4
$\begingroup$

I have a V5000 antenna and I’m tuning it but can’t get the Reactance (X) on my MFJ 259B down. Resistance is at 50 ohms, SWR reading 1.2 to 1.3 (I don’t give that much credence) but the Reactance (X) is around 15. Any ideas how to get that down? LMR 400 coax if it matters.

That's what I'm still trying to get a grip on! From what I've been told it has to do with the capacitance of the antenna which is affected by the proximity of the driven element to ground plane radials or whatever your ground plane is. I'm still figuring that one out!

$\endgroup$
1
  • 1
    $\begingroup$ Welcome to this corner of StackExchange. Please, tell us: 1) the length of your LMR400 feedline; 2) your operating frequency. $\endgroup$
    – Brian K1LI
    Dec 23 '20 at 23:25
6
$\begingroup$

Reactance is the imaginary part of impedance. Together with the real part, resistance, impedance describes how a given load will respond to an AC voltage or current source. While for DC analysis resistance alone is sufficient for this purpose, for AC analysis we need to know not only the magnitude of the voltage or current, but the phase angle between them, which is given by:

$$ \theta = \tan^{-1}\left( X \over R \right) $$

Inductors and capacitors are examples of discrete components that have reactance but no resistance. In an antenna, the reactance comes from the timing of the wave reflected from the ends of the antenna. At some length, the reflections align such that voltage and current are in phase at the feedpoint terminals, which means reactance is zero, and the antenna is said to be resonant.

I should note there's no particular reason reactance needs to be zero. An SWR of 1.3 is perfectly acceptable, and there's likely no practical performance gain to be realized by reducing it further. Furthermore, reactance will be zero only at one particular frequency (and likely harmonics thereof) but it is impossible for the SWR to be 1:1 or for the reactance to be zero across the entire band.

However if you want to get reactance to zero you can do that by adjusting the length of the antenna or whatever other mechanisms your antenna has for tuning it. If you have an antenna analyzer then measure the reactance over a range of frequencies:

  • A little below the resonant frequency, the antenna is "too short" and reactance will be negative.
  • A little above the resonant frequency, the antenna is "too long" and reactance will be positive.
  • Precisely at the resonant frequency, reactance will be zero.

If you get very far away from the resonant frequency these rules will not hold, but you would not ordinarily operate an antenna this far away from resonance.

Make the necessary adjustments to your antenna and repeat your measurements until you have the resonant frequency where you want it.

$\endgroup$
1
  • $\begingroup$ Tks, Always wonder about that!!! Tks again!!! $\endgroup$
    – booksbenji
    Dec 22 '20 at 23:31
0
$\begingroup$

Any physical object where there are potential voltage or current reflections off of impedance mismatches or open circuits (such as the ends of a wire antenna) can change the AC waveform of the current to be different from that of the voltage at the feedpoint. And wherever the current doesn't exactly follow the voltage (or vice versa), that is defined in linear electrical systems as a non-zero reactance.

Nothing has to be imaginary about either the current or the voltage. Just certain mathematical forms of the description of such, often to save chalk on the chalkboard.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.