The EFHW Is a Dipole… But The EFOC Definitely Isn’t

Here’s Why The Pattern Goes Weird

Short version: if you hang a classic half-wave dipole and a well-behaved EFHW in exactly the same way, they put RF in the same directions. Move the feedpoint off-center (EFOC/OCF), and suddenly one leg carries most of the RF, the pattern leans toward that side, and things get weird… especially on the higher bands.

The “no-math” story

Imagine three antennas made from the same piece of wire, all installed identically:

• Same total length (for example an 80 m ½-wave, about 40 m / 132 ft).
• Same height above ground.
• Same “flat-top” shape.
• Same surroundings.

The only thing you change is where you hook the feedline.

We compare:

• A classic ½-wave dipole (center-fed).
• An EFHW – End-Fed Half-Wave.
• An EFOC / OCF – End-Fed Off-Center / Off-Center-Fed.

Dipole vs. EFHW: same wire, same pattern

Start with a regular 80 m half-wave dipole:

• About 40 m (132 ft) total length.
• Fed in the middle with 50 Ω coax through a 1:1 current balun.

Now take the same wire, at the same height and in the same shape, but:

• Move the feedpoint to one end.
• Add a 49:1 transformer so your 50 Ω coax is happy.
• Add a good common-mode choke 0.05 lambda from the box so the coax does not become part of the antenna.

What happens to the RF current in the wire?

• The current is largest near the center of the wire.
• It tapers to almost zero toward the physical ends.

As long as you excite the same basic ½-wave resonance, the current distribution along the wire is effectively the same for the center-fed and the properly-choked end-fed. Same wire, same height, same shape, same current hump → same far-field pattern.

The real differences are not in the pattern, but in what the rig “sees”:

• Dipole: lowish impedance (~70 Ω), moderate voltage, relatively high current.
• EFHW: very high impedance (~2–4 kΩ), very high voltage, lower current.
• The EFHW’s transformer and the insulation at the end are heavily stressed.
• It is harder to keep the coax and surroundings from becoming “part of the antenna” because that end sits at very high RF voltage.

But if you tame the coax with a proper choke right at the feed, the dipole and EFHW behave like twins in the sky. The difference is mainly down at the feedpoint, not up in the pattern.

Enter the EFOC / OCF: one leg gets “more RF”

Now take the same 40 m of wire and move the feedpoint off-center instead of at the end.

Classic example, roughly 27 % / 73 % split:

• Short leg ≈ 12 m.
• Long leg ≈ 29 m.
• Feedpoint at that junction, usually via a 4:1 unun (impedance around 200 Ω on 80 m).

The current along the whole wire still looks like a ½-wave “hump” – largest near the center of the total length, small near the physical ends. But look at where you placed the feedpoint:

• The short leg lives mostly out where the current is already small.
• The long leg includes the high-current region near the middle of the wire.

Seen from the feedpoint:

• The long side carries most of the radiated power.
• The short side still radiates, but much less – it behaves partly like a counterpoise or tuning stub.

So the antenna behaves more like “a slightly long ⅜- to ½-wave radiator on the long leg plus a helper stub” than like a symmetric dipole. That imbalance is what tilts and distorts the pattern.

Why the EFOC pattern is not the same as the dipole or EFHW

Back to our three antennas, all the same height and geometry:

• Center-fed dipole: currents are mirrored left/right → clean broadside pattern.
• EFHW with good choke: same current distribution as dipole → pattern basically the same as the dipole.
• EFOC / OCF: much more RF on the long leg than the short leg → pattern leans toward the long side and is no longer symmetric.

On a multi-band 80–10 m EFOC/OCF, this gets more dramatic:

• On 80 m, the pattern is still somewhat “dipole-ish”, but skewed.
• On 40, 20, 17, 15, 12, 10 m the wire is multiple wavelengths long. The off-center feed taps into different parts of the standing wave, so you get odd lobes and nulls that do not match a simple center-fed or well-behaved EFHW.

However, there is an important twist on the upper bands (17–15–12–10 m).

Take a typical 80–10 m EFHW made from about 40 m of wire:

• On 17 m that full wire is already around 2½ λ long.
• On 15 m it is roughly 3 λ.
• On 12 m it grows to about 3½ λ.
• On 10 m it is very close to a full 4 λ radiator.

A 3–4 λ long wire produces a dense “picket fence” of narrow lobes and deep nulls. The pattern is still symmetric left/right, but it can look quite busy and erratic: lots of hot directions with very sharp dead spots in between.

In the 80–10 m EFOC/OCF with a 27 % / 73 % split, the wire length is the same, but the part that really does most of the radiating is the long leg. That long leg is only about 73 % of the total, so on the same high bands it is much shorter in electrical terms:

• On 17 m, the long leg is roughly around 1¾–2 λ.
• On 15 m it is just over 2 λ.
• On 12 m it is a bit above 2½ λ.
• On 10 m it ends up around 2.7–2.9 λ, not 4 λ.

