BoSL Board V0.5 Debug

This page contains a list of observations from using the BoSL v0.5.x, they may be items which need to be kept in mind when developing and improving code for the BoSL v0.5

Issue List

ATmega issue

Description:
On the code ported from v0.4.x, there is sometimes no response on the USB serial ports. This may be because the ATmega has crashed. This issue has been observed two times
Test:
This has been observed on both when using the RTC for timed-sleep and when using delay().

SIM7000 issue

Description:
The sim7000 may be unresponsive on power up.
Test:
The sim7000 was powered up and the "AT" command was sent. No response was received, even when the baud rate was adjusted.
This issue was resolved once the BoSLboard reset button was pressed.

Network register issue

Description:
Network registration was not achieved on some boards
Test:
Of 6 new boards tested with the same code, 4 were not able to register to the network. These boards gave "+CREG: 0,2" meaning that they were searching for operators to register to.
This difficulty persisted across resetting the sim7000 and reconnecting the battery.

Debug history

6th Feb 2024

Yesterday (cloudy) six BoSL v0.5 boards were tested. Four of these were not able to register to the network.
Today (clear day) the four board which were not able to register to the network were retested using the same sim card and same code.
2 of them registered to the network, no idea why, the weather?
1 of them still could not register to the network, even after "sendATcmd(F("AT+COPS=0"), "OK", 150000, 1);".
but after reconnecting to the battery 2 or 3 times, it was able to register to network.
There is still one board cannot register to network, reconnecting to the battery or await "AT+COPS" for 150s not work.

Adding:
The last board which cannot register to the network started to logging after being connected to battery for about 1 hour,
it looks like the Sim7000 module needs a warm up before work properly.

8th Feb 2024

Stephen said he got a BoSL V0.5 board that cannot register to the network neither.
He tried to fix it and found that sending "AT+CEREG=1" ATCommand to SIM7000 solved the problem.
It is a good finding. We can test this ATCommand on more boards that have network register issues.
Once we confirm that this ATCommand always works, we can add this ATCommand to our BoSL logger code.

19 Feb 2024

B5_TurbDepth_3 recovered after stopping logging for seven days.

20 Feb 2024

B5_DT_SD_2 is not logging for three days after being set up. It starts to log today when connected to the PC USB port.
It looks like a board quality issue.

B5_TurbDepth_3 stopped logging again.
The debug showed it keeps resetting.
It is a hardware bug and has already been fixed on V0.5.3.
For V0.5.1 to V0.5.2, we should avoid boards working at a voltage lower than 3.7V.

26 Feb 2024

B5_DT_SD_2 stopped logging on 23 Feb when the battery voltage dropped below 3.38V.
It is good, the performance is better than expected (above 3.6V).
However, the low-voltage protection module does not satisfy our expectations.
The VBAT pin is expected to be above at least 1V but tested at 0.82V.
It logs again after the battery has been replaced.