New "TSDZ2 Torque Sensor Central Motor"

Discussions related to motors other than hub motors.
This includes R/C motors, botttom bracket, roller and geared drives.
Kisazul   100 mW

100 mW
Posts: 44
Joined: Jun 18 2018 10:32pm

Re: New "TSDZ2 Torque Sensor Central Motor"

Post by Kisazul » Aug 13 2018 7:37pm

Hi guys. I wrote above that I had a problem with the motor. It looked like this. If I pedal without the support of a motor, then the bike perfectly rides off my leg strength. So the clutch that transmits the torque from the pedals to the wheel works. But when I turn on the motor support, volume when the pedals turn with effort I hear a slipping clutch. Probably those who have a blue gear with excessive load will fail the teeth of the gear. But I have a metal gear. Inside this gear is a clutch that transmits the torque from the motor to the wheel. And also when we do not use support, this clutch opens the pedals and the motor so that we do not waste energy on its rotation. I replaced the Chinese clutch and another bearing on a quality Japanese. The replacement is not difficult! As a result, I did it, the engine perfectly digests 800-900 watts. I use a 52v battery. My weight is 90 kg.
I hope my information will be useful to someone!
Attachments
IMG_20180814_103019.jpg
IMG_20180814_103035.jpg
IMG_20180814_103049.jpg

KD5ZXG   10 W

10 W
Posts: 66
Joined: Jul 13 2018 10:02pm

Re: New "TSDZ2 Torque Sensor Central Motor"

Post by KD5ZXG » Aug 13 2018 7:55pm

I get whats going in the vid with torque low pass.
Schottky and series resistor in parallel with resistor
already on the board might let come down faster.

So resistor has different value coming than going.
But still slow enough to damp whatever problem
they made it slow to solve in the first place.

Don't yet know how accessible that circuit might or
not be. Even if it were easy to get at, not everyone
has solder skills and equipment. No problem for me,
but not for everyone.

Or as you noted, pedal letoff and/or coaster braking
might be quickly determined by a sudden decrease of
cadence. And that method needs no hardware change.

So to avoid strange start/stop misbehaviors:
If cadence is slow or increasing: torque rules.
If cadence is fast or decreasing: cadence rules.
Can't wrap my mind round weather what I'm
suggesting actually makes a stable loop.

Non-linear Bode plots aren't my specialty, and most
of my quadrature feedback audio amplifier schemes
have unresolved oscillation issues.

User avatar
jbalat   10 kW

10 kW
Posts: 531
Joined: Nov 10 2012 11:50pm
Location: Melbourne, Australia

Re: New "TSDZ2 Torque Sensor Central Motor"

Post by jbalat » Aug 13 2018 11:56pm

Just wanted to share my experience this morning on the first ride in to work after I had calibrated the torque sensor using the new firmware.

The bike is very responsive in fact it doesn't feel like the old bike at all
I am now riding in Level 2 and I was able to back off the main power to 300 W max and happily rode in most of the way at around 200W and 32km/hr with not too much effort. When starting off and with some headwind it would go to 280W but not for long
I kept a good eye on my cadence with was between 60 and 67 most of the time
Pushing a little harder I could ride at 75-78 cadence and it felt really comfortable. Easy to change gears by just backing off the torque a little bit, you don’t need to stop pedalling.
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. (Update the cadence did top out at just over 80 for my 36v motor with 39v but I wouldn’t have hit 80 without trying hard)
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.

What could be improved...
So I kept my feet off the pedals or had pedals in 6-12oclock position when stopped at lights so that the motor would not engage. It’s not that big a deal when you get used to it although it would be nice if I had cutoff brakes installed.
I felt like it was too sensitive sometimes as it did not require me to press hard before it gave max assist. I thought if I could bump up the max torque setting just a little it would be a little more natural. This is pretty easy to do.

At very lower power and cadence at the start of my journey I experienced some weird noises from the motor, almost like a resonance or electrical noise. I could not figure out what it was but it went away and did not come back again ?? Weird. This did not happen on my way home.

