In many youtube videos about building low power radio transmeters , I have seen a piece of regular wire or extension cord used as a antenna. But , it seems like telecommunication antennas have specific geometry (like dipole antenna , dish antenna etc).But why do we have to follow specific geometry for designing antenna?

Is it because using an extension cord as a antenna have much more power losses? If it does have , how do we calculate it?


A wire antenna that doesn't have a specific geometry may be called a random wire antenna. Such an antenna will correspondingly have a random performance with a random radiation pattern. In these cases, typically we care more about getting any signal out at all, rather than doing it efficiently or in a specific direction. A random wire antenna might be cut to length for a specific frequency, or it may be a more or less random length with an antenna tuner between it and the radio to match the impedance. (Even a random length wire may need to be cut within a range of lengths for some tuners to be able to tune it.)

A dipole's most important characteristic is that it is close to a half wave long. It's geometry isn't critical beyond that, but variations in geometry will change characteristic impedance and radiation pattern (and possibly polarization) at least. Typically a dipole is depicted as two quarter wave segments that are colinear. Practically, we may want to actually have them in an inverted V pattern for best impedance, but other patterns work as well, and may or may not affect radiation pattern significantly. A dipole has some directionality, but typically the concern with a dipole is polarization (for vhf) and elevation angle of the radiation pattern (for vhf and hf). For HF dipoles, height above ground may affect elevation angle of the radiation pattern more than the antenna's internal geometry.

A dish is not an antenna, but may be part of an antenna system. A dish is a reflector, and typically has an antenna near the focus of the dish with a radiation pattern that covers the dish; the dish shape is critical to focusing RF energy on the antenna, and the antenna's shape is important for optimizing the aiming of the energy at the dish. The dish is typically parabolic but may be spherical or hyperbolic, and the curve may be in 2 or 3 dimensions. The ideal antenna for the dish can be any number of models with a hemispherical or one sided conical radiation pattern. (Although, a 2d parabolic dish could use a dipole exactly at the focus, parallel to the axis of curvature.)

Other antennas, for example, yagi, moxon, patch, etc., are directional antennas, and their geometry is critical to forming their directionality. The directionality also is coupled with making the antenna high gain.

  • $\begingroup$ I was sloppy in this answer about if this is for transmit or receive. Reciprocity says it doesn't matter, but there are still minor differences. For transmit, a directional antenna sends more power in the correct direction. For receive, a directional antenna might have better signal to noise ratio. Similarly, resonance matters differently for the two. $\endgroup$ – user10489 Apr 27 at 1:46
  • $\begingroup$ I think the important thing is that you can use a wire coat hanger and get some signals. But if you transmit into it, you’re probably going to blow your finals. $\endgroup$ – Scott Earle May 4 at 14:39
  • $\begingroup$ Actually, the wire coat hanger probably has a higher resistance, and thus more loss, and less reflection. So I'd assume you'd be less likely to blow stuff with a coat hanger than with an identically shaped and sized piece of copper, not that I'd want to do such a horrible thing to a piece of copper. :) $\endgroup$ – user10489 May 5 at 5:20

An antenna that is small in relation to the wavelength always has a dipole radiation pattern. Electric dipole or magnetic dipole or a combination of both. For transmit small antennas have a narrow bandwidth because they have to be power matched to the transmitter. Currents become very large in transmit so efficiency suffers. Small antennas may have a wide bandwidth for receive by use of special low noise amplifiers. As size grows antennas may get all sorts of interesting properties. Directivity (yagis) wide bandwidth (logperiodic) or both (parabolas) There are also interesting antennas with very long wires. There is no need to make an antenna "resonant" other than that the matching network needed to match it to a 50 ohm transmitter may become simpler or even superfluous.

A non resonant extension cord as a antenna will need an antenna tuner for transmit because the impedance will be far from 50 ohms. In case the wire is thin there will be ohmic losses and if the wire is covered with thick PVC there will be dielectric losses. On receive losses do usually not matter at all since external noise is attenuated by losses to the same extent as the desired signals. External noise is typically much weaker than the receiver noise below 30 MHz. With high losses in the upper HF range one might need an antenna tuner also on receive.

