TSDZ2 open source firmware only for KT-LCD3 (v0.19.0 / v.0.20.0beta1)

mctubster said:
The 850c code is still missing the polish of the LCD3, lots of missing bits, odo, soc, light flashing and so on.
You need to read the documentation!! Odometer and battery SOC are surely available and working very well.

Light flashing is not implemented because I do not consider a priority comparing to make SW102 firmware at pair with 850C.
 
mctubster said:
1 - Measured voltage is too high. Likely due to a change buba made to improve the voltage measured but appears there is a constant that is wrong

Can you elaborate on this please?
 
The 0.20 works perfectly with the lcd3! Thanks for the great software! The only thing I don't like is the wrong volt display.

Mfg Michael
 
michih. said:
The 0.20 works perfectly with the lcd3! Thanks for the great software! The only thing I don't like is the wrong volt display.

Mfg Michael

What are you findings and how have you measured them?

OSF is reporting 0.5V higher than what my bluetooth BMS reads, so if yours is the same then try these settings below.

The correction to BATTERY_CURRENT_PER_10_BIT_ADC_STEP_X512 was proposed by Stempelo on Github.
https://github.com/stempelo/TSDZ2-Smart-EBike/commit/5269e27b2b86d0a1a45fd02dc78b11a158532d4e

/src/controller/main.h
Code:
// ADC battery voltage measurement
#define BATTERY_VOLTAGE_PER_10_BIT_ADC_STEP_X512           44
#define BATTERY_VOLTAGE_PER_10_BIT_ADC_STEP_X1000          86

// ADC battery current measurement
#define BATTERY_CURRENT_PER_10_BIT_ADC_STEP_X512           87
#define BATTERY_CURRENT_PER_10_BIT_ADC_STEP_X100           17
 
version 20 beta 1 is practically perfect on lcd03. I think it would be interesting to make it compatible with lcd05, it's cheaper, small but not too much and easy to read and leaves room to put a smartphone on the handlebar.
On lcd03 the voltage is 0.3v higher than the real but it is not a real problem.
 
 
famichiki said:
michih. said:
The 0.20 works perfectly with the lcd3! Thanks for the great software! The only thing I don't like is the wrong volt display.

Mfg Michael

What are you findings and how have you measured them?

OSF is reporting 0.5V higher than what my bluetooth BMS reads, so if yours is the same then try these settings below.

The correction to BATTERY_CURRENT_PER_10_BIT_ADC_STEP_X512 was proposed by Stempelo on Github.
https://github.com/stempelo/TSDZ2-Smart-EBike/commit/5269e27b2b86d0a1a45fd02dc78b11a158532d4e

/src/controller/main.h
Code:
// ADC battery voltage measurement
#define BATTERY_VOLTAGE_PER_10_BIT_ADC_STEP_X512           44
#define BATTERY_VOLTAGE_PER_10_BIT_ADC_STEP_X1000          86

// ADC battery current measurement
#define BATTERY_CURRENT_PER_10_BIT_ADC_STEP_X512           87
#define BATTERY_CURRENT_PER_10_BIT_ADC_STEP_X100           17

Thanks, unfortunately I have no idea what to do with it. On 13s is 1.1v higher.....
Can you send me a finished file?


Kt LCD 4 would be interesting


Mfg Michael
 
Open the firmware folder and navigate to the /src/controller folder.

Open main.h in a text editor.

Find these lines in particular:

Code:
// ADC battery voltage measurement
#define BATTERY_VOLTAGE_PER_10_BIT_ADC_STEP_X512         44 
#define BATTERY_VOLTAGE_PER_10_BIT_ADC_STEP_X1000        87  // conversion value verified with a cheap power meter

I'm using 52V and reducing the X1000 value by 1 equals approximately -0.5V.

(87 / OSF reported voltage) * actual voltage = new X1000 value (round this up or down to get closest to actual voltage)

But at this stage I don't know if the discrepancy is the same for everyone, depends on battery type or varies individually.

So you will need to experiment with changing that 87 value, saving the file, re-compiling and flashing.

Note that BATTERY_VOLTAGE_PER_10_BIT_ADC_STEP_X512 is related, and calculated at..

(BATTERY_VOLTAGE_PER_10_BIT_ADC_STEP_X1000 / 1000) * 512

Manually calculate that and disregard the decimal places. For example, if you change X1000 to 85 then X512 will need to be..

(85 / 1000) * 512 = 43

In addition to there was a miscalculation with the ADC battery current values identified after the beta release.

BATTERY_CURRENT_PER_10_BIT_ADC_STEP_X512 should be 87 (not 102), being....

(BATTERY_CURRENT_PER_10_BIT_ADC_STEP_X100 / 100) * 512 where X100 = 17.

So, your values for those need to be as below.

Code:
// ADC battery current measurement
#define BATTERY_CURRENT_PER_10_BIT_ADC_STEP_X512         86
#define BATTERY_CURRENT_PER_10_BIT_ADC_STEP_X100         17  // conversion value verified with a cheap power meter
 
I've made good progress with the motor overrun issue, the second video below is what I was previously experiencing in v0.19 and v0.20 initially had similar characteristics. Perhaps some motors are affected more than others, or maybe it just doesn't bother some people as much as me. But this is now a big improvement and I'm very happy. :D

[youtube]vHQeYvUv09s[/youtube]

For those who haven't seen what I was experiencing before...