On the whole I'm pretty happy and have to thank casainho and all that have contributed to this awesome project :)
Please subscribe to my Youtube home page for Quadcopters, FPV and Bike Rides http://www.youtube.com/jbalatutube
My build log here http://endless-sphere.com/forums/viewto ... =6&t=55352
For some cool ebike talk join us on The Best ebike https://www.facebook.com/groups/193581737807033/
Tech talk TSDZ2 Opensource firmware https://endless-sphere.com/forums/viewt ... &start=150
Opensource TSDZ2 Wiki https://github.com/OpenSource-EBike-fir ... _wiki/wiki

casainho   100 MW

100 MW
Posts: 2793
Joined: Feb 14 2011 2:43pm

Re: New "TSDZ2 Torque Sensor Central Motor"

Post by casainho » Aug 14 2018 6:10am

jbalat wrote:
Aug 13 2018 11:56pm
Just wanted to share my experience this morning on the first ride in to work after I had calibrated the torque sensor using the new firmware.

The bike is very responsive in fact it doesn't feel like the old bike at all
I am now riding in Level 2 and I was able to back off the main power to 300 W max and happily rode in most of the way at around 200W and 32km/hr with not too much effort. When starting off and with some headwind it would go to 280W but not for long
I kept a good eye on my cadence with was between 60 and 67 most of the time
Pushing a little harder I could ride at 75-78 cadence and it felt really comfortable. Easy to change gears by just backing off the torque a little bit, you don’t need to stop pedalling.
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. (Update the cadence did top out at just over 80 for my 36v motor with 39v but I wouldn’t have hit 80 without trying hard)
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.
So this is a very positive feedback and I interpret that with our OpenSource firmware the TSDZ2 motor runs more efficient/uses less battery energy, while the motor also has more torque!!

The true is that I am not sure that original firmware implements FOC while I did on our OpenSource firmware!! With FOC we get the max torque possible per battery amp!! This means the motor is more efficient/uses less battery energy and has more torque.
jbalat wrote:
Aug 13 2018 11:56pm
What could be improved...
So I kept my feet off the pedals or had pedals in 6-12oclock position when stopped at lights so that the motor would not engage. It’s not that big a deal when you get used to it although it would be nice if I had cutoff brakes installed.
I felt like it was too sensitive sometimes as it did not require me to press hard before it gave max assist. I thought if I could bump up the max torque setting just a little it would be a little more natural. This is pretty easy to do.

At very lower power and cadence at the start of my journey I experienced some weird noises from the motor, almost like a resonance or electrical noise. I could not figure out what it was but it went away and did not come back again ?? Weird. This did not happen on my way home.

On the whole I'm pretty happy and have to thank casainho and all that have contributed to this awesome project :)
1. Pedals torque sensor are very sensitive
This is very good to know!! This means for hand bikes application like the ones used by quadriplegic or even for little kids like at age of 6, TSDZ2 is great piece of technology due to his very sensitive pedal torque sensor.

For your needs and the ones alike that don't have bake sensor installed, I understand that the initial sensitivity should be reduced and as you say, should be easy to implement and I would like to do it using an option on config.h file for that value. Some questions to you:

1.1 - what is the minimum value? can you please test and see the value on LCD3: 3: advanced technical data 0: --> 3: pedal torque sensor.

1.2 - I am thinking in put that limitation only when you press the pedals and ebike wheel speed is 0, is that ok?

2. Motor vibration at very low speed
Thanks for this feedback. I am well aware of this and there is not a problem to the motor, only it is different from original firmware. Both original firmware of TSDZ2 and KT motor controllers start with 6 step/block commutation and after switches to Space Vector Modulation. When I did the 6 steps implementation, the switch to Space Vector Modulation had an issue with the current controller... we did discuss this on KT OpenSource firmware and we weren't able to improve this... maybe I can revisit this later.

The switch happens at a specific motor speed ERS and this value is defined on config.h:

Code: Select all

#define MOTOR_ROTOR_ERPS_START_INTERPOLATION_60_DEGREES 10
Other users reported that they did lower the value to 1 or even 0 and that vibration disappears and the motor works well as expected. Maybe you want to try that to see if you like.
Developer of the Flexible OpenSource firmware for EBike motor controllers (TSDZ2 and KT) and LCDs (KT-LCD3 and Bafang 850C color LCD).

If you like my work, please consider making a donation. I am being using the donations to buy needed resources for my developments. My paypal: casainho AT gmail.com.

