STM32U585RIT6 - Power supply the MCU at 3.3V act as a shortcircuit

[calugarfinias]

Hello all,

I developed a board based on the STM32U585RIT6 (without SMPS package) based on the STM32U5 guidelines.
I've encountered a problem during tests: When I turn on the MCU block on my board, the MCU draws >600mA of current.
My first reaction was to test continuity between GND and 3.3V. Result: no continuity and insulation around 49kOhms. It was very strange. I also tested the possibility of a short between the MCU pins, but nothing.
However, I still manage to connect to the MCU with a STLINK. I also used a thermal camera which shows me that the problem is with the MCU.

I continued by gradually removing the components around the MCU to exclude a potential external short circuit.
=> I tested the MCU in 2 main configs:
1) All GPIOs are not connected to anything, but I kept all decoupling capacitors
Results--> Abnormal current  >600 mA consumed by the MCU.

2) I removed everything off the board around the MCU (GPIOs connected to anything)

I directly feed the MCU power pins at 3.3V (and GND connected too). Results --> same results, the MCU draws >600mA

I didn't have any components around the MCU, so I'm wondering if I'm missing something regarding the power supply.
I read the STM system power management presentation, but nothing was wrong for me.

Anyone have any idea why I can't power this MCU with 3.3V?

Any help will be great :)
Thank you!

[TDK · Accepted solution]

VCAP should not be shorted to 3.3V. It should be bypassed to GND with a 4.7uF cap per the guidelines you linked.

[Tesla DeLorean]

VCAP typically being the output of the internal regulator, with a suitably large bulk capacitance.

Didn't check the U5, but others this pin is notionally 1.2 - 1.25 V

[calugarfinias]

Thank you, a new view on the situation was helpfull!!