New "TSDZ2 Torque Sensor Central Motor"

[casainho]

See page 11 of this manual https://drive.google.com/open?id=1fr27O ... mRNKkLJ-HU

Error code E02 is hall error or motor short circuit

Maybe a false hall sensor? This makes sense on lower speed (walk) and higher speed (speed and turbo).

EndlessCadence, Try to reconnect to the board the hall sensors cable! Once I had to do that because my motor was failing.

 

[EndlessCadence]

Thanks guys. Where is this board located? I didn't open up the casing yet. When the hall sensor is failing, how can it be that it works perfect in the first two assist levels? Also, overvolting from 36V to 48V isn't causing this issue is it? I have programmed the data memory accordingly.

 

[jbalat]

Endless caidence...
I was thinking the same thing. Your BMS must be cutting out the battery.
How many amps can the battery deliver ? What is the rating of the BMS ?
Although sounds like you may have a bad connection ?

For the motor vibration issue ??
I use a thick piece of rubber (car heater hose) between the motor and my frame. I rotate the motor up so it is firm on the rubber and tighten it up. I use a plastic washer between the locking nut and the bottom bracket to stop the nut coming undone.

 

[playvltk03]

I bought this motor 48v 500w couple of week ago and it is a great disappointment. I was excited at first seeing an affordable product with great features. It turn out far worse than a bbs02 36v 500w. Especially with the climbing hill or even doing simply riding. The funniest thing is the motor cut off at 43-42v with my 48v battery, hence, I can only use half of my battery capacity 13ah. I changed 2 battery, both at 48v 13ah and none of them working as intended. I stuck yesterday at the middle of a hill myself yesterday when my battery is still at 43v.

The setting doesn't seem to work, whatever A I set. I really don't think it was 80nm as they advertised, a BBS01 with 80nm is far more powerful than this when climbing.

Given a choice again, I will go for a BBS, I think this is a great product, but the variant and quality control is really required attention to compare with bbs, not to mention bosch and such.

Such disappointment.

 

[John and Cecil]

Such disappointment.

Yes, 48v controller firmware cuts battery power out at 42-43v. However that is not "half" the capacity of the battery. You are using just about all the "safe" capacity. It is not good to drain you battery below 3v per cell and it can cause it irreparable harm to do so. A 48v battery has 13 cells, so 42v is about 3.2v per cell. That seems to be a reasonable lvc to preserve battery longevity.

The motor performs better at 52v and with 750w amp setting. If you need more power climbing hills you may be able to flash your controller for a higher amp setting. 4.2v to 3.2v is about all the battery power you can use without reducing the longevity of your battery. Perhaps you could drain it down to 39v, but that would be the absolute minimum. Maybe you are losing 15% capacity.

The amp setting in the menu does not work, whatever it is set at by the factory is what you have no matter what amps you select in the menu. I think 80nm is for 750w too, not 500w.

If you want the motor to work at higher speeds you need to run a higher voltage battery. A 52v battery and reflash of the controller will increase motor speed by maybe 10%. Some people have pulled the 48v motor winding and replaced it with a 36v winding. That increases cadence like 30%, but you will lose some torque.

Actually the Lingbei motor is looking like it might have some potential. It will handle a little more power than the tsdz2, plus it is light and has a torque sensor and is also fairly reasonably priced. It is new to the market and still being revised but it looks promising!

 

[jbalat]

Or use the open-source firmware. I bet my 36v with stock battery can blow away the 52v motor

 

[casainho]

The funniest thing is the motor cut off at 43-42v with my 48v battery, hence, I can only use half of my battery capacity 13ah. I changed 2 battery, both at 48v 13ah and none of them working as intended. I stuck yesterday at the middle of a hill myself yesterday when my battery is still at 43v.

Unlock and get more power by using our OpenSource firmware, where you can configure on the LCD the battery low voltage cut off.

Here the feedback from jbalat, a long time user of TSDZ2 that already did more than 5000kms on it:

I think the new firmware has given the motor new life. (...) it feels much more responsive and actually gives you more power as you spin out instead of dropping off like the original firmware. That is what I wanted so it now feels more like the Bosch motor.

In fact the old firmware would feel like you are hitting a brick wall the faster you peddle but with this firmware it feels like a breath of fresh air. The faster you want to go the more it helps you. When I got right near the end and still had some battery left I pumped the power up to 550W and was able to ride comfortably at 38km/hr with little effort. I would not have been able to do this with the old firmware with a battery that was so low.