User avatar
John and Cecil   1 kW

1 kW
Posts: 473
Joined: Mar 14 2018 7:26pm

Re: New "TSDZ2 Torque Sensor Central Motor"

Post by John and Cecil » Aug 14 2018 11:10am

Timelord wrote:
Aug 12 2018 9:45pm
Sigh. I give up. Not going to read this topic anymore since it's 99% hacking talk that's of no interest to most purchasers of the unit.
Speady wrote:
Aug 13 2018 12:02pm
Me too.
Hi. There is another option available, it is not perfect but it is the only method that I can think of that might be of help to you. You can go into your settings (user control panel) and add the names of the people that are discussing the hacked firmware in this forum into your "foe" list (blocked members). You will no longer be able to see their posts in the forum and instead of a big box of text/pics/etc it will only show a a very short line of text stating that the blocked user made a post. I think if you block ten or so people the hacking posts in this thread will be reduced by 80-90%. This way you can still have access to the thread and add to the collective knowledge here and hopefully it won't be so annoying for you anymore. I will endeavor to make less posts regarding the hacked firmware in this thread, however you are welcome to block me as well and I will not take offense. Have a nice week!
Last edited by John and Cecil on Aug 14 2018 11:23am, edited 1 time in total.

User avatar
John and Cecil   1 kW

1 kW
Posts: 473
Joined: Mar 14 2018 7:26pm

Re: New "TSDZ2 Torque Sensor Central Motor"

Post by John and Cecil » Aug 14 2018 11:21am

Kisazul wrote:
Aug 13 2018 7:37pm
Hi guys. I wrote above that I had a problem with the motor. It looked like this. If I pedal without the support of a motor, then the bike perfectly rides off my leg strength. So the clutch that transmits the torque from the pedals to the wheel works. But when I turn on the motor support, volume when the pedals turn with effort I hear a slipping clutch. Probably those who have a blue gear with excessive load will fail the teeth of the gear. But I have a metal gear. Inside this gear is a clutch that transmits the torque from the motor to the wheel. And also when we do not use support, this clutch opens the pedals and the motor so that we do not waste energy on its rotation. I replaced the Chinese clutch and another bearing on a quality Japanese. The replacement is not difficult! As a result, I did it, the engine perfectly digests 800-900 watts. I use a 52v battery. My weight is 90 kg.
I hope my information will be useful to someone!

Do the made in Japan parts use the same part numbers, and if not do you have the part numbers for the more robust parts? I am still running the plastic gear but depending on when and how it fails I may upgrade to a metal gear. If I can get 4 - 5k miles out of the original before failure then it is probably worthwhile to replace it with another plastic gear, but if it only lasts less than 3k miles I will replace it with a metal gear. So far we only have 750 miles on it so we have a long way to go!

casainho   100 MW

100 MW
Posts: 2793
Joined: Feb 14 2011 2:43pm

Re: New "TSDZ2 Torque Sensor Central Motor"

Post by casainho » Aug 14 2018 11:50am

John and Cecil wrote:
Aug 14 2018 11:10am
the people that are discussing the hacked firmware in this forum
I wish you guys could correctly refer to our efforts: we are developing just like Tongsheng engineers did. I am also an engineer and working professionaly as a firmware developer, for a Germany company that develops top technology for ebikes, scooters, etc. Our OpenSource firmware is not a hack but instead is a development of people like me, just like the Tongsheng engineers.
Developer of the Flexible OpenSource firmware for EBike motor controllers (TSDZ2 and KT) and LCDs (KT-LCD3 and Bafang 850C color LCD).

If you like my work, please consider making a donation. I am being using the donations to buy needed resources for my developments. My paypal: casainho AT gmail.com.

User avatar
feketehegyi   1 W

1 W
Posts: 53
Joined: Jul 11 2018 6:18am

Re: New "TSDZ2 Torque Sensor Central Motor"

Post by feketehegyi » Aug 14 2018 1:27pm

This post is a bluetooth android display hack. Skip it if you don't like hacking TSDZ2.

The parts for the bluetooth module have arrived. I've built them together and put into a box.
Please see my video about. This is still for stock factory firmware.

User avatar
John and Cecil   1 kW

1 kW
Posts: 473
Joined: Mar 14 2018 7:26pm

Re: New "TSDZ2 Torque Sensor Central Motor"

Post by John and Cecil » Aug 14 2018 1:32pm

casainho wrote:
Aug 14 2018 11:50am
John and Cecil wrote:
Aug 14 2018 11:10am
the people that are discussing the hacked firmware in this forum
I wish you guys could correctly refer to our efforts: we are developing just like Tongsheng engineers did. I am also an engineer and working professionaly as a firmware developer, for a Germany company that develops top technology for ebikes, scooters, etc. Our OpenSource firmware is not a hack but instead is a development of people like me, just like the Tongsheng engineers.
Some of us are not programmers, so when I say hacked you should accept that it also includes development. To those that do not program or manufacture firmware, hacking and development could have the same meaning. As far as I know you are not associated with the engineers that built the original hardware, and therefore we see you as more of hackers. Now if you were part of the original motor development team it would appear different, at least to me, but when you develop software for something without having access to all the design data/etc hacking seems appropriate. It is not meant to have a negative connotation or that I (we) do not appreciate your efforts, it just reflects the assumption that you are doing so without oversight and guidance from the manufacturer.

