New "TSDZ2 Torque Sensor Central Motor"

Andrew707 said:
casainho said:
Then you should go with our OpenSource firmware that has high efficiency when compared to other firmware. I would go with a 250Wh battery and with our firmware you can fine tune directly on display for long travels to optimize range, also battery SOC with very good estimation that includes the parameter of battery internal resistance. 250Wh battery is pulling 250W during one hour and maybe you will only pull about 75W average per hour, so, enough for 4 hours ride - or you can have an extra 250W battery to use on the long travels and for daily riding use only one.
For display I would go with SW102 because is very small and cheap and looks cheap also, without much value.
I would like to go with your firmware and the features you'd like to implement sounds like a very good idea!

I just checked my dropbars diameter and it's 31.8mm while looks like SW102 is only 22.2mm. Do you think the clamp would still fit with a longer screw? SW102 screen looks really stealth, I like it!
I just measure and maybe it can go up to 26mm max. In that case, I would 3D print something to hold it in place or find a creative way to do it.

Andrew707 said:
Talking about the batteries I probably forgot to mention where I'm planning to use my bike is a very hilly area. Constantly up and down, up and down. I'm not sure if 250Wh would be enough. I already used a hub motor with two LiGo batteries which were almost 200Wh and they were okay for shorter rides but not an option for a bikepacking. Ideally, I would like to have around 50km of range. That's why I was thinking about either 36V 16.5Ah or 48V 16.5Ah battery because looks like they are a good capacity to price ratio (e.g. the same capacity LiGo batteries would cost twice as much).

50 kms range? -- let's compare:

You say you are pretty fit and I weight 105kgs with 173cms, I am far from fit. My bicycle is this one, tires 2.2:

casainho_MTB-half.jpg


Recently a did a trip of 43kms and used 166Wh (36%) of my 460Wh battery (that has with 200 milliohms resistance):

2020-04-12-19-05-26-1.jpg
2020-04-12-19-06-13-1.jpg


This trip was done half in road and other half on gravel, on flat roads near the ocean. And my average pedal human power was about 150Wh. Since you are pretty fit compared to me and probably have a bicycle with slim tires, I would say you would use less than the 166Wh I used - now the question is how many hills you will get on that 50kms trips?

I usually record the power I use for trips so I can optimize the battery I will take on next trip. One of the last MTB trip, with the same bicycle as you see on the picture (including the rear panniers), 45kms and +1400m, I took about 5h and I used only 320Wh. So, since you are very fit compared to me and probably with a bicycle with slim tires, I would say a 250Wh battery would be enough for your 50 kms bikepacking trips.

image.png


Note that the very latest firmware version was optimized to use low power at values like below 150W - torque sensor has 4x more sensitivity and in addition the motor current is controlled with 4x more resolution.

For bikepacking and having a small and light battery pack, I would go with cells that have high capacity like 18650 I use with 3600 mAh but they have high internal resistance which is not a issue when pulling low current / power, which will be the case anyway for the long trips.

For a 36V motor, maybe you could go with a 10S2P battery pack: 3.6 amp each cell * 3.7 volts each cells * 10 cells in series * 2 parallel = 266Wh. This battery pack would need to be hand made (my were done on a local shop, mainly soldering the tabs and then I wired the BMS and put tape with insulating spongy plastic) and would be half in size of that one of 36V 16.5Ah you did mention.

About the gear, the blue gear is very robust when your use our OpenSource firmware that has a feature to limit the current ramp / motor current torque ramp and this saves a lot this gear. The metal one is clearly more noisy and heavy.
The magnetic brake sensors are very easy to install and you do not have them, you will have serious risk to damage the blue gear!!

About the drag, I would keep using always the motor assisting since you can have a very small and light battery and configure to use the motor max power, like you can quick configure on the display main screen the max power of 25, 50, 75 or more Watts and that will probably compensate the drag.