 

[bart1006]

Thanks guys. Where is this board located? I didn't open up the casing yet. When the hall sensor is failing, how can it be that it works perfect in the first two assist levels? Also, overvolting from 36V to 48V isn't causing this issue is it? I have programmed the data memory accordingly.

Sorry I didn't read your post right. I see that the motor completely stops I thought it was running less smooth with higher speed.
What do you exactly mean with overvolting from 36v to 48v? if you are running a higher voltage it can cut-off the motor. so maybe try to drain you battery a little bit more and see if it works?

 

[EndlessCadence]

What do you exactly mean with overvolting from 36v to 48v? if you are running a higher voltage it can cut-off the motor. so maybe try to drain you battery a little bit more and see if it works?

I'm using a 48v battery with a 36v motor to get assist while having higher cadence. Because I'm usually a road bike rider I tend to be in the higher RPM range (90-110). The motor will cut off in this range when using a 36v motor with a 36v battery as I've learned from jbalat and others. I've programmed the data memory to allow the 48v battery.

 

[emr]

Anyone else have some situational data on range or ah consumption w/ the 52v 750w motors?

I have 52v 750w and 14s 52v battery.
I achieved 80km from fully charged to fully empty using original firmware.

emr, sorry but 14s, 52v doesnt tell us what the battery capacity is...Ah or Whr is the relavent figure .
FWIW...my 750/48v standard firmware used approx 6ah over a 40km rough trail ride
That was using "Turbo" mode all the time !
I suspect these torque sensing systems are more economical than others methods of controller operation.?

Sorry, battery is 14Ah

 

[jbalat]

What do you exactly mean with overvolting from 36v to 48v? if you are running a higher voltage it can cut-off the motor. so maybe try to drain you battery a little bit more and see if it works?

I'm using a 48v battery with a 36v motor to get assist while having higher cadence. Because I'm usually a road bike rider I tend to be in the higher RPM range (90-110). The motor will cut off in this range when using a 36v motor with a 36v battery as I've learned from jbalat and others. I've programmed the data memory to allow the 48v battery.

So did that work on the standard firmware? I have heard cases where programming the memory doesn’t actually do anything.

I’ve also noticed that going to 48v on a 36v motor doesn’t give any extra cadence. It doesn’t really make sense to me because it works on my hub motor. Maybe I just need to try it again to confirm ?

I managed to get my 36v tsdz2 from 80 cadence to 112 cadence with the normal 10s battery by changing the over-speed value and the FOC advance in the Opensource firmware. If the FOC advance is not right the motor will lose timing. Perhaps that’s why you are getting the error code ?

What cadence are you getting with the normal firmware ?

 

[EndlessCadence]

I believe it does work. Currently I don't have a cadence sensor mounted but will do that soon. Does the opensource firmware display cadence as well? It feels like the motor assists easily up to around 90-100 RPM, will get back to this with real figures soon.

You could be right in that I experience some timing issues. I also want to inspect the hall sensors, to do this I need to open the housing but I have to buy a crank puller first.

Btw, the battery is 13s 48v, Panasonic NCR18650BD cells (seem to be genuine cells, as requested from the supplier), BMS 30A.
I've also tried to lower the amperage on the display, doesn't seem to do anything.

 

[jbalat]

Yes Opensource displays the cadence and no the amps does not work on the original display.
My friend was able to make a cadence meter use an old speedo and mounting the magnet to his cranks

 

[mauri_78]

I want to thank jbala for the very usefull video about the open firmware installation.
Seeing your videos i have only one doubt regarding the connector that you have soldered on the LCD3 display cable in order to avoid to cut the original motor to display cable.
In which way did you abtained this connector starting from a speed sensor cable? Could you better detail this step?
Thank you very much for your videos and you suggestions.
The actual open firmware has also the odo and trip function implemented?

Maurizio

 

[jbalat]

It appears to have a trip function on bottom right and the odo function is not implemented yet but I have raised it as a request on github.
Now that the firmware is working all these little things can be implemented in the near future.

So I ordered a normal speed cable from pswpower for $5 but they ran out and sent me the y splitter cable which is supposed to run the 6v lights. I got it for $5. I was upset at first but in the end I’m glad I got it so I didn’t have to cut my display cable. I would not spend $25 for it since all you really need are 2 connectors. The connectors don’t contain enough wires so you need to strip them all the way back carefully and solder to the back of them. Would be nice to find a cheaper source for these.

Also note that the y splitter cable only has a 6 pin connector so it does not work if you have a throttle which is an 8 pin.

