• State
  • Locked Locked
  • Replies 6 replies
  • Subscribers 56 subscribers
  • Views 421 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.

RF section problem on CC2538

sankar Swaminathan
Prodigy 30 points

Other Parts Discussed in Thread: CC2538

Hi,

    As a hobbyist, I designed a board based on the CC2538 Eval board. I've retained the RF section design and layout as is. All other parts of the design is working except the RF section. I tried debugging the issue with SmartRF studio.

I found that the Receiver of my design is not sensing any signal when the eval board transmits in continuous TX mode.

Please provide some pointer or checklist to debug this.

-Sankar

  

  • 0
    TI__Mastermind 27145 points

    Hi,

    First make sure that the Crysal oscillator on the reciver board is running at the correct frequecy.

    Make sue all the Balun components are correct in value (as per the ref design) and no cold joins on the board whch includes RF Connector.

    Use the board as tranmitter (in CW MODE) and compare the response with the EV board. If the assembly is good then it should match the performance with the EV board.

    Then try it as a Receiver (in CW MODE -  Monitor RSSI).

    Thanks,

    PM

     

  • 0 in reply to PM
    Prodigy 30 points

    Hi PM,

         Attached is the document describing the testing senarios. Please let me know what we missed/fine tune.

    3124.SMART_RF_CC2538_Captures.doc

    Thanks,

    Sankar

  • 0 in reply to sankar Swaminathan
    TI__Mastermind 27145 points

    Hi Sankar,

     

    I  reviewed your RSSI - CW mode plots and observed the following.

    Your board has a loss of around 35dB in the RF path (may be cold joints or bad components or crystal offset) and spurious at the RF input. First better to check the crysatl frequency on your board. If it is not at the center then you need to tune the shunt caps.

    If you observe the RSSI (custom board as Rx) on your board, it shows around -50dBm without any input and around -45dBm with the input. This shows that there is spurious on the custom board. Whereas with CC2538EM, it shows around -100dBm without any input signal and around -35dBm witht the signal. 

    When you use cutom board as a transmitter, the receiver (CC2538EM) shows RSSI of around -70dBm, where as when CC2538EMs used as Tx and Rx, it shows RSSI of around -35dBm. This shows there is a loss of 35dB is the transmitter. The loss could be due to the offset in the transmit frequency or due to the broken RF path. If this is due the the offset in the transmit frequency then you need check your crystal frequency on the board. If it is not at the center then you need to tune the shunt capacitors of the crystal to correct the offset. 

    So, you need to debug the board by using Spectrum Analyzer.

    Thanks,

    PM

  • 0 in reply to PM
    Prodigy 30 points

    Hi PM,

         I managed to find the issue. The RBIAS resistor was not mounted properly. RF is working now.

    Thanks for your support.

    Thanks,

    Sankar

  • 0 in reply to sankar Swaminathan
    Intellectual 730 points

    hi all, excuse me if I write in this topic, I have also problems on the RF section of a custom board with cc2538, i'm focused on the crystal used in my board: a kyocera with 8pF load capacitors:

    http://global.kyocera.com/prdct/electro/pdf/xtal/cx2016db_wl_e.pdf

    could these be the problem (the ones on the EM board are 12pF) ?

  • 0 in reply to Giuseppe Giancola1
    TI__Mastermind 27145 points

    Hi,

    First measure the transmit frequency or crystal oscillator frequecy with spectrum analyzer or conuter. If  the freq offset is within the Crystal spec ( <10ppm) then it is OK. If the freq offset is more than the spec (>10ppm) then you need to tune the shunt caps (in your case 8pF) to bring the frequecy to the center.

    Thanks,

    PM