New "TSDZ2 Torque Sensor Central Motor"

I really don't think anything needs to change with this thread. People can follow along as much as they are able and give input when they feel like it. A small group of really smart people are doing the deep thinking and its good that this thread brings them together for brainstorming. We have to expect not everyone can keep up with the technical details. It is still useful for these "smart" guys to hear how the rest of us dummies are using our bikes and what we expect of them. Development is a difficult process and its working very well here.

Anyone can and should start another thread titled "TZDS2 installation and use",... or. " TZDS2 mechanical issues" or even,.. TZDS2 for dummies.

I'm very interested in the open source firmware and it's one of the reasons I bought the tsdz2.I want only to thank the developer for their wonderful and free work! I ordered the various pieces for the installation and as soon as they arrive I will try to install it even if I I'm a noob.
thanks all and scuse for my bad english!!!

Since too many are throwing themselves into an intersection
for want of non-firmware topics: I need to install mine on an
Electra cruiser with large diameter bottom bracket. The width
of my bracket is OK close to 68, just a bigger hole than 30mm.

So, your experience is that it snugs up anyway and not worry?
Do I need an offset bracket adaptor, or fill the empty cavity
by pounding in a bunch of chopsticks?

My TSDZ2 is still in the box for lack of compatible rear brake.
Sturmy 5 speed drum was never delivered. Motor hasn't been
priority. My technical comments come from prior experience
testing motor controllers for TI, not yet hands-on with TSDZ2.
Don't even know the proper way assemble this damn thing.

I may test on my 7 speed derailed Mongoose for now, as the
bracket seems close to proper fit. My Mongoose has V brakes.
Don't much care for the Mongoose as it's just way too normal.
Dumbed down for the consumption of $99 Wallmart normals.

But no, eventually my Ratrod must have this motor. The pic
with motor is someone else's rat. The rat with Sunlite Double
Springer and no motor is mine. I'm thinking 36V battery in a
low slung ammo box pannier on the non-chain side, strangely
the box doesn't want to fit either hole of my frame...

Just out of curiosity is anyone else using the opensource firmware ? We need your input to get this spot on

My thoughts yesterday I had a small runaway..
It was moving at snail pace after I took my feet off after stopping at lights. I was able to arrest it with brakes. To stop it i had to peddle backwards. Downside if you had a blue gear then you probably would have munched it !

Moral of the story.
The firmware is still beta. Only use it for yourself to be able to provide feedback to casainho !!
If you want to be super safe then only use it with cutoff brakes.

New firmware with option that waits for you to move pedals.
This does not work and is very frustrating. I had to peddle 3 revolutions to get the motor to cut in. I'm thinking it was either on a timer or waiting for a speed input from the wheel sensor ?
Suggestion: Use the pas magnet pulse in combination with torque to take off from rest. This way you have to peddle at least 15 degrees before the bike moves. It will also allow a smaller or no torque offset to get the bike moving. I like a torque offset of 2. I find the current offset of 4 too high.

Cadence
I have confirmed that cadence reading on the display is too low by a factor of 1.3

KD5ZXG
I would make sure it is snug in the BB. Make up a sleeve so it doesnt move around

I've riden quite a few ebikes and currently have my TSDZ2 converted hardtail and a haibik Sduro yamaha pw-se bike. The yamaha Is great with an almost instant on power when the pedals rotate. Reading the Yamaha blurb it uses the three sensor to determine powe levels. The three sensors being cadence, speed and torque. I'm not sure the algorythm but happy to contribute any obesrevationa to help development of the open source firmWare. I'd really like to be able to make the TSDZ2 feel similar to the yamaha.

I don't have any bad feelings about you dude. I just don't particularly care for your firmware, it is nothing personal. I am sure your firmware is wonderful for dirt racers and the like, but perhaps not so good for average riders and youngsters that might end up in a wheelchair if they are not careful.