The short leg, being only about 27 % of the total, tends to behave more like a 1 λ-ish helper stub / counterpoise on those bands. The net effect is:

• On 17–10 m, the EFOC/OCF still has a multi-lobe pattern, but with fewer and broader lobes than a full 4 λ EFHW.
• The lobes are pulled strongly toward the long side (because that is where most of the current lives), instead of staying nicely centered.
• So the EFOC/OCF pattern is skewed but a bit less “porcupine-like” than the high-band EFHW pattern.

In other words, on 17–15–12–10 m the EFOC/OCF trades the EFHW’s very finely sliced, symmetric lobe structure for a pattern that is:

• somewhat simpler and less erratic in terms of the number of lobes,
• but clearly tilted toward the long leg instead of being centered over the wire.

This is what people really mean when they say things like:

“A half-wave dipole and an EFHW have the same radiation pattern if they’re the same height and shape, but an EFOC does not, because of where it sits relative to resonance.”

Exactly: the off-center feed taps into the standing wave so that currents and phases on the two legs are not equal, especially on harmonics. That unbalance is what bends the pattern and moves the hot directions on different bands.

The “beer-table” explanation you can use on the air

If you want to explain it in one minute to another ham:

If you hang a center-fed dipole and an EFHW at the same height in the same shape, and you keep the coax from radiating, they put the same RF in the same directions — they’re just fed differently.

But once you feed that same wire off-center, one leg carries most of the RF and the other behaves more like a counterpoise. That unbalances the currents and shifts where the antenna “thinks” its half-wave is, so the pattern is no longer the same as a classic dipole or EFHW.

The nerdy tech-talk with formulas

Current distribution on a thin ½-wave wire

Take a straight, thin wire of length L in free space, resonant at about ½ λ. Put the origin at the center of the wire so that:

• z = 0 at the center.
• z = ±L/2 at the ends.

For the fundamental ½-wave mode, the current distribution is approximately:

I(z) ≈ I0 · sin(k · (L/2 − |z|))

where:

• k = 2π/λ is the wavenumber.
• I0 is the peak current at the center.

Key points:

• I(0) = I0 (maximum at the center).
• I(±L/2) = 0 (zero at the open ends).
• The curve is symmetric around the center.

Important: this ½-wave “hump” is set by the length and boundary conditions (open ends), not by where you feed it. Changing the feedpoint mostly changes the input impedance, not the basic shape of I(z) – as long as you excite this same mode.

In ideal free space, a ½-wave thin wire has the same current distribution and far-field pattern whether you feed it:

• in the center (dipole),
• at the end (EFHW),
• or off-center (EFOC/OCF).

That statement is mathematically correct for a perfectly isolated wire in a vacuum with no feedline, no mast, no roof, no trees and no ground. But that is not where our antennas live.

In the real world your 80–10 m wire is hung over lossy ground, often bent around buildings or trees, and it is fed by a real coax line that can and will carry common-mode current if you let it. Once you include ground, supports and feedline in the picture, the “same wire, different feedpoint” ideal breaks down: the EFHW becomes very sensitive to choking and routing, and an off-center feed tends to drive the long leg plus coax as an unbalanced radiator that tilts the pattern.

So you can safely use the free-space picture to understand the basic standing wave along the wire, but you must remember that the actual pattern and noise behavior are set by the entire installation: wire + feedline + ground + surroundings. That is where the EFHW vs. EFOC vs. dipole really diverge.

The real-world differences come from what the feedline and counterpoise do with displacement current and common-mode current.

Why dipole and EFHW can share the same pattern

For the center-fed dipole:

• We drive the center to some feed current Ifeed.
• The wire supports the ½-wave standing wave described above.

For the EFHW, if we use:

• a 49:1 transformer (e.g., ~3 kΩ down to ~50 Ω), and
• a good common-mode choke so the coax shield current is negligible,

then we are simply exciting the same standing-wave pattern as the dipole, but from the end instead of from the center.

The far-field of a thin ½-wave along the z-axis has the familiar elevation pattern:

Eθ(θ) ∝ cos( (π/2) · cosθ ) / sinθ

This depends on the current distribution along the wire – not on exactly where the feedpoint sits. So:

Same wire + same current I(z) ⇒ same Eθ(θ) ⇒ same pattern.

Feedpoint impedance and voltage/current levels

Now look at the three antennas from the rig’s point of view on 80 m. Typical resistive feedpoint impedances:

• Center-fed dipole: Rdip ≈ 70 Ω.
• EFOC/OCF (27 % / 73 %): REFOC ≈ 200 Ω.
• EFHW end: REFHW ≈ 2000–4000 Ω.

Assume 100 W of RF power. Use:

P = I² · R ⇒ I = √(P / R)
P = V² / R ⇒ V = √(P · R)

Take representative values:

• Dipole: R = 70 Ω.
• EFOC: R = 200 Ω.
• EFHW: R = 3000 Ω.

Then:

• Dipole: I ≈ √(100/70) ≈ 1.2 A; V ≈ √(100·70) ≈ 84 V.
• EFOC: I ≈ √(100/200) ≈ 0.71 A; V ≈ √(100·200) ≈ 140 V.
• EFHW: I ≈ √(100/3000) ≈ 0.18 A; V ≈ √(100·3000) ≈ 550 V.