Waynemarlow   100 W

100 W
Posts: 223
Joined: Jul 09 2016 7:22am
Location: Bucks, England

Re: New "TSDZ2 Torque Sensor Central Motor"

Post by Waynemarlow » Aug 14 2018 2:04pm

Guys,

Reprogramming and development can be the very reason those who cannot understand this side of Ebike technology, will switch off from this thread. Perhaps one of us should just set up a new thread titled perhaps "TSDZ2 development, software and hardware " and that would leave the basic long thread with the more mundane stuff the less nerdy among us, can understand.

PS I think we can develop this little engine into something that is very much better than the standard, but we are going to need to have a supply of made up leads and software a little more user friendly than we have at present for the less knowledgeable to really participate, I think we must be getting close to that situation and it reminds me of the early days on the Bafang threads.

casainho   100 MW

100 MW
Posts: 2793
Joined: Feb 14 2011 2:43pm

Re: New "TSDZ2 Torque Sensor Central Motor"

Post by casainho » Aug 14 2018 2:49pm

feketehegyi wrote:
Aug 14 2018 1:27pm
The parts for the bluetooth module have arrived. I've built them together and put into a box.
Please see my video about. This is still for stock factory firmware.
Nice box!!

Look, I have one idea: LCD3 + Bluetooth.

1. LCD3 already has 5V power supply, so would simplify your hardware.
2. LCD3 already connects to the motor controller.
3. LCD3 already receives and stores system variables, the same you are doing.
4. LCD3 enclosure seems to have space inside for the Bluetooth module.
5. LCD3 already turns on/off the power supply/motor controller.
Developer of the Flexible OpenSource firmware for EBike motor controllers (TSDZ2 and KT) and LCDs (KT-LCD3 and Bafang 850C color LCD).

If you like my work, please consider making a donation. I am being using the donations to buy needed resources for my developments. My paypal: casainho AT gmail.com.

AZUR   10 W

10 W
Posts: 83
Joined: Apr 05 2018 6:52am

Re: New "TSDZ2 Torque Sensor Central Motor"

Post by AZUR » Aug 14 2018 3:56pm

feketehegyi wrote:
Aug 14 2018 1:27pm
This post is a bluetooth android display hack. Skip it if you don't like hacking TSDZ2.

The parts for the bluetooth module have arrived. I've built them together and put into a box.
Please see my video about. This is still for stock factory firmware.
Very good feketehegyi.

Nice work. With Casainho firmware, that Jbalat says its far better then the original firmware, we have, maybe, the perfect marriage.

Look that Jbalat did already ride more then 5000 km with the TSZD2 original firmware and he is maybe the better Casainho firmware tester.

And jbalat has not only 5000 km done with TSDZ2.

For comparing he has already, since May, an Trek Powerfly 5FS bike, with a good Bosch motor.
trek2.jpg
Thanks

Kisazul   100 mW

100 mW
Posts: 44
Joined: Jun 18 2018 10:32pm

Re: New "TSDZ2 Torque Sensor Central Motor"

Post by Kisazul » Aug 14 2018 5:47pm

John and Cecil wrote:
Aug 14 2018 11:21am
Kisazul wrote:
Aug 13 2018 7:37pm
Hi guys. I wrote above that I had a problem with the motor. It looked like this. If I pedal without the support of a motor, then the bike perfectly rides off my leg strength. So the clutch that transmits the torque from the pedals to the wheel works. But when I turn on the motor support, volume when the pedals turn with effort I hear a slipping clutch. Probably those who have a blue gear with excessive load will fail the teeth of the gear. But I have a metal gear. Inside this gear is a clutch that transmits the torque from the motor to the wheel. And also when we do not use support, this clutch opens the pedals and the motor so that we do not waste energy on its rotation. I replaced the Chinese clutch and another bearing on a quality Japanese. The replacement is not difficult! As a result, I did it, the engine perfectly digests 800-900 watts. I use a 52v battery. My weight is 90 kg.
I hope my information will be useful to someone!