See the TSDZ2 wiki FAQ here for all information: https://github.com/OpenSource-EBike-firmware/TSDZ2_wiki/wiki/FAQ
 
Tsdz2 creaking sound
So guys, i asked about this earlier but those tips did not work, my motor does this loud creaking/cracking when under load, i evenI disassembled the whole motor and checked every gear, cleaned and re greased them, still same sound.
What would be causing this?
Only thing i left checking was the torquesensor because i did not have tools to remove axle from me motor.
Here is link to youtube video about it: https://www.youtube.com/watch?v=7YkTqahKT-o
Thank you guys.
 
casainho said:
RicMcK said:
I don't understand why one has to put specific weights on the peddles if the ACD values don't change when I change the weight or put the different weights on the peddle.
They change!! The ADC values for torque sensor change for different weights!!

I understand that they should change. However when I put the different weights on the pedal the ACD values are static (don't change on the display). Should the ACD values be dynamically displayed? If yes, and mine are static, do you have any suggestions?

Thanks
 
volcamin said:
Tsdz2 creaking sound
So guys, i asked about this earlier but those tips did not work, my motor does this loud creaking/cracking when under load, i evenI disassembled the whole motor and checked every gear, cleaned and re greased them, still same sound.
What would be causing this?
Only thing i left checking was the torquesensor because i did not have tools to remove axle from me motor.
Here is link to youtube video about it: https://www.youtube.com/watch?v=7YkTqahKT-o
Thank you guys.
I can't understand the reason because you already checked and saw to issue. I would say something in the blue gear and his axle, etc. See the wiki FAQ about the common issues, maybe it will help you.

RicMcK said:
casainho said:
RicMcK said:
I don't understand why one has to put specific weights on the peddles if the ACD values don't change when I change the weight or put the different weights on the peddle.
They change!! The ADC values for torque sensor change for different weights!!

I understand that they should change. However when I put the different weights on the pedal the ACD values are static (don't change on the display). Should the ACD values be dynamically displayed? If yes, and mine are static, do you have any suggestions?
Yes, they are dynamically displayed, just like you should see the cadence value change while you rotate the pedals.

If the torque sensor ADC value do not change then I would say the torque sensor is not working. See the wiki TSDZ2 FAQ, maybe it can help you.
 
TSDZ2 at 5475 RPMs / cadence of 135 RPM

I am developing a version of our firmware to run the motor at high cadence without sacrificing the system efficiency. I am increasing the PWM frequency and it seems to be working on very early tests and I would like to know what is the desired max cadence in real world as the TSDZ2 max cadence with stock firmware is 90 RPM.

Well, I must say this increase of cadence without sacrificing the efficiency will only work when overvoltage the motor, like using a 36V motor with 48V battery. Still a 52V battery can be used and some users did some tests without problems so far on 55V batteries.
 
Andrew707 said:
jeff.page.rides said:
You definitely want to buy a 52 Volt battery.
https://www.electrifybike.com/store/p102/52v_40-AMP_14Ah_Super-Shark_E-bike_Battery.html#/
https://www.electrifybike.com/store/p103/52v_40-AMP_17.5_Ah_Jumbo_Shark_E-bike_Battery_.html#/
What's the main reason you're advising a 52V battery? Will it work with 36V motor?
 
jeff.page.rides said:
Andrew707 said:
jeff.page.rides said:
You definitely want to buy a 52 Volt battery.
https://www.electrifybike.com/store/p102/52v_40-AMP_14Ah_Super-Shark_E-bike_Battery.html#/
https://www.electrifybike.com/store/p103/52v_40-AMP_17.5_Ah_Jumbo_Shark_E-bike_Battery_.html#/
What's the main reason you're advising a 52V battery? Will it work with 36V motor?

With the higher voltage, you get more power and more longevity without creating more heat. Higher watts and amps creates more heat.
 
Calibrating torque sensor Cont.

I reflashed both the display and the motor controller (P860).
The bike registers both human and motor assist watts and it seems to ride OK. And the motor power varies with applied peddle force. This is a brand new motor that seemed to ride OK with the stock VLCD5.

Calibrating the torque sensor process.
Boot system with left peddle down (per torque sensor menu).
Set assist level to 1
Select, the torque sensor calibration menu.
select "enable".
There is default data in both the weight & ACD values. Left "0" value=192, Right "0" value=190
Input all the test weights for both sides.
When I put weights on either peddle the ACD values stay fixed?

