  284, 21,
  292, 63,
  340, 160,
};

uint16_t ui16_torque_sensor_linearize_left[TORQUE_SENSOR_LINEARIZE_NR_POINTS * 2] =
{
  // ADC 10 bits step, steps_per_kg_x100
  0, 19,
  292, 19,
  304, 42,
  308, 125,
  328, 360,
};

Table ui16_torque_sensor_linearize_left is for left pedal and the other for the right one.

The values on left column are the ADC steps that you will read on the LCD but you will need to multiply them by 4 to get the ones to put on the table. The last point is your weight and the first one at resting but you must put 0.
On the right column are the steps in kgs X 100, like the first 19 means that each ADC step is 0.19 kgs * 100 = 19, because each ADC step that I did read on LCD was about 0.8 kgs and so for the 4x more I had to divide to 4.

Also note that Pedal torque sensor, menu 9.3 on LCD3 now shows the weight on the pedals so you should check here how is your torque sensor measuring the weigh - please check before doing the calibration as my values may be near the ones for your sensor, who knows??
You check the ADC value on 9.2 and see the measured kgs in 9.3.
 
casainho said:
famichiki said:
Will doing the calibration process mean we don't need to keep our feet off the pedals at startup any more? Does it need to be re-calibrated before a different person rides the bike?
No, it should work for everyone and you need to keep your feet out of pedals and the right pedal down in direction to the ground!!

So every time we switch the bike on we still need to keep feet off pedals and now also ensure the right pedal is down in direction to the ground?

I was hoping maybe after the new calibration process that wouldn't matter any more. So we could just switch on with a foot on pedal and go.
 
famichiki said:
casainho said:
famichiki said:
Will doing the calibration process mean we don't need to keep our feet off the pedals at startup any more? Does it need to be re-calibrated before a different person rides the bike?
No, it should work for everyone and you need to keep your feet out of pedals and the right pedal down in direction to the ground!!

So every time we switch the bike on we still need to keep feet off pedals and now also ensure the right pedal is down in direction to the ground?

I was hoping maybe after the new calibration process that wouldn't matter any more. So we could just switch on with a foot on pedal and go.
That is not possible. Note that we are doing advanced features, improving the motor, that the original motor/firmware don't have, it misses hardware for this specific features and so there is this workarounds.

I am not sure users will prefere this or simple use the torque sensor not calibrated.
 
I am trying to put a TDSZ2 on my tandem,and I need a longer spindle on the non drive side (the side where I motor is. There is not enough clearance for the linking chain ring spider. I need a spindle that is longer on the motor side, by about 1/4" or so. See photos.

Does anyone have any suggestions?
 

Attachments

  • 19-07-27_5.19.29PM-2124x1195.jpg
    19-07-27_5.19.29PM-2124x1195.jpg
    138.3 KB · Views: 1,349
  • 19-07-27_5.19.55PM-2124x1195.jpg
    19-07-27_5.19.55PM-2124x1195.jpg
    158.3 KB · Views: 1,349
  • 19-07-27_5.20.17PM-2124x1195.jpg
    19-07-27_5.20.17PM-2124x1195.jpg
    155.4 KB · Views: 1,349
Back
Top