Fractal Antenna (Koch Dipole)
"Fractal antenna" covers a family of designs that fold a self-similar, repeating geometric pattern into an element to pack more wire into a shorter physical span — the same basic goal as a loading coil or a trap, achieved with geometry instead of electrical components. This guide builds the most practical ham version: a Koch-curve dipole, where each leg is bent into a single zigzag iteration instead of running straight. The size reduction is real but modest, and this guide says so plainly rather than repeating the more dramatic claims sometimes made about fractal antennas.
What a Koch curve does to a dipole leg
Instead of a straight wire, each dipole leg is bent into a single repeating zigzag (the first iteration of the classic Koch curve): the wire's actual physical path is shorter tip-to-tip than its total length, because it folds back and forth along the way. That extra folded wire length still contributes to the electrical length needed for resonance, so the antenna can resonate at a lower frequency than a straight wire of the same physical span — which is exactly the shortening effect a builder wants.
Koch-folded dipole: physical span shorter than electrical half-wavelength, extra length "hidden" in the folds
Why the size reduction is modest, not dramatic
A single iteration of the Koch fold on a practical ham antenna gives a real but modest size reduction — commonly cited figures land in the 15-20% range versus a straight dipole of the same resonant frequency, not the 50-80% reductions sometimes claimed in popular fractal-antenna writing. Adding further iterations packs in more fold detail but delivers rapidly diminishing additional shortening while adding more bends, more construction complexity, and typically some additional loss — there's a real point of diminishing returns here, and it comes quickly.
- 1st iteration: modest, worthwhile shortening for a manageable amount of extra construction complexity.
- 2nd+ iterations: rapidly diminishing extra shortening for meaningfully more bends and build time — rarely worth it for a homebrew HF dipole.
What fractal geometry doesn't do
Folding a wire into a fractal shape doesn't repeal the basic physics that governs any electrically shortened antenna: bandwidth narrows and efficiency drops somewhat compared to a full-size element, the same tradeoff you'd see from a loading coil or a trap. Some other fractal shapes (like the Sierpinski gasket, more common on printed/PCB antennas than wire HF antennas) can show multiple related resonances, but that's a different specific geometry and use case from the shortened Koch dipole this guide builds.
Honest bottom line
Build this for a real, modest space savings over a straight dipole, or as a genuinely interesting construction project — not expecting to dramatically shrink an antenna's footprint while keeping full-size performance. That expectation-setting matters more here than on most designs, since fractal antennas have attracted more marketing enthusiasm than some of the more conservative claims in this guide reflect.
Installation options
- Flat-top between two supports: the standard install, taking advantage of the modest length reduction in a slightly tighter yard than a straight dipole would need.
- Inverted-V from a single mast: workable the same as any dipole variant, with the usual inverted-V pattern and impedance shift.
- Attic or stealth install: the shortened span can be the difference between fitting in an attic run or not, for builders working within a fixed physical constraint.
| Band | Straight-dipole length | Koch-folded physical span | Notes |
|---|---|---|---|
| 40m | ~65.5 ft (20.0 m) | ~54-56 ft (16.5-17.1 m) | ~15-18% shorter tip-to-tip than a straight dipole |
| 20m | ~33.1 ft (10.1 m) | ~27-28 ft (8.2-8.5 m) | Same proportional shortening applies |
| Fold indentation depth | — | ~13% of each straight segment | Standard first-iteration Koch construction ratio |
Koch Fractal Dipole Dimension Calculator
Materials for Fractal Antenna (Koch Dipole)
Building the Fractal (Koch) Dipole
The wire is longer than the physical span it occupies — the extra length gets folded into the zigzag pattern along each leg.
Choose your design frequency and get both wire lengths
Use the calculator above to get the total wire length per leg (longer than the physical span) and the target physical span itself.
Build a former to hold the zigzag shape
Build or mark a non-conductive former (a light lattice or a series of standoff points) that will hold each leg's wire in the repeating zigzag pattern along its physical span.
Route each leg's wire in the Koch zigzag
Cut each leg to its calculated total wire length and route it along the former in the repeating fold pattern, keeping the fold depth and spacing consistent along the whole leg.
Install the center feedpoint
Connect both folded legs to the center feedpoint bracket, and install the choke balun.
Attach end insulators and raise the antenna
Tie off both formers' far ends to egg insulators and hoist the antenna flat-top or as an inverted-V between supports.
Connect coax and sweep SWR
Connect your feedline and sweep for the resonant dip near your design frequency.
Trim if needed
If the resonant dip is off target, trim a small, equal amount of wire from the outermost fold on each leg and re-sweep — trim from the folded section, not by shortening the physical span directly.
| Symptom | Most likely cause | Diagnosis | Fix |
|---|---|---|---|
| Resonance is noticeably off from the calculated frequency | Fold depth or spacing isn't consistent between the two legs | Compare both legs' fold pattern against each other and against the design | Correct the fold geometry to be symmetric on both legs |
| Bandwidth seems narrower than a straight dipole on the same band | Expected behavior for any electrically shortened antenna | Compare SWR bandwidth against a straight dipole's known figures | Not a fault — this is the normal shortened-antenna bandwidth tradeoff, same as a loading coil would cause |
| Antenna seems to perform somewhat below a straight dipole of the same electrical length | Modest efficiency cost inherent to any shortened element, fractal or otherwise | Compare against realistic shortened-antenna expectations, not full-size performance | Expected; this is the honest tradeoff for the physical size reduction |
How much smaller does this really make my dipole?
Realistically 15-20% shorter tip-to-tip than a straight dipole resonant at the same frequency, using a practical single-iteration Koch fold — a real, worthwhile reduction, but not the dramatic size cuts sometimes claimed for fractal antennas in general.
Should I use more fold iterations for more shortening?
Generally not worth it for a homebrew HF dipole — each additional iteration adds real construction complexity for rapidly diminishing extra shortening, plus some additional loss. One iteration is the practical sweet spot for most builders.
Does this perform as well as a full-size dipole?
Not quite — like any electrically shortened antenna, expect somewhat narrower bandwidth and modestly lower efficiency than a full-size straight dipole. The tradeoff is real but comparable to other shortening methods, not worse.
Is this the same as a Sierpinski fractal antenna?
No — the Sierpinski gasket is a different fractal geometry, more commonly used on compact printed/PCB antennas for multiband behavior, not typically built as a wire HF dipole. This guide covers the Koch-curve-folded dipole specifically.
Do fractal antennas really give multiband coverage automatically?
Some fractal geometries show multiple related resonances, but that's shape-specific and not automatic just from "being a fractal." The Koch dipole in this guide is a single-band shortened design, not a multiband one.
Is precision in the folds really necessary?
Reasonably so — keeping the fold depth and spacing consistent on both legs matters for a clean, predictable resonance and balanced pattern, more so than on a plain straight-wire dipole.