Ground Systems Demystified — Efficiency vs Myth

What rigorous experiments (Rudy Severns N6LF) and classic radio-science actually say about verticals, radials, and “RF ground.”

The one-minute takeaway

• Efficiency lives—or dies—in the near field. Loss happens when your vertical’s fields drive current into lossy soil. A ground system works by intercepting and returning that current in metal instead of dirt.
• “More wire somewhere” is not a strategy. Where and how you place radials matters far more than any single magic length or wire gauge.
• The big myths below keep getting repeated. The data from Rudy Severns (N6LF) and earlier scientists like Brown, Lewis & Epstein are surprisingly clear—and often at odds with ham-lore.

Myth 1 — A ground rod (or two) is plenty for RF

For an HF vertical, a lone ground stake is a terrible RF return. Near the base, soil loss can equal—or exceed—your radiated power. Radials reduce that loss by carrying return current in copper, not dirt. Severns’ HF analyses and measurements show exactly this behavior and why it happens.

Myth 2 — Radials must be exactly ¼-wave long

Buried or on-ground radials are not resonant in the dipole sense; length mainly determines how much of the lossy near field they “shield.” Classic broadcast experiments (Brown–Lewis–Epstein, 1937) and W2DU’s summary show diminishing returns around 0.4–0.5 λ. Beyond that, added length yields little additional field strength.

Myth 3 — A few long radials beat many short ones

With only a few radials, the outer ends do very little—radial spacing grows with radius, the lattice impedance rises, and more current is forced back into the soil where it produces loss. Severns measured and modeled the current distribution: density near the base (lots of conductors close together) is what matters most for cutting loss.

Design tip: If you’re building a big system, Severns shows you can save wire by adding radials in rings—denser near the base, sparser farther out—with near-identical performance to a uniform 128-radial field.

Myth 4 — You need 120 radials, period

That rule comes from AM-broadcast practice at ≈3 MHz. The BLE measurements give a more nuanced picture of diminishing returns (vertical over average ground, ¼-λ radials):

• 2 radials ≈ 4.2 dB below “perfect ground”
• 15 radials ≈ 2.1 dB down
• 30 radials ≈ 1.6 dB down
• 60 radials ≈ 0.87 dB down
• ≈ 113 radials ≈ 0.67 dB down

Performance improves substantially up to a few dozen radials; after ≈ 60, each extra wire buys only tenths of a dB. At HF, soil behaves more like a complex dielectric, so losses can be higher than BLE would predict—but the same diminishing-returns trend holds.

Myth 5 — Elevated radials always beat ground radials

For single-band verticals, 2–4 properly tuned, elevated ¼-λ radials (a few feet or ≈ 1 m above ground, symmetric and clear of metal) can rival a very large on-ground system—if they’re truly resonant and balanced. This shows up in modeling and in Severns’ follow-on experiments; the catch is that multiband use or asymmetry erodes the advantage quickly.

Myth 6 — Radials are still crucial over saltwater

Saltwater is such an excellent ground that a vertical very close to the water line may change only ≈ 0.1 dB when you add a big radial field. (You still need a safe RF/surge return; this is about efficiency.)

What the science actually measured

Brown, Lewis & Epstein (1937) ran the landmark field-strength tests that established how number and length of radials affect efficiency. Their curves underpin broadcast standards and quantify the “diminishing returns” that hams often hear about without seeing the numbers.

Rudy Severns (N6LF) revisited the problem for HF, where soil is not just “resistive.” Using the more general Abbott formulation for power loss in complex ground, he showed why BLE’s low-frequency assumptions underestimate HF ground loss—and why design emphasis should move toward radial density near the base. His measurements confirm the theory.

Jerry C. Sevick (W2FMI) added decades of practical data on short verticals and short/loaded radial systems, demonstrating how much efficiency can be clawed back with thoughtful ground systems even when the radiator itself is compromised.

Data-backed build recipes

• If you can’t dig (rooftop, rock, rented): Use 4 elevated, tuned ¼-λ radials (monoband), 1–3 m above ground, symmetric and clear of metal. Expect performance within ≈ 1 dB of a very large on-ground field—if everything is truly resonant and balanced.
• If you can lay wire on the ground: Aim for 16–32 radials of 0.1–0.25 λ each. That usually buys the biggest step up from “bare earth.” More is better, but expect diminishing returns; don’t stress about exact ¼-wave resonance.
• If you want broadcast-style performance: 60 radials ≈ ¼ λ each is a sweet spot (< 1 dB from “perfect ground”). If you go bigger, use ring-and-radials to put copper where it matters most and save wire.
• Special case — by the sea: Place the vertical as close to saltwater as practical; extra copper is mainly for lightning/safety, not efficiency.

Bottom line

If you’re optimizing a vertical for efficiency, stop chasing folklore. Put copper where the near-field current actually flows—dense radials near the base or a small, tuned elevated counterpoise—and you’ll get most of the performance without mythical wire counts or hand-waving about “RF ground.”

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.