Showing results for 
Search instead for 
Did you mean: 

STM32H745IIT6 - is power wired right?

Associate III

Ok here is the situation

Did a board design, when plugged in i noticed 3.3v was not outputting correctly. Realized I put the wrong 3.3v regulator, maxes out at 100ma. Removed and put in a 3.3v regulator that goes up to 800ma which i thought would be fine for the STM32H745IIT6. Now when I plug it in i get 2v and its trying to pull over 1000ma which is where my meter maxes out. 
Obviously there could be many issues that could cause this but I want to start at the chip and just make sure I have it wired correctly for power and I didn't miss something on its power section. 

If you can give any thoughts on if any of the power pins are not supplied correctly would be super helpful as i'm very new to this specific chip. What MA should I expect with this chip? If i did anything wrong with this post just let me know! 

Thank you! 





100mA is enough for the chip to boot up at its default (erased) state. Expect maybe 10 mA when erased.

You have 3 VCAP pins, two of them should have a 2.2uF to ground (not a 0.1uF). They are recommended to be tied together, although this is not strictly necessary. VCAP3 needs tied to VCAP1/2.

Can't see anything else wrong.

Are you sure part orientation is correct? Perhaps post an image of the board.

Hardware design guide for reference:


How are DGND and GND connected? Better be shorted, otherwise that's your problem.

If you feel a post has answered your question, please click "Accept as Solution".
Lead II

At first glance your power lines look plausible, although I'm not sure of the distinction between GND/3.3B and GND/17B. (Unless there's an error in your symbol so the pin assignments aren't what you expect).

Typical power-consumption is shown in the Data Sheet, in Electrical Characteristics:Operating Conditions:Supply Current Characteristics, and I wouldn't expect it to exceed 100 mA at startup.

Is the 2V what you measure on the input or the output of your 3.3V regulator?

One possibility is that you have a short somewhere on your pcb. Perhaps a solder-splash between two adjacent pins on the stm32, or a manufacturing fault on the pcb itself. (I've had both in the past). What resistance do you measure between Vdd and Vss whan not applying power?

Are there other components on the pcb that might be drawing power? I see you have nets bearing names including "bypass relay". Until your program is running as it should, the GPIO pins might be floating or in whatever state the ST bootloader has them in hoping for a program-download by a specified interface. And that might put peripherals in high-current state.

Do you have a layer specifically for Vdd? If not, you might be able to follow where the current is going by looking for voltage-drops along the Vdd path with a sensitive voltmeter. (Horiwitz and Hill 2nd Edition - the version I have - described such a tool as as a "stuck Node Tracer")