TSDZ2 OSF for all displays, VLCD5-VLCD6-XH18, LCD3, 860C-850C-SW102.

perryscope said:
James Broadhurst said:
I know its technically correct to display imperial temperature in Fahrenheit but Centigrade is so much easier and the KT-LCD3 always reads Centigrade?

:thumb: Would defiantly be nice to be able to choose the temperature in Centigrade when setting all other settings to 'imperial'

Can I be really difficult, and ask for the ability to mix MPH and KPH in there too?
 
martin.shane said:
I will have few of them after CNC process. Cheap. Only Cost of the material for members of this forum.
How long do you think it will last in the 3d printed nylon/carbon ?
I find the 42 a good size for on and off road.. For road only one of my friends has gone for a 46 and the other is using a 52. They would have an improved chain line
 
jbalat said:
martin.shane said:
I will have few of them after CNC process. Cheap. Only Cost of the material for members of this forum.
How long do you think it will last in the 3d printed nylon/carbon ?
I find the 42 a good size for on and off road.. For road only one of my friends has gone for a 46 and the other is using a 52. They would have an improved chain line

3D prints are only for prototyping in this case. Is all about the structure and precision. Of course U can use it but it will last only up to first hard push to the pedals.
Chain line and size.
Size.
Its all depends what wheels are U using in U bike. As I have pointed in the diagram it plays big role. U can achieve the same speed with lower chainring if U wheels are going bigger. Less stress to components and to U legs. Only downside is that if U use 38 chainring U are limited by the construction of the motor case but still U have 6-7 mm offset so same as the 42 from TSDZ2 ( it has offset 6mm)
If U use 42, 46, 52 from aftermarket suppliers U will have 10 mm offset . BUT chain will be very very close to the plastic cover of the motor. If motor is New then ok, but after some time wobble will start as we all know and rubbing will be bigger and bigger.
On 52 or 50 U put a lot of stress to the components. All off them as there is a big high ratio. Look at the diagrams bellow. All for 26' wheels. On bigger wheels it would be much more higer. Look at the last speed for cadence ( I have change cadence for 80 and 90 with 52 chainring and 29+ tires and 90 cadence would probably burn motor very fast )
So , those all all things what U need to consider.
Gain_ratio_42.jpgGain_ratio_46.jpgGain_ratio_52.jpg29wheels_52Speedforcadence.jpg
 
Does Emtb mode or torque sensor mode provide correction from bike speed? If yes, then how I can the correction factor be increased or decreased from bike speed?
 
After using the bike for a long time on various routes, I came to the conclusion that cadence control is not the best solution. A more correct strategy is pedaling torque control plus additional assistance when driving at high speed. With this strategy, you can have one universal mode that will not be too harsh at low speeds and at the same time be able to provide sufficient help during long driving at high speeds.
 
Lii said:
After using the bike for a long time on various routes, I came to the conclusion that cadence control is not the best solution. A more correct strategy is pedaling torque control plus additional assistance when driving at high speed.

What you're looking for is the hybrid mode. Or the power mode with boost.
 
JohnAnanda said:
Lii said:
After using the bike for a long time on various routes, I came to the conclusion that cadence control is not the best solution. A more correct strategy is pedaling torque control plus additional assistance when driving at high speed.

What you're looking for is the hybrid mode. Or the power mode with boost.

Hybrid mode is independent of bike speed. It depends on the pedaling speed, which is not the same thing. The speed of rotation of the pedals cannot depend on the strength of the oncoming air and the load that has increased from this. And the boost function is not suitable for me at all, as it creates dangerous jerks. As I wrote earlier, my wife uses this bike and she gets scared when a light bike behaves like a powerful motorcycle))
 
I am trying to start learning programming for stm8, but I feel that it will take me a lot of time until I can figure out the peculiarities of programming for this controller and in the code. But maybe there is already a ready-made solution? Ready code that uses torque sensor + correction from bike speed?
 
