PCA9450AAHNY disables power rail VDD_SOC

cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

PCA9450AAHNY disables power rail VDD_SOC

1,004 Views
sveitola
Contributor II

Hello,

we're designing a board with an MIMX8MM6CVTKZAA processor that is powered from the PMIC PCA9450AAHNY. While we got one prototype to start booting, we see problems with one of the power rails of the second prototype and I hope someone can point me in a direction where to look for the cause of it.

All of the supply rails are coming up, and the processor transmits messages on a debug UART_2 port. But after about 600 ms the PMIC supply output VDD_SOC switches of. The duration of how the supply is maintained is slightly varying. On the scope the supply voltage is rectangular (it's not a capacitive discharge but the rail seems actively disabled by the PMIC). All other supply rails remain up.

We do not tune the supply voltage. And the other prototype works well. Right now I don't know where to start looking.

0 Kudos
4 Replies

990 Views
sveitola
Contributor II

Thank you for your reply. The BUCK1 supplies solely iMX8 rails and no other components. I will check if maybe an MLCC on these rails is leaking.

0 Kudos

980 Views
sveitola
Contributor II

For now, we're booting the processor from SD card. The voltage rails remain stable at the initial levels when there is no SD card inserted. The processor sets PMIC_ON to HIGH and all BUCK outputs work just fine. I can exclude assembly problems on the voltage rails (like MLCCs or solder shorts).

If I do start the board with SD card inserted, the processor starts booting and manages to send a few log messages out of UART2. But now, the processor asserts PMIC_STBY_REQ. (I have swapped the buck inductor in the meantime to ensure that no problems come from there). The standby signal goes sporadically LOW.

SD card is powered from VDD_3V3 of buck 4. 

0 Kudos

968 Views
art
NXP Employee
NXP Employee

It looks like a software problem (the software manages to set the PMIC_STBY_REQ signal High for some reason). Please try another version of the software.

Best Regards,

Artur

0 Kudos

996 Views
art
NXP Employee
NXP Employee

The situation you describe looks like the overcurrent protection triggering on
the SW1 regulator. So, first, please check for some PCB assembly level issues,
e.g. some short-circuits or extra current loads.

Best Regards,
Artur

0 Kudos