ad: elecraft

Why does this NanoVNA SWR sweep of a 70cm antenna show small waves within the larger wave?

Discussion in 'VHF/UHF - 50Mhz and Beyond' started by MRBOMBASTIC, Jul 5, 2021.

ad: L-HROutlet
ad: l-rl
ad: l-BCInc
ad: Left-3
ad: L-Geochron
ad: abrind-2
ad: Left-2
ad: L-MFJ


    Per the image below, I've just re-calibrated and then connected my NanoVNA to analyze the SWR of a 1/4 wave antenna tuned for 70cm. The horizontal axis is 100mhz - 500mhz.

    Why would the resulting graph have these small recurring waves within the larger (and understandable to me) trend of an SWR decrease with increasing frequency? I've seen this before, though rarely, while using the NanoVNA but I have never seen this amount of deviation from the larger trend line.

    Would it make sense for a 1/4 wave 70cm antenna to exhibit this fluctuating high and low SWR? Or is this indicative of some error in my measurement method?

  2. KP4SX

    KP4SX Premium Subscriber QRZ Page

    Too wide of a scan.
  3. K6CLS

    K6CLS Ham Member QRZ Page

    Look at the data plotted on a Smith chart.

    And yeah, too wide of a scan.
  4. KM3F

    KM3F Ham Member QRZ Page

    Out of band the impedance variations are due to the resonance/ impedance combinations of feed line and antenna reactance values that occur with frequency change.
    That's why you should set the sweep width to just below and just above the antenna design frequency.
    Then you don't see those variations that your not interested in.
    If you were to sweep a dual band antenna over the width of both bands, you would tend to see the same swings>between< the bands and would not care about that either.
    Also testing an antenna system from the radio end of a long coax with a regular Analyzer and see swings, that would tend to suggest the coax termination at the other end has a fault like a short or an open for the same reasons instead of a good low SWR at intended frequencies..
    A useful diagnostic function when testing for problem before taking things apart.
    For example, I had a problem with 7/8 inch feed line test that gave me a reading like that.
    First thought was the feed line cannot take in moisture just from it's construction so it must be in the UHF adapters.
    Sure enough that was it.
    Cleared the issue , places a flat dummy load, retested as a perfect 1 to 1 SWR with the same Analyzer on the other end.
    Lessen learned is seal the connections better for the long term.
    Another observation you can see on your display is the peaks and valleys from low frequency to high frequency get closer together as you would expect because the wave length gets shorter as F goes up.
    Good luck.
    Last edited: Jul 7, 2021
    KA0HCP likes this.
  5. SM0AOM

    SM0AOM Ham Member QRZ Page

    What you are seeing are the combined effects of the imperfect directivity in the directional bridge in the VNA, and multiple small reflections along the cable run.

    These add and subtract together as a function of cable length and frequency, and form a ripple on the average reflection coefficient curve. Usually the magnitude of this ripple is insignificant compared to the main curve amplitude.

    W6KCS and KO4LZ like this.


    KM3F, SM0AOM - Your points are very helpful.

    It doesn't seem to be the width I choose to analyze as some suggested but more likely something along the lines of your responses. When I use a short jumper to connect the NanoVNA to the antenna, the ripples completely disappear!

    Once I put a realistic amount of coaxial cable for my application (6' to 20', and whether self-assembled or pre-assembled by a manufacturer like MPD Digital) in between the antenna and the NanoVNA, the ripples return. What does that say?

    Is it that there is a small fault in the connections that gets magnified by the longer lengths of coax? Or that the fault is itself introduced by the longer sections of coax themselves?
  7. SM0AOM

    SM0AOM Ham Member QRZ Page

    This is not a "fault" but instead a consequence of that the directional bridge in the VNA and the cable matching are
    not perfect.

    As long as the bridge directivity and the residual return loss of the cable are better than 20 dB or so, it should be possible to "calibrate away" the ripples by making a SOL calibration at the end of the cable.

  8. KA9JLM

    KA9JLM Ham Member QRZ Page

    The display can not make a straight line at that resolution.


    I calibrate it at 420 - 480 mhz and connect to the antenna via length of coax. The line is full of oscillations. Then, I plug it into the antenna via short jumper. The curve is perfectly smooth. Yes, it can make a smooth line at this resolution if calibrated properly.


    So, I can assume that the hypothetical trendline that would be drawn through the oscillations is the accurate SWR curve and ignore the oscillations for purposes of evaluating the antenna? I am not sure how to perform an SOL calibration but will research.

Share This Page

ad: Radclub22-1