Bafang (or the hackers/ secondary programmers) did it right, they built an easy to use interface so people could set their own settings in a simple interface. If you notice that firmware modification does not come preloaded with dangerous settings that make the slightest pressure on the pedals make the bike take off - especially on pas only models without e-brakes. The only way to make them dangerous (if it is even possible) is to download the interface with safe settings and then play around with the settings yourself.

I am not a programmer, but I did manage an engineering and surveying office for a few years and I have seen my share of lawsuits. When someone gets hurt, especially a minor, they will pull everyone, and I mean everyone, into court. It happens all the time, especially in multi million dollar wrongful death suits. If the settings are dangerous to begin with I think an attorney would have a difficult time defending a lawsuit, especially since Tongsheng will likely be named as well and their defense will be their motor is safe but the modification made it dangerous.

Hi John and Cecil,
I think you are being too harsh.
The new firmware is totally unbelievable. I cant be happier. If you ride with it it will give you a smile on your face for sure.

There are some bugs to make it a bit safer but we are very close to sorting that out..

What Im saying is don't say you dont like it until you have tried it

Hi guys,I'm from Czech republic but I can't speak English so I use translator and I would like to use your firmware for motor 48v 350w, but I'd like to keep the LCDV 5.Can you please fix

ri53hu said:

I would like to use your firmware for motor 48v 350w, but I'd like to keep the LCDV 5.Can you please fix

Click to expand...

Is not possible to use original LCD.

Hi casahino,

I understand the current beta firmware does not support other the lcd, but what would be necessary to add support?
Is my understanding correct, that the LCDV5 simply sends and receives serial data at 9600 baud?

DL2ZZ said:

Hi casahino,

I understand the current beta firmware does not support other the lcd, but what would be necessary to add support?
Is my understanding correct, that the LCDV5 simply sends and receives serial data at 9600 baud?

Click to expand...

Original LCD can't be programmed and is also weak in terms of features. Would never support the standard features (like showing the battery voltage) or the new advanced features like adjusting battey low voltage cut off, battery power limit, etc.

I should put this question on FAQ because I think is a permanent question for new users.

jbalat said:

Hi John and Cecil,
I think you are being too harsh.
The new firmware is totally unbelievable. I cant be happier. If you ride with it it will give you a smile on your face for sure.

There are some bugs to make it a bit safer but we are very close to sorting that out..

What Im saying is don't say you dont like it until you have tried it

Click to expand...

I will wait and see but the direction the firmware has taken does not sound like something I want anymore. At first it was adjustable voltage and amps and added display features which was nice, but wanting power to come up with just some weight on the pedals - I will pass. If anything I would like to ease back the responsiveness of the stock unit. I am a 50 yo man riding around with his 25lb dog on the back of the bike. When in doubt I would prefer the motor to not kick on, and when it does I would prefer the power to ramp up more gradually. I would prefer to protect the blue gear (motor defect) by reducing low to mid power while maintaining upper rpm power and increasing max rpms. The firmware direction is the exact opposite of what I would prefer.

I am watching the Lingbei thread. I think that motor will be much better once they get the firmware sorted out and the reliability tested. Hopefully it is not designed with defect(s) too. It is able to handle more power (and they are adding additional mosfets in the future), it has a torque sensor, and perhaps a properly meshed gear drive.

John and Cecil said:

jbalat said:

Hi John and Cecil,
I think you are being too harsh.
The new firmware is totally unbelievable. I cant be happier. If you ride with it it will give you a smile on your face for sure.

There are some bugs to make it a bit safer but we are very close to sorting that out..

What Im saying is don't say you dont like it until you have tried it

Click to expand...