 

[EndlessCadence]

Yes Opensource displays the cadence and no the amps does not work on the original display.
My friend was able to make a cadence meter use an old speedo and mounting the magnet to his cranks

Thanks. I do have a Garmin device with cadence sensor. Will install and try this tonight if I find the time.

 

[John and Cecil]

I believe it does work. Currently I don't have a cadence sensor mounted but will do that soon. Does the opensource firmware display cadence as well? It feels like the motor assists easily up to around 90-100 RPM, will get back to this with real figures soon.

You could be right in that I experience some timing issues. I also want to inspect the hall sensors, to do this I need to open the housing but I have to buy a crank puller first.

Btw, the battery is 13s 48v, Panasonic NCR18650BD cells (seem to be genuine cells, as requested from the supplier), BMS 30A.
I've also tried to lower the amperage on the display, doesn't seem to do anything.

You can verify motor cadence without a sensor. Just rev out first gear on a flat surface and use a tire and gearing calculator to estimate your max motor cadence.

http://www.bikecalc.com/gear_speed

Just plug in your tire size and width and your gearing and then use trial and error to adjust cadence until your 1st gear max speed matches the speed on the chart.

 

[playvltk03]

Such disappointment.

Yes, 48v controller firmware cuts battery power out at 42-43v. However that is not "half" the capacity of the battery. You are using just about all the "safe" capacity. It is not good to drain you battery below 3v per cell and it can cause it irreparable harm to do so. A 48v battery has 13 cells, so 42v is about 3.2v per cell. That seems to be a reasonable lvc to preserve battery longevity.

The motor performs better at 52v and with 750w amp setting. If you need more power climbing hills you may be able to flash your controller for a higher amp setting. 4.2v to 3.2v is about all the battery power you can use without reducing the longevity of your battery. Perhaps you could drain it down to 39v, but that would be the absolute minimum. Maybe you are losing 15% capacity.

The amp setting in the menu does not work, whatever it is set at by the factory is what you have no matter what amps you select in the menu. I think 80nm is for 750w too, not 500w.

If you want the motor to work at higher speeds you need to run a higher voltage battery. A 52v battery and reflash of the controller will increase motor speed by maybe 10%. Some people have pulled the 48v motor winding and replaced it with a 36v winding. That increases cadence like 30%, but you will lose some torque.

Actually the Lingbei motor is looking like it might have some potential. It will handle a little more power than the tsdz2, plus it is light and has a torque sensor and is also fairly reasonably priced. It is new to the market and still being revised but it looks promising!

I agree with you on this, however with me it look like half baked product. Your valuable time and material has helped make this product great but I don't know why the manufacturer wouldn't do this in the first place or at least have some decent firmware rather than expect people to flash and do their work. I bought a motor, I expected it to work as its specification, this 500w motor operated like my 350w rear hub and it's not acceptable, in which later I found out, there are 48v version, 48v overlock version, 52v version, 52v overlock version, tsdz2b version and so on, I also found out not all of TSDZ2 has throttle enable, that means some work with throttle, some doesn't and you either have to replace the controllers and or change the display. This gives no different than having a hub drive and just change the controllers (in fact this might be the easier way).

Flashing and change the controller seems like a great way to increase this product but It doesn't make much sense for me, for 2 reasons, where I'm at (Western Canada), it's outrageous shipping import and duty to buy all the items required (cable, display) for the flashing, 2nd reason is that even if i'm willing to spend money for 1st, i prob have to wait like 2-3 weeks and when the winter came, it is going to be for next year riding.

I agree with you it look promising, but until they standardlized their product at least and make the setting works, I would not touch it. Some experiment was fun, this wasn't, a good product required firmware to make it better, this make me felt like the firmware make a bad product become bearable.

Again, no pun intended, i think you and other worked on the open firmware has did a great job, the manufacturer should have bought your time and the software code making their product better. As far as they didn't do that or even not having the functionality that you guys spend time and efforts on, I'll stay away from this.

 

[John and Cecil]

I agree with you on this, however with me it look like half baked product. Your valuable time and material has helped make this product great but I don't know why the manufacturer wouldn't do this in the first place or at least have some decent firmware rather than expect people to flash and do their work. I bought a motor, I expected it to work as its specification, this 500w motor operated like my 350w rear hub and it's not acceptable, in which later I found out, there are 48v version, 48v overlock version, 52v version, 52v overlock version, tsdz2b version and so on, I also found out not all of TSDZ2 has throttle enable, that means some work with throttle, some doesn't and you either have to replace the controllers and or change the display. This gives no different than having a hub drive and just change the controllers (in fact this might be the easier way).

