• State Resolved
  • Locked Locked
  • Replies 10 replies
  • Subscribers 53 subscribers
  • Views 1663 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.

question on the RF front end for CC2640

Shardul Bhosale95
Prodigy 160 points

Other Parts Discussed in Thread: CC2640, CC2520

I have some question on the RF front end for CC2640. The application circuit shows a T network before the 50 Ohm antenna.
The T is followed by a Balun.

1. I would like to know the output impedance of the chip.
2. The output impedance of balun.
3. The impedance conversion value of T network.
4. Is the T part of antenna matching to 50 Ohms? Or is it for matching the balun to 50 Ohms antenna?

This information is not available in any TI document as I know.

  • 0
    TI__Expert 6995 points

    Hi Shardul, 

    We don't generally publish impedance numbers since they are generally not sufficient to make a good matching  network for 2.4 GHz RF. We have rather focused on designing and publishing several different matching networks and front-end configurations that we have tested and characterized. Perhaps we have something that could suit your requirements already tested. Can you share some information on your requirements?

    Peder

  • 0 in reply to Peder
    Prodigy 160 points
    Hi Peder,

    Thanks for the reply. I have visited all TI configuration. They are generally good if the implementation is on a single flat PCB. In our case the product is more like a cube with sides of 30mm. Its quite compact and there are lots of components around. The over all dielectrics is high around the antenna which makes it electrically bit longer. The antenna is PIFA type but the ground plane is moved further away to reduce capacitance and increase bandwidth. The results are quite good and antenna has much better bandwidth and return loss. The antenna is electrically longer then L/4 (inductive) but is tuned in using external matching with good antenna aperture, radiation impedance and bandwidth. Its not long enough to produce significant side lobes. Typical values RL are -19.4 dB at 2.4GHz, -30.18 dB at 2.45 GHz and -19.3 dB at 2.5 GHz.
    When connected to CC2640 with all the network provided by TI the range/RSSI is not good. The T network was found to be quite lossy. We removed the T network and the range/RSSI was very good as compared to third party products.
    TI do publish Chip impedance for other ICs like CC2520 etc. We are also interested in chip balun which might answer more questions.

    Regards

    Shardul
  • 0
    TI__Guru** 106705 points
    Hi Shardul,

    1) processors.wiki.ti.com/.../CC26xx_Optimal_Load_Impedance

    2), 3) and 4) The output of the T-filter and balun is close to 50 ohm. The T-filter is mainly for filtering and does not shift the impedance of the balun a lot.

    If you change or remove the T-filter I highly recommend that you verify that radiated 2nd and 3rd harmonics are within regulatory requirements.

    You say that the T-filter is quite lossy, that should not be the case. It can be a result of changes to layout and ground connections of the filter compared to the reference design.

    Cheers,
    Fredrik
  • 0 in reply to Fredrik K
    Prodigy 160 points

    Hi Fredrik,

    Thanks for you help. This answers all the questions. The T filter needs to be investigated more and may be the inductors are not from correct series as used by reference design. If compliance is the issue then we would include the filter in the design.

    Regards

    Shardul

  • 0 in reply to Fredrik K
    Prodigy 170 points

    Hello, 

    Balun network has impedance of 50 ohm right?

    And do I have to design T filter of 50 ohm impedance? or  I should care about only cut off frequency of T filter regardless of its impedance ?

  • 0 in reply to Ameya Gambhir
    TI__Guru** 106705 points

    The output of the balun is 50 ohm, yes. The filter should thus have 50 ohm impedance.

    Again, I would highly recommend just following the reference design.

  • 0 in reply to Fredrik K
    Prodigy 170 points
    Hello,
    1. The filter contains only L and C components then how could be its impedance real with 50 ohm?

    2.My understanding was if balun impedance is 50 Ohm and filter has impedance of +j10 then it would shift matching point to 50+j10, with VSWR of 1.2 which is acceptable. Is not it correct?
  • 0 in reply to Ameya Gambhir
    TI__Mastermind 18780 points
    Both filter and balun is close to 50 ohm.
  • 0 in reply to CHS
    Prodigy 170 points
    Hello,
    Still my both questions are unanswered.
    1. If filter has only L,C components then how could be its impedance 50 Ohm?

    2. My understanding was if balun impedance is 50 Ohm and filter has impedance of +j10 then it would shift matching point to 50+j10, with VSWR of 1.2 which is acceptable. Is not it correct?
  • 0 in reply to Ameya Gambhir
    TI__Guru** 106705 points
    It is not exactly 50 ohm, but close enough. You can see for yourself here: cgi.www.telestrian.co.uk/.../smiths.pl