NRF9160 reset loop prevention never seems to actually trigger

[Campbell Young]

Hi,

We've got a board with a fairly limited power supply in production and it's expected to brown out under certain conditions. As such, I need to verify the reset loop prevention described in this Nordic Semiconductor white paper.

I'm using mfw_nrf9160_1.3.5 and SDK 2.0, so I expect the prevention to trigger after 7 resets. To test this I unconditionally trigger a connection during the boot sequence and have our power supply limited to ~30 mA to trigger a brown out. I can see +CEREG: 2 and +CSCON: 1 just before each reset, so the device appears to be browning out during active comms, but it never triggers the reset loop prevention. I let it reset approx 20 times before stopping the test.

I never call +CFUN=0 and can't it doesn't appear to be called from any sdk code as if I breakpoint in `lte_lc_func_mode_set` it never triggers.

Any help would be appreciated

Thanks

Campbell

[Didrik Rokhaug]

Hi,

The reset loop restriction counts soft-resets, i.e. resets caused from the application SW (and crashes).

It does not count power-on resets, which you will get after a brown out.

Best regards,

Didrik

[Campbell Young]

Ah, that would certainly explain it. Thanks for the fast reply!