Why I Prefer EFOC29 over an All-Band EFHW or Long Wire

Why did we choose the EFOC29 approach instead of another EFHW 80–10 or a 37 m long wire? My answer starts before the SWR meter. For a practical multiband station, I prefer to avoid an unnecessarily extreme impedance transformation, give the antenna a defined return branch and arrange the conductors to suit the garden. Those are design choices with useful consequences—not just three names for a wire.

Start With the Antenna, Not the Promise on the Box

An EFHW 80–10 uses a wire near a half wavelength on its lowest intended band and higher-order modes above it. Its feed is near a current minimum and voltage maximum, so a large impedance transformation is part of the job. The familiar 49:1 unit is not there by accident.

The EFOC29 takes a different route. Its two unequal branches make an off-centre-fed current system, even though the installer feeds it at the end of the main wire. We design around a 4:1 UNUN rather than treating the extreme end of a half-wave radiator as the only convenient place to feed. That does not make the impedance exactly 200 Ω on every band; it makes a less extreme transformation the design target.

The EFLW37 comparison is the other familiar choice: a 37 m long-wire arrangement using a 9:1 matching unit and a tuner to accommodate its band-dependent load. That flexibility can be useful. But 9:1 is a fixed ratio, not an instruction telling every complex antenna impedance to become 50 Ω. I prefer to reduce the matching problem through the antenna and return-path design before handing the remaining job to a tuner.

A Smaller Transformation Is a Real Engineering Advantage

Impedance ratio and turns ratio are not the same thing. Ideally, impedance transforms as the square of the turns ratio: a 4:1 impedance transformation corresponds to 2:1 voltage/turns, while 49:1 corresponds to 7:1. The distinction is set out in Mini-Circuits' RF-transformer application note.

Here is the voltage consequence without any antenna marketing. Deliver the same 100 W to two purely resistive loads:

These are circuit calculations, using V = √(PR) and I = √(P/R), not measured EFOC29 or EFHW feedpoint values. At this equal delivered power, the higher-resistance load requires 3.5 times the terminal voltage. The moderate-impedance option trades that voltage for more load current, so conductor and contact resistance still matter.

That lower required step-up is the advantage I am choosing. It reduces the high-side terminal-voltage burden and gives the designer a less extreme winding ratio to realise. Depending on topology, it can also reduce the need for a long high-side winding and the associated parasitic complications. It is a better starting problem for our multiband design—not a claim that every 4:1 transformer must beat every 49:1 transformer.

Fewer Turns Are Useful Only When the Core Still Has the Right Job

Simply removing turns is not the recipe. From Faraday's transformer relationship, for the same sinusoidal winding voltage, frequency and core area, fewer turns mean more magnetic flux density. Lower terminal voltage on the antenna side must not be confused with automatically lower flux in the core. Winding topology, volts per turn, material and frequency remain part of the design.

At the low-frequency end, sufficient magnetising inductance matters. At the high-frequency end, leakage inductance, winding capacitance and core/conductor losses become increasingly important. Mini-Circuits' matching-device overview explains why an ideal turns ratio alone cannot describe a real RF transformer. Reducing the transformation burden gives us design room; choosing and constructing the complete transformer is how we use it.

This is also why RF.Guru uses selected, traceable Würth ferrite materials in its 4:1 and 9:1 matching transformers. The material is part of an engineered assembly, not a generic mix number standing in for a design. A manufacturer's name cannot tell you the loss of a finished transformer.

Choosing between mix 43 and mix 52 does not settle the question either. Fair-Rite's 43 material data and 52 material data describe different permeability and loss behaviour with frequency. Their material characteristics and suppression applications are not completed high-power EFHW transformer ratings.

Why Stretching an EFHW from 80 to 10 m Bothers Me

From the 80 m band to the top of 10 m is more than three octaves. The antenna can provide useful separated operating windows, but the transformer still has to deal with widely separated frequencies and changing loads. More low-frequency inductance and manageable high-frequency parasitics are competing design demands. Calling the wire “resonant on many bands” does not remove that compromise.

