cancel
Showing results for 
Search instead for 
Did you mean: 

Power brown-out on NUCLEO-H755 at max clock (480 MHz)??

paatx
Associate II

Has anyone else experienced a power supply brown-out/crash when running the NUCLEO-H755 (or any other NUCLEO) at the maximum clock speed with quite a few peripherals enabled?  I am suspicious that the 5V/0.5A power provided by USB is not enough to power everything at full speed when using both cores, 100Mb ethernet, and various counters and serial interfaces.  Reducing the clock to 400 MHz seems to resolve the issue.

From looking at the board schematics, it appears that using the 5V EXT power supply input will feed the main 3.3V LDO regulator directly which supports 1300mA of current.  However, I don't think this LDO can handle the additional thermal power loss (1.7V x 1300mA =~2.2 Watts of heat) without adding a large heat sink...

1 ACCEPTED SOLUTION

Accepted Solutions
mƎALLEm
ST Employee

Hello,

You are not allowed to exceed 400MHz with NUCLEO-H755 as the power is configured in SMPS mode in this board by default unless if you modified the hardware to be in LDO mode.

In SMPS you cannot exceed 400MHz at VOS1.

This is mentioned in the datasheet of the product: the boosted mode (VOS0) is available only with LDO power config:

mALLEm_0-1760711606042.png

To give better visibility on the answered topics, please click on "Accept as Solution" on the reply which solved your issue or answered your question.

View solution in original post

8 REPLIES 8
TDK
Super User

Nothing on the board will pull 1.3 A.

The user manual lists all the powering options. Max allowed is 800 mA through VIN.

TDK_0-1760661846652.png

USB connector is limited to 500 mA.

 

If you try to draw more, you will eventually either brown out or thermally limit somewhere.

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

That 800 mA maximum at 7V is to prevent the Vin linear low-dropout (LDO) regulator from generating too much heat and melting.  From the schematic, if the 5V EXT input is used instead, that Vin regulator is bypassed.  The separate 3.3V linear LDO from the 5V EXT is rated for 1300 mA max, but would likely need a much better heat sink to not melt.  My question is:  Can the MCU with all the circuitry on the full board actually draw more power than is available from the USB when everything is running and fully loaded?  I've read in one of the App notes that the USB only requests 300mA of power when initializing.  Has anyone actually used this NUCLEO board with numerous serial ports and counters running with a heavy software processing load and with the 100 Mb ethernet nearly saturated with UDP packets?  I seem to have reliability problems at high processing load at 480 MHz, but it seems to work well when configured for 400 MHz max.  Usually these type of problems are due to a power supply that cannot provide enough surge current when needed..

mƎALLEm
ST Employee

Hello,

You are not allowed to exceed 400MHz with NUCLEO-H755 as the power is configured in SMPS mode in this board by default unless if you modified the hardware to be in LDO mode.

In SMPS you cannot exceed 400MHz at VOS1.

This is mentioned in the datasheet of the product: the boosted mode (VOS0) is available only with LDO power config:

mALLEm_0-1760711606042.png

To give better visibility on the answered topics, please click on "Accept as Solution" on the reply which solved your issue or answered your question.
TDK
Super User

@mƎALLEm likely has the right answer here.

 

Also note that you can make modifications as given in the user manual to convert the board to use LDO.

TDK_0-1760712576341.png

 

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

Indeed.

It's also indicated in the board's schematic:

mALLEm_0-1760716073140.png

 

To give better visibility on the answered topics, please click on "Accept as Solution" on the reply which solved your issue or answered your question.

That may definitely explain my problem.  I guess I was fooled when running a lightly loaded test application that seemed to work well at 480 MHz.  I will make the required board modifications and test again at the higher frequency.

Thank you very much for the help!


@paatx wrote:

That may definitely explain my problem.  I guess I was fooled when running a lightly loaded test application that seemed to work well at 480 MHz.  I will make the required board modifications and test again at the higher frequency.

Thank you very much for the help!


Glad to hear that my previous answer helped you. So please mark it as solution.

To give better visibility on the answered topics, please click on "Accept as Solution" on the reply which solved your issue or answered your question.

@paatx wrote:

I will make the required board modifications and test again at the higher frequency.


And don't forget to modify the power config in your software SMPS -> LDO.

To give better visibility on the answered topics, please click on "Accept as Solution" on the reply which solved your issue or answered your question.