And, according to the docs, it can sustain a minimum of 100K program/erase cycles on each sector or block
And, according to the docs, it can sustain a minimum of 100K program/erase cycles on each sector or block
Czy wolisz polską wersję strony elektroda?
Nie, dziękuję Przekieruj mnie tamsebastian48 wrote:
- Running one device seems to be issue free. I've had it running for a week without a single issue in response.
- On the same day I power cycle the other device (which then hooks up to the system, both devices working) the first one crashes and stops responding. Physical button still working.
This feels more low level than I initially thought. Or it might work out after a reconfigure in HomeAssistant. I find it weird that HA shows both as switches but with different icons: one is a lighting bolt and one is a typical switch button.
p.kaczmarek2 wrote:most likely I will not be wasting time to recreate the thing we already have, but if someone is willing to contribute it himself for OpenBeken, then it would be great
p.kaczmarek2 wrote:I will not be wasting time to recreate the thing we already have
p.kaczmarek2 wrote:@opfer15 that's a useful information, I would like to support this platform, but the whole strip costs a lot. Did you do a transplant for ESP12F? Do you still have WB3R left? I don't have any Realtek chips at hand so my Realtek development has come to halt.
kuba2k2 wrote:Did you know that Tuya has official MCU protocol documentation? With everything explained in English in a .PDF file...
startDriver TuyaMCU
setChannelType 1 toggle
setChannelType 2 dimmer
tuyaMcu_setDimmerRange 0 1000
linkTuyaMCUOutputToChannel 1 1 1
linkTuyaMCUOutputToChannel 2 2 2
setPinRole 10 Btn
setPinChannel 10 0
setPinRole 11 Btn
setPinChannel 11 0
setPinRole 26 PWM
setPinChannel 26 1
setEventHandler OnClick 10 setChannel 1 100
setEventHandler OnHold 10 addChannel 1 10
setEventHandler OnClick 11 setChannel 1 0
setEventHandler OnHold 11 addChannel 1 -10
kuba2k2 wrote:
Sure, I think I can do this. Yes, GitHub workflows would need to be recreated, but I bet there are many PIO workflows readily available. Also, we would need only one workflow for all platforms.
opfer15 wrote:
Still have the WBR3 Modules. As they're most likley not useable to any of us without a HA compatible FW i have to swap them at a given time and probably loose some of the feature set. There is also the possibility to swap it with an esp32-C3, which i just ordered 2 modules of. But as you statet (on github i think) it's a waste of a good mcu and i don't like the idea to waste them too![]()
If you wanna work on it, i'll send you a parcel with one or two WB3 modules. GER to PL is just 5€. Or if you want to support the strip in total, i'll send you 10 bucks so it ain't that expensive anymore.
p.kaczmarek2 wrote:Haha I am not sure if you are referring to me or to the guy who wanted to copy Tasmota
p.kaczmarek2 wrote:I have to understand the protocol first in order to use it the best I can and I am sorry if you think that's a wrong approach and would rather just copy & paste code.
p.kaczmarek2 wrote:It's better to understand the functionality and implement it clearly, instead of copying it blindly like some high school student.
p.kaczmarek2 wrote:It would be better to get modules, as you don't have a use of them anyway. I also don't trust aliexpress much because in last year it started having more common tax issues. I am not sure if "over 20$ has no customs" rule still applies, I am not up to speed with the import tax duties. I buy my stuff on Bg with shipping from Czech.
kuba2k2 wrote:@opfer15
This chip is AmebaZ2. Other SDKs like ambd_sdk are not compatible and will not work. I'm preparing to better support these chips in my project.
opfer15 wrote:too bad this didn't work out
sebastian48 wrote:
- On the same day I power cycle the other device (which then hooks up to the system, both devices working) the first one crashes and stops responding. Physical button still working.
p.kaczmarek2 wrote:What is written on those "extra MCUs", are they by any chance I2C controlled?
p.kaczmarek2 wrote:it can sustain a minimum of 100K program/erase cycles on each sector or block
TL;DR: 100 % flash-success reported on BK7231 T/N modules [Elektroda, p.kaczmarek2, post #19906676]; “OpenBK7231T now boots on three chip families” [Elektroda, p.kaczmarek2, post #19883071] Use QIO image for BK7231N, UA for BK7231T; erase only 0×11000-0×1EF000. Why it matters: a single workflow now replaces vendor firmware on 60 + low-cost IoT boards.
• Supported MCUs: BK7231T, BK7231N, XR809 (Wi-Fi + BLE) [Elektroda, 19883071] • Recommended UART speed: 921 600 bps for write, 115 200 bps for read [Elektroda, 19857664] • Flash sizes: Typical 2 MB on-chip; config area starts @ 0×1EF000 [Elektroda, post #19893493] • OTA format: .rbl (gzip + AES, served by HTTP) [Elektroda, btsimonh, post #19880525] • Typical power: 5 V @ 500 mA to AMS1117 input for safe programming [Elektroda, ExploWare, post #19853546]
python bk7231tools.py read_flash -d COMx --no-verify-checksum -s 0 -c 512 dump.bin for T/N chips (2 MB, 512 × 4 KB) [GitHub bk7231tools].http-here --host 0.0.0.0 8000). In WebApp → OTA tab → enter http://<IP>:8000/firmware.rbl; device backs up filesystem, flashes, restores settings [Elektroda, btsimonh, post #19880525]OpenBK7231T, then OpenBK7231T_App into /apps. Build: ./build_app.sh apps/OpenBK7231T_App OpenBK7231T_App git (Linux) or Cygwin on Windows; ensure build-essential and gcc-arm-none-eabi installed [Elektroda, boozeman, post #19885620]