So at the feedpoint we have:

• Dipole → moderate voltage, relatively high current.
• EFOC → mid-voltage, mid-current.
• EFHW → very high voltage, low current.

That huge EFHW voltage is where displacement current and “mystery currents” into the surroundings become important.

Conduction vs. displacement current at the feedpoint

Maxwell’s corrected Ampère’s law in simple form:

∇ × H = Jcond + ∂D/∂t

where:

• Jcond is the conduction current (electrons in metal).
• ∂D/∂t is the displacement current (changing electric field in dielectrics/air).

In a wire antenna:

• Conduction current flows along the conductor.
• Displacement current “jumps the gaps” at the feedpoint, ends, insulators and into nearby structures.

At the EFHW feed:

• Because V is so large, the E-field is large.
• Displacement current into nearby objects (coax shield, mast, roof, tree) is strong.
• That displacement current must close the loop somewhere, so it forces a return path through the coax and surroundings → the coax becomes part of the antenna.

At the EFOC feedpoint:

• The voltage is much lower (for the same power).
• Most displacement current crosses the small feed gap between the two legs, not into the feedline.
• The return current mostly stays in the short leg and near fields, rather than wandering down the coax (assuming a decent choke).

This is why, in practice:

• Dipole + EFHW with a good choke can share essentially the same pattern (wire dominates, coax quiet).
• EFOC/OCF, even when choked, has an inherently unbalanced pair of radiating arms (long vs. short leg), so the pattern is different by design.

How much more does the long leg radiate?

Let’s make the “long leg radiates more” argument more concrete. Let:

• Total length L = λ/2.
• Coordinate z ∈ [−L/2, +L/2], center at z = 0.
• Feedpoint at z = a (off-center).

Then:

• Short leg = from −L/2 to a.
• Long leg = from a to +L/2.

Radiated power from each segment is roughly proportional to:

Psegment ∝ ∫ I²(z) dz over that segment.

Using the ½-wave current approximation:

I(z) = I0 · sin(k · (L/2 − |z|))

and choosing a so that the lengths are about 27 % / 73 % of L, you find that:

• The integral of I² over the long leg can be several times larger than over the short leg.

In other words, for the same feed current Ifeed, most of the radiated power comes from the long leg. The “effective” ½-wave is pulled toward that side.

Mathematically the ½-wave mode of the entire wire is still symmetric in free space, but in practical terms the power contribution from each side of the feedpoint is very unequal. That’s the sense in which:

• The effective λ of the EFOC is “oriented” toward the long piece.
• A center-fed dipole or well-behaved EFHW has its λ centered on the full length, not skewed to one leg.

Why this matters more on 80–10 m EFOC/OCF antennas

On a typical 80–10 m EFOC/OCF:

• The feedpoint position is chosen so that the impedance is reasonable on several harmonic bands.
• On each harmonic, the standing-wave pattern along the wire has multiple peaks and nulls.
• The off-center feedpoint lands at different relative positions (near a peak on some bands, offset on others).

Consequences:

• On some bands the long leg dominates even more, pulling the main lobes toward that side.
• On others you get strong current near the ends, creating higher-angle or skewed lobes that are a poor match to the classic broadside dipole pattern.

So the 80–10 m EFOC/OCF gives you useful SWR on many bands, but the pattern varies a lot from band to band and is generally not the same as “a dipole on each harmonic”. You have to treat it like its own animal.

By contrast, an 80 m EFHW is also harmonic-friendly, but if you control the common-mode currents, each harmonic still behaves like a “centered” wire where the whole length participates more evenly, not dominated by one leg as in an EFOC. On the very highest bands, that full-length radiator can be close to 4 λ, which explains why the EFHW pattern often shows more, narrower lobes than the shorter effective radiator of the EFOC/OCF.

TL;DR – what to remember when choosing between dipole, EFHW and EFOC

• Dipole vs. EFHW: same height + same shape + properly choked → almost the same current along the wire → same radiation pattern. The EFHW is essentially a dipole fed at a high-impedance end, with higher voltage stress and more risk of common-mode trouble if you skimp on the choke.
• EFOC / OCF: the off-center feed delivers more RF to the long leg and less to the short leg, especially on harmonics. The currents and phases no longer mirror left/right, the effective ½-wave shifts toward the long side, and the pattern becomes asymmetric and complex, changing strongly from band to band. On 17–10 m it is often a bit less erratic than a full-length 80–10 m EFHW (shorter effective radiator, fewer lobes), but also more skewed toward the long leg.

If your main goal is a predictable pattern, a classic dipole or a well-choked EFHW is easier to reason about. If you accept a more “creative” pattern in return for harmonic SWR convenience, the EFOC/OCF can still be a very useful tool – as long as you know that it is not “just a dipole fed somewhere else”.

Interested in more technical content? Subscribe to our updates for deep-dive RF articles and lab notes.

Questions or experiences to share? Contact RF.Guru and let us know what you are seeing on your own station.