Non unique unique device ID or counterfeit parts?

[Innomatic]

Hi,

I've encountered with weird situation with STM32L053C8T6 devices, namely all the devices share the same unique device ID.

I use LL_GetUID_WordX() functions. And for the third word (LL_GetUID_Word2), I tried both offsets, 8U as current Cube library specifies or 14U as reference manual specifies. But what I get is (4,71,56,48), (50,54,50,56), (00,17,00,00) same over multiple devices I have tested.

These batch of L053 has other weird things. Pull-up level of GPIO drops from 3V3 to 1V without warning. And the DAC output swings only around 1V.

We strictly source our parts either from Digikey or Mouser. So it is very unlikely that those parts are fake. Still the situation is not explainable.

Any idea would be greatly appreciated.

[Innomatic]

I think I found the answer for the unique device ID issue. First of all the CUBE LL API has error in reading the third word of the unique device ID. In case of L053 the offset of third word is 14 and not 8, according to the reference manual.

Problem is that by simply fixing the offset from 8 to 14 is not enough because this third word is located across double word boundary of the memory. You have to read the first two bytes from the last word of 0x1ff80050 line and read the last two bytes from the first word of 0x1ff80060 line, separately. Otherwise your last two bytes of the last word are always zero, which makes non-unique unique device ID.

[Imen.D]

Hello @Community member​ ,

Thank you for bringing this issue to our attention.

We raised this issue internally for check and investigation.

With Regards

Imen.

[henry.dick]

The uid offset is specified in the datasheet. So just put a pointer there and you are done.

Much simpler and fool proof than some function call's from a questionable source.

[T J]

DigiKey has guaranteed that all parts sourced are from the primary manufacturer.

DigiKey has this guarantee in place because of the Defense force requirements about counterfeit product.