Lii said:
...trying to start learning programming for stm8, but I feel that it will take me a lot of time until I can figure out the peculiarities of programming for this controller ....
On the wiki is a development page for tsdz2 and inside "ebike_app.c" you find the assist modes and some readable comments about the used formula's.
 
Lii, it sounds like what you are after is constant human power mode !!!

So if you are happy to put in 100w of human power then the firmware will keep varying the motor power to maintain 100w at the peddles ?

I wonder how practical this is ? It may need some modifier for lower speeds but would be a fun experiment
 
jbalat said:
Lii, it sounds like what you are after is constant human power mode !!!

So if you are happy to put in 100w of human power then the firmware will keep varying the motor power to maintain 100w at the peddles ?

I wonder how practical this is ? It may need some modifier for lower speeds but would be a fun experiment
Unfortunately, in the existing strategy of modes, I can't get what I need. Because it is not possible to get an increase in power as the speed of the bike increases.
 
Hi guys, I just wanted to share a couple of observations on the TSDZ2 that I've been riding actively for a week now on a full-suspension MTB.

I'm on OSF v20.1C.2 and I love it, I very much appreciate the work done by all the enthusiasts. I did the calibration of the torque sensor - no load, 25kg and full load and the respective ADC values were 142, 238 and 272. While cycling uphill yesterday, at a short rest in eMTB mode I noticed that the motor was providing power with me off the bike. I went into "Technical" and saw the no-load ADC at 153. Decided to try full load and it was 302 - very strange. I can't have gained that much weight overnight. Although I've done the heat dissipation mods with quality pads and paste and temperature is in control, the motor had warmed up quite a bit. I adjusted the offset ADC and the max ADC and went on cycling. At another rest, with the motor having cooled down quite a bit now, I checked the ADC values again and they were close to the initial ones. So temperature does influence materials' properties quite a bit, no surprise. But this greatly defeats the purpose of the calibration, not sure what to suggest. Perhaps brands like Bosch have temperature compensation for this, who knows.

I'm on the SW102, but I've ordered the 860c as the SW102 freezes randomly far too often in the calibrations menus. Happens randomly and can be circumvented by going out of the menu often to save settings, but it happens every single time I go to Torque Calibration and try to scroll past the 25-kg weight ADC value (default 250). This also happens on OSF 1.1.0 randomly. The SW102 for me is a frustration. I've searched for these freezes and it seems I'm not the only one with the SW102 having them, so that's a relief.

Also, this is probably not the best place to share this, but I saw on another board a user who had replaced the oil seal on the chainring adapter with a bearing of the same size - I tried that and it fits perfectly. I think this helps prevent play at the big clutch bearing as it reduces lateral forces on it caused by the imperfect chainline, It may prevent failure, we'll see. It might also help offload some of the shaft bending forces, I'm guessing.

Anyway, I love the TSDZ2 and the OSF, I'd be gladly making a modest donation. Thank you!
 
thepowerof4 said:
Hi guys, I just wanted to share a couple of observations on the TSDZ2 that I've been riding actively for a week now on a full-suspension MTB.
Also, this is probably not the best place to share this, but I saw on another board a user who had replaced the oil seal on the chainring adapter with a bearing of the same size - I tried that and it fits perfectly. I think this helps prevent play at the big clutch bearing as it reduces lateral forces on it caused by the imperfect chainline, It may prevent failure, we'll see. It might also help offload some of the shaft bending forces, I'm guessing.

Interesting, Please can you link to this post? :thumb:
 
perryscope said:
Interesting, Please can you link to this post? :thumb:

Well, I just saw this picture and did the same, it's by mbrusa himself:

http://www.jobike.it/Public/data/mbrusa/202092013442_IMG-20200602-WA0018.jpg

Get e 15x28x7 mm bearing, push out the seal, put the new bearing in - it fits easily into the adapter. Now it's not very tight and what I did was slide an o-ring on the shaft after sliding in the bearing - the pedal pushes on the o-ring and it pushes onto the inner race of the bearing. The other side of the inner race rests solidly on a washer on the shaft. This works great. Since then I've painted my 104BCD chainring adapter and the paint made the bearing fit very tightly.