I will wait and see but the direction the firmware has taken does not sound like something I want anymore. At first it was adjustable voltage and amps and added display features which was nice, but wanting power to come up with just some weight on the pedals - I will pass. If anything I would like to ease back the responsiveness of the stock unit. I am a 50 yo man riding around with his 25lb dog on the back of the bike. When in doubt I would prefer the motor to not kick on, and when it does I would prefer the power to ramp up more gradually. I would prefer to protect the blue gear (motor defect) by reducing low to mid power while maintaining upper rpm power and increasing max rpms. The firmware direction is the exact opposite of what I would prefer.

Click to expand...

Seems you didn't understand, but I am trying to make the firmware flexible were all that key parameters that you mention can be adjusted by users.

Do you think you will get that freedom from any manufacturer?

casainho said:

Seems you didn't understand, but I am trying to make the firmware flexible were all that key parameters that you mention can be adjusted by users.

Do you think you will get that freedom from any manufacturer?

Click to expand...

Keep up the good work casainho
The default TSDZ2 setup is ok, but your improvements sound like they will greatly improve the riding experience and provide the ability to monitor all the power related values.

Is one able to use the XH18 with the LCD3?

Hopefully I'll be able to order the parts necessary next month

aja said:

casainho said:

Seems you didn't understand, but I am trying to make the firmware flexible were all that key parameters that you mention can be adjusted by users.

Do you think you will get that freedom from any manufacturer?

Click to expand...

Keep up the good work casainho
The default TSDZ2 setup is ok, but your improvements sound like they will greatly improve the riding experience and provide the ability to monitor all the power related values.

Is one able to use the XH18 with the LCD3?

Hopefully I'll be able to order the parts necessary next month

Click to expand...

Only LCD3.

You guys may remember when I told that my girlfriend was using half of the watts hour on the same trip as me. On that time I always thought that was strange... Now, after I exchanged my motor for a new one because the old one was demagnetized, our watts hours for the same trips are almost equal!! So, my conclusion is that my motor was demagnetized already some time ago!!

While I don't develop the automatic motor max temperature management, I am limiting the current to 10 amps continuously and I see the motor just get warm.

Just a little bit of feedback from yesterday when I swapped bikes with my friend. He also has 36v motor like mine. He couldn’t believe the difference and now wants to convert his motor too but he likes his xh18 so I don’t know if he is willing to part with it.

We spent a long time looking through the code after we got home to see how we can fix the problem ( that you can’t rest your leg on the peddles when you are stopped) and it looks like there is only one line of code we need to change so we will be testing over the next couple of days !!

When I jumped on his bike I could immediately feel that the power seemed very restricted, it was dropping off with higher cadence and it forced me to change gears way before I had to. He was powering up ahead pulling away with a smile on his face.

His comments were ‘this thing goes better than your Trek* !!

Jbalat, you've got me really interested now!!! I think I'm going to put aside my prejudice of the KT-LCD3 for a bit I gotta try it (as soon as I'm headed, I'm laid up right now).

Casianho, so I wonder, did the slight burning of the motor actually cause any power loss? What in the past could have caused the demagnetization? Too bad there was not a thermal readout to determine exactly how hot it got... I don't know what would cause it to be demagnetized ober than the heat. Any ideas?

Good, I'm trying to tell you guys how good it is but you are not going to know unless you try it yourself.

eyebyesickle said:

Jbalat, you've got me really interested now!!! I think I'm going to put aside my prejudice of the KT-LCD3 for a bit I gotta try it (as soon as I'm headed, I'm laid up right now).

Casianho, so I wonder, did the slight burning of the motor actually cause any power loss? What in the past could have caused the demagnetization? Too bad there was not a thermal readout to determine exactly how hot it got... I don't know what would cause it to be demagnetized ober than the heat. Any ideas?

Click to expand...

I wrote in that page what I think caused the demagnetization.

I always did run my motor at max power and highest gear on my ebike, for long rides like 1h.