[youtube]20iHVS-A5dQ[/youtube]
 
famichiki said:
I've made good progress with the motor overrun issue, .......
Nice to see that you have found at last a working solution and it isn't a hardware problem.
 
famichiki said:
I've made good progress with the motor overrun issue, the second video below is what I was previously experiencing in v0.19 and v0.20 initially had similar characteristics.
Great!! Can you please explain what was the issue and what was your solution??
 
michih. said:
Cool, we need the software including the volt adjustment.😁😁😁

Mfg Michael

I will test this for a while to make sure there are no problems, but I've posted all the info about doing the volt adjustment because that will possibly vary for everyone. Once we get some reports back from people about their adjustment then we can determine if the code should be updated.
 
famichiki said:
michih. said:
Cool, we need the software including the volt adjustment.😁😁😁

Mfg Michael

I will test this for a while to make sure there are no problems, but I've posted all the info about doing the volt adjustment because that will possibly vary for everyone. Once we get some reports back from people about their adjustment then we can determine if the code should be updated.

I saw the volt adjustment. but unfortunately I can't do it.

mfg michael
 
now it works with this value:

#define BATTERY_VOLTAGE_PER_10_BIT_ADC_STEP_X512 43
# definiere BATTERY_VOLTAGE_PER_10_BIT_ADC_STEP_X1000 85

#define BATTERY_CURRENT_PER_10_BIT_ADC_STEP_X512 87


big thx!!!!!!


mfg michael
 
@famichiki, can you please explain what was your solution for the overrun issue?

casainho said:
famichiki said:
I've made good progress with the motor overrun issue, the second video below is what I was previously experiencing in v0.19 and v0.20 initially had similar characteristics.
Great!! Can you please explain what was the issue and what was your solution??
 
Hello, quick question on v.20 beta1 (buba) + LCD3 display.
Is the cruise working on your bike?
This is a function I practically never use, but I remember during the debug of buba's alfa versions the cruise was working just pressing the DOWN button while the bike is 10+ km/h, in that case the bike was keeping on going the actual speed even without the pedal rotation.

Now on beta1 (never tried the cruise) I am not able to switch the cruise on. Walk assist is working as intended, I'm saying that because the procedure to activate walk assist and cruise is exactly the same, depending on the bike speed.
Thanks for the feedback.
 
Waynemarlow said:
In the V20 you have to activate the cruise function in the setup screen, I think it’s default is off in the initial setup.

Of course it is activated. The relevant submenu 0 has value 1.
Do you confirm that after you press DOWN for a second or so your cruise starts working?
 
thineight said:
Waynemarlow said:
In the V20 you have to activate the cruise function in the setup screen, I think it’s default is off in the initial setup.

Of course it is activated. The relevant submenu 0 has value 1.
Do you confirm that after you press DOWN for a second or so your cruise starts working?

Sorry I have never used the cruise function and have mine disabled. Unfortunately my motor is US at the moment ( water ingress from a damaged plastic side cover ) and awaiting delivery of a controller from PSW ) so can't test for you.
 
thineight said:
Waynemarlow said:
In the V20 you have to activate the cruise function in the setup screen, I think it’s default is off in the initial setup.

Of course it is activated. The relevant submenu 0 has value 1.
Do you confirm that after you press DOWN for a second or so your cruise starts working?

Yes my cruise works. Beta 1. Double check the config including target speed
 
mctubster said:
thineight said:
Waynemarlow said:
In the V20 you have to activate the cruise function in the setup screen, I think it’s default is off in the initial setup.

Of course it is activated. The relevant submenu 0 has value 1.
Do you confirm that after you press DOWN for a second or so your cruise starts working?

Yes my cruise works. Beta 1. Double check the config including target speed
Thanks, will check. To activate it, you simply press and hold the DOWN button, right?
Just for comparison, what are your settings for the cruise submenu?
0: 1
1: ?
2: ?
3: ?
Thanks a lot
 
thineight said:
Thanks, will check. To activate it, you simply press and hold the DOWN button, right?
Just for comparison, what are your settings for the cruise submenu?
0: 1
1: ?
2: ?
3: ?
Thanks a lot

Yes hold the DOWN button over 10? km/h

0:1
1:1
2:25 km/h
3:1
 
0.20, makes every ride an early christmas. Can't wait to see what sw102 and 850c when its a bit more mature around the curve! I have seen the awesome torque calibration work that has been done, though my torquesensor is not health personified, I'm holding off a bit. Thanks devs and testers again!!!
 
Long time, no buba... :wink:

Just wanted to say that I am alive and well. Just very busy. But also very grateful.

Am hoping to catch up with everything but it will take some time. Luckily, my wonderful father is an avid reader of everything TSDZ2. Therefore, the last couple of months he has kept an eye on my Endless Sphere account while at the same time trying to keep me somewhat up to date. Hopefully everything will slow down the closer we get to summer and it will be possible for me to personally engage more.

I highly appreciate the kind words and the positive vibes I have received from all of you! Sorry for not being able to respond. Hopefully you understand. Will get back as soon as I have more time!

Thank you!
 
thanks for posting back, good to hear from you again !
 
Back
Top