Do the made in Japan parts use the same part numbers, and if not do you have the part numbers for the more robust parts? I am still running the plastic gear but depending on when and how it fails I may upgrade to a metal gear. If I can get 4 - 5k miles out of the original before failure then it is probably worthwhile to replace it with another plastic gear, but if it only lasts less than 3k miles I will replace it with a metal gear. So far we only have 750 miles on it so we have a long way to go!
Hi. Replacement of the clutch probably only makes sense for bronze gear. In a plastic gear faster fails teeth. I ordered new details on the same numbers that were taken. But visually can see the difference in the manufacture, bearing and clutch have more dense landing on the shaft. Number of clutch to replace: (HF1216), number bearing to replace:(609)

User avatar
feketehegyi   1 W

1 W
Posts: 53
Joined: Jul 11 2018 6:18am

Re: New "TSDZ2 Torque Sensor Central Motor"

Post by feketehegyi » Aug 14 2018 11:51pm

casainho wrote:
Aug 14 2018 2:49pm
feketehegyi wrote:
Aug 14 2018 1:27pm
The parts for the bluetooth module have arrived. I've built them together and put into a box.
Please see my video about. This is still for stock factory firmware.
Nice box!!

Look, I have one idea: LCD3 + Bluetooth.

1. LCD3 already has 5V power supply, so would simplify your hardware.
2. LCD3 already connects to the motor controller.
3. LCD3 already receives and stores system variables, the same you are doing.
4. LCD3 enclosure seems to have space inside for the Bluetooth module.
5. LCD3 already turns on/off the power supply/motor controller.
Thanks! Well, actually in my plan it should work without LCD3.
Of course it can work alongside with the LCD3 as well if someone prefers that way.

Rydon   100 W

100 W
Posts: 129
Joined: Aug 11 2016 5:26pm

Re: New "TSDZ2 Torque Sensor Central Motor"

Post by Rydon » Aug 15 2018 2:10am

casainho wrote:
Aug 14 2018 6:10am

1.2 - I am thinking in put that limitation only when you press the pedals and ebike wheel speed is 0, is that ok?
Wheel speed is not ideal because there is only one pulse per revolution and that is highly variable with pedaling based on what gear you are in. There are a lot more PAS pulses. Why not use that? I may be missing something.

KD5ZXG   10 W

10 W
Posts: 66
Joined: Jul 13 2018 10:02pm

Re: New "TSDZ2 Torque Sensor Central Motor"

Post by KD5ZXG » Aug 15 2018 7:06am

Pedal cadence alone won't reveal when the bike is cruising on inertia
vs stopped. To tell the difference you must also observe the spokes.

Battery Power is Volts * Amps. Volts are steady, we modulate Amps...
Human Power is Torque * Cadence. Both may vary over a wide range.

Trouble is: Zero times anything is zero. Can't scale up pedal power
proportionately from a dead stop, when cadence is zero. Nor while
cruising at zero torque with pedals turning slower than the motor.

With a legitimate human input indicating the need for assistance,
but little or no calculated human power, assistance drops to zero.
Unless we fudge those rules somewhat.

There have to be assumed minimums for each factor. Perhaps not at
the same time, cause it would never agree to stop. But a sufficient
measurement of either should imply at least a small amount of the
other missing factor.

And that's where we need a little extra safety to prevent run-away.
We would like to stop and rest with one foot on the pedal, without
it mistaking that for GO. Spokes tell when to ignore a small torque.

Not saying anything works that way right now, only suggesting a
simple and safe way to get what we are after. Or I could be the
one missing something, it happens a lot...

User avatar
John and Cecil   1 kW

1 kW
Posts: 473
Joined: Mar 14 2018 7:26pm

Re: New "TSDZ2 Torque Sensor Central Motor"

Post by John and Cecil » Aug 15 2018 9:14am

What completely eludes me is why so many people seem to feel that the motor needs to be more responsive with the assist mode. I would think that anyone caring about precise and immediate responsiveness should be using a throttle, which is in fact a device designed to be responsive and offer control in all situations. Reprogramming the assist to take off from a stop is like trying to remove a torque converter from a car with an automatic transmission, it is not responsive for a reason and that reason is safety and reliability/durability.

The key word here is assist: verb - To help (someone), typically by doing a share of the work.

The motor was designed by the engineers to "assist" a pedaling rider. The rider is doing the work with his/her legs (or hands for people with disabilities), and the motor is "helping". It seems to me that many wish to convert the assist into a throttle, but the torque sensor is not an accurate tool. If you want an accurate and responsive tool for powering the motor then you should consider using a throttle.

Perhaps I too am missing something, but some of this (to me) defies common sense.

