• State
  • Locked Locked
  • Replies 15 replies
  • Subscribers 40 subscribers
  • Views 2083 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.

CC3100: SAR Testing FCC cc3100 Fails

Sandro Burn
Intellectual 315 points
Part Number: CC3100

Hi all,

I'm doing now at the moment the FCC SAR Certification Tests. And actually the WiFi cc3100 chips fails for the "Average conducted power measurements (CPM)".

For these measurements we created a tune up table, shown all power levels for the given modes of the wifi. Now the theoretical value shown in the datasheet is 18dBm for 1DSSS. The measured value at the output of the wifi chip is 11.50dBm. We have a bandpassfilter at the output which gives us 1dB difference but still, there are 5.5dB to less at our output.

Does anyone has an idea why this happen?


- We are running with the newest Service Pack and the newest Radio Tool. We ve created our own CTI for the Radio tool, but the software on the mcu is exaclty the same as in the radio tool.

What we also saw is that the duty cycle is very unstable! What could be the reason of such an unstable duty cycle?

Thank you for your help.

  • 0
    TI__Guru 64405 points

    Hello,

    Just to verify, are you using the max TX power (i.e. 0 and not 15)? This number stands for the backoff in dB from maximum power.

    With Radio Tool, are you using packetized mode with delay or continuous?

    With continuous, you cannot control the delay between packets as it is all drived internally by the device where with packetized mode, you control the delay.

    If you are using contiuous mode, can you try packetized instead and make sure the duty cycle is stable?

    Thanks,

    Shlomi

  • 0 in reply to Shlomi Itzhak
    Intellectual 315 points
    Hi Shlomi


    We are using max Power (so value 0). We are using continuous mode. For the Carrier Wave only we had the desired average output power, but the power of the continuous is much lower, maybe because of the unstable duty cycle.

    Do we have some other chances to get a stable continuous mode? Because If I understand the radio tool description correctly, we should use this mode for FCC certification.


    Thanks,
    Sandro
  • 0 in reply to Sandro Burn
    Intellectual 315 points
    Hi Shlomi

    Ok, I've done some measurements now.

    - Mode: Packetized, Channel: 3, Rate: 1M, DataPattern: ALL1, Power: 0, CCA: 1, Size: 1400, Amount: 0
    — duty cycle: 97% (almost constant)
    — peak 18.54dBm
    — bandwith 14.4MHz
    — average Power: 10.9 dBm


    So the average Power is still to low. We should have around 17dBm for the b-mode, so that means around 6dB to low.

    Thanks,

    Sandro
  • 0 in reply to Sandro Burn
    Intellectual 315 points
    Hi Shlomi

    Ok, I've done some measurements now.

    - Mode: Packetized, Channel: 3, Rate: 1M, DataPattern: ALL1, Power: 0, CCA: 1, Size: 1400, Amount: 0
    — duty cycle: 97% (almost constant)
    — peak 18.54dBm
    — bandwith 14.4MHz
    — average Power: 10.9 dBm


    So the average Power is still to low. We should have around 17dBm for the b-mode, so that means around 6dB to low.

    Thanks,

    Sandro
  • 0 in reply to Sandro Burn
    TI__Guru 64405 points

    Hello Sandro,

    Continuous may load the internal PA. This is why I suggested packetized in which duty cycle is stable and can be compensated for.

    In any case, looping in an hardware expert to provide feedback.

    Regards,

    Shlomi

  • 0 in reply to Shlomi Itzhak
    Intellectual 315 points
    Hi Shlomi

    That would be great if you can as a hardware expert.

    Maybe there is still another issue which influences, that we do not have the wanted average output power.

    Thanks and looking forward to your answer.

    BR
    Sandro
  • 0 in reply to Sandro Burn
    TI__Genius 12740 points

    Hi Sandro,

    Can you tell me how you are measuring the average power? is it will the power meter or spectrum analyzer. Please provide some details of the measurement setup so I can try and understand the issue. Also have you checked this with our version of the RadioTool? What about validation on our CC3100 BP?

    Thanks,

    Riz

  • 0 in reply to Rizwan Murji
    Intellectual 315 points
    Hi Riz,

    Actually the Certification Laboratory is measuring the average power with a power meter (wide bandwith). There we had around 11.50dBm instead of 17dBm.

    Afterwards we have done some software changes, meaning we have changed from continuous mode to packetized. These measurements were done with a spectrum analyzer.

