This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

MMWCAS-RF-EVM: Role of NERROR LED in the cascade board

Part Number: MMWCAS-RF-EVM
Other Parts Discussed in Thread: AWR2243

Hi,
We've noticed that sometimes the NERROR LED on the cascade board turns on during normal operation. Upon looking through the datasheet of 2243/1243 and the cascade board, we found out that it turns on when this condition occurs.

"Open-drain fail-safe output signal. Connected to PMIC/Processor/MCU to indicate that some severe criticality fault has happened. Recovery would be through reset."

I'm not able to find out what does the datasheet means by "severe critically fault". Does the LED turn on during power-on failures? What else does this LED indicate? 

EDIT: We ran the demo on MMwave studio -2G on 2243 cascade board and as soon as the Firmware was done flashing on the Master, the red nError light came on. We used the stock LUA script(Cascade_Configuration_MIMO.lua) provided with the installation to do it, which sits under C:\ti\mmwave_studio_03_00_00_07\mmWaveStudio\Scripts\Cascade. The firmware Image that it is pointing to is xwr22xx_metaImage.bin under C:\ti\mmwave_dfp_02_01_05_03\firmware\

  • Hello Asher,

    The nError LED indicates some hardware fault in one of the AWr devices. Could you elaborate on what instance does the LED glow? Is there any particular operation that causes the LED to glow?

    Regards,

    Vivek

  • Hi Vivek,

    on our 1243 Cascade board, we were unable to reproduce what causes the nError LED to turn on and we're still looking into it. However on our 2243 Cascade board, the nError LED is always on as soon as AWR firmware is downloaded via processor_SDK. We can reproduce this issue with our custom usecase as well as the Stock usecase on PROCESSOR_SDK_RADAR v(03.08). 

    We know that the SDK on it's own is not compatible with 2243 cascade board, Piyali from TI has given us patch files to make the SDK compatible with 2243.

  • Edited the main post to add more findings. 

  • Hi  

    Any updates on this post?

  • Hi Asher,  

    I am able to replicate this issue on my AWR2243 cascade EVM with mmwave_dfp_02_01_05_03 and mmWave Studio 3.0.0.7 (mysecuresw release). Have you noticed any functional or performance problems issues associated with this issue?

    There is a lua output log in mmWave Studio which appears during the start-up of each device: [DeviceId-0] MSS CPU Fault error status Async. The error out signal seems to be synchronous to this message. This may be triggering the error logic to fire. I will look into this more with our firmware team. 

    Thank you,

    -Randy

  • Hi Randy,

    I'm also seeing the same error on my Lua output log of mmWave Studio. With some hiccups, everything seems to run fine. although I'm not sure how exactly it is affecting the performance of the board. If there is an underlying problem, it hasn't made itself apparent yet.

    One issue that we see on the board is when we use performance SDK, the Firmware is not able to download properly as detailed in the forum thread here.

    https://e2e.ti.com/support/processors/f/791/t/913305?tisearch=e2e-sitesearch&keymatch=%2520user%253A379156

    A clarification on what exactly is causing the error LED to turn on and the implications it causes would be really helpful, as it would allow us to know exactly what to trust and what not with the board.

  • Hi ,

    Any updates on this?

  • Hi Randy,

    After performing the same test on the AWR2243 cascade imaging radar, I also found that the NERROR LED will light up.

    After single-step debugging the mmwavelink code of mmwave_dfp_02_01_05_03, it is found that after the Master device is configured, the resetDevice() is executed on the Slave device, and the LED will light up, eventually causing TDACreateApplication() to fail.

    At first I thought it was because the program didn't have enough delay. But after introducing the delay, it still didn't change. But the strange thing is that if I use mmwave studio to complete the data collection, there will be no errors. It makes me very confused.