Rydon   100 W

100 W
Posts: 129
Joined: Aug 11 2016 5:26pm

Re: New "TSDZ2 Torque Sensor Central Motor"

Post by Rydon » Aug 15 2018 12:19pm

John and Cecil wrote:
Aug 15 2018 9:14am
What completely eludes me is why so many people seem to feel that the motor needs to be more responsive with the assist mode. I would think that anyone caring about precise and immediate responsiveness should be using a throttle
What completely eludes me is that someone who would ask that question would even be riding a TSDZ2 and participating in a TSDZ2 forum. That is a question I would expect from someone that is overvolting a hub motor or buying a Pedego or ebikekit hub motor kit (that until recently did not even have pedal assist) or from a motorcyclist but not from someone that loves to ride bikes.

The TSDZ2 is for people that love cycling and like the feel of pedaling a bike. Whether they want to go further faster, flatten hills or need to keep up with stronger cyclists they really don't want to feel like they are riding a motorcycle, which is what a throttle does. They also typically like the natural feel of a torque sensing motor because it responds to the effort applied to the pedals and just feels like a bike on steroids. The important thing is that it feels like a bike.

I actually like having a throttle but I only use it in emergencies when I need immediate power to be safe around cars or when I need instant power and I am in a low pedal assist mode. I rarely use turbo mode.

I also think you underestimate the responsiveness of the TSDZ2 torque sensor. The problems are in the firmware and Casainho et al are taking care of that. :)

User avatar
John and Cecil   1 kW

1 kW
Posts: 473
Joined: Mar 14 2018 7:26pm

Re: New "TSDZ2 Torque Sensor Central Motor"

Post by John and Cecil » Aug 15 2018 1:08pm

Rydon wrote:
Aug 15 2018 12:19pm
John and Cecil wrote:
Aug 15 2018 9:14am
What completely eludes me is why so many people seem to feel that the motor needs to be more responsive with the assist mode. I would think that anyone caring about precise and immediate responsiveness should be using a throttle
What completely eludes me is that someone who would ask that question would even be riding a TSDZ2 and participating in a TSDZ2 forum. That is a question I would expect from someone that is overvolting a hub motor or buying a Pedego or ebikekit hub motor kit (that until recently did not even have pedal assist) or from a motorcyclist but not from someone that loves to ride bikes.

The TSDZ2 is for people that love cycling and like the feel of pedaling a bike. Whether they want to go further faster, flatten hills or need to keep up with stronger cyclists they really don't want to feel like they are riding a motorcycle, which is what a throttle does. They also typically like the natural feel of a torque sensing motor because it responds to the effort applied to the pedals and just feels like a bike on steroids. The important thing is that it feels like a bike.

I actually like having a throttle but I only use it in emergencies when I need immediate power to be safe around cars or when I need instant power and I am in a low pedal assist mode. I rarely use turbo mode.

I also think you underestimate the responsiveness of the TSDZ2 torque sensor. The problems are in the firmware and Casainho et al are taking care of that. :)
Hub motors do not climb steep hills, perhaps you did not know that. A lot of the participants here are more concerned with safety and reliability, thus why so many are leaving this thread. (Perhaps only a handful have said they were leaving, but I wonder how many more left without saying a word.) Perhaps you also did not know but the tsdz2 is "defective" due to improper gear mesh, and also how these performance "fixes" actually are making the defects worse.

I would suggest that those that wish to remove the safety features from this device get a really good liability waiver and put it on their download links, although even that cannot protect oneself completely from a lawsuit. When some 14yo kid decides to upgrade his firmware and the bike tosses him out into traffic due to a lack of pedal safety threshold and he is killed the lawsuit could be for a lifetime's worth of earnings...

casainho   100 MW

100 MW
Posts: 2793
Joined: Feb 14 2011 2:43pm

Re: New "TSDZ2 Torque Sensor Central Motor"

Post by casainho » Aug 15 2018 2:23pm

KD5ZXG wrote:
Aug 15 2018 7:06am
With a legitimate human input indicating the need for assistance,
but little or no calculated human power, assistance drops to zero.
Unless we fudge those rules somewhat.
Good explanation, it did help to me review all this.

So, when pedal cadence is < 25 I am only using torque sensor signal -- this is the way I "fudge those rules".
KD5ZXG wrote:
Aug 15 2018 7:06am
And that's where we need a little extra safety to prevent run-away.
We would like to stop and rest with one foot on the pedal, without
it mistaking that for GO. Spokes tell when to ignore a small torque.
Ok, I am looking at the spokes (wheel speed) to ignore when I am resting on pedals and yes, I think is a good idea also to ignore that initial torque for safety reasons. BUT I will put that as optional and maybe as default.

