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.

IWR1443BOOST: Can't communicate with IWR1443

Part Number: IWR1443BOOST
Other Parts Discussed in Thread: IWR1642

I just programmed the IWR1443BOOST board with the prebuilt binary for water-ground segmentation (v2.3.0) but cannot get data from it.

If using the Demo Visualizer I get an irregular behaviour (I got it to work once, other times the PC sends the configuration but fails at the last command "Error: Invalid usage of the CLI command", others simply doesn't send anything)

If connecting to the board with a console (Teraterm) I get acknowledge of all commands except sensorStart: there is no answer, as below. the commands entered here are those in the demo config file.

If using the Matlab GUI it does a similar thing, seems to hang at sensorStart:

Things I've checked:

- The ports are correct (it worked once, I can communicate with console -but not start it-), baud rate as well (115k, 921k)

- For the Demo Visualizer I'm selecting the right board and right SDK (1.0) as well as Antenna Config (4Rx, 3Tx, 15 Deg + Elevation)

- For the Matlab GUI, I have the MCR installed

Any clues?

  • Hello Juan,

    Typically, when the mmWave device does not return a "Done" response to the "sensorStart" command, it means either that the configuration was not valid for the device/firmware or the device crashed when it attempted to start the sensing and data processing. As long as you are using the firmware and GUI from the Water vs. Ground lab then the configuration should be valid.

    There are a couple issues that can cause the device to crash when the sensorStart command is issued:

    1. The EVM power supply must be able to supply at least 2.5 Amps at 5V. Even if your current power supply spec mentions at least 2.5 Amps, you may want to try a different one to see if the current one has an issue or is not supplying the required current.

    2. Please make sure you are using the USB cable that came with the EVM to connect the PC directly to the EVM. Using a very long cable or adding a USB extension cable can cause failures.

    One other note is that the mmWave Demo Visualizer GUI is not intended to work with the Water vs. Ground lab firmware and will likely not work properly. The mmWave Demo Visualizer is intended to work with the mmWave SDK out-of-box (mmw) demo. You may want to try flashing the out-of-box (mmw) firmware from the desired mmWave SDK version (after downloading and installing it) and try it out with the mmWave Demo Visualizer to verify that the EVM is working properly. The instructions for flashing and running the demo are found in the mmWave SDK user guide. Please make sure to select the correct SDK version in the Visualizer.

    Please mark the thread as answered if this resolves the issue or reply if more support is required.

    Regards,
    John
  • I am using the unmodified binaries and configuration file from the example (just double checked the date)

    I tried now with a power supply setting the current limit to the max (3 A), it does the same. 

    I'm using the original, short USB cable. I also have a IWR1642 that works fine with the same PC and even longer (5 m) USB cable.

    Certainly, the Demo visualizer expects some format for the data, I just wanted to show there is no acknowledge after the sensorStart.

    Another thing I've observed is when I power the board Led DS1 is ON and goes off if I press the RESET button.

    I thought that LED meant the RF is on but is a HW error, isn't it? Does this mean there is a hw issue?

    Pictures below. After power on, three LEDS are on (DS1, DS2 and DS4)

    And after reset only DS2 and DS4 are on:

  • I have flashed the mmw_mss_14xx  application (locally compiled from original source files) and the Demo Visualizer cannot even send the very first command...

    If I connect the console cannot even type. Power cycling the board produces the following response in the console (done three times)

  • Hello,

    Sometimes the LED labeled "NERR_OUT" will come on after powering on the device even though the device is running the flashed firmware successfully. You can press the NRST button as you mentioned to reset/reboot the device and restart the firmware, which typically clears the NERR_OUT LED if there is really no error.

    Can you try flashing the prebuilt binaries (both the MSS binary and RadarSS binary) that come with the mmWave SDK? (This is described in section "3. 2. 1. Demonstration Mode" in the mmWave SDK version 1.2 user guide.)

    Also, which mmWave SDK version are you using? Can you try using the mmw binaries from mmWave SDK version 1.2 (the latest SDK that supports the 14xx EVM)?

    Regards,
    John
  • Oh, it was that! I have to flash two binaries. I thought that was only for the 16xx as it has the DSS, the 14xx as far as I know has a hardware accelerator that needs no software.

    Anyways, happy the board is fine and I can carry on.

    I have both 1.0 and 1.2 SDK, btw.