New "TSDZ2 Torque Sensor Central Motor"

[Waynemarlow]

Yup I have a theory about building EBikes, like for like motor Watts, for every 1kg under 22kgs will cost £ 1000.00 more.

The really funny thing is that the older 2000 - 2005 era bikes where 27.5 wheels were just arriving, converted with a TSDZ2 engine, will all be under 20kgs, nearer 18kgs some of them ( best we have seen was 17.4 ). Now 18kgs is the holly grail at present of the EBike manufacturers and will come at a premium of around £ 8 - 10K.

[casainho]

TSDZ2 wireless display and wireless remote (including the brake signal)

Yesterday I did the very first ride using a wireless remote and a wireless display!! The advantage is that I can have a clean EBike if I decide to not use the remote and/or the display as they are optional, there are no cables to connect. I can always keep my handle bar clean and use my mobile phone instead of the remote and the display, as I always carry it my pocket.
I also do not install the wheel speed sensor as the EBike speed can be seen on the GPS display because it uses the GPS for calculating the speed.

I am using the 860C display remote buttons but I could also use the VLCD5 remote buttons - and they are connected to the wireless remote box, that blinks that green light when I change the assist level - this is a prototype, so it is not clean as I want on my final installation.
Note that I am also using the Garmin Edge GPS wireless display and I value a lot to see the graph of my heart rate while I am ridding as also the EBike battery charge level as also the assist level.

Project page: https://opensourceebike.github.io



[youtube]v_xF5_GnSXQ[/youtube]

[Manbeer]

would be cool to make some type of wireless interface that can incorporate brake sensors and throttle into it, but then it would be need power i guess. still, a little box up front communicating wirelessly with a couple 14500s or something would still be better than cables run up the frame. with a usb out could even charge up the display in an emergency situation

[casainho]

would be cool to make some type of wireless interface that can incorporate brake sensors and throttle into it, but then it would be need power i guess. still, a little box up front communicating wirelessly with a couple 14500s or something would still be better than cables run up the frame. with a usb out could even charge up the display in an emergency situation

As I wrote, brakes are connected to the wireless remote as they should be very near the wireless remote on the handle bar - I use magnetic brake sensor. And throttle should be ok to be wireless also.

[Manbeer]

would be cool to make some type of wireless interface that can incorporate brake sensors and throttle into it, but then it would be need power i guess. still, a little box up front communicating wirelessly with a couple 14500s or something would still be better than cables run up the frame. with a usb out could even charge up the display in an emergency situation

As I wrote, brakes are connected to the wireless remote as they should be very near the wireless remote on the handle bar - I use magnetic brake sensor. And throttle should be ok to be wireless also.

Oh, I guess I must've missed that part about the brakes. I just figured a lot of times people omit the brake sensors. That's freaking awesome

[Wapous]

Just for references!

It is your design? If so, can you share the source files? because I am pretty sure they can be very useful for others.

And I wish some mechanical guy could improve the motor by reducing the noise - I think the noise comes from the gears, maybe from the main gear and the pinion gear.

I did industrial design for several years. I have been using Solidworks software since 2000.
Obviously I would be happy to offer the reverse engineering drawings free of charge but NOT in the native form.
Those interested only have to ask for the desired format.
Like STEP, DXF, SAT, IGES etc.
Note: The CAD files you see in my mail are not for production but for viewing. But these CAD drawings are faithful to the components of the TSDZ2 engine.

Can you please please share in STEP files as also that images, so I can add to our github and to the TSDZ2 FAQ repair page?

I am sending a step file of an assembly. If you want individual files let me know. Witch program are you using to open the step files?
https://drive.google.com/file/d/1QTW4yI ... sp=sharing
Animations will follow soon.

[Elinx]

I am sending a step file of an assembly.....
Witch program are you using to open the step files?.....

Thank you for this file. For viewing I have used FreeCad.

[casainho]

Thanks, I shared here the files: https://github.com/OpenSourceEBike/TSDZ2_wiki/wiki/FAQ

I want to add the big STEP file but there are some issues with big file storage on github that I will solve. At least there you can see the images.

I am not a mechanical guy but yes, I would use FreeCAD to view the STEP files.

[Elinx]

Thanks, I shared here the files: https://github.com/OpenSourceEBike/TSDZ2_wiki/wiki/FAQ....
..... I would use FreeCAD to view the STEP files.

There are some small (mechanical) differences between colored pictures and step file.
The step file missed the bearings on the axle and torque sensor, but cranks are added.

[casainho]

STLinkV2 different versions: some don´t work for flashing TSDZ2 motor controller firmware

You can buy STLinkV2 on EBay or Aliexpress for very cheap price like 2€ (shipping included) but unfortunately there are different versions of them and some simple do not work well (at least for flashing the TSDZ2 motor firmware).

More information here: https://opensourceebike.github.io/stlinkv2.html

See two version side by side:

[casainho]

I would like to show my TSDZ2 fully wireless and the best wireless display for TSDZ2

I build my own wireless remote and I decided to use a 860C display keypad. As seen on the picture, one cable is the keypad and the other is for the brake sensors (yes, the brake sensors are also wireless!!).
Also note the CR2032 coin cell that gives the power to the wireless remote for about 2.5 years.