Is this a bug? Am I doing something incorrectly?

Thanks
 
casainho said:
TSDZ2 at 5475 RPMs / cadence of 135 RPM

I am developing a version of our firmware to run the motor at high cadence without sacrificing the system efficiency. I am increasing the PWM frequency and it seems to be working on very early tests and I would like to know what is the desired max cadence in real world as the TSDZ2 max cadence with stock firmware is 90 RPM.

Well, I must say this increase of cadence without sacrificing the efficiency will only work when overvoltage the motor, like using a 36V motor with 48V battery. Still a 52V battery can be used and some users did some tests without problems so far on 55V batteries.

Casainho, 135rpm is way above what I would expect to see as an off road rider, I'm reasonably fit and spin in the 95 max 100 region when pushing it and normally in that 80 - 95 range. On the V20 with the standard config but with 52 volts the motor behaves pretty nicely and still puts in some torque at 95rpm. Do people really pull into the 100's routinely, I would doubt that.
 
Waynemarlow said:
casainho said:
TSDZ2 at 5475 RPMs / cadence of 135 RPM

I am developing a version of our firmware to run the motor at high cadence without sacrificing the system efficiency. I am increasing the PWM frequency and it seems to be working on very early tests and I would like to know what is the desired max cadence in real world as the TSDZ2 max cadence with stock firmware is 90 RPM.

Well, I must say this increase of cadence without sacrificing the efficiency will only work when overvoltage the motor, like using a 36V motor with 48V battery. Still a 52V battery can be used and some users did some tests without problems so far on 55V batteries.

Casainho, 135rpm is way above what I would expect to see as an off road rider, I'm reasonably fit and spin in the 95 max 100 region when pushing it and normally in that 80 - 95 range. On the V20 with the standard config but with 52 volts the motor behaves pretty nicely and still puts in some torque at 95rpm. Do people really pull into the 100's routinely, I would doubt that.

I try to ride at 90-110 RPM. I have a 48 motor and 48 battery so I am stuck with a cadence around 80 max. Being able to use a 48/48 at a hight cadence would be great.
 
volcamin said:
Tsdz2 creaking sound
So guys, i asked about this earlier but those tips did not work, my motor does this loud creaking/cracking when under load, i evenI disassembled the whole motor and checked every gear, cleaned and re greased them, still same sound.
What would be causing this?
Only thing i left checking was the torquesensor because i did not have tools to remove axle from me motor.
Here is link to youtube video about it: https://www.youtube.com/watch?v=7YkTqahKT-o
Thank you guys.
Take a big screwdriver and push the handle against your ear.
Push the other end against the motor on different places while making the motor run.
This will help you find the location of the sound, it will be louder closer to the source.
 
New SW102 Firmware: Hey guys, I'd like to but I can't find this in the masses of information. I just updated my SW102's to the latest firmware via bluetooth. Now neither motors work. I went through the settings and I don't see anything I am missing. They just don't engage and the speed sensor does not work. It worked perfectly fine before I upgraded. I am thinking there is just something I am missing? I did turn on the actual assist using the up arrow.

Firmware version 0.8 on the LCD, and I didn't update the motor since the firmware hasn't changed.
 
casainho said:
Zamzoo said:
Just tried to update from 0.68 to 0.80 and I get the error e:firmware when booting it up (sw102 display, nrf toolbox 4.5.1 on iOS 13)

Any ideas what I need to check?

I've downgraded to 0.69 (I had downloaded it but never installed) now which works.

As an aside, any way to keep your configuration saved between firmware upgrades?
Probably you didn't flashed the correct motor firmware for the same display firmware - follow the instructions.

There seems to be only one motor firmware? Nothing has changed since August 2019. I just did a new install and there is no mention of firmware matching the 0.8.0 version for the SW102. For the record I am having the same problem (see post above).

Edit: Ok yesterday the wiki seemed to be linking to the old firmware. My fault for not double checking I guess, but now I see the new version. Time to redo the motors! If anyone else gets the e:firmware error this is the problem.

https://github.com/OpenSource-EBike-firmware/TSDZ2-Smart-EBike/releases
 
New firmware v0.8.0-beta.1 -- finally, TSDZ2 cadence limit were increase to 120 RPM (mostly without sacrificing the energy efficiency) while previous was 90 RPM.

