NRF9160 POFWARN

[Thomas]

Hi,

I was investigating the registers of the nrf9160 using segger embedded studio and found POFCON, which is not mentioned in the documentation.

Can you provide the documentation for this? Also, what is the best way to get access to the POFWARN interrupt? From what I see nrf_power_clock.c is using the interrupt for the clock without allowing access to the power interrupts? 

Kind regards,

Thomas

[Thomas]

Right I didn’t see that note in the manual, I am using B0 which would explain it. So a way to differentiate brown out resets on both hardware revision would be ideal. 

Kind regards,

Thomas

[Simonr]

Hi Thomas

Revision 1 and 2 are different in this respect:

• Rev. 1 does not really have a POF mechanism, but you can monitor battery voltage "manually" by requesting info and notifications dynamically from modem via the AT command %XVBAT, There's no POF specific reset generated.
• Rev. 2 has an additional POF service feature through the modem AT command %XPOFWARN. There's still no specific POF reset generated though. 
• Both Rev. 1 and 2 will generate a BOR when the SiP's internal voltage drops below the BOR limit (PS chapter "5.1.3.4.2 Power supply supervisor")
• PS chapter "5.3.1.1.14 RESETREAS" states that "If none of the reset sources are flagged, this indicates that the chip was reset from the on-chip reset generator, which will indicate a power-on reset or a brownout reset." This goes for both revisions.

Best regards.

Simon

[Thomas]

Hi Simon,

I have discovered you can set your own clock driver using CONFIG_CLOCK_CONTROL_NRF_FORCE_ALT. So I now have access to the interrupt.

Can I please get confirmation on the layout of the register? From what I can gather:

1) It is at location 0x40004510

2) Bit 0 is used to turn the it on/off

3) Bits 1-4 control the threshold with 0 being 1.3V and the LSB being 0.1V 

4) When the threshold is crossed the POFWARN event is triggered

5) Internally the chip runs at 2.2V so the level must be set lower than that to work correctly

I am using this on the B0 version and it is running correctly. There is no difference with the B1 version correct? We will be building tens of thousands at a time so I need to ensure the hardware versions will behave the same.

Kind regards,

Thomas

[Simonr]

Hi Thomas

We would not suggest doing anything not defined in the specification itself. I have to get word back internally regarding this register, so please stand by until I get back to you. 

Best regards,

Simon

[Thomas]

Okay thanks Simon. I’m confused as to why it isn’t in the specification as the POFWARN event that is triggered by this is in the specification but this not.