How Long Is Too Long?

Counterpoises, Radials, and Return Currents in End-Fed HF Systems

This topic looks simple until you actually install an end-fed outside of a textbook environment. Then the “rules” quickly turn into “it depends”. The main takeaway is straightforward:

There is no single counterpoise / radial approach that is “the right one” for every multiband end-fed installation. The only approach that works every time is measuring what your specific installation is doing, then deliberately controlling where RF current is allowed to flow.

What an end-fed system really is

With an end-fed wire and a 4:1, 9:1, 49:1 (or other) transformer, you are not feeding a “standalone antenna”. You are exciting an RF system:

• the radiator wire,
• the transformer and its “ground” terminal,
• any counterpoise or radials connected there,
• the coax shield outer surface (unwanted current path),
• and everything nearby that the system couples to (soil, gutters, fences, mast, balcony rails, wiring, etc.).

The transformer ratio changes impedance. It does not decide where the return current flows. Geometry, coupling, and environment decide that… unless you deliberately control it.

NEC and simulations can show trends, but at HF (roughly 2–30 MHz) local environment often dominates: soil, nearby metal, building wiring, height, coax routing, and even moisture can shift the outcome dramatically.

What “counterpoise” actually means (and why the word causes confusion)

In end-fed discussions, “counterpoise” gets used for multiple different jobs at once:

• Return path / completing the circuit: RF current must return somehow. If you don’t provide a defined return conductor, the return path will happily become the coax outer surface, station wiring, house wiring, and any convenient metalwork.
• Loss management: Soil is usually lossy. If the return current couples strongly into ground loss (especially on lower HF bands), efficiency suffers. Radials and counterpoises are often about moving current away from lossy paths.
• Controlling where RF flows (pattern + RFI + safety): If feedline and nearby conductors carry significant current, pattern becomes unpredictable, RF-in-the-shack becomes more likely, and near-field exposure assumptions break.

The “reflector” mental model is helpful for a classic vertical over a ground plane. But for practical multiband end-feds, the more useful question is: Where is the current actually flowing, and how much of it is flowing in places you did not intend?

Why the 5%–10% guideline exists (and why it isn’t a law)

The “short elevated counterpoise around 5%–10% of the lowest band” guideline is popular because it often gives the transformer “ground” terminal something that is:

• not purely soil loss,
• long enough to provide meaningful capacitance and a return path,
• short enough that it may be less likely to behave like a strong second radiator on the lowest band.

But it’s still only a starting point. In multiband operation, any “short” wire becomes a significant fraction of a wavelength somewhere. So a wire that is 0.05–0.10 λ on the lowest band can become 0.25 λ (or more) on a higher band… and then it’s no longer “just a counterpoise” in practice.

When is “too long” actually too long?

There is no universal percentage that marks a hard boundary. A counterpoise becomes a meaningful co-radiator when it carries significant RF current. That typically happens when its electrical length lands near a resonant condition on one of the bands (or when coupling to the environment makes it behave resonantly).

In a clean, symmetric setup (classic elevated radials on a vertical), radial currents can largely cancel in the far field. But end-fed systems are rarely symmetric, and real environments are rarely symmetric. One counterpoise wire, a sloping coax, a gutter on one side, a tree on the other… the cancellation assumptions break quickly.

So instead of chasing a magic percentage, think of it this way:

• If you want the return conductor to radiate as little as possible: keep unwanted current low on the feedline outer surface and arrange return conductors so they don’t become the easiest radiator.
• If you accept that the return conductor will radiate (often perfectly fine): route and place it like part of the antenna… away from electronics, away from people, away from random metalwork, and with geometry that gives you predictable behavior.

A systematic way to choose a counterpoise/radial strategy

Rather than ranking counterpoise types globally, choose based on what you want to optimize:

Predictable behavior and a “quiet” coax

• Install a high-quality common-mode choke right at (or very near) the transformer.
• Provide a deliberate return path: one or more counterpoise wires/radials connected at the transformer ground terminal.
• Measure current on the coax outer surface. The goal is low net current on the feedline’s outside.

Maximum efficiency on the lower HF bands

• If return current couples into soil, loss is your enemy.
• A proper radial system generally wins here.
• Many on-ground radials (even shorter than quarter-wave) can be very effective.
• Elevated quarter-wave radials can also be excellent, sometimes with surprisingly few radials, but they demand space, tuning, and careful geometry.

Portable / quick to deploy

• A few elevated quarter-wave radials for the lowest band you care about can be a strong practical solution (if you can keep them reasonably symmetric and off the ground).
• If that’s not possible: a short counterpoise plus careful feedline routing can work well enough, but expect more variability.

Minimum RF in the shack, even if efficiency isn’t perfect

• Treat the coax as something you want to keep RF off.
• Choke near the feedpoint; if needed, a second choke closer to the station.
• Route the feedline away from the radiator for the first few meters when possible; avoid running it parallel to the antenna.

“I have a big metal object available” (gutter, fence, foil strap, etc.)

• Sometimes these make an excellent return path. Sometimes they create a huge unpredictable radiator next to your house.
• This is exactly where measurement matters most: if it reduces unwanted feedline current and improves far-field performance without causing RFI, it’s a win. If not, disconnect it.

Why combinations sometimes work (and sometimes confuse)

Combining systems (short ground radials + elevated quarter-wave radials) can make sense because you’re addressing two different things:

• ground radials reduce soil loss and provide broadband capacitive coupling,
• elevated tuned radials can provide a lower-impedance return on specific bands and help shape current distribution.

But combining systems also increases the number of ways the installation can “choose” an unintended current path. The moment you add options, the system may take the one you didn’t expect.

Safety distance calculations

Any conductor carrying RF current contributes to near fields. In real end-fed installations, the counterpoise and/or feedline often carry some RF current… so they matter for exposure and safety distances.

Many simplified calculators assume a representative geometry and cannot cover every real-world routing of counterpoise wires and feedlines. The safest practical mindset is to treat the entire system as potentially radiating (wire + return conductors + any feedline section carrying unwanted current), keep conservative distances, and reduce unwanted feedline current where possible.

The one approach that works every time: measure

Recipes are tempting, but multiband end-feds are often “plug-and-pray” unless you measure. Two tools are worth more than dozens of rules of thumb:

Clamp-on RF current probe (feedline outer-surface current)

• This tells you immediately whether your “counterpoise strategy” is actually keeping the feedline quiet.
• If you change counterpoise/radial routing or length and feedline current drops, you are moving toward a more controlled system.

Far-field comparison method

• Compare in the far field (not “how noisy it is in the shack” and not SWR alone).
• Practical options: a field strength meter at a known distance, a remote SDR, consistent WSPR/FT8 reporter comparisons, or A/B tests with a fixed receiving station.

SWR is not efficiency. A pretty match can hide a very lossy return path. And a messy match can still radiate well.

A short practical starting recipe (multiband end-fed with 4:1, 9:1, 49:1, ...)

• Install a good common-mode choke right at the transformer output (or as close as physically possible).
• Add one dedicated counterpoise wire as a first test (5%–10% of the lowest band is fine as a starting point, not a promise).
• Route the feedline away from the radiator for a few meters.
• Measure feedline outer-surface current and adjust counterpoise length/routing.
• If low-band efficiency is still poor, move toward a real radial solution (several ground radials, or a small set of elevated radials for the lowest band you care about).
• Verify improvement with far-field comparisons.

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

Questions or experiences to share? Feel free to contact RF.Guru.