Flashing and change the controller seems like a great way to increase this product but It doesn't make much sense for me, for 2 reasons, where I'm at (Western Canada), it's outrageous shipping import and duty to buy all the items required (cable, display) for the flashing, 2nd reason is that even if i'm willing to spend money for 1st, i prob have to wait like 2-3 weeks and when the winter came, it is going to be for next year riding.

I agree with you it look promising, but until they standardlized their product at least and make the setting works, I would not touch it. Some experiment was fun, this wasn't, a good product required firmware to make it better, this make me felt like the firmware make a bad product become bearable.

Again, no pun intended, i think you and other worked on the open firmware has did a great job, the manufacturer should have bought your time and the software code making their product better. As far as they didn't do that or even not having the functionality that you guys spend time and efforts on, I'll stay away from this.

Hi, just to be clear I am not working on the open firmware. Casainho and some others are working on that, I am not using that nor do I endorse it at the moment. I have not even tried it so I really have no opinion on it other than I would wait until it is tested and deemed safe. I too do not wish to use it based on cutting cables, rewiring, using an aftermarket display etc. I have not done much, if anything, on this motor and I am just a user and not modifying it in any way except for battery voltage. I only post from my experiences and try to help others when possible.

I completely agree, the manufacturer has dropped the ball on this motor. Furthermore the motor is defective and the gears do not mesh properly thus making premature failure of the nylon drive gear almost commonplace. The display functioning does not work correctly either, and they can't even get their quality control to ensure the motors are flashed for the correct voltage before shipping.

I think the Lingbei might be a very good motor, but you are 100% correct that it is not completed yet nor tested. It only has potential to be good. Only time will tell if it lives up to it's potential.

There is a way to flash the motor with the original firmware. The user Eyebyesickle is working on program and memory files to flash so it will work with the original equipment. If you download the 750w 48w firmware files and flash it should at least have more power.

I was not endorsing that you buy this motor. I was only trying to help you because I assumed you owned it and were keeping it so I tried to give you some options. If you have not bought it you are correct that Bafang is more reliable and a more polished product, and perhaps the Lingbei will be good if a torque sensor is desired but only after the finished product is released and tested thoroughly.

I hope this clarifies my position You do not want to read back through 130 pages but I have been vocal about this motor being defective and the manufacturer not doing anything to correct the issues.

 

[playvltk03]

I agree with you on this, however with me it look like half baked product. Your valuable time and material has helped make this product great but I don't know why the manufacturer wouldn't do this in the first place or at least have some decent firmware rather than expect people to flash and do their work. I bought a motor, I expected it to work as its specification, this 500w motor operated like my 350w rear hub and it's not acceptable, in which later I found out, there are 48v version, 48v overlock version, 52v version, 52v overlock version, tsdz2b version and so on, I also found out not all of TSDZ2 has throttle enable, that means some work with throttle, some doesn't and you either have to replace the controllers and or change the display. This gives no different than having a hub drive and just change the controllers (in fact this might be the easier way).

Flashing and change the controller seems like a great way to increase this product but It doesn't make much sense for me, for 2 reasons, where I'm at (Western Canada), it's outrageous shipping import and duty to buy all the items required (cable, display) for the flashing, 2nd reason is that even if i'm willing to spend money for 1st, i prob have to wait like 2-3 weeks and when the winter came, it is going to be for next year riding.

I agree with you it look promising, but until they standardlized their product at least and make the setting works, I would not touch it. Some experiment was fun, this wasn't, a good product required firmware to make it better, this make me felt like the firmware make a bad product become bearable.

Again, no pun intended, i think you and other worked on the open firmware has did a great job, the manufacturer should have bought your time and the software code making their product better. As far as they didn't do that or even not having the functionality that you guys spend time and efforts on, I'll stay away from this.

Hi, just to be clear I am not working on the open firmware. Casainho and some others are working on that, I am not using that nor do I endorse it at the moment. I have not even tried it so I really have no opinion on it other than I would wait until it is tested and deemed safe. I too do not wish to use it based on cutting cables, rewiring, using an aftermarket display etc. I have not done much, if anything, on this motor and I am just a user and not modifying it in any way except for battery voltage. I only post from my experiences and try to help others when possible.

I completely agree, the manufacturer has dropped the ball on this motor. Furthermore the motor is defective and the gears do not mesh properly thus making premature failure of the nylon drive gear almost commonplace. The display functioning does not work correctly either, and they can't even get their quality control to ensure the motors are flashed for the correct voltage before shipping.