I hope we can keep all this flexibility in terms of options on the firmware, so it can be easily configured for different applications.
Developer of the Flexible OpenSource firmware for EBike motor controllers (TSDZ2 and KT) and LCDs (KT-LCD3 and Bafang 850C color LCD).

If you like my work, please consider making a donation. I am being using the donations to buy needed resources for my developments. My paypal: casainho AT gmail.com.

casainho   100 MW

100 MW
Posts: 2793
Joined: Feb 14 2011 2:43pm

Re: New "TSDZ2 Torque Sensor Central Motor"

Post by casainho » Aug 15 2018 2:41pm

John and Cecil wrote:
Aug 15 2018 9:14am
What completely eludes me is why so many people seem to feel that the motor needs to be more responsive with the assist mode. I would think that anyone caring about precise and immediate responsiveness should be using a throttle, which is in fact a device designed to be responsive and offer control in all situations. Reprogramming the assist to take off from a stop is like trying to remove a torque converter from a car with an automatic transmission, it is not responsive for a reason and that reason is safety and reliability/durability.

The key word here is assist: verb - To help (someone), typically by doing a share of the work.

The motor was designed by the engineers to "assist" a pedaling rider. The rider is doing the work with his/her legs (or hands for people with disabilities), and the motor is "helping". It seems to me that many wish to convert the assist into a throttle, but the torque sensor is not an accurate tool. If you want an accurate and responsive tool for powering the motor then you should consider using a throttle.

Perhaps I too am missing something, but some of this (to me) defies common sense.
Was hard for me to interpret your message. This is my view after reading your message a few times:

1. this motor and system is ready to be used with a fast control/be responsive, as you say, with the use of throttle so I see no problem to try get the same fast control/responsive using the torque sensor

2. your are right the torque sensor is different from the throttle -- from the tecnhical information we have, torque sensor electric signal is slower to change compared to throttle electric signal. But also, torque sensor includes the PAS signal that throttle do not has.

3. you say that "the torque sensor is not an accurate tool. If you want an accurate and responsive tool for powering the motor then you should consider using a throttle." and again, this is true that torque sensor is less accurate and responsive than the throttle but the true is that it seems to not matter because our legs are not very accurate, the level of accuracy of this torque sensor is very good and we are even talking in reducing it because of safety. We know that with our OpenSource firmware we can get the full range of the torque sensor, we know that is very sensitivity.
With throttles I have, I can't be very accurate with my thumb even if the throttle has that accuracy.

4. "The motor was designed by the engineers to "assist" a pedaling rider." I think we know by now that Tongsheng (as many other companies) put this product on the market probably very fast without developing some parts that were always on EBikes and we miss them -- see the LCD missing battery voltage indication for instance. So I think we can't say the engineers and the company did the development that some of us need/want. But no one stop us to do it by our self :-)
Developer of the Flexible OpenSource firmware for EBike motor controllers (TSDZ2 and KT) and LCDs (KT-LCD3 and Bafang 850C color LCD).

If you like my work, please consider making a donation. I am being using the donations to buy needed resources for my developments. My paypal: casainho AT gmail.com.

User avatar
John and Cecil   1 kW

1 kW
Posts: 473
Joined: Mar 14 2018 7:26pm

Re: New "TSDZ2 Torque Sensor Central Motor"

Post by John and Cecil » Aug 15 2018 2:51pm

Without a doubt Tongsheng rushed this motor to market without enough testing, revisions, and re-design work! I wish the Lingbei motor was available at the time, I think I would have purchased that motor instead had it been out at the beginning of the year.

However, jest because a throttle is fast does not mean that a torque sensor should be just as fast. Try to remember that some countries have banned throttles. The reason why throttles are banned is partially because they react too fast and are deemed to be dangerous.

Drum   100 W

100 W
Posts: 247
Joined: Dec 29 2013 1:20pm
Location: Otaki New Zealand

Re: New "TSDZ2 Torque Sensor Central Motor"

Post by Drum » Aug 15 2018 3:32pm

Hi John & Cecil,

It is obvious that different people ride different bikes in different terrain, and the requirements of each will be different.
Just because I want instant response to an increase in pedal torque, and somebody else thinks a gentler / slower ramp up of response is better doesn't make me right and them wrong (or vice-versa).
We are both right .. for our personal requirements.