Will be an advantage for the ones that have a 36V motor and a 48V battery, or for 48V motor with 15S batteries.

Changelog:

- TSDZ2 max pedal cadence increased from 90 to 120 RPM (mostly without sacrificing the energy efficiency)
-- increased PWM frequency from 15.6kHz to 19 kHz
-- UART communications frequency is now twice to save processing time
- ebike control loop now runs at twice the frequency in the hope to make system a bit more responsive (tested by vshitikov)

https://github.com/OpenSource-EBike-firmware/Color_LCD/releases/tag/860C_850C_SW102_v0.8.0-beta.1
 
casainho said:
Semogonif said:
casainho said:
I updated the wiki on how to install TSDZ2 on cargo bike -- now I am looking for notes and pictures of installation on a fat bike, I think this is the last category left - anyone available to share his own notes and pictures?

image.png

Well, I built an eMTB for my wife, with an 92mm BB. It's not a fat bike, but the challenge is similar:

IMG_20191224_115038_a.jpg
If you share with me photos and notes about that build, I can put on the wiki. Thanks.
Ok, I have to collect some information, and I will send it to you as soon as possible.
 
jeff.page.rides said:
With the higher voltage, you get more power and more longevity without creating more heat. Higher watts and amps creates more heat.
Makes sense. I guess extra 180W (36V*15A vs 48V*15A) might be useful sometimes climbing steeper hills with heavy panniers.

casainho said:
Will be an advantage for the ones that have a 36V motor and a 48V battery
I'm really considering ordering 48V battery for my 36V motor. I think it'll take a while until new mini screen arrives.
Will my motor work with 48V battery out of the box with VLCD5 display?
Is it possible to install OS firmware with VLCD5 display?

casainho said:
<...>
Recently a did a trip of 43kms and used 166Wh (36%) of my 460Wh battery (that has with 200 milliohms resistance):
<...>
One of the last MTB trip, with the same bicycle as you see on the picture (including the rear panniers), 45kms and +1400m, I took about 5h and I used only 320Wh. So, since you are very fit compared to me and probably with a bicycle with slim tires, I would say a 250Wh battery would be enough for your 50 kms bikepacking trips.
<...>
For a 36V motor, maybe you could go with a 10S2P battery pack: 3.6 amp each cell * 3.7 volts each cells * 10 cells in series * 2 parallel = 266Wh. This battery pack would need to be hand made (my were done on a local shop, mainly soldering the tabs and then I wired the BMS and put tape with insulating spongy plastic) and would be half in size of that one of 36V 16.5Ah you did mention.
Thank you for such an informative response. I can't believe that this motor is so efficient! :shock:

Previously I had Bafang G311 motor with 2 x LiGo 36V 2.7Ah 98W batteries and a throttle (no pedal assist) and I could do maybe ~20km. I guess throttle is not that efficient :D

I live in Canada and I don't know any bike shops around who could build me a custom pack. Also I don't want to order from USA because of poor $ exchange rate, shipping and possible duties. It's much more easier just to pick up the battery locally from Grin (https://www.ebikes.ca/shop/electric-bicycle-parts/batteries.html). However they don't have too much selection.

I would like to have light and stealthy setup and could possible use 3 or 4 Ligo batteries but that would cost me more or the same as 36V 16.5Ah or 48V 16.5Ah. So I'm thinking just sacrificing weight for more capacity. And if with such battery I could do 100km instead of 50 - that's even better.
 
Andrew707 said:
.......
I'm really considering ordering 48V battery for my 36V motor. I think it'll take a while until new mini screen arrives.
Will my motor work with 48V battery out of the box with VLCD5 display?
Is it possible to install OS firmware with VLCD5 display?.....
If you have a 36V motor with 48V battery and stock firmware, that would not work without changing some bits in that firmware.

