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.

CC2651R3SIPA: EMI Peaks at 2.4GHz above FCC EMI Limits

Part Number: CC2651R3SIPA

Tool/software:


During our pre-scan of CC2651R3SIPAT0MOUR - there are peak spikes during normal operation (broadcast for pairing) and also during charging.

(1) Normal Operation : V-Scan : Fail, -0.86dB@2.426211GHz
(2) Unit Charging       : H-Scan : Fail, -1.79dB@2.401739GHz
                                     V-Scan : Fail, -1.99dB@2.426GHz
Q1. This module is already has FCC Certification, are these peaks acceptable?

Q2. Do we need to reduce the transmit power during pairing and how to configure in software?

Q3. If we reduce the advertisement interval time - what is the recommeded interval? 


  • Hi Mar,

    Q1. I will need help from the FCC certification experts currently out of the office until next week, I will come back to you as soon I have answers.

    Q2. Reducing the transmit power will reduce the EMI, take a look at HCI_EXT_SetTxPowerCmd() in the API documentation for information.

    Q3. The recommended advertisement interval time should be defined according to your use case. Take a look at the Advertising Parameters in the Scanning and Advertising Basics SLA for more information.

    Regards,
    Tanguy

  • Hello Mar,

    If the device is used according to the requirements noted in https://www.ti.com/lit/an/swra748/swra748.pdf then there should be no issue. I am curious about your test results as you haven't included the limits you are using, the test setup, or the actual measured values. It's not really feasible to evaluate the results without that information. The data provided doesn't really make sense as the FCC limit for the 2.4GHz – 2.4835GHz band is +21dBm and the maximum transmit power for CC2651R3SIPAT0MOUR is +5dBm. See https://www.ti.com/lit/an/swra048/swra048.pdf for more information on FCC limits.

    Thanks,

    Jake

  • Tanguy and Jake,

    Thank you.


    Settings/Set-up

    (1) The DUT is set for an Advertisement of for 20ms interval without link partner.
    (2) Test Limit settings : Peak+Limit = 74dBuV/m and AVG Limit = 54dBuV/M

    Do you mean on the band 2.4GHz – 2.4835GHz : FCC limit is +21dBm equivalent to  116.23 dBuV/m? Is my understanding correct?

  • Hi Mar,

    Your advertising channels line up with the information you provided above:

    2426MHz = Channel 38

    2402MHz = Channel 37

    Channel 37, 38 are the advertising channels so I would expect activity on them.

    For output power a conducted (connected to test equipment via coaxial cable) measurement is performed rather than a radiated measurement.

    What are you referencing when setting those limits? Those look like the limits for band edge testing but wouldn't be applicable in the intentional radiator frequency band.

    For my understanding, what specific FCC requirement are you trying to evaluate or test? For example the band edge evaluation might look something like:

    where any spurious emissions or intended emissions cannot have energy out past the band edge up to the limit provided. This image is from our certification report for CC2651R3SIPA (https://dr-download.ti.com/design-tools-simulation/certification/MD-gej3rHqrPC/7.20.00.00/CC2651R3SIPAT0MOUR_Cert_Reports_7_20_00_00.zip). Maybe it would be good to look at those reports and compare them against your results. These test reports describe the test conditions and the limits. If you're seeing similar results then you should be good.

    In particular maybe look at FR242614A_R01_Part15C_TI_CC2651R3SIPA.pdf which is the Bluetooth low energy report. Appendix E and F show the band edge limits. 

    BR,

    Jake

  • Hi Jake,

    Thank you for the insights, during the prescan the LAB just added the limits for spurious emission.

    On the further scan - the lab added rejection filter for the intentional radiator band(BLE advertisement channels) and we are passing.