It seems to me that Casainho and the others are trying to make the control of the TSDZ2 unit adjustable, to allow a wide range of users to tune the unit to suit their requirements, so we can all have a torque sensing mid-drive kit that is close to "right" for each of us.
For those who are happy with the unit as delivered, great!

Regarding you comment "I would suggest that those that wish to remove the safety features from this device ..", I think that's a bit harsh and a bit previous.
What we are watching is a "warts and all" development process, which is something we don't usually get to see.
Usually we buy a finished product, and do not get to know about all the failed prototypes, modifications, reworks etc that went on in the lead-up to the finished product. There would be very few design / development projects where the finished product was exactly the same as the first drawing.
Some of the suggestions that have been made could perhaps lead to "interesting" situations if applied exactly as first suggested, but comments / suggestions from others refine / modify the idea and it moves forward in a better direction.

Sure, don't put an early version of the open source firmware on a bike to be ridden by a young, elderly or inexperienced rider, but I would expect the finished product to be acceptably safe, given the competence of the people working on the project.

I think it would be great if this thread could be split into "Issues with the standard TSDZ2"; and "Improving the Firmware", and I can understand people who are not interested in the development side getting frustrated. Does anyone know whether it is possible to retrospectively split a thread?

casainho   100 MW

100 MW
Posts: 2793
Joined: Feb 14 2011 2:43pm

Re: New "TSDZ2 Torque Sensor Central Motor"

Post by casainho » Aug 15 2018 4:23pm

John and Cecil wrote:
Aug 15 2018 1:08pm
A lot of the participants here are more concerned with safety and reliability, thus why so many are leaving this thread. (Perhaps only a handful have said they were leaving, but I wonder how many more left without saying a word.)
Do you have numbers or is a feeling?? And how many new participants are joining??
John and Cecil wrote:
Aug 15 2018 1:08pm
Perhaps you also did not know but the tsdz2 is "defective" due to improper gear mesh, and also how these performance "fixes" actually are making the defects worse.
I understand you are not a professional developer of electric motor control as me and that is why you don't understand FOC. Also you didn't view the TSDZ2 original firmware as I did and I think original firmware does not implement FOC. Maybe that is why you are so negative about all this and call "performance fixes". But I hope that at least you can understand that with our OpenSource firmware, unlike on "locked" original firmware, we can control very important things like max motor phase current, max battery current and PWM duty_cycle ramp up/down and that may let us reduce that defects or make worse as you say -- I think it depends on what we decide to do.
John and Cecil wrote:
Aug 15 2018 1:08pm
I would suggest that those that wish to remove the safety features from this device get a really good liability waiver and put it on their download links, although even that cannot protect oneself completely from a lawsuit. When some 14yo kid decides to upgrade his firmware and the bike tosses him out into traffic due to a lack of pedal safety threshold and he is killed the lawsuit could be for a lifetime's worth of earnings...
You are right, I also think we should put warnings explaining the risks and things like that it voids the warranty, etc. Sure, I will also want to write the strong advantages of our flexible OpenSource firmware over the original one.

All you message seems a bit negative. I prefer to be positive. I hope all this discussion don't make bad feelings between us. And that is why I prefer to do more and talk less.
Developer of the Flexible OpenSource firmware for EBike motor controllers (TSDZ2 and KT) and LCDs (KT-LCD3 and Bafang 850C color LCD).

If you like my work, please consider making a donation. I am being using the donations to buy needed resources for my developments. My paypal: casainho AT gmail.com.

casainho   100 MW

100 MW
Posts: 2793
Joined: Feb 14 2011 2:43pm

Re: New "TSDZ2 Torque Sensor Central Motor"

Post by casainho » Aug 15 2018 4:33pm

Drum wrote:
Aug 15 2018 3:32pm
I think it would be great if this thread could be split into "Issues with the standard TSDZ2"; and "Improving the Firmware", and I can understand people who are not interested in the development side getting frustrated.
A thread TSDZ2 mid drive -- Flexible OpenSource firmware for TongSheng TSDZ2 mid drive motor already exists.

Also development can't happen without opinions/experiences/feedback of regular TSDZ2 users and that is why I like to announce new ideas here, etc. I try to keep deep details to the other thread. Also, sometimes people ask here and so I also answer here.
Developer of the Flexible OpenSource firmware for EBike motor controllers (TSDZ2 and KT) and LCDs (KT-LCD3 and Bafang 850C color LCD).

If you like my work, please consider making a donation. I am being using the donations to buy needed resources for my developments. My paypal: casainho AT gmail.com.

Post Reply