• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 2 answers
  • Subscribers 31 subscribers
  • Views 4120 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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: uniflash not able to connect serial port, and the chip heats badly

Yicheng Fan
Prodigy 30 points
Part Number: IWR1443BOOST
Other Parts Discussed in Thread: UNIFLASH, IWR1443,

Hi, I had been able to flash the binary and run against the demo without any problem for three days.

But yesterday when I do an experiments, I observed my mac alerts me "usb accessories disabled for too much power".
My USB port1 can no longer detect the board, so I switched to port2 to continue. (BTW, it does not work when I connect back to port 1 after SMC reset)

Uniflash alerts "not able to connect serial port"
although I could see two com ports are shown on device manager.
SOP is set correctly.

The chip stores SDK demo binaries before not working,
so I also tried the SDK demo visualizer.
After click configure, it shows correct com port, and can get to the state of hardware connected.

But when I send settings to device, the console does not print anything.

Meanwhile, I notice even when USB is not connected, the chip heats badly, way more heated than before.
With the mac alerts, I was wondering if this is because the board or the chip is short-circuited somehow?

How can I verify if the hardware is broken?

  • 0
    TI__Guru*** 148974 points
    Hi,

    So, you are not able to flash any binaries to the EVM?

    thank you
    Cesar
  • 0 in reply to Cesar
    TI__Mastermind 37500 points
    Hello,
    If this is a question for the EVM.
    The IWR1443 BOOST EVM schematic is provided in the technical documents section.

    a) there is an external 5v power supply, in normal use cases the 5v power draw < 500ma @ 5v

    b) there are several power supplies on the board. The PMIC on the back of the board, has 4 output inductors, we expect the voltages to be within 5% for 3.3v, 2.3v, 1.8v, and 1.2v.

    c) there is a separate 3.3v board power supply that is used for the internal XDS110. It is interlocked with the (b) 3.3v OK.

    d) when you want to use Uniflash there is an SOP jumper code that is needed SOP2 jumper connected, SOP0 jumper connected.

    e) Uniflash communicates with the mmave sensor using the Control UART Tx and Rx. Uniflash needs to know the Control UART number.
    Turning the mmwave sensor power on and off, the appearance of the USB mapped UART (2) should show the Tiva processor XDS110 is available.

    f) if the two COM ports are not available, and the 3.3v board power supply is present, the TIVA processor can be reflashed, with its boot code.

    g) the 2nd UART port is the MSS Data Logger Emulation, this is used by the Out of Box demo as a data port.

    h) if the EVM is heating up, the mmwave sensor can be hot while running, the PMIC and LDOs provide up to 1A of current for each rail, these can be hot while running. If you push the RESET button on the EVM board, under normal conditions (unless you have flashed automatically booting code) the mmwave sensor should be in idle.

    Regards,
    Joe Quintal
  • 0 in reply to Radio-Joe
    Prodigy 30 points

    1. About my operations: 

    I can confirm from "device manager", com3(UART) and com4(data aux) are displayed

    I connected SOP0 and SOP2, and update com3 in uniflash settings, and send binaries

    Uniflash stuck on "Connect to com port com3", and fails with
    [6/7/2018, 9:39:09 PM] [ERROR] Cortex_R4_0: Initial response from the device was not received. Please power cycle device before re-flashing.
    [6/7/2018, 9:39:09 PM] [ERROR] Cortex_R4_0: Not able to connect to serial port. Recheck COM port selected and/or permissions.

    I used to be able to flash the binary, so I don't think my operation is wrong here.

    2. Power supply:

    I'm using 5V 2.5A  490-SMI36-5-V-P5.
    Not sure how to use 3.3V power supply. But I don't think it causes UBS power alert, as I've been able to use without problem for three days.

    3. Heating

    What is normal condition? I didn't flash automatical booting code, I just flashed the SDK demo and vital sign lab.

    I pressed RESET button, and yellow light turned off.
    Regardless press it once, or press-and-hold, mmwave chip, and the metal area in the back are still heating badly.

    I understand the chip will turned warmer when in use, but I feel the temperature is much higher compared with the working status I experienced a couple of days ago.

    I know there's don't-touch sign on the board, but I accidentally touched the area between FB1 and u10, that's very hot that can almost scald me.

    4. Assumption

    If there's possibility of short-circuit?
    Consider these two things happen consecutively
       I was using the metal brackets to hold the board horizontally which may connected some parts with metal
       alert of USB accessory using too much battery from mac while the board is connected with 5V-2.5A power supplier
       The board is heated worse than days ago
       UART and Data aux port shown on device manager, "hardware connected" on demo visualization
       demo visualization console does not show anything when sending configures
       uniflash can no longer open com3 port.

  • 0 in reply to Cesar
    Prodigy 30 points
    I tried SDK demo and vital sign lab, so I assume yes.
    Uniflash fails to open com3 port, before actually send binaries to EVM.
    So content of binaries do not matter here I assume
  • 0 in reply to Yicheng Fan
    TI__Mastermind 37500 points

    Hello,

    If you used to have a working board, and the board now fails to flash, and is very warm.

    The board is damaged.   If you follow the device schematic from the  " www.ti.com/.../IWR1443BOOST " under Design files.

    The schematic shows the 5v input power, and 5 power supplies.  If you can identify or have someone help you identify the faulty part, you may be able to

    have it replaced, otherwise you will need another EVM.   

    Using the IWR1443BOOST_RevA_Schematics.pdf, IWR1443-BOOST-Rev-A_Assembly.pdf, IWR1443BOOST_REV-A_BOM.xls, the 6 main radar supplies and 3v3board are shown in the figure on sheet 2 of the Schematic drawing.

    a) 5v input - 4.5 to 5.5vdc, P5 to DC Gnd - front of board

    b) PMIC DCDC converter output (4) - back of board

          L1 to GND - 1.2v (1.14 - 1.26) (*)

          L2 to GND -  3,3v (3.14 - 3.46) (*)

          L3 to GND - 2.3v (2.18 - 2.42)

          L4 to GND - 1.8v (1.71 - 1.89)

    c) RF LDOs 

         U4-1 to GND - 1.8vA (1.71 - 1.89) - front of board  (*)

         U5-15 to GND - 1.3vA (1.24 - 1.36) 1.3A RF1 - back of board (*)

         U5-11 to GND - 1.3vA (1.24 - 1.36) 1.3A RF2 - back of board (*)

    If (*) rails have power voltage + .4v higher, the EVM needs to be replaced, as the power failure has failed the mmwave radar device on EOS (Electrical OverStress)

    if U8 PMIC is hot with no active radar chirping, device needs to be replaced.

    d) 3.3v Board  -  power supply will not be correct output if L2 to GND measurement is incorrect.

         U9-3 to GND - (3.14 - 3.46)

    if metal contacted the board while it had DC power, the EVM is damaged, its possible if you can isolate the bad power supply you could repair it.

    if the USB com-ports are not seen when the USB is attached, the XDS110, may need to be reflashed.   The TM4C device has a PC based USB flash download utility

    Good Luck in your repair effort,

    Joe Quintal