nrf528333 held in reset (hardware BOD)

[eb12345]

I am using an nrf52833 with MCU boot. In an abnormal situation, the Vcc hoovers around the BOD threshold of 1.6V-1.7V. The Vcc in this situation is noisy (with 100/150mv swings). Once this happens, the processor latches up in reset and never comes out (even when the Vcc recovers to 3.0V). I have checked this using a gpio output and the nrf52833 is definitely latched. The only thing that will recover this latch up is to take the Vcc below the BOD threshold again.

I thought the BOD hardware is supposed to alleviate this type of scenario ?

[mathiaso]

Hi, Eran.

So, if I understand you correctly, Vcc goes down to a value below the BOR voltage, it goes up to 3.0V, but still stays in the reset state. But, if you move back down below the BOR threshold voltage again, you exit the reset state? Or what voltage do you have to get up to to exit the reset state after having been below the BOR threshold voltage two times?

Best regards,
Mathias

[eb12345]

Yes. With a lab supply I move from 3.0V and to 1.58V, then to 1.6V  and it comes out of reset.

Eran

[mathiaso]

Sorry for the delay, Eran.

I've created an internal ticket on the issue.

You don't have any measurements showing the relationship between VDD and the reset state of the device, do you?

[mathiaso]

Do you have a plot or similar of how you ramp up the SoC? Are you within the max. supply rise time given in the recommended operating conditions?

[eb12345]

I don't have a plot, but I believe it might be slower to rise than the recommended operating conditions. It also might stay around that 1.7V depending on the battery stage of life. Is there a way to alleviate any of the BOD issue we're seeing ?

[mathiaso]

I see. If it's outside the recommendations, our suggestion would be to try to modify the ramp-up profile, in order to get within the recommended operating conditions.

eb12345 said:

It also might stay around that 1.7V depending on the battery stage of life.

This sounds like it might be related to the ramp-up speed. Would you be able to try removing/reducing some decoupling caps on the supply?