+ Post New Thread
Results 1 to 5 of 5
  1. #1
    Member level 1
    Points: 843, Level: 6

    Join Date
    Apr 2011
    Posts
    40
    Helped
    1 / 1
    Points
    843
    Level
    6

    Improve design performance Balun+Pi+Antenna

    Hello,

    We have a design with CC2520 texas instruments RF chip (2.4Ghz 802.15.4). I have to improve the output of this RF component. The out is a balun+Pi+Antenna.

    In balun and Pi network inductor are fixed (based on microstrip).

    Ok, so these are the steps iīm going to follow:

    1) mesasure VSWR of antenna in our enclosure. This way i know Z0 of antena.

    2) measure S11 of Chip+Balun+Pi configuring chip as receiver and using a attenuator.

    3) Modify Pi to match Chip+Balun+Pi>>>>Antenna

    Another thing that i can do is try to maximize the output power changing capacitors and then make all them make all the rest. I have measure the output power to the antenna and here are the results:

    Frequency Power
    2405 Mhz 2.88 dBm
    2445 Mhz 2.54 dBm
    2480 Mhz 1.96 dBm

    Do you think that i can improve the performance making another thing?

    I have also a question about measuring S11 of Chip+Balun+Pi. I canīt configure the attenuator of VNA so i bought an external one (20dB). Today i try to measure S11 but i canīt calibrate the equipment with the attenuator (if i do this, when i connect load/oc/sc it goes crazy!) and if i attach it after calibration S11 of the attenuator is taken into account and the measure is not valid. How can make this measurement?

    Click image for larger version. 

Name:	PCB.jpg 
Views:	26 
Size:	255.2 KB 
ID:	55765

    •   AltAdvertisement

        
       

  2. #2
    Advanced Member level 2
    Points: 6,438, Level: 19

    Join Date
    Jan 2007
    Location
    Sweden
    Posts
    610
    Helped
    270 / 270
    Points
    6,438
    Level
    19

    Re: Improve design performance Balun+Pi+Antenna

    Pout is relative low, why that big attenuator then? 20dB attenuation => 20 dB reduced dynamic range of VNA. How much power can your VNA handle?
    Measure TX impedance requires that the chip can be set in CW mode, I assume?
    Measuring VSWR of antenna do not result in impedance values? What type of VNA do you use?
    Measured TX out power level can be improved, in optimal case by 2 dB, which probably not will give the user any noticeable extended range.
    Do you plan to improve RX in similar way?
    If the circuit and PCB is designed in a acceptable way from RF view, but improvements in coverage still is needed, had I started with measuring the antenna efficiency. Improvements of the antenna is often the were most amount of dB's can be found. Improves both TX and RX.



    •   AltAdvertisement

        
       

  3. #3
    Super Moderator
    Points: 260,203, Level: 100
    Awards:
    1st Helpful Member

    Join Date
    Jan 2008
    Location
    Bochum, Germany
    Posts
    45,441
    Helped
    13826 / 13826
    Points
    260,203
    Level
    100

    Re: Improve design performance Balun+Pi+Antenna

    Some details are unclear, e.g. the attenuator point. A S11 (one-port) measurement with a 20 dB attenuator will be effectively useless.

    1) You would wan't to measure complex Z0 respectively S11 of the antenna rather than a scalar VSWR number to adjust the matching.

    2) The most effective method to optimize the balun matching would be load pulling.



    •   AltAdvertisement

        
       

  4. #4
    Member level 1
    Points: 843, Level: 6

    Join Date
    Apr 2011
    Posts
    40
    Helped
    1 / 1
    Points
    843
    Level
    6

    Re: Improve design performance Balun+Pi+Antenna

    Hi all! Thanks for the response.

    E Kafeman:

    Some details are unclear, e.g. the attenuator point. A S11 (one-port) measurement with a 20 dB attenuator will be effectively useless.
    Acording to Texas Instrumets AN068 (my VNAīs output power is -10dBm):

    10.1 Balun / Matching Network Measurements
    Measurements assure that the design is operating correctly. Placing the Radio in RX mode and connecting a VNA to the antenna port allows one to plot S11 for the balun and filter matching network. Make sure the VNA power output is turned down to -30 or -40 dBm so not to saturate the LNA and change its input impedance.
    E Kafeman:

    If the circuit and PCB is designed in a acceptable way from RF view, but improvements in coverage still is needed, had I started with measuring the antenna efficiency. Improvements of the antenna is often the were most amount of dB's can be found. Improves both TX and RX.
    How can i measure PCB PIFA antenna effeciency? If in another design i use chip antenna i canīt do anything about efficiency.

    FvM:

    A S11 (one-port) measurement with a 20 dB attenuator will be effectively useless. You would wan't to measure complex Z0 respectively S11 of the antenna rather than a scalar VSWR number to adjust the matching.
    So how can i measure Z0 with the attenuator? Yes iīm measuring S11. Sorry!

    FvM:
    2) The most effective method to optimize the balun matching would be load pulling.
    How can i make load pulling? What do i need?



  5. #5
    Advanced Member level 2
    Points: 6,438, Level: 19

    Join Date
    Jan 2007
    Location
    Sweden
    Posts
    610
    Helped
    270 / 270
    Points
    6,438
    Level
    19

    Re: Improve design performance Balun+Pi+Antenna

    Are you sure that it is a VNA you have? It sounds more like a SNA (Scalar Network Analyzer). With a VNA can you measure complex impedance and at these instruments is it much a standard function to be able to adjust output power. With a SNA, and at these frequencies can RX impedance matching be done by trial and error. Solder SMD inductors and caps at measurement point until reactive part is canceled, remaining mismatch is then resistive or do the other way and start with resistors. Bridges and adjustable components is hard to use due to its size or leg connections lengths which becomes a part of the impedance network.
    You must reduce your VNA output power or get hold of a VNA with such adjustments as you else will measure at a saturated RX chip. In worst case can RX chip be fried with -10 dBm power.
    Antenna efficiency is commonly measured in an anechoic chamber or free field. Chip antenna is a good start as there are a lot that can be improved. For this kind of antenna is the antenna itself not a good radiator. Main radiator is the ground plane. Start with checking antenna center resonance frequency and if needed adjust with matching components or distance between ground and antenna. Depending on ground plane size can total antenna performance also be very affected on where on PCB the antenna is placed. Best improvement for a chip antenna, if there is enough space, is to replace it with a real antenna (personal opinion).
    If you not have any tools to do required impedance measurements, can some matching be done on lab bench. Set up a relative reflection free test range of 1 meter and do practical RSSI tests. Replace a matching component and do a new test. Repeat until you is satisfied.



--[[ ]]--