A shunt capacitor can improve the impedance response of a particular transformer-and-load combination. It also changes susceptance and circulating current; it does not create free bandwidth or certify low loss. It is not a series capacitor, and it is not equivalent to designing a complete matching network for each band. Our shunt-capacitor discussion follows that distinction in detail.

My preference for EFOC29 is therefore not an objection to half-wave antennas. A purpose-designed monoband or bounded dual-band EFHW has a much narrower job to solve. My objection is making the all-band convenience claim do the work of transformer engineering. With EFOC29, we choose a less extreme matching target instead.

The Return Path Is Deliberate, Not an Accidental Length of Coax

The EFOC29 configuration uses a 29 m radiator starting dimension. In the coax-return version, approximately the first 12 m of coax exterior is an intended return branch, ending at the first choke. The supplied package has approximately 26 m of coax and two large chokes. The alternative M6-terminal version uses a separate 13 m wire counterpoise and does not include that coax/choke package.

This is the practical benefit of treating the matching unit and choke as separate functions. The 4:1 UNUN handles impedance transformation. The choke establishes a boundary between the intended exterior-current section and the feedline continuing toward the station. Moving that first choke directly to the feedpoint would change this particular antenna, not simply “improve the choking”. Follow the installation guide for the chosen configuration.

A defined boundary can reduce unwanted coupling between the antenna and station wiring. That is a useful engineering reason to prefer a controlled return path over an unexplained one. It is not an intrinsic noise advantage of the letters EFOC: an EFHW or long wire can also have a deliberate counterpoise and effective common-mode control. Tom Rauch, W8JI, explains the reciprocal noise and interference paths in Common Mode Currents.

The Garden Still Decides Where the Signal Goes

A 29 m main wire can be easier to accommodate than an approximately 40 m EFHW radiator. If that lets me keep the main span higher or avoid wrapping it around several obstacles, it is a practical advantage worth choosing. But the return branch remains part of the EFOC29 antenna. Twenty-nine metres is not the total length of every conductor carrying antenna current.

On the upper bands, both arrangements can develop multiple lobes and nulls. Those patterns are consequences of current distribution, not random behaviour and not something a 4:1 transformer can straighten out. The EFOC29 return routing, wire bends, height and surroundings all contribute. I choose a layout that suits the wanted paths; I do not claim that the shorter main wire automatically produces a lower take-off angle or wins in every direction.

That leaves a meaningful installation choice: if the EFOC29 layout fits the available supports well, use that freedom to keep its geometry deliberate. If a properly placed monoband EFHW or a feedpoint-tuned long wire better fits the operating target, use that advantage instead. A tuner at the feedpoint can keep a highly mismatched long-wire load off most of the coax; a tuner in the shack cannot remove loss already incurred along a mismatched feedline.

Why EFOC29 Remains My Multiband Preference

For the station this design is meant to serve—several HF bands, a practical main-wire span and room for the specified return path—I prefer EFOC29 because it starts with a manageable transformation target and a complete current-path plan. Less voltage step-up, a less extreme transformer ratio and an intentional station-side boundary are useful advantages before any claim about a particular DX contact.

It is not a tuner-free promise. The configuration requires a tuner on 60 m; 30 m and 15 m may also need one, depending on the installation. The supplied radiator is deliberately longer than its nominal starting dimension: measure it, mark it and fold back surplus before cutting.

The point is to choose the compromise, not disguise it. An all-band EFHW asks a high-ratio transformer to span a demanding frequency and load range. A long wire gives a tuner more of the job. EFOC29 is our choice to make the antenna geometry, moderate-ratio matching and defined return path work together. That is why I prefer it for this multiband job.

Questions, antenna-factor records or height trials to share? Contact RF.Guru.