Folded Dipole

Home ] Up ]


The folded dipole is claimed to offer increased bandwidth, and to closely match 300-ohm line. This is much like the double bazooka antenna, or coaxial dipole antenna. Inaccurate double bazooka and coaxial dipole claims have been debunked in numerous forums and texts, including in QST and Reflections by Walter Maxwell.

While double bazooka or folded dipole antennas can offer significantly more bandwidth than a thin wire dipole, it is important to understand why certain antennas have more bandwidth. Let's look at a folded dipole and see how it really behaves.

Folded Dipole Myths

Myth #1:

-The advantage of using a folded dipole is wide bandwidth.

False: On a single band, bandwidth is essentially no different than just tying the legs in parallel, splitting it in the center, and feeding it with coax. We will see this in the models below. A folded dipole, like any dipole, is wider primarily from being made thicker. It does not get significantly wider not from the stub effect or folding, and not from the higher feed line impedance.

Myth #2

-When used on other bands, the higher natural impedance tends to swamp the impedance from being non-resonant, which makes for easier matching.

False: The folded dipole is only useable with reasonable feed efficiency on the fundamental and on odd harmonics. The folded dipole is much worse than a regular ladder line fed dipole dipole for harmonic or out-of-band operation. On even harmonic bands, it appears as a nearly dead short. On odd harmonic bands, the resonance shifts upwards, typically about 4-6 %. It is not a good multiband antenna, and actually has excellent radiation suppression of even harmonics.


Myth #3

-A twinlead folded dipole fed with twinlead will have an impedance around 300 ohms on the designed band. Variations in folded dipole SWR is less with changes in antenna length or frequency are less than in a regular coax fed dipole.
False: SWR variations are essentially the same. A regular coaxial fed dipole varies from ~50 ohms  to ~70 ohms or so over normal height ranges. This means 50-ohm SWR can be as high as ~1.5:1 with normal height variations

With a folded dipole, impedance varies between ~200 ohms to about 300 ohms. This also means SWR can be as high as ~1.5:1 over normal height ranges!

Myth #4

-The folded dipole has the advantage of using balanced line feed, and balanced line has low-loss at high SWR.  

False: Loss has little to do with being balanced, except the lack of a shield increases surge impedance. 7/8th inch diameter heliax coax, for example, has less loss than 1 inch wide "window" or "ladder" line. The real reason balanced lines can operate with reasonable SWR on multiband doublets is the antenna impedance. A dipole's impedance is about 4000 ohms or less on even harmonics, so SWR is only 10:1 on a typical 400-ohm balanced line. If the feeder was a 50 ohm line, SWR would be 80:1!

On the fundamental and odd harmonics, the dipole impedance is about 50 to 100 ohms. That impedance is near perfect for coax, and the ladder line is still only 8:1 or less.


Don't be fooled by claims wide bandwidth means high loss, or that narrow bandwidth indicates high efficiency. Bandwidth is related to system Q, but system Q is not determined solely by resistive or heat losses. System Q is really a function of stored energy to energy moved through the system. Energy can be moved through the system and exit as wasteful heat, or it can exit as useful work. An antenna 5 kHz wide can be far less than 1% efficient, while an antenna 500 kHz wide can be almost 100% efficient. Without knowing where energy is going, we should never assume bandwidth means anything other than bandwidth.

Most antennas store a majority of energy in an electric field, although small "magnetic" loops have a lot of energy in a strong, very local, magnetic field.

The folded dipole is no exception to this generalization. The folded dipole, like almost any antenna, has an electric field boundary at the wire to space transition. Current flowing into the antenna "travels" through space via displacement currents. The displacement currents are a function of voltage and capacitance along the conductor to the outside world.

The electric field boundary is most intense at the open ends of the antenna, and this gives rise to most of the antenna's energy storage. As we spread this boundary region out, reducing electric field concentration,  stored energy is reduced. Reducing stored energy reduces Q, and increases bandwidth. At the same time, it might actually reduce loss.  

When any antenna is made thicker, the energy storage boundary is more evenly distributed. This is especially true near the open ends of an antenna, such as the end of a dipole.

The folded dipole obtains virtually all of any bandwidth improvement by virtue of being thicker, especially near the outer ends. We will see this demonstrated below.

The fact the arms also form "stubs", as with the coaxial dipole, has almost no bearing on bandwidth. The folded dipole is wider bandwidth than thinner dipoles, and narrower than thicker dipoles, when loss is constant. Any bandwidth improvement mostly comes from being thicker, but some actually comes from slightly increased loss. 

Basic Wire Table of Folded Dipole Model




I constructed a model that allows a relatively easy change from folded dipole to dipole.


VSWR Curve of Folded Dipole with 300 Ohm feed line

2:1 VSWR at 3.65 and 3.9 MHz. 2:1 bandwidth using 300-ohm line is 250 KHz






VSWR Curve of Folded Dipole with 200 Ohm feed line

2:1 VSWR with 200 ohm line at 3.635 and 3.86 MHz. 2:1 VSWR bandwidth is 225 kHz



Fed as regular dipole, no other changes:




SWR Curve 50-ohm coaxial fed regular dipole made with same size and materials

2:1 VSWR at 3.635 and 3.85 MHz, VSWR Bandwidth 215 kHz



SWR Curve 75-ohm coaxial fed regular dipole made with same size and materials

2:1 VSWR at 3.64 and 3.89 MHz. 2:1 VSWR BW is 250 kHz  


#14 (regular) insulated Wire Dipole

2:1 VSWR 3.63 - 3.8 MHz     2:1 VSWR BW = 170 kHz




Type feed line Lowest SWR 2:1 SWR BW
Folded 300 1.4:1 250 kHz
Folded 200 1.05:1 225 kHz
Regular same wire 75 1.4:1 250 kHz
Regular same wire 50 1.05:1 215 kHz
Regular single  #14 wire 50 1.25 170 kHz


We can see there are negligible differences between folded dipole and regular dipole systems unless we change the effective wire diameter!.