STMWLE5CC (RAK3172) Radio End-of-life flag not correctly resetting

[IFant.1]

Greetings, i am developing a application that will run on a RAK3172 module and i have been attempting to use the radio End-Of-Life flag to detect when the battery supply is too depleted to operate the radio safely.

The base application is the LoRaWAN_End_Node sample with modifications to make it run on a RAK3172 package mounted on a custom PCB.

As the first thing upon entering the main function, i activate the end of life detection enable register with the function (HAL_PWREx_SetRadioEOL) provided to me by the sdk. Then i connected the board to a power supply to simulate different battery charge levels.

To make a long story short, for some reason the EOL flag (in PWR-SR2) remains always set to 1, regardless of:

-enabling or disabling of the EOL enable register

-state of the radio

-supply voltage

Unfortunately this makes the end-of-life detector useless as i cannot discriminate whether or not the battery is sufficiently full to operate the radio safely.

Thanks in advance

[Louis AUDOLY]

Hello,

Thank you for your message

We are investigating. Note anyway that a more accurate and more power-efficient way to detect voltage to enable or not sub-Ghz radio is to use the PVD feature.

Please look at chapter 5.2.2 of RM0461

 

We keep you inform

Best Regards

[IFant.1]

Hello, has there been any news in that regard?

Minor addendum, further tests on the MCU in our possession have shown that:

• the EOL flag behaves in the opposite manner as shown in the datasheet (ie: the flag is 0 while the supply voltage is below the radio end of life operating low level rather than 1 as it is written in the datasheet)
• in a test using a pair of partially discharged AA batteries, both the PVD (level 0) and the EOL flag indicated that supply voltage was sufficient to operate the radio safely. However when the join procedure was initiated, the current drain was large enough to cause the voltage to dip below the threshold level for BOR level 0 and thus cause a reset of the MCU

[Louis AUDOLY]

Hello @IFant.1​ ,

Thank you for your precisions !

We are still working on this issue, we will let you know when we found a fix.

Hope you are not stuck because of this issue

Best Regards

[Benoit MARCHAND]

Hello,

Thank you for your investigations.

We confirm this feature works properly but there is indeed an error in the user manual as you state it.

We will correct it in the next release.

Best Regard,

Benoit