    This gaves me the following results:
    - Mode: Packetized, Channel: 3, Rate: 11M, DataPattern: ALL1, Power: 0, CCA: 1, Size: 1400, Amount: 0
    — duty cycle: 97% (almost constant)
    — peak 18.54dBm
    — bandwith 14.4MHz
    — average Power: 10.9 dBm



    Just for understanding our hardware. The antenna is no longer connected. This means, we have directly connected some cables behind the bandpass filter (insertion loss 1.02dB). The bandpass filter is then directly connected to the output of the cc3100 (RF_BG Pin 31) and to GND.

    In the datasheet (cc3100 page 11 "4.8 WLAN Transmitter Characteristics") all output power parameters are shown. So our tune up table which is required for SAR testing is based on this table. Because they want to measure the worst case, we assumed a tolerance of the chips from around 2dB (can you give us exact max and min values, because only the typ. is shown in the datasheet).

    Moreover the SAR certification says, we have to take the highest possible value. Then our EUT (testing sample) has to be in the range of 2dB from this max value.


    For example: 1 DSSS -> 18dBm + 2dB tolerance - 1dB (Bandpassfilter ) = 19dBm

    So our EUT mus be between 17dBm and 19dBm.

    And finally with both measurements, power meter (11.50dBm) and spectrum analyzer (10.9dBm) we are much lower. For the measured values from the power meter and spectrum analyzer, the cable damping and the duty cycle is already taken into account.

    Actually we are not sure if there is really a hardware or software issue. Maybe we are doing just an error during the measurements. Because as I explained before. The CW (Carrier Wave) measurements are showing the correct power. But actually this is not allowed for the SAR certification.


    Thank you very much for your help. We were able to move the final measurements to next Monday. But actually then we have to show a solution.

    BR
    Sandro
  • 0 in reply to Sandro Burn
    TI__Genius 12740 points

    Hi Sandro,

    Overall I don't see an issue with your technique. The maximum power out will be on Channel 7 as that is where the calibration is done. We see around +/-1.5dBb variation from part to part. In addition CH 1 and CH11 are reduced around 1dB to meet spectral mask requirements.

    The difference between the peak and average should be approximately 2dB for 11b modulations. The fact that you are not getting this (your are seeing 8dB PAR) does not seem correct. The settings of the radio seem ok. I would suggest you try the following:

    1) ensure that when you are in packetized mode that the power measurements are taking into account duty cycle correction. As an example, if you have a 50% duty cycle, you will actually be 3dB higher.

    2) Have the lab measure the CCDF or PAR of the 1DSSS or 11CCK signal to ensure a proper PAR. If theyare not getting ~2dB then the signal being transmitted may no b correct.

    3) We have done measurements using the 1QView and we see a 2dB difference between average and peak.

    I hope this helps. I will keep dinging and if I come up with any other ideas I will post them here. Feel free to also ping me off-line at rizwan.murji@ti.com and send me additional data of the FCC measurements for review.

    Thanks,

    Riz

  • 0 in reply to Rizwan Murji
    Intellectual 315 points

    Hi Riz

    Thank you very much. I have sent you a mail with some small questions.

    Kind Regards
    Sandro

  • 0 in reply to Sandro Burn
    Intellectual 315 points

    Hi Riz,

    Just one more question about the reduced power level for the top and bottom channel. If we want to do the CE certification as well, is then channel 11, 12 and 13 1dB reduced? Because
     with our testing software we are not able to differentiate between FCC and CE. So I think all channels from 11 to 13 are then reduced. Is this correct?

    Thank you very much.

    BR

    Sandro

  • 0 in reply to Sandro Burn
    TI__Genius 12740 points
    Hi Sandro,

    12 and 13 are not reduced. These channels are not valid in the US (only channels 1 to 11). Thus power was not reduced on these channels to pass FCC requirements.

    Thanks,
    Riz
  • 0 in reply to Rizwan Murji
    Intellectual 315 points

    Hi Riz

    Yes, I know, but channel 12 and 13 are used for the CE certification and there we have to do SAR testing as well. So are then all channels reduced or just the channel 13? Because we can't differentiate between US and EU in the RadioTool

  • 0 in reply to Sandro Burn
    TI__Genius 12740 points
    Hi Sandro,

    The radio tool does not allow control over the region. It cannot distinguish between the different regions. So channels 12 and 13 will be active on the radio tool. The FW itself is where you will set the region code. The FW defaults to US region thus when running an application channels 12 and 13 will be disabled.

    I hope this helps.

    thanks,
    Riz
  • 0 in reply to Rizwan Murji
    TI__Genius 12740 points
    Hi Sandro,

    I will go ahead and close this e2e thread as we are discussing over e-mail.

    Thanks,
    Riz