So I was observing torque sensor ADC values today and as I mentioned, heating up the TSDZ2 results in totally different ADC values. When at room temperature, those were 142, 238 and 272. I can't measure the 25kg ADC while cycling, but the other two at their extremes today were 163 and 311 when I pushed it as hard as I could during a sunny trail hill climb. To me it seems that calibration is pointless, I don't know if it's my unit only, but I doubt that.

Also, I don't think I can actually achieve any calibration, because every time I set the values and exit the config menu, Error 2 pops up - torque sensor. I have to turn off the system and back on again for the error to disappear and I thought that this would actually set the values, since I left the "Calibration-enabled" setting. Wrong, because the default value of the ADC offset is 150 and when there is a push on the pedals above that, unwanted start of the motor occurs, even if the value is set to something much higher like say 163. Which is also strange, because the "no assistance without pedaling" setting is also on.

I really hope that all this is due to the SW102 display. I can only say that at the moment the original 1.1.0 version with the SW102 is a much, much more polished firmware in my case. It also doesn't exhibit the power glitch that has been mentioned here.

I'll be trying v20.1C on the 860c when it's delivered. Again, thanks everyone for all the effort put into this.
 
This is a simple, unassuming calibration.
The purpose of the calibration is to adapt the sensitivity and range of the torque sensor to optimal values.
I have three motors, out of two the maximum difference in torque values ​​between cold and hot is 5, I have no problem with calibration, enabled or disabled it's always good.
On the third motor the difference is similar to yours 22, I had to try several times before finding satisfactory values.
Calibration is not useless, it is precisely in these cases that it is needed, but the data obtained are not to be considered definitive, they can and must be corrected if necessary.

Error 2 occurs if you set a value beyond the threshold of 25 compared to the real value. It is for security reasons, a wrong value of Torque offset, can lead to an undesired start with high power.
Nothing changes between SW102 and 860C, the only difference is that the calibration procedure can be affected by the problems of using the SW102 display.
 
mbrusa said:
Error 2 occurs if you set a value beyond the threshold of 25 compared to the real value. It is for security reasons, a wrong value of Torque offset, can lead to an undesired start with high power.

Well on my system (SW102 / 20.1C2) the error 2 occurs every time I exit the configuration menu, even though the torque ADC offset is set at the real (no load) value. Turning off and on the display clears the error and the values I have set are saved.
It's not a big annoyance since the booting time is so short now..

Are there other values than Torque offset that may trigger this error 2 ?
I have Torque ADC threshold set to 30 for a more responsive start, but even if I lower it to 5 (the minimum available), the error is still there.
If I disable torque sensor calibration, the error does not occur.
 
Beli said:
After applying pop cc at the right place and 10km it seems to be okay :thumb:
Is there a compiled version to test with this last change!
 
JohnAnanda said:
Well on my system (SW102 / 20.1C2) the error 2 occurs every time I exit the configuration menu, even though the torque ADC offset is set at the real (no load) value. Turning off and on the display clears the error and the values I have set are saved.
If I disable torque sensor calibration, the error does not occur.

Yeah, exactly the same for me as well, but I don't think the values are actually saved, because, as I mentioned, the ADC offset setting doesn't seem to take effect. If I set it to say 163, with the default being 150, I still start getting assistance once I push the pedals beyond 150 while still well below 163.

I am not trying to complain and I am extremely appreciative for all the effort put into this firmware. I'll be trying the 860c display, which doesn't seem to cause any issues, or at least I haven't seen such feedback yet.

