STM32L432 Nucleo-32 External Power i.e. 7V Boot up Issue

[Dave001]

I am struggling with an intermittent boot issue with Nucleo STM32L432.

 

What happens:

When the board boots up from VIN of 7V, sometimes (I would say 50% of the time) STM32 struggles to run the code, it's almost like it stuck during the boot up. When i power it up from USB to check what is the issue, problem goes away and board is running fine. 

My application involves ~40 of these boards and i am seeing this issue in 5 boards and it happens 50% of the time in those 5 boards. 

 

Only mods i have done to the nucleo board are - 

1) I have taken a suggestion made in this post: Solved: STM32L432 Nucleo-32 External Power - STMicroelectronics Community i.e. to remove SB9 to be able to power it from VIN of 7V. 

2) Connected a solder bridge at SB17 for HSC

 

Anyone faced this issue or any troubleshoot tips anyone can share.

Thanks in Advance

 

 

[mƎALLEm]

Hello,

First, I think this is a continuation of this post.

Second, it could be a power issue. Are you sure that 7V power supply could deliver sufficient current to the Nucleo board?

From the board's user manual UM1956:

If you are using controlled power supply, ensure the current limit is not set below to what the Nucleo board can drain as current. Set it at 1A for example.

And what do you mean by HSC here?:

---

@Dave001 wrote:

2) Connected a solder bridge at SB17 for HSC

[Dave001]

---

@mƎALLEm wrote:

Hello,

First, I think this is a continuation of this post.

Second, it could be a power issue. Are you sure that 7V power supply could deliver sufficient current to the Nucleo board?

Reply:

I think you are right in saying it is some Power issue but i fail to see how. I am using "LMR33620BDDA" which is capable of providing 2A to the load.

Few more new things i have discovered which makes me think it is the power issue as well now - 

1) Pressing reset button on the nucleo stm32l432 solves the issue. I tried pressing reset button right after i boot up and board booted up alright. so, it has something to do with the first boot up.

What i did is 

turn ON the power -> checked that boards are not booted up correctly (there is a led on the board which did not turn ON) -> pressed reset -> checked that boards are fine now

2) I tried powering up the board from an external 7v and not use the on board "LMR33620BDDA" and boot up was successful 100% of the time. so, it has to be the power side of things.

3) Why i don't think it is "LMR33620BDDA" problem on those 5 problematic boards - problem is following only those 5 stm32 boards. I have 40 of these connected via headers on a PCB board and 35 boards are fine and only these 5 are problematic.

 

What to do now: 

It has to do with second boot up somehow. Could it be a clock sync issue? when MCU boots up maybe the HSE clock is not fully stable?

Next thing i was thinking was probing the voltage and then try increasing the BOR levels unless you think i can try something else. 

 

From the board's user manual UM1956:

If you are using controlled power supply, ensure the current limit is not set below to what the Nucleo board can drain as current. Set it at 1A for example.

And what do you mean by HSC here?:

---

@Dave001 wrote:

2) Connected a solder bridge at SB17 for HSC

 

---

"HSC" is a typo for HSE (High-Speed External) clock i meant i have added SB17 which connects MCO output from the on-board STLink chip ('F103) to PA0 of the target i.e. 'L432KC

[MM..1]

Try explain more this 2) I tried powering up the board from an external 7v and not use the on board "LMR33620BDDA" and boot up was successful 100% of the time. so, it has to be the power side of things.

In Nucleo schematics not exist an LMR ...