New batch STM32F051 LSE don't start

[passafabio2]

A new batch of electronic board mount an STM32F05R8T6

AA170 9U

TWN AA 131 marked.

All of this devices don't start up LSE clock.

The code is the same for about five years and makes the LSE clock start up till this days correctly. The initialization was made using example code bundled with discovery stm32F051.

I create with CubeIDE Version: 1.9.0 a minimal setup (CubeMX) for initializa internal RC but the code hangs on

while(__HAL_RCC_GET_FLAG(RCC_FLAG_LSERDY) == RESET)
 {
 if((HAL_GetTick() - tickstart ) > RCC_LSE_TIMEOUT_VALUE)
 {
 return HAL_TIMEOUT;
 }
 }

The quartz is the same sice five years and i try to modify the capacitor from 15nF to 6,8nF with no success.

Same board with STM32F05R8T6

GQ272 9U

CHN GQ 944 marked

run perfectly.

Any ideas?

Thanks 

Fabio

[TDK]

If you increase RCC_LSE_TIMEOUT_VALUE, does it work?

If you scope LSE, do you see a signal?

[passafabio2]

Doubled the value from 5000 to 10000 but nothing change...

A good board programmed with the test code for LSE created with CubeMx corectly goes over the init and chek functionality of LSE, light on a led done by a successive code..

[waclawek.jan]

> 15nF to 6,8nF

Double-check the capacitors, they surely must not be in the nF range.

Also, how exactly is VBAT connected?

Write a simple test program, which does nothing but

• enables backup domain access by setting PWR_CR.DBP (note that PWR clock has to be set in RCC prior to this)
• resets backup domain using RCC_BDCR.BDRST
• enables LSE by setting RCC_BDCR.LSEON
• waits undefinitely until RCC_BDCR.LSERDY is set
• lights up a LED or indicates success in some other suitable way

Note, that LSE startup times in several seconds (say up to 10s) is completely normal.

In this program, try also different RCC_BDCR.LSEDRV settings.

JW

[passafabio2]

Sorry my mistake, I wanted to say 15pF to 6,8pF

[Tesla DeLorean]

STM32F05R8T6 ?? or STM32F051R8T6 STM32F058R8T6

Can you read the device ID from the part, to be sure of the die/stepping ?

DBGMCU->IDCODE

Might need lower capacitance.

There should be a means to measure/export the LSE clock without probing directly.

[passafabio2]

STM32F051R8T6

AA170 9U

TWN AA 131

IDCODE 0x20006440

while(__HAL_RCC_GET_FLAG(RCC_FLAG_LSERDY) == RESET)

Is the flag which manage the running/halt state ol LSE clock.

[Tesla DeLorean]

I mean with an oscilloscope or TIM peripheral vs HSE clock​

[passafabio2]

Today I tried using 3,9pF cap (the only I have in my warehouse) with no success....

[waclawek.jan]

Have you tried any of what I wrote above, e.g. playing with LSEDRV?

JW

[passafabio2]

Yes changing LSEDRV nothing happen.

And untill this batch of micro we made about one thousand of board with LSE Ok...

[TDK]

Try replacing the STM32 chip on a new board with one from an old board.

[waclawek.jan]

At this point you may want to contact ST directly, through FAE or the web contact form.

JW

[Amel NASRI]

Hi @passafabio2​ ,

As suggested by @Community member​ , I'm checking internally if there is any noted change regarding LSE as I'm not aware of any update.

In such case, a failure analysis request (FAR) will be required and needs to be initiated by your FAE.

-Amel

[Tesla DeLorean]

Hi @Amel NASRI​ 

I'd assume there's a new RM0091 somewhere covering the new ICs, the one publicly available is over 5 years old, and clearly not up to date.

Even if it is a new/duplicate mask-set for a new fab.

"9U" is the TSMC USA Fab 11

"98" is the TSMC Taiwan Fab 8

And in 2019 some was moved to "9C" TSMC Singapore pursuant to PCN MDG/19/10803

[passafabio2]

Yes nothing change.