Because of the holidays, TI E2E™ design support forum responses will be delayed from Dec. 25 through Jan. 2. Thank you for your patience.

  • State Resolved
  • Locked Locked
  • Replies 7 replies
  • Subscribers 55 subscribers
  • Views 1902 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

CC1310: PCB trace antenna impedance measuring and matching

Devin Murphy
Intellectual 430 points
Part Number: CC1310

We have done a CC1310 design that uses the miniature helical antenna from DN038 and I am really struggling to get it tuned properly at 916MHz. I have followed AN058 to set up my board for measuring using a semi-rigid coax cable. I have calibrated my VNA (Keysight FieldFox) to move the point of reference to the end of the semi-rigid coax. The measurements look great with the open, short, and 50 ohm resistive load.

When I move to my board, I take the initial measurement using a 0 ohm series resistor to connect the antenna and get a value around 65 - j270 ohms @916MHz. I use some software to come up with some theoretical matches and when I start by adding a parallel 0.1pF capacitor the impedance change makes sense (49.5 - j237 ohms @916MHz). When I swap the 0 ohm series resistor for a 39nH inductor the impedance change makes no sense (1.1 + j85 ohms @916MHz).... In fact if I remove the parallel component the impedance with just the series inductor makes no sense. It doesn't really matter what capacitance or inductance value I use for the series component, the measurement never makes sense.

I tried a different cable and re-calibrating with no luck. I have tried a sanity check by measuring the impedance of the meandering monopole antenna on the LaunchXL-CC1310 and I am not sure that is making any sense either. I measured a starting impedance of 13.9 + j89 ohms @916MHz on the LaunchXL-CC1310 (R12 = 0 ohms). That impedance changed to 22 + j59 ohms @916MHz when I changed R12 to 2.2pF.

I am just hoping for some kind of explanation for the huge discrepancy between the change I was expecting and what I actually measured and/or suggestions for how I can verify my setup and calibration so that I know I can actually believe my measurements.

Thanks!

  • 0
    TI__Mastermind 36385 points
    Devin,

    Antennas are going to be very dependent on the surrounding, have you made sure that you have fixed the PCB during your testing to a common fixture and that this fixture is mostly "air". I put air in quotes, but typically I would use foam as a backing for antenna measurements because foam is very close to the dielectric constant of air.

    I write, this because as you say something does not make sense. Its typically the fixtures that influences the results and causes confusion.

    Regards,
    /TA
  • 0
    TI__Mastermind 31610 points
    Hi Devin,

    Agree with your matching values based upon the impedances. When the antenna matching and smith chart are not in alignment this is usually due to an inaccurate impedance measurement or the final casing is loading the antenna structure.

    Do you have the shielding GND connection of your semi-rigid cable close to the antenna point ?

    The antenna used on the CC1310 LP is not so dependent on the antenna match values since this has a natural resonance around 868/916 MHz. The helical antenna is sensitive to antenna match values since the natural resonance is significantly higher than 868 MHz. The resonance is basically set by the antenna match values.
  • 0 in reply to RGW
    Intellectual 430 points

    Thank you for the feedback, it does seem like I underestimated the impact of the enclosure / environment. I have just about got these values pinned down for my board, just waiting on some new components I ordered.

    Yes, I have the semi-rigid shielding connected to GND close to the antenna point.

  • 0 in reply to Devin Murphy
    TI__Mastermind 31610 points
    For example, the sensortag rubber casing we have has a significant effect on the resonance of the antenna. Even though this is only rubber / plastic this still effects the RF and loading of the antenna. If the casing is very close to the PCB antenna or even touching the antenna, this will act as a load and then the theoretical values obtained from the smith chart do not always align when measuring on the VNA.
  • 0 in reply to RGW
    Intellectual 345 points
    Hi Richard,
    what is the suggestion /options in such cases where plastic cases are planned to be used ?

    thanks
  • 0 in reply to Cristiano Gavião
    TI__Guru**** 317180 points
    Antenna tuning should always be done with the final casing in place to take into account the potential frequency shift due to the plastic/ material used.