  • $\begingroup$ As antennas get smaller, their pattern starts to resemble isotropic more than dipole. Small loops have an opposite radiation pattern to a dipole. So the first sentence doesn't really make sense. Extension cords are typically 12 gauge, which is more than enough for HF and ohmic losses will be minimal. And even an extension cord could get lucky and be 50 ohms. The resonance has little to do with 50 ohms. So many things wrong with this answer. $\endgroup$ – user10489 Apr 29 at 4:47
  • $\begingroup$ Telephone extension cables may have two wires of 0.22 mm. =.44 ohms/meter with both in parallel. Depending on current, which might become high at some points, losses can be significant. Each wire may have 0.5 mm PVC around it with an additional layer of 1 mm PVC on the pair. At places where the current has a minimum and hence the voltage is high dielectric losses may become significant. $\endgroup$ – sm5bsz Apr 30 at 12:07
  • $\begingroup$ Small loops are NOT isotropic. They have the radiation pattern of a magnetic dipole with a null in the plane of the loop. A doughnut pattern. $\endgroup$ – sm5bsz Apr 30 at 12:19
  • $\begingroup$ I didn't say small loops are isotropic. I said they have an opposite pattern from a dipole -- as in, different direction. $\endgroup$ – user10489 Apr 30 at 15:49
  • $\begingroup$ He didn't say telephone cord. Usually "extension cord" describes household electrical cords, which are sometimes called zipline, and are typically rated for 13A. Zipline is frequently used for antennas. I've never heard of someone using telephone wires for that, probably because it'd be a pain to separate the wires enough to make it useful. $\endgroup$ – user10489 Apr 30 at 15:55

There are at least 3 reasons for using "special geometries" for RF antennas: resonant frequency, impedance match, and pattern.

  • If the antenna is of a length and geometry that is resonant at the frequency of interest, then the antenna will be more efficient capturing energy from the airwaves at the frequency of interest.

  • If the antenna geometry has a matched feedpoint impedance at the frequency of interest, then it can pass power to the feedline or feedpoint with a lower likelihood of resistive losses (depending on the feed line or wire type).

  • If the antenna geometry at the frequency of interest is sized, shaped, and angled to better capture signals in the polarization and direction of interest, and potentially less well positioned and angled to capture RF noise, then the signal transferred to the receiver will have a better S/N ratio (less noise that could completely cover up the signal(s) of interest). Very short wires (relative to wavelengths) might capture more household RF noise than any interesting HF radio signals.

  • 1
    $\begingroup$ I was under the impression that resonant antennas are no more efficient, just easier to match. $\endgroup$ – rclocher3 Apr 26 at 16:40
  • $\begingroup$ Assume no feedpoint connection. Will a resonant conductive object present a better impedance match to free space at that frequency? $\endgroup$ – hotpaw2 Apr 26 at 17:39
  • 1
    $\begingroup$ I think you're off-base on this one. The group consensus of the answers to these two questions shows that the users of this site clearly think that resonant antennas are not more efficient just because they're resonant. I could be wrong, but I don't think there is impedance-matching happening when a disconnected conductive object encounters a radio wave in a vacuum. $\endgroup$ – rclocher3 Apr 26 at 22:42
  • $\begingroup$ Sloppy use of the term resonance frequently means "low SWR" -- it's not exactly right, but it is commonly used that way. Efficiency is also slippery. If by efficiency, you mean maximum power from the radio is radiated, then yes, 50+0i is the most efficient. If by efficiency you mean least power lost to heat via loss resistance in the antenna, then yes, efficiency isn't important. If by resonant, you mean X+0i rather than 50+0i, then yes, resonance isn't the most important. $\endgroup$ – user10489 Apr 27 at 1:35
  • $\begingroup$ Also, not that sometimes you cant get to 50+0i, in which case there's a compromise between best impedance match (which might be 45+20i) and closest resonance (which might be 75+0i) where neither is perfect but radiation is maximized. $\endgroup$ – user10489 Apr 27 at 1:41

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.