New B-G431b-esc1 ($AU3) can't connect with a CAN bus.

[MSzym.1]

I have some drivers left from a few months ago, and they are working with no problem. This same project loaded to the new boards and everything works O'K except that the packets don't flow (neither way).

I've tried to reset and set the CAN_SHDN (PC11) signal with no change.

Is there any change in the latest bootloader regarding the use of the CAN bus?

How to fix the problem?

[Imen.D]

Hello @MSzym.1​ ,

Make sure that your firmware & tools are up to date.

Check this article: How to connect to the STM32 CAN Bootloader interface? this may help you on CAN bootloader.

Imen

[MSzym.1]

Imen.D You probably have skipped the part when I was telling that this same project have worked and is still working (after loading again) into the older drivers, so the problem is not that I can't use CAN bus or have outdated tools.

To confirm this I've checked that the CAN driver IC itself is different on new boards - the old one was marked "330" and the new one is marked "AEKF" the other odd thing I've noticed is that the new IC is oriented probably the wrong way - the pin nr1 mark is in the opposite way from the previous generation ICs, but the PCB paths vaven't changed Is this possible that there's an assembly error?  To verify I need specifications for both chips - the one marked "330" and the one marked "AEKF" - I can't find them.

[MSzym.1]

Here you have the pictures under the microscope to illustrate the issue.

    

        Pic1: The old version - working                                                 Pic2: The new version - not working

Please give me the specs for both ICs and verify this issue also on your side.

[MSzym.1]

Or is the "AEFK" marking on the not working chip an anagram of "FAKE"?  ;)

[LCE]

google:

can transceiver ic marking "aekf"

=> https://www.analog.com/media/en/technical-documentation/data-sheets/max3051.pdf

same with "330":

https://www.ti.com/lit/ds/symlink/tcan332.pdf?ts=1689251768566&ref_url=https%253A%252F%252Fwww.google.com%252F

same pinout, so yes, the assembly is screwed up...

[MSzym.1]

So, I've checked in the production files of all board variants (in BOM files) that there should be the U2 => TCAN330DCNT (hence "330" marking on the IC), therefor since there's a different IC in the newest generation of the board, it is a mistake of some kind in the assembly process - a different IC (out of the specification) and/or a wrong orientation of the IC (pin 1 on the opposite side).

[LCE]

> and/or a wrong orientation of the IC (pin 1 on the opposite side).

That's what I just said:

> same pinout, so yes, the assembly is screwed up...

The Maxim and the TI transceivers are pin compatible, but the assembler rotated the new on by 180°. Or someone gave them the wrong assembly files, or whatever...

You could simply unsolder the IC, turn and resolder it.

Or better, give that PCB back and ask for a replacement.

[MSzym.1]

@LCE Thank you for the confirmation. Sorry for the redundant posts, but the browser didn't refresh the thread, so I haven't seen your answer. 

I don't suppose our problem can be solved by reclamation - We've bought 10 drivers from Mouser lately, from TME before - all with this same MAXIM chip (and it's orientation that is problematic - not the IC itself) and so my guess is all the volume of these drivers from ST's last production order is like that. You (ST) should redraw the faulty products probably. ST may need also to verify how the Chinese manufacturer mount these chips on your other products - the error seam to be of the systemic nature, so (to be sure to be sure)...

I'll manage to unsolder the chips by myself (I've got the microscope and a hot air soldering station in the workshop, but other guys (RC enthusiasts the product is dedicated to) are in the worse position. 

ST, Please, announce when the good drivers are back on the market and maybe indicate this new corrected series by some sticker or just a series/version number easy to find on the package so we can order a proper version in the future (not the faulty ones from a stock).

Since the post about the problem is already here, maybe this is the place for you to announce your decisions and actions upon the problem.

BR

[JKong.3]

Here's my b-g431-esc1 board purchased around May for comparison.