• State TI Thinks Resolved
  • Locked Locked
  • Replies 10 replies
  • Answers 1 answer
  • Subscribers 34 subscribers
  • Views 1187 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.

AWR1642: Run-time periodical calibration is enable but the Tx power variances across different temperatures are still large

James Austrial
Expert 4170 points
Part Number: AWR1642


Hi,

SW: mmWave SDK 2.0.0.4

HW: AWR1642 ES2.0 customer board

We use mmWave demo in mmWave SDK 2.0.0.4 to test run-time periodical calibration and observe the Tx power variances across different temperatures.

The result is like:

AWR1642 temperature Start Freq.(GHz) Start(dBm) Stop Freq.(GHz) Stop(dBm)
-8℃ 76.118 -24.21 76.3 -23.85
0℃ 76.116 -24.34 76.304 -23.64
13℃ 76.116 -24.41 76.3 -23.29
54℃ 76.108 -25.45 76.221 -25.97
70℃ 76.1082 -28.72 76.2203 -28.63
75℃ 76.1079 -28.51 76.2206 -28.53
80℃ 76.107 -28.04 76.22 -29.06
85℃ 76.1085 -29.19 76.2197 -29.1
90℃ 76.108 -28.88 76.22 -29.32
115℃ 76.109 -29.41 76.221 -30.64

As you can see, the Tx power variance across temperature is pretty large (5~6 dBm) and the result is very different from "self-calibration" application note.

We're pretty sure the run-time periodical calibration in mmw demo is working since we can get the calibration report periodically.

Do we miss anything?

  • 0
    Intellectual 580 points

    Hi, 

        In TI doc. spracf4 Self-Calibration in TI's mmWave Radar Device.

    What's the test setup in Figure 2? Could I duplicate the test setup?

  • 0 in reply to liang yu chou
    TI__Genius 15005 points
    Hi Cheng,

    The plots in the application note are only representative plots - they may not reflect actual device performance.

    The TX power will degrade significantly if calibration is disabled. Even with calibration enabled, it is not always possible to maintain the full TX power at very high temperatures.

    Best Regards,
    Anand
  • 0 in reply to Anand Gadiyar
    Intellectual 425 points

    Sir, 

    May I know our current  result is reasonable? 

    If it is possbile to offer your EVM's related test result for reference?

    By the way, we also found the case at 2 way (TX & RX), we found the S/N is almost the same at different temperature. Is it also reasonable? If yes, can you explain why?

  • 0 in reply to JY Su
    TI__Genius 15005 points
    Hi JY,

    Could you please let us know how you are measuring the TX power - is it with continuous wave (CW) mode or with regular chirp mode?

    Thanks,
    Anand
  • 0 in reply to Anand Gadiyar
    Intellectual 425 points

    It is based on regular chirp mode. We apply different temperature condition with fixed location to our radar & spectrum receiver to check its TX output radiation difference.

  • 0 in reply to JY Su
    TI__Guru 58035 points
    Hello JY,
    The variation you are seeing is much higher than expected. The TX o/p power from the AWR device varies <1dB across temperature if the run time calibrations are enabled. 5-6 dB variation is not expected. Could there be some other path loss variation in the setup across temperature?

    >>By the way, we also found the case at 2 way (TX & RX), we found the S/N is almost the same at different temperature. Is it also >>reasonable? If yes, can you explain why?
    The S/N can remain same only if the TX o/p is actually similar. If that reduces then S/N would degrade. This indicates there could be some other issue in the TX o/p measurement setup.

    Regards,
    Vivek
  • 0 in reply to Vivek Dham
    Intellectual 425 points

    Sir,

    May I know if AWR1642 need power on for a while (such as 2min) to ready for actually S/N measurement?

    From TX power, you can find the package top temperature range from 70~115 degree with near the same TX power. 

    The other temperature range from 13~-8 degree are also similar. May I know if it is possible caused by using wrong calibration period?

     

  • 0 in reply to JY Su
    TI__Guru 58035 points
    Hello JY Su,
    The thermal heating will depend on the PCB and heat sink design on the customer board. If you want to do the measurement after the temperature has settled then you can wait for few mins before taking the reading, but its not mandatory to do that.
    This temperature increase is not caused by calibration, its just the average power consumption causing the heating of the module. You can reduce the temperature increase by reducing the duty cycle of the frames (ie increase the frame periodicity).

    Regards,
    Vivek
  • 0 in reply to Vivek Dham
    Intellectual 425 points

    Sir,

    My question is why the TX power output are different as power on vs. for a while, such as 10min, by using spectrum to read tx output level & the AWR1642 autocalibration is enabled?

    JY

     

  • 0 in reply to JY Su
    TI__Guru 58035 points
    Hello JY,
    In the OOB demo application the run time calibrations periodicity is set to 10frames, so the calibration will take place once every 10 frame.

    Have they isolated if the change in power is coming from the device and not from the setup? If it is radiated setup then I assume they would have come kind of radome material around the temperature chamber to allow for radiation. Some times the material itself could have a change in attenuation over temperature.

    Regards,
    Vivek