And some tape to make it water prof and the water prof Higo connector to connect to the magnetic brake sensors that are installed on the brake levers:



And the remote LED with red light showing that I am using the brakes:



And the GPS wireless display showing the TSDZ2 EBike battery level as also the current assist level number. Also note my heart rate graph, where it is being measured on my pulse by my watch and sent wireless to the GPS wireless display:



If you want to have a clean EBike and/or cheap EBike, the Garmin GPS wireless display and the wireless remote are optional as you can use you mobile phone to see the EBike data and control everything on the TSDZ2.

The cost of materials to make TSDZ2 fully wireless is about 25€ only!! See how to build it for free on the project page: https://opensourceebike.github.io/

The mobile app:

[Wapous]

TSDZ2 motion study video
Hope I did it easier to understand!


https://youtu.be/EsMlBj6ogVQ

[casainho]

TSDZ2 motion study video
Hope I did it easier to understand!

[youtube]EsMlBj6ogVQ[/youtube]

That is really great!! I added to the TSDZ2 FAQ: https://github.com/OpenSourceEBike/TSDZ2_wiki/wiki/FAQ

What would be nice was if you or someone else could develop a modification to reduce the noise from the metal gears!!

[youtube]Pz4yXvuMhXo[/youtube]

[jbalat]

Awesome work Wapous !!

Now to identify any improvements...
Turning the heavy main gear could be a good start... Most of it could be hollowed out on a lathe

[dameri]

Great work Wapous

[Waynemarlow]

Awesome work Wapous !!

Now to identify any improvements...
Turning the heavy main gear could be a good start... Most of it could be hollowed out on a lathe

Somewhere on this thread is a picture of a very hollowed out main gear. Nice but very much on a CNC mill.

I am not sure but there may well be enough room for a belt drive of the final stage, the latest kevlar belts would be able to withstand the power but a lot of work unless we could get a preconfigured gear we could simply shrink on once the teeth had been faced off on a lathe.

[Wapous]

The timing gear (gold color) is an existing model with around 16 teeth. To me, it is obvious that the main gear needs to be replaced with a new machined part. I will study this in more depth this month. Ultimately this modification would be similar to the Bose motor drive mechanism. This modification would make it a quieter engine, surely. One of the constraints I see is to install a mechanism for adjusting the tension of the timing belt. A V belt is to be avoided given the tension on the bearings. An interesting challenge with lots of twists and turns.

[Waynemarlow]

One of the constraints I see is to install a mechanism for adjusting the tension of the timing belt.

I'm not sure you do need an adjuster. If you can get a carbon / Kevlar toothed belt of near enough the correct length, then the amount of stretch over the lifetime of the belt is negliable plus you can ease the belt over the teeth with sufficient tension on a very large gear like we have, to be well within tension limits ( actually a modern toothed belt has very little tension to prevent damage to the shaft bearing ).

When I looked at doing it I worked out you would need to drop 1 tooth on the motor and that would allow a much smaller main gear. I didn't follow up though on any calcs to find the right size of belt.

[jbalat]

Even though the belt will reduce the noise you are still pushing around the big gear and the smaller gear and relying on the freewheel on the nylon gear

Im still looking at a way to run a standard crankset and maybe a centrigigul clutch with a freewheel to engage the cranks ?

just throwing some ideas out there...

[Waynemarlow]

You may want to look at photos of the internals of the Brose engine, not unsimilar to the TSDZ2 in its use of steel and plastic internal gears with just the belt on the final drive, its probably the quietest out there.

[Wapous]

It is easy to understand the problem. I don't know if we can reverse the direction of rotation of the 3-phase motor. The one-way bearing of the blue gear should be reversed.
So my question is: Can we reverse the rotation of the 3 phases motor?

[jbalat]

With sensorless brushless motors you just swap any 2 of the phase wires to reverse I’m guessing you would also need to swap the same 2 sensor wires

[casainho]

It is easy to understand the problem. I don't know if we can reverse the direction of rotation of the 3-phase motor. The one-way bearing of the blue gear should be reversed.
So my question is: Can we reverse the rotation of the 3 phases motor?

Sure we can, with our OpenSource firmware we have full control of the system: motor; sensor inputs and display!!

So, for what I understand, will be needed a machined main gear and pinion gear. The main gear will need to be smaller, so the motor will have less torque. And I hope the inversion of the clutch /one-way bearing on the blue gear can be done.

[Waynemarlow]

Interesting, I hadn’t considered the reversal of the main gear. Torque at the wheel is just gear ratio, as the motor will always start off with X amount of torque. For what you would loose on the main gear ratio would easily be compensated on the outside of the motor.

[casainho]

Interesting, I hadn’t considered the reversal of the main gear. Torque at the wheel is just gear ratio, as the motor will always start off with X amount of torque. For what you would loose on the main gear ratio would easily be compensated on the outside of the motor.

Yes, I do not think that small torque difference will be relevant and reducing the TSDZ2 noise on the gears are my top wish!! -- TSDZ2 still is the best mid drive motor and will continue to be!!