The vanilla 1.1.0 OSF lacks the eMTB mode, which I love very much for being extremely responsive. To circumvent this, I set a couple of settings to their highest/lowest values for a more responsive assistance reaction of the motor. Works OK. On 1.1.0 I still get the random config menu freeze, but I can go to all menus eventually. That's just life with the SW102. I would not recommend this display, even if one loves the minimalist design as I do.
 
thepowerof4 said:
JohnAnanda said:
Well on my system (SW102 / 20.1C2) the error 2 occurs every time I exit the configuration menu, even though the torque ADC offset is set at the real (no load) value. Turning off and on the display clears the error and the values I have set are saved.
If I disable torque sensor calibration, the error does not occur.

Yeah, exactly the same for me as well, but I don't think the values are actually saved, because, as I mentioned, the ADC offset setting doesn't seem to take effect. If I set it to say 163, with the default being 150, I still start getting assistance once I push the pedals beyond 150 while still well below 163.

If I understood well, the Torque ADC offset is not meant to be the value from which the motor will start. It's simply the "weight" of the pedals alone, so the system has a baseline.
You can fine tune the pressure level from which you will have assist with the Torque ADC threshold value : the higher you set it, the most sensitive the response will be. For instance on my system, if I set it to 60 the motor will start as soon as I put my feet on the pedals. I've set it to 20 so I can comfortably rest my foot on the pedal at at stop and still have a quick assist as soon as I put a little pressure (I have assit without pedalling enabled).

I admit the SW102 is a bit the problem child of the family, but I like it anyway for its uniqueness as it is unobtrusive but still allows exented acces to datas and settings.
 
JohnAnanda said:
If I understood well, the Torque ADC offset is not meant to be the value from which the motor will start. It's simply the "weight" of the pedals alone, so the system has a baseline.
You can fine tune the pressure level from which you will have assist with the Torque ADC threshold value : the higher you set it, the most sensitive the response will be. For instance on my system, if I set it to 60 the motor will start as soon as I put my feet on the pedals. I've set it to 20 so I can comfortably rest my foot on the pedal at at stop and still have a quick assist as soon as I put a little pressure (I have assit without pedalling enabled).

I admit the SW102 is a bit the problem child of the family, but I like it anyway for its uniqueness as it is unobtrusive but still allows exented acces to datas and settings.

I'm currently tinkering with 1.1.0 and I could swear that the threshold works this way - the lower you set it, the faster the motor responds upon a push. I have it at the minimum at the moment - 5, but I did try the higher extremes as well, like 60, 80, and the max of 100. Ill test again to make sure. My understanding is that this is the ADC increase, above which the system responds. So if the idle ADC is 150 and threshold is 20, then you'd have to push to 170+ for the system to respond. If threshold is 5 - then just 155+. But I'll check again.

On v20.1C.2 eMTB mode is very responsive and I just like it sooo much. But it ignores the setting which says that no assistance should be provided without spinning the pedals. That would have been fine if my unit didn't have so varying ADC offset values when cold and hot. Or if calibration actually worked and I could set the ADC offset a little higher to prevent the unwanted start. Also, in this mode, the ADC threshold value doesn't seem to do anything.
 
Hello everyone. I have a VLCD-5 display, I cannot enter the calibration of the torque sensor. I press the light button 6 times, nothing happens. Help me please.
 
@ thepowerof4 @JohnAnanda, from what you write the calibration with SW102 does not work correctly.
Unfortunately, not being able to try, I have to rely on the feedback of those who use it.
It appears that the Torque offset value is saved, but at power up, the default value is always used. I'll try to understand why.
 
mbrusa said:
Atrihalov said:
Hello everyone. I have a VLCD-5 display, I cannot enter the calibration of the torque sensor. I press the light button 6 times, nothing happens. Help me please.
Is "Set parameters on startup" enabled?

I do not know for sure. Does this parameter affect something?
 
Back
Top