Skip to main content
chuck guzis
Associate II
November 29, 2022
Question

Chinese STM32F407VET6 Fakes?

  • November 29, 2022
  • 8 replies
  • 5349 views

I've got a few of the Aliexpress Chinese STM32F407VET6 development boards that arrived recently. I became suspicious when the content of 0x1FFF7A22 is reported as 0xffff, when it should be the flash size in KB. The MCU chip itself reads:

STM32f407

VET6

991ES 9R

MYS 99 145 4

ST e3 ARM

04

Note that there's no revision number, which immediately made me suspicious. The MCU at first blush, seems to work okay, but what should I be aware of if I should try to use it?

Thanks!

Chuck

This topic has been closed for replies.

8 replies

Tesla DeLorean
Guru
November 29, 2022

A picture would help, high resolution, in focus

The ES is suggestive that it is an Engineering Sample, which would explain the OTP not being programmed.

It is not out of the French Fab

The "4" is a valid marking, and consistent with the mask set / fab

Tips, Buy me a coffee, or three.. PayPal VenmoUp vote any posts that you find helpful, it shows what's working..
chuck guzis
Associate II
November 29, 2022

Here's a photo. I also have some with the 06 marking. What perplexes me is the 0xffff read from the flash size register. It also seems to confuse OpenOCD, which defaults to 1024K flash size. A known-valid STM32407VET6 reads 0x200 in the same location.0693W00000WJhoAQAT.jpg

Tesla DeLorean
Guru
November 29, 2022

This has different markings than you mentioned earlier.

Does this have the boot loader and Flash size as 0xFF ?

The OpenOCD app should identify the Core, and that's usually different for CKS vs ST

Tips, Buy me a coffee, or three.. PayPal VenmoUp vote any posts that you find helpful, it shows what's working..
chuck guzis
Associate II
November 29, 2022

On the one I initially reported:

openocd reports Info : device id = 0x10076413

memory at 0x1FFFF000: 0xffffffff

flash size = 0xffff

On the one in the photo, same device ID, but flash size register is 0x200.

I'll attach a photo of the one I initially reported a bit later...

I should note that both came from the same shipment.

chuck guzis
Associate II
November 29, 2022
Tesla DeLorean
Guru
November 29, 2022

That link doesn't translate/work here.

Try https://www.sekorm.com/news/42433110.html

In STM32 Cube Programmer, what does the memory at 0x1FFFF000 look like?

Tips, Buy me a coffee, or three.. PayPal VenmoUp vote any posts that you find helpful, it shows what's working..
chuck guzis
Associate II
November 29, 2022

Don't use STM32 Cube, but I'll jot down some C code and get back to you.

AScha.3
Super User
November 29, 2022

just try ... i had a CKS32F103 -- and made a test loop on ADC sampling; surprise: it was better than the "original" STM32F103 . (otherwise was same as STM , as far, as i tested.)

so not all "fakes" are fakes, maybe just "pin and function compatible" chips from another source.

(my chips also labeled CKSxx .)

"If you feel a post has answered your question, please click ""Accept as Solution""."
chuck guzis
Associate II
November 29, 2022

The type font looks more like what one sees on the Gigadevice GD32F407VET6--sort of an OCR-A.

What with the chip shortage, I wonder how many "clone" manufacturers are taking advantage of the situation...

AScha.3
Super User
November 29, 2022

maybe. i think, the "chinese" are not "bad", it is just - business.

if you can get some money, when a big company is lazy - why not ?

just -a "clone" - to make a chip, maybe invest 200000 $ , then you want some back...

nothing wrong, so far. if chip is marked as ... "CKS32Fxxx" , ok. nothing wrong.

if label is "STM32Fxxx" , but not by STM -- thats wrong.

but - if CKS didnt re-label, it will be difficult, to find out, who did this.

so first, you have to proof, this chip is not by STM. then find out, where it coming from...

to find the source of who changes label - then try to start a process against that someone - in chinese. :)

good luck.

"If you feel a post has answered your question, please click ""Accept as Solution""."
chuck guzis
Associate II
November 29, 2022

That's my problem--if the chips were "GD32F4" or "CKS32F4" labeled, at least I'd know what I was getting.

So, to answer my own question "Are there STM32F407 fakes out there?" The answer is clearly "Yes!"

chuck guzis
Associate II
November 29, 2022

Here's a photo of the MCU that shows 0xffff for the flash size. While it seems to work, I'm going to swap it out for one that reports flash correctly. I don't need another thing to worry about. What's notable on this one is how "ARM" is quite visible from a distance, but the rest of the chip labeling is a bit faded.

0693W00000WJnE3QAL.jpg

Douglas MILLER
ST Employee
December 13, 2022

This forum thread was marked by the moderator as needing a little more investigation, so a Support case was created in your name and will be handled off-line. Should anyone have any similar questions, please feel free to open a Support case directly at your myST portal: https://st.com/ols