Ethernet PHY Not Resetting Properly on S32K344-Q172

[gormae1]

Hello,

I have an NXP S32K3X4EVB-Q172 (SCH-51972 REV A) and I'm trying to test out Ethernet functionality and the lwIP stack using the NXP ping example project: lwip_s32k344. However, I am encountering some strange behaviour which is making the example not work.

Upon powering up the board, I am able to plug in a standard Ethernet RJ45 cable from my laptop to the jack on the board and see both status lights pulse (as well as seeing some MDNS traffic coming from the board's IP address on Wireshark). However, upon starting the debug session, both status lights go out and no further Ethernet traffic appears on Wireshark (and ping doesn't work). Occasionally, but not consistently, at some point during the debug session I see the orange/yellow status light on the board's port flicker once but that's it.

I did a little debugging into the example project's device_init() function and discovered that it does not seem to find the PHY's address when searching for it, see not_executing.png. (The PHY on the board is the TLK110). However, it just presses on with initialization and the rest of the example runs fine.

I'm not sure if this is a hardware or software issue, considering that the PHY seems to sort of work upon power up. I've attached my version info in case it is of any use (see ver_info.txt).

Of course, please let me know if any more information is needed.

Thanks,

Elliot.

[PetrS]

Hi,

as far as I know there was an issue on some of the boards(SCH-51972 REV A) and one of the main reasons is the soldering of the level shifter between the S32K3 and ENET PHY. Could check the level shifters(U8,U10,U11,U14,U19) first?

BR, Petr

[gormae1]

Hi Petr,

Thank you for the quick response.

I did indeed probe the various level shifters as you suggested. I couldn't really find anything directly amiss with them, they all seemed to have their supply voltage set at 3.3V (in other words the VCCB = VCCA). Was there a specific ENETPHY output on the shifters you were interested in?

I also probed the MDIO line on U8 pin 2 (going out of the level shifter and into the PHY) and I did see what looked like MDIO signals (see the attached picture scope_output.png). However, when I probed the MDIO pin on the actual PHY, I saw no traffic whatsoever, only a constant 3.3V. I assume this is an issue since they should be the same right?

Thanks again,
Elliot.

[PetrS]

Hi,

I rather thought to check soldering, alignment of level shifters etc. If you see signal on U8.2 then it must be too on PHY MDIO pad, as you wrote. There are just 2 resistors (0Ohm and 22Ohm) in trace between.
This really looks as soldering issue or similar.

BR, Petr