• State TI Thinks Resolved
  • Locked Locked
  • Replies 17 replies
  • Answers 4 answers
  • Subscribers 51 subscribers
  • Views 859 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.

CC2650: KC certification issue related to SMART RF STUDIO 7

Sejin Park
Intellectual 430 points
Part Number: CC2650

We faced a difficulty for KC certification in Korea.

We developed a cc2650 based Bluetooth enabled product, and need the KC certificate, which is mandatory in Korea.

We send the product to test department, and got a feedback like below video clip.

Please refer the link below for the test result.

 

https://drive.google.com/file/d/1T1ujZFzYzpzlZ1oFuBIHOhFKgGIaU8QF/view?usp=sharing

 

In the video clip, the central frequency is keep changing, and because of this, we could not meet the test procedure of KC certificate.

 

 

We used SMART RF STUDIO 7 and cc devpack debug board for the setup.

The problem is the generated signal is not periodic. We even tested the same procedure with cc2650STK (original one from TI) and it could not pass the test.

 

Please help us to resolve this issue.

 

Best regards,

>Sejin

  • 0
    TI__Guru** 106705 points
    Hi Sejin,

    The spectrum analyzer is configured for zero-span. You are not looking at frequency offset, but transmitted packets over time.

    Regards,
    Fredrik
  • 0 in reply to Fredrik K
    Intellectual 430 points

    Hello

    Thank you for your support.

    Can SMART RF STUDIO 7 generate periodic signal with default setting?

    If we use continous TX, I think the generated signal might not be periodic.  That is why the spectrum analyzer could not find the central frequency.

    Please let me know your opinion.

  • 0 in reply to Sejin Park
    TI__Guru** 106705 points
    Hi Sejin,

    Yes it can if you configure it to send packets. Based on the video you posted, it seems to be that you have configured Studio correctly for this mode (since you are looking at the time domain, what you see are the packets being sent at a regular interval).

    If the figure shows what you want to measure, then you are pretty close. You should just zoom in a bit and perhaps trigger on a certain power level.

    If you want to look at the central frequency you have to look at the frequency domain. Since you are sending packets and not a continuous signal you need to turn on max hold and let the device run for a while to see the frequency output.

    Regards,
    Fredrik
  • 0 in reply to Fredrik K
    Intellectual 430 points

    Hello

    Please see the attached image.

    We found a spike signal for every consecutive waveform.

    We just used the test command with the guideline. Please let us know if we can remove the spike.

    Best regards,

    >Sejin

  • 0 in reply to Sejin Park
    TI__Guru** 106705 points
    Hi Sejin,

    According to what you posted before you are supposed to measure the average power. Does the spike matter?

    Regards,
    Fredrik
  • 0 in reply to Fredrik K
    Intellectual 430 points

    Hello Fredrik

    Thank you for your kind assistance.

    Please let us know what is the root cause of this spike signal.

    The cc2650 sensortag shows same situation.

    I don't know exactly which part of the KC certification affected by this issue. I will keep update later.

    Best regards,

    >Sejin

  • 0 in reply to Sejin Park
    TI__Guru** 106705 points
    Hi Sejin,

    I belive what you are seeing is just a little overshoot from the PA turning on.

    Regards,
    Fredrik
  • 0 in reply to Fredrik K
    Intellectual 430 points
    Hello Fredrik
    Thank you for your quick response.
    So, is it common for most bluetooth RF circuit?
    I don't know why our test department raised this issue.
    Is there any guideline for this kind overshoot?
    If we analyze this signal with spectrum analyzer, it might affect harmonic distortion factor.
    Best regards,
    >Sejin
  • 0 in reply to Sejin Park
    TI__Guru** 106705 points
    Hi Sejin,

    It is common and it is not a problem. Our evaluation boards all pass the most common regulatory limits with margins.

    Regards,
    Fredrik
  • 0 in reply to Fredrik K
    Intellectual 430 points

    Hello Fredrik

    We faced difficulty for this issue with KC certification.

    According to test lab, the shape of the signal is overshooting signal. with this kind of  signal, they cannot start the test

    For KC certification, they want normal operating signal that does not have over shooting signal for test.

    Please provide us any evidence, that the spike or overshoot is not a issue for the certification.

    Best regards,

    Sejin

  • 0 in reply to Sejin Park
    TI__Guru** 106705 points
    Hi Sejin,

    If you provide me the requirements for the test specification you are failing, it will be more clear to me how this is a problem.

    There are several products with CC26xx in the market certified for the Korean market. What you are asking about has never come up as an issue before.

    Regards,
    Fredrik
  • 0 in reply to Fredrik K
    Intellectual 430 points

    Hello

    Thank you for your support.

    The issue is described below.

    1. If we use Smart RF studio 7, the generated signal is not periodic.

    2. If we use "HCI_EXT_ModemTestTxCmd", the generated signal has spike at the every risiing part. Actually, the test case needs actual bluetooth signal not test waveform. 

    3. If possible, please let us know the contact in TI for KC certification related issue in Korea.  (This certification revised recently)

    Best regards,

    >Sejin

  • 0 in reply to Sejin Park
    TI__Guru** 106705 points
    Hi Sejin,

    1. SmartRF Studio can send both continuous Tx and packets. Refer to the documentation.
    2. HCI_EXT_ModemTestTxCmd only transmits continuous signals. For packets you need to use HCI_LE_TransmitterTestCmd. - You still have not explained why the peak is a problem though.
    3. I can help you, but you need to explain what regulation you are violating and why.

    Regards,
    Fredrik
  • 0 in reply to Fredrik K
    Intellectual 430 points

    Hello

    Thank you for the assistance.

    Please refer the belows.

    1. Does SmartRF Studio generated continuous Tx and packets are periodic?
    2. The feedback from test department is that, In normal operation (e.g. A smartphone communicate with module via bluetooth), peaks is not observed. 
    The test case assume a real communication scenario not simulated test command.

    Best regards,
    >Sejin

  • 0 in reply to Sejin Park
    TI__Guru** 106705 points
    Hi Sejin,

    1. As I wrote in the previous post, SmartRF Studio can generate both continuous TX, and transmit actual data packets.
    2. To transmit "real" packets, you can either use SmartRF Studio or HCI_LE_TransmitterTestCmd.

    Regards,
    Fredrik
  • 0 in reply to Fredrik K
    Intellectual 430 points

    Hello

    Thank you for your support.

    Please refer below.

    1.  Please refer the video clip I posted above.  I tested Smart RF Studio, but we have issue with central frequency change issue. 

    https://drive.google.com/file/d/1T1ujZFzYzpzlZ1oFuBIHOhFKgGIaU8QF/view?usp=sharing

    This is Ok in previous KC certification, However recent  changed procedure describes it is mandatory to fix central frequency, I think this is caused by unperiodic signal generated by Smart RF Studio.

    2, The feedback from test department is that, all other vender supplied Bluetooth test suite does not have spike signal.

    I think they might have little experience with TI product.  I am consider to change the KC certification counter part, but again we do not have much information or document to pursade them.

    Please help us to provide detail technical information.  We already purchased 1st production chipset, but could not start production due to this KC certification issue.

    Best regards,

    >Sejin

  • 0 in reply to Sejin Park
    TI__Guru** 106705 points
    Hi Sejin,

    I already commented on the video. You are not showing frequency changing. What you are seeing is RF power vs. time and the trigger is set such that the packets start at different positions in the display.

    The frequency is not jumping in this video!

    To be very honest, it does not sound like the test lab know what they are doing. I would recommend you to take your certification elsewhere.

    Regards,
    Fredrik