i have a Custome pcb using rak3172 and RUI3 firmware. the bord works for some time and randomly stop booting after power cycle. i have this failure happened to 3 out of 5 samples. 2 of them also stopped responding to ST debugger and 1 still responding. i downloaded the hex file from Download Centre and after uploading it the current consumption stays at 7mA. when i upload my FW, the same thing happens and the board doesn’t try to join the network.
i would like to know how to solve this problem.
Welcome to the forum @ahmedmansoub96
Without the schematic of our custom board we cannot say anything.
Schematics look ok.
How much current are your pressure sensor and TOF sensor drawing?
On the devices that do not respond, what is the voltage on 2V8?
What version of RUI3 do you use?
Did you try to pull BOOT0 to 2V8 and try to do chip erase and upload firmware with STM32CubeProgrammer?
1- the current doesn’t exceed 20mA total and below 100mA during transmission. the board is sleeping mode all the time and wakes up to send data and goes back to sleep. sleep current 60uA.
2- the voltage drops to 1.7v and other devices to 0.4V
3- latest version.
4- i didn’t try that option
That is below the supported voltage range:
Why is it going so low. Looks like you have a short somewhere on your PCB.
when i remove the module from the board, the voltage goes back to normal
i would like to ask if there is any suggestions because one more sample also died but without power cycle or even touching it. it is very concerning and i delayed launching 4 products because of this issue. please help
I have no idea what is happening on your custom PCB.
What is this chip doing?
this is bq27441 Fuel Gauge to read battery. let me ask a question, what happens if the interrupt is triggered during initialization of the lorawan? can the memory be corrupted?
what is the best way to manage sleep mode and wake up the device using interrupt? if i didn’t manage it properly , would it cause this issue?
During normal operation, there is no write access to the flash, an interrupt cannot cause a memory corruption.
Exception, if you use RUI3 functions to save data in the flash memory there is write access. But I have not seen any problem with this function.
RAM memory corruption can be caused if you have stack overflows or unassigned pointers to memory.
But none of them would cause to drop the supply voltage to a level that would cause a problem.
i will try to investigate the problem more next week and if i didn’t find a solution, is there any possibility to have a zoom meeting with a RAK representative to show the issue and the data to find a solution?
I advise you to change the RT9078-28GJ5 because there may be a problem although given for 300mA if possible put a 500mA