• State Answer Suggested
  • Locked Locked
  • Replies 14 replies
  • Answers 1 answer
  • Subscribers 52 subscribers
  • Views 5968 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.

CC2640: CE/FCC certification & testing issue with FCC test sample code

Ankit Channa
Intellectual 590 points

Part Number: CC2640

Hi,

For CE/FCC certification we have used code from below link. 

http://processors.wiki.ti.com/index.php/FCC_ETSI_Test

We have checked the channel bandwidth and got 400KHz; which needs to be more than 500KHz using above code but it did not provide the expected result.  

We require all 1's test pattern to be transmitted with modulated.  Currently, we are using below commands for the same.

HCI_LE_TransmitterTestCmd(10,1, HCI_DIRECT_TEST_PAYLOAD_0xFF );  //This is only for modulated.

This API's working and got occupied BW more than 500KHz.  We did not find code/API's for un-modulated signal transmission. OR there isn't  a parameter to use same API (HCI_LE_TrasmitterTestCmd) to transmit the raw data (Un-modulated). 

Can we use above API's for CE/FCC certification? Also please share the source code for un-modulated signal.  

Please suggest as soon as possible.

Thanks,

Ankit Channa

  • 0
    TI__Guru** 106705 points
    Hi Ankit,

    Are you saying that HCI_EXT_ModemTestTxCmd(HCI_EXT_TX_MODULATED_CARRIER,19); did not give you OBW of more than 400 kHz? That does not sound right. CC2640 OBW for EN 300 328 should be around 1 MHz, and 6 dB BW for FCC 15.247 should be above 700 kHz.

    It is also not correct that you should transmit all 1s, you need to transmit random data to get the correct BW. If you are using HCI_LE_TransmitterTestCmd you must use this payload: HCI_DIRECT_TEST_PAYLOAD_PRBS9.

    You can use HCI_LE_TransmitterTestCmd, but note that this will send bursts of packets, not a continuous modulated signal like HCI_EXT_ModemTestTxCmd. This makes the latter easier to use in many cases. Also, there is no reason for these commands to have different BW, they are transmitting the same way.

    For unmodulated TX you should use: HCI_EXT_ModemTestTxCmd(HCI_EXT_TX_MODULATED_CARRIER,39);

    Regards,
    Fredrik
  • 0 in reply to Fredrik K
    Intellectual 590 points

    Hi Fredrik,

    Thank you for the reply.

    We have verified it again with HC_EXT_ModemTestTxCmd() and seems like we got OCW more
    than 500K for both modulated and unmodulated tx.

    The earlier requirement was to transmit all 1's by the lab. Please check image for OCW of modulated tx and let me know if the settings are correct or not.


    The issue with FCC test image was that the output power was getting reduced after two freq. viz 2440 & 2480 Mhz. In the init we have set the power to 4 dBm so for the first key press the output power was around -3.7 dbm but when pressed again it was getting around -9.8 dbm. So we did set the power upon each key press & after HCI_EXT_ModemTestEndCmd() still the output power is observed to be less than power at 2402 Mhz frequency for other two frequencies .

    We expect you to please verify the settings OCW and confirm the same so that we could proceed with CE certification.

    Also please do let us know what could be the solution for setting the power properly? Currently we are using HCI_EXT_SetTxPowerCmd(HCI_EXT_TX_POWER_4_DBM) command to set power.

    Regards,

    Ankit Channa

  • 0 in reply to Ankit Channa
    TI__Guru** 106705 points
    Hi Ankit,

    Good to hear that you got the expected results. Your statement about getting OBW > 500 kHz for unmodulated TX does not make sense, but I guess this is just a typo or some kind.

    For ETSI EN 300 328 OBW you should measure the 99% OBW. For FCC it is 6 dB bandwidth, in other words, your current 75% configuration is wrong.

    As you can see from the typical characteristics in the datasheet, the output power is slightly lower at higher frequencies, so your observations are expected (as long as the difference is small).

    The TX power command you are using is correct. As HCI_EXT_ModemTestTxCmd() defaults to 0 dBm, you have to call this before every test as you are now doing.

    Regards,
    Fredrik
  • 0 in reply to Fredrik K
    Intellectual 590 points

    Hi Fredrik,

    As mentioned, we have measured the OBW and 6 dB bandwidth for CE and FCC verification. Below are the two images capture for 2402 MHz freq testing for modulated signal.

    And also as you mention that the output power is slightly lower at higher frequencies so what should be the maximum difference in dBm ?

    Can you please confirm that the CE and FCC testing will get pass with below results?  

    Thanks & regards,

    Ankit Channa

  • 0 in reply to Ankit Channa
    TI__Guru** 106705 points

    These measurements look to be within the requirements.

    For details on TX power variation refer to the datasheet.

    Cheers,

    Fredrik

  • 0 in reply to Fredrik K
    Intellectual 590 points
    Can you please confirm that the CE and FCC testing will get pass with above results?
  • 0 in reply to Fredrik K
    Intellectual 590 points

    Hi Fredrik,

    Using standard sample code simplebleperipheral and using advertisments enabled we are getting below output power on spectrum. But after using FCC test the output power is less than that. 

    Below are the test results for output power.

    Using simpleBLEperipheral image & standard advertising with interval 100 ms

    2402.6 MHz= -4.5 dbm

    2426.12 MHz= -4.7 dbm

    2479.76 MHz= -5.9 dbm

    FCC test image

    2401.76 MHz= -7.7 dbm

    2439.756 MHz= -8.7 dbm

    2479.743 MHz= -10.0 dbm

    Why there is so much differencce in output power for HCI_EXT_MODEM_TestTxCmd()? and the power was set to 4 dBm for every end test command.

    Regards,

    Ankit Channa

  • 0 in reply to Ankit Channa
    TI__Guru** 106705 points

    This can be caused by issues with you HW design, or by how you do the actual measurements. For the difference between SimplePeripheral and the test code, are you sure you are setting the output power to 4 dBm in the latter case?

  • 0 in reply to Fredrik K
    Intellectual 590 points
    The hardware is same and output power set to 4 dBm for both sample code & test code.
  • 0 in reply to Fredrik K
    Genius 3630 points

    Hi Fredrik,

    As I understand, for unmodulated Tx, HCI_EXT_ModemTestTxCmd(HCI_EXT_TX_UNMODULATED_CARRIER,39);

    But, your comment mentioned "For unmodulated TX you should use: HCI_EXT_ModemTestTxCmd(HCI_EXT_TX_MODULATED_CARRIER,39);"

    it looks like typo. right?

    And the above command will generate the continuous Tx on 39 channel with no packet. right?

    Ji Won

  • 0 in reply to Ji-won Lee
    TI__Guru** 106705 points
    Hi Ji Won,

    Yes, that is a typo.

    HCI_EXT_ModemTestTxCmd(HCI_EXT_TX_MODULATED_CARRIER,39); will generate continuous, modulated, TX on PHY channel 39. No packets.

    Regards,
    Fredrik
  • 0 in reply to Fredrik K
    Genius 3630 points
    still typo... ^^

    Anyway, thanks a lot for your supports.
  • 0 in reply to Ji-won Lee
    TI__Guru** 106705 points
    No typo this time :-)

    _MODULATED_ will generate ...modulated...
  • 0 in reply to Fredrik K
    Genius 3630 points
    got it. it will generate modulated Tx, but no packets.