If anyone is interested I just finished Part 4
[youtube]0nXtaJniZQc[/youtube]

Guys, we need to be a bit more pragmatic about what power we can continuously pull over time extended periods out of the TDSZ2. Most manufacturers of electric engines have a max power period and a working power maximum. Often engines can be specced at 50% or even less of max power as their working power. Those engines that can work continuously at or near max power typically have huge fans and heat sinks and are typically horrendously expensive.

The TDSZ2 is not in that game, has no effective heat sinks and no internal fans and to make it worse, is fully waterproof meaning there is no exchange of air trapped inside ( warm air is quite destructive to the internal windings ), if you are going to run these little engines at full power for extended periods then you are going to get these occasional burnt out windings. By all means use full power but do limit the time that is happening. This is no different than probably 95% of all electric engines out there.

We probably have a pretty good grasp already of what the engines can handle power wise and it would also be relatively easy for the likes of Casainho's program to monitor Watts against time and put some sort of cut back to safer wattage after a few minutes of say full power.

Waynemarlow said:

Casainho's program to monitor Watts against time and put some sort of cut back to safer wattage after a few minutes of say full power.

Click to expand...

I am being writing about this on the firmware development thread.

I want to go by reading directly the motor case temperature, gluing a temperature sensor inside the TSDZ2 case and outside of the motor case.
I want to reduce linearly the electric current when temperature rises (also ahow in real time on LCD3). This way we may even push the motor to higher values than the stock 18 amps limit.

Yes, that means more wire connections to the motor controller but that is the way to get top performance using top technology. I wish shops can sell this upgrades to users. But also temperature hardware and measurements should be optional.

You should be able to monitor your temp rises against Watts output and also see the time for the casing to cool once the power has been reduced, from that you must be able to work out what most units can handle and build into your software some sort of rough approx. safety factor.

By the way Casainho thanks for all the time you are putting in.

Do we know any young software students amongst us who perhaps we could all chuck a few dollars at, to create an interface program for the less computer literate. Certainly the one that I know of for the Bafang works extremely well.

Waynemarlow said:

Do we know any young software students amongst us who perhaps we could all chuck a few dollars at, to create an interface program for the less computer literate. Certainly the one that I know of for the Bafang works extremely well.

Click to expand...

That sounds like a winner right there!!! I've already got the tentacles out here locally for someone to make this in an APP version, but the computer software may be even more universal for now.

What I would really like is all the capabilities and ability to switch settings, with the ability to use any display, similar to the BBS... That would be nice...

I realize that as of now the displays need to be modded to fit the opensource firmware, and not all displays could even support all the features, but I wonder how we could modify the opensource to be compatible with all of the stock displays... (Keeping same code for the communication with the display, is this possible?)

I would settle for the stock display screens with no additional settings in sight, if we could keep all the customize able settings function in the background, even if they had to be accessed with a programming cable/PC. I'm sure that would include hefty modifications to the code, but then all anyone would need for the upgrade would be a programming cable and the software...

Sure, we could not see anything more on the screen than usual, but as long as we have the improved motor function and the ability to still customize the new settings in the background with a programming cable and a PC with software comparable to bafangs, then almost anyone who wanted to could use the new firmware/upgrades. As of now, the KT-LCD3 flashing process severely limits the user base.

I'm assuming this is a possibility with the right coding? To keep the function of the new firmware in he background, but have compatibility with any display? So one would only have to flash the new code to the motor, and it keeps the same communication with the displays as the original firmware?

If the above is a possibility, I don't mind putting someone on it... I just don't have the resources to pay someone only for them to conclude this is not possible - resources squandered...

So, is this worth pursuing?

Also casianho, FYI, this is not in any way critical of your work, I only support it - my end game is just to make this more available to the masses, it really is a game changer for this motor, I know.


Also, side note... I looked but I don't see if you ever found where the motor cuts the temp from, or if you were sure... Because I do notice it does cut the temperature when too hot (maybe not soon enough, although mine didn't seem to suffer after cutout) and there is a overtemp protection fault code... So just wondering about that as I see more temp protection talk..