MSP430FR2311: EMC-Trouble, maybe clock-issues?

Power suppy not output I wanted to write

Hi Thomas,

What's kind of emc test do you test so far?

Could you can make a UART port to output some debug information?

Dose the device reset?(You can use a GPIO toggle event to detect reset event) Or just hang up at some where?

Everything is fine except fast transient burst test. The debug interface is totally stripped out and I have no idea how we would get another firmware on without rebaking hole device.The device seems to hang in main routine but interruptroutine seems to work. Can it be? Without EMC-Testsignal everything seems fine. And there is a strange other phenomena, the input is triggered, it should not be...

Thomas Schumacher said:

totally stripped out

So do you mean you can't get the SBW interface on the hardware even to use some jump ware to soldering at some points that can connect TEST and RST pin?

Thomas Schumacher said:

The device seems to hang in main routine but interruptroutine seems to work. Can it be?

If the ISR can work, that means the CPU it not hang up but it maybe running at some where that not expected that may cause the input be triggered unexpected. 

If you can't get any information on the failed device that will be hard to debug to move this issue forward, please help figure out a way to access SBW.

Yes, we totally tossed everything out to tht is not neccesary to get keep traces small as it is single layer. I don't think it can be still a software bug. The code is double reviewed and security certified. The IC looks too small to get a needle adaptor on?. The only alternative I Idea other than stablizing input is flshing another firmware and using the internal clock source. The firmware and hardware is coded and changed external. Do you have code to output failurecode on pin if UVLO or something like that triggers?

Could you help to share the schematic here? If no way to get information from the MCU, you just have to make some protection components on the hardware level.

For now we don't think it is a good idea to put the schematic in public place. We tried TVDS, ferrite bead, different capacitor-ceramics and additional Z-diode without having success with.

Have you try add the Common mode chokes on the noise transmission path? 

Have you make a good connection between your GND to the ground?

Few document here maybe help with this case

https://www.ti.com/lit/an/slva833d/slva833d.pdf 

https://www.infineon.com/dgdl/Infineon-AN80994_Design_Considerations_for_Electrical_Fast_Transient_(EFT)_Immunity-ApplicationNotes-v09_00-EN.pdf?fileId=8ac78c8c7cdc391c017d072b38ea5000