I think the Lingbei might be a very good motor, but you are 100% correct that it is not completed yet nor tested. It only has potential to be good. Only time will tell if it lives up to it's potential.

There is a way to flash the motor with the original firmware. The user Eyebyesickle is working on program and memory files to flash so it will work with the original equipment. If you download the 750w 48w firmware files and flash it should at least have more power.

I was not endorsing that you buy this motor. I was only trying to help you because I assumed you owned it and were keeping it so I tried to give you some options. If you have not bought it you are correct that Bafang is more reliable and a more polished product, and perhaps the Lingbei will be good if a torque sensor is desired but only after the finished product is released and tested thoroughly.

I hope this clarifies my position You do not want to read back through 130 pages but I have been vocal about this motor being defective and the manufacturer not doing anything to correct the issues.

It's ok John and Cecil, i bought this purely on curiousity and positive feedback I got from reddit and some youtube video, I don't blame anyone for that. Thank for chipping the voice, I got no other options to keep this motor, until better option arrived.

Can't believe I'm a completed fool to sell a bafang max 350w 48v for this thinking this is a better bet based on motor rating.

Just voice my concern for this motor and nothing against anyone else.

Thanks !

 

[EndlessCadence]

Alright, I've tried almost all your suggestions and it's not helping unfortunately. I've also reconnected the connectors in the motor, amongst them the Hall sensor cable. Could it be that programming the data memory (changing values) of the 36v controller to 48v simply doesn't work at all? Like some have said before (they were using 52v I believe)?

I've also noticed that the walk assist is actually working when holding the rear wheel of the ground (no load), when I drop the wheel two things can happen: either the motor cuts off or the bike tries to run away from me for a short time (is this 6 km/h? feels maybe too fast?).

 

[John and Cecil]

Hey dude, you were interested in my 15s battery test so I have some preliminary results for you. So far it looks very promising. I rec'd my small 21v 1a charger and charged my 3 batteries individually (takes forever to charge the batteries too). I just got back from a quick test run. I tried 61.1v today and it worked right out of the garage without the need to "kick it" with the throttle. I also ran some test runs down a very slight incline and used the bike calc to calculate cadence, my results are below. The 61v result was done with the new battery, the others were run with my 52v battery. It appears there may not be any need to make a custom firmware for an under-volted 15s battery, and it should get exceptional range due to the 45v low voltage cutoff.

61v - 12.6 mph in 1st gear - 119 calc rpm
58v - 12.0 mph in 1st gear - 113 calc rpm
51v - 11.0 mph in 1st gear - 104 calc rpm
47v - 10.0 mph in 1st gear - 94 calc rpm

These number may be off a little due to differences in tire sizes, but it shows how voltage increases cadence. The difference between 48v and 62v is about 25 rpms. Hopefully over the next few days I can determine the upper voltage cutoff for the 52v firmware, but so far 61v works fine. Bear in mind my controller may have worn and can handle slightly higher voltages, perhaps 0.5 - 1.0 v higher. I will attempt 62v next time, however I do not think I want to try any higher than that I suspect I will be be able to throttle kick 62v with the current firmware. Hopefully the charger I ordered last week will be able to reach 61.5 - 62v max charge.

Cecil and I are going to go run it for 5 miles and see how it works, just wanted to let you know! Thanks man

Edit - we just did a 3 mile run, works wonderful. I definitely feel the power and speed increase. Since it can go down to 45v everything over 58.5v is added range and added power. I threw the batteries back on the charger and we will try 61.5-62v either tonight or tomorrow.

 

[eyebyesickle]

Nice! Thanks for sharing!

 

[jbalat]

Alright, I've tried almost all your suggestions and it's not helping unfortunately. I've also reconnected the connectors in the motor, amongst them the Hall sensor cable. Could it be that programming the data memory (changing values) of the 36v controller to 48v simply doesn't work at all? Like some have said before (they were using 52v I believe)?

I've also noticed that the walk assist is actually working when holding the rear wheel of the ground (no load), when I drop the wheel two things can happen: either the motor cuts off or the bike tries to run away from me for a short time (is this 6 km/h? feels maybe too fast?).

Only one way to find out. Try going back to 36v firmware and use a battery charged to 41v
I can post my backup of the 36v firmware, memory and option bytes if you need it.

If it does work then yes it means because the 36v has different windings your motor is losing timing using the 48v firmware.

 

[jbalat]

Nice! Thanks for sharing!

You made me curious again. I will retry going from 10s to 14s on my 36v and get some real data and maybe some video.