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.

IWR6843AOPEVM: Noise profile jumping

Part Number: IWR6843AOPEVM
Other Parts Discussed in Thread: IWR6843AOP, IWR6843

Hello,

I just received the IWR6843AOPEVM and did the out of the box test. The device is directly connected to the PC via micro USB (no boost board). When staring the "mmWave_Demo_Visualizer_IWR6843AOP" connection is established and I can access the device.

My problem is that the noise level is jumping from low to high periodically and in the point could I only see single points e.g. when pointing to a wall I just see 1 point and it is not constantly visible. I have try all different configurations but no change. Displaying the Doppler heat map I also observe this "jumps" in the back ground (from dark blue to yellow).

PS: the chip gets rather hot

Thanks for helping

Rainer

  • Hi Rainer,

    I've asked an expert to look into this and we should have an answer for you sometime next week.

     

    Cheers,

    Akash

  • Rainer,

       Could you please post the picture of your observation?

    Thanks and regards,

    CHETHAN KUMAR Y.B. 

  • Hello Chethan,

    here the required images. In the mean time I have installed the IWR6843 ISK with booth board - besides some problem with the COM connection I do not see this behavior. 

    Device "cold" just plugged to USB port.

    Device running for 1 - 2 minutes and the "jumping" of noise starts image 1 "low noise" image 2 "high noise" environment not changed

  • Rainer,

         You could try, lowering the duty-cycle (Reducing the frame rate) or Cooling the device would help. 

    Are you using the heatsink? 

    Thanks and regards,

    CHETHAN KUMAR Y.B.

  • Dear Chethan,

    lowering the frame rate only has no effect the chip gets hot very fast. I avoided cooling (a simple fan does not change the situation and ice spray may damage the chip). 

    Heat sink:

    1. There was no heat sink delivered with the EVM (only a metal holder which I use)

    2. The thermal contact between the chip and the back side of the PCB is very poor there is a foam inserted between the PCB and the black metal plate at the back of the PCB

    3. I would expect that a EVM works i.e. that the design was tested and that I do not have to take care about thermal management

    To be very frank I believe the device is either defective of not usable for us. I will try to get a replacement and see if the performance is better.

    Rainer

  • Rainer, 

            Yes, EVM has been tested. However, there is restriction on the duty-cycle on this EVM due to small form-factor.

    Please refer to below FAQ. 

    https://e2e.ti.com/support/sensors/f/1023/t/816332

    Yes, it would be good to check on a replacement unit. 

    Thanks and regards,

    CHETHAN KUMAR Y.B.

     

  • Hi Rainer,

    Thanks for the feedback, please see response below on the issues you point out.

    Rainer Nebosis said:
    1. There was no heat sink delivered with the EVM (only a metal holder which I use)

    The black metal plate on the back of the board is intended to be the heat sink

    Rainer Nebosis said:
    2. The thermal contact between the chip and the back side of the PCB is very poor there is a foam inserted between the PCB and the black metal plate at the back of the PCB

    The "foam" in the middle between the PCB and the heat sink is a thermal interface material used to transfer the heat from the back of the board to the heat sink.

    Rainer Nebosis said:
    3. I would expect that a EVM works i.e. that the design was tested and that I do not have to take care about thermal management

    The EVM was tested. Considering the size of the board it is expected to heat up compared to other larger EVMs in the mmWave portfolio. We ran duty cycle vs temperature test with and without the heat sink and confirmed that the temp stays below the junction temperature limit up to 50% duty cycle. Other extensive performance test / verification was also done.

    The noise issue you are experiencing is not common across all the IWR6843AOPEVM. Can you please try another EVM for your immediate development while we troubleshoot the faulty EVM to find the root cause?

    Regards,

    Charles O

  •    Could you please post the picture of your observation?

  • Hello,

         Below is the image of the noise profile. 

    Thanks and regards,

    CHETHAN KUMAR Y.B.