Casainho has always developed OSF for complete tsdz2 configuration with the display. (first KT-lcd3 and now 850C, 860C-SW102).
It is possible to use an OSF fork with vlcd5 or other stock display too, but in that case you must do the configuration before flashing the controller.
So if you want to try OSF before buying another display you can use that release (also with 48V battery)
 
Elinx said:
Andrew707 said:
.......
I'm really considering ordering 48V battery for my 36V motor. I think it'll take a while until new mini screen arrives.
Will my motor work with 48V battery out of the box with VLCD5 display?
Is it possible to install OS firmware with VLCD5 display?.....
If you have a 36V motor with 48V battery and stock firmware, that would not work without changing some bits in that firmware.

Casainho has always developed OSF for complete tsdz2 configuration with the display. (first KT-lcd3 and now 850C, 860C-SW102).
It is possible to use an OSF fork with vlcd5 or other stock display too, but in that case you must do the configuration before flashing the controller.
So if you want to try OSF before buying another display you can use that release (also with 48V battery)
You should go with latest firmware version that unlocks the TSDZ2 cadence from 90 to 120 RPM, should be perfect for 36V motor with 48V battery!!
 
RicMcK said:
Calibrating torque sensor Cont.

I reflashed both the display and the motor controller (P860).
The bike registers both human and motor assist watts and it seems to ride OK. And the motor power varies with applied peddle force. This is a brand new motor that seemed to ride OK with the stock VLCD5.

Calibrating the torque sensor process.
Boot system with left peddle down (per torque sensor menu).
Set assist level to 1
Select, the torque sensor calibration menu.
select "enable".
There is default data in both the weight & ACD values. Left "0" value=192, Right "0" value=190
Input all the test weights for both sides.
When I put weights on either peddle the ACD values stay fixed?

Is this a bug? Am I doing something incorrectly?

Thanks

Does anyone have any suggestions for this issue?
 
RicMcK said:
Does anyone have any suggestions for this issue?
So, your torque sensor is not working. I would start to see if the connector inside the TSDZ2 is ok... You can try see if the values are different with the connector on place and unconnected.
 
Andrew707 said:
.......
I'm really considering ordering 48V battery for my 36V motor. I think it'll take a while until new mini screen arrives.
....
casainho said:
.....
You should go with latest firmware version that unlocks the TSDZ2 cadence from 90 to 120 RPM, .....
Ofcourse, if that is the main reason you want to go for OSF, but in that case there is no possibility to use the vlcd5 and you have to wait for your new screen.
 
Where do you guys buy SW102 Bluetooth version? I can only find the ones without Bluetooth, e.g. this one.
Do I understand correctly that I could still install OS firmware to this display via ST-Link V2 connector?
 
casainho said:
New firmware v0.8.0-beta.1 -- finally, TSDZ2 cadence limit were increase to 120 RPM (mostly without sacrificing the energy efficiency) while previous was 90 RPM.

Will be an advantage for the ones that have a 36V motor and a 48V battery, or for 48V motor with 15S batteries.

Changelog:

- TSDZ2 max pedal cadence increased from 90 to 120 RPM (mostly without sacrificing the energy efficiency)
-- increased PWM frequency from 15.6kHz to 19 kHz
-- UART communications frequency is now twice to save processing time
- ebike control loop now runs at twice the frequency in the hope to make system a bit more responsive (tested by vshitikov)

https://github.com/OpenSource-EBike-firmware/Color_LCD/releases/tag/860C_850C_SW102_v0.8.0-beta.1

Seems I'm getting the motor at a perfect time! Really excited and can't wait to get all the components delivered and start testing the open source firmware. Thanks for the development Casainho and others involved!
 
Andrew707 said:
Where do you guys buy SW102 Bluetooth version? I can only find the ones without Bluetooth, e.g. this one.
Do I understand correctly that I could still install OS firmware to this display via ST-Link V2 connector?
That is a good question and I need to put this info on the wiki. Although the sellers do not mention Bluetooth, maybe because they would need to pay license fees to Bluetooth organization, the hardware is just the same and is ready for Bluetooth which is implemented on firmware.

Yes, that unit will mostly sure work as expected, just follow the wiki instructions.
 
Back
Top