For standard 1/2" copper sweat fittings.
| Measurement (C-to-C) | Feet & Inches | Inches | Metric |
|---|---|---|---|
| (A) Long Element (Radiator) | - | - | - |
| (B) Short Element (Stub) | - | - | - |
| (C) Feedpoint Height (hot jumper tap, on Long Element) | - | - | - |
| (D) Spacing | - | - | - |
| (E) Ground Mount Center Height (connector shell, on Short Element) | - | - | - |
| Cut | Feet & Inches | Inches | Metric |
|---|---|---|---|
| Long Pipe Cut | - | - | - |
| Short Pipe Cut | - | - | - |
Note: For soldered fittings, the insertion depth into the cup (~0.5") roughly cancels out the centerline offset of the T-fitting. Cut lengths are equal to electrical lengths.
(E) Connector Mount: Mount the SO-239 (or N-female) directly on the short element (B), centered on its mounting hole at height (E) above — that's calculated to land at the same height as the feedpoint tap (C), not an eyeballed guess. Its shell is the ground/return connection, so soldering or bolting the connector body straight to the pipe there is the ground bond — no separate ground wire needed. Run one short jumper (solid copper wire — even a scrap of Romex/NM conductor works fine) from the center pin across to the long element (A) at height (C). Keep the shell on the short element rather than the long one: since the coax shield is also your equipment ground, bonding it to the short (shorted-stub) side keeps stray current off the actual radiator, which is what stops the feedline itself from becoming part of the antenna. A short gray strut braced back to the long element gives the connector body extra rigidity without changing that wiring. A PL-259 is not mounted on the antenna — it's the male plug on the end of your coax that plugs into the SO-239.
(F) Mast Mount: This lower stub is purely mechanical, plumbed straight down from the long element (A) for a sturdier mounting line — electrically it doesn't matter which pipe it comes off of, since both elements are already tied together at the base. It carries no feed connection of its own.
For standard 1/2" compression plumbing fittings.
| Measurement (C-to-C) | Feet & Inches | Inches | Metric |
|---|---|---|---|
| (A) Long Element (Radiator) | - | - | - |
| (B) Short Element (Stub) | - | - | - |
| (C) Feedpoint Height (hot jumper tap, on Long Element) | - | - | - |
| (D) Spacing | - | - | - |
| (E) Ground Mount Center Height (connector shell, on Short Element) | - | - | - |
| Cut | Feet & Inches | Inches | Metric |
|---|---|---|---|
| Long Pipe Cut | - | - | - |
| Short Pipe Cut | - | - | - |
Note: Compression fittings sit higher on the pipe. We deduct 0.875" from the cut length to account for the gap between the fitting centerline and where the pipe stops inside.
(E) Connector Mount: Mount the SO-239 (or N-female) directly on the short element (B), centered on its mounting hole at height (E) above — that's calculated to land at the same height as the feedpoint tap (C), not an eyeballed guess. Its shell is the ground/return connection, so soldering or bolting the connector body straight to the pipe there is the ground bond — no separate ground wire needed. Run one short jumper (solid copper wire — even a scrap of Romex/NM conductor works fine) from the center pin across to the long element (A) at height (C). Keep the shell on the short element rather than the long one: since the coax shield is also your equipment ground, bonding it to the short (shorted-stub) side keeps stray current off the actual radiator, which is what stops the feedline itself from becoming part of the antenna. A short gray strut braced back to the long element gives the connector body extra rigidity without changing that wiring. A PL-259 is not mounted on the antenna — it's the male plug on the end of your coax that plugs into the SO-239.
(F) Mast Mount: This lower stub is purely mechanical, plumbed straight down from the long element (A) for a sturdier mounting line — electrically it doesn't matter which pipe it comes off of, since both elements are already tied together at the base. It carries no feed connection of its own.
This is an illustrative model, not a simulation or measurement — there's no EM solver behind this page, so it can't know the antenna's real impedance across frequency. It approximates the design as a single resonant circuit using a commonly-cited ~4% 2:1-SWR bandwidth for a copper J-Pole, centered on your entered frequency. Unlike a plain dipole, a J-Pole doesn't generically work on odd harmonics — but it shows one narrower (~2%) peak at 3× your frequency, since that's specifically where 70cm sits relative to 2m (146 MHz × 3 ≈ 438 MHz), which is exactly why dual-band 2m/70cm copper J-Poles are a well-known homebrew design. Purpose-built dual-banders often tweak dimensions for both bands rather than relying on a single-band design's harmonic alone, so treat that peak as a bonus, not a guarantee. Real bandwidth depends on pipe size, height, and nearby objects — use this for intuition, and an actual antenna analyzer for the real answer.
Scroll sideways to see the full 1 MHz–5 GHz ruler — hash marks every 5 MHz (1 MHz–1 GHz) and every 100 MHz (1 GHz–5 GHz).
Same illustrative model, zoomed to a linear scale around your design frequency so you can actually see the shape of the notch (the log-scale chart above compresses it to a sliver).
Scroll sideways to see the full ±50 MHz window — hash marks every 1 MHz, labeled every 5 MHz.
A standard J-Pole radiates equally well in every horizontal direction but spreads some of that energy up and down toward the sky and ground — wasted for most VHF/UHF work. A Super J-Pole adds a second half-wave radiating section above the base, connected through a phasing coil that keeps the current in both sections pushing together in phase. That collinear stacking squeezes more of the radiated energy toward the horizon, commonly cited at around +3 dB gain (roughly double the effective radiated power in the favored direction) over a single-section J-Pole — the same principle behind commercial stacked/collinear base antennas.
Same proven base J-Pole (long element, shorted stub, and feedpoint tap) as the standard tab, extended with a phasing coil and a second half-wave radiator above it.
| Measurement | Feet & Inches | Inches | Metric |
|---|---|---|---|
| (A) Base Long Element (Radiator) | - | - | - |
| (B) Short Element (Stub) | - | - | - |
| (C) Feedpoint Height | - | - | - |
| (D) Spacing | - | - | - |
| (F) Upper Gain Radiator (half-wave) | - | - | - |
| Approx. Total Height (A+F, plus a few inches for the coil) | - | - | - |
(C) Feedpoint / connector mount: wired exactly like the standard J-Pole tab — the SO-239/N shell bonds directly to the short element (B) at height (C), with a single jumper from the center pin across to the long element (A) at that same height. See the J-Pole tab for the full reasoning and mast-mount guidance; it's unchanged here.
(E) Phasing coil — the one dimension on this page that isn't computed: the coil's job is to add roughly a half-wavelength's worth of electrical delay in a compact space, flipping the current in the upper section back in phase with the base so the two sections radiate together instead of partially canceling. Unlike everything else here, there's no simple formula for it — published designs commonly land around 4–8 turns of stiff wire or rod (often the same copper as your elements) wound into a coil roughly 1/2"–3/4" in diameter and 1–2" long, inserted in series at the top of the base section. Start on the low end, check SWR and (ideally) field strength/pattern, and add turns or stretch/compress the coil until it peaks — this component is tuned empirically, not calculated.
(F) Upper radiator: a plain half-wavelength section (same pipe-diameter correction as the base element), mounted above the coil. Because harmonic behavior for a phased, multi-section antenna is far less predictable than a simple single-element design, this page doesn't attempt to estimate 2m/70cm dual-band behavior for the Super J-Pole the way it does for the standard J-Pole — treat this as a single-band design.
This is an illustrative model, not a simulation or measurement. It uses a narrower ~3% assumed 2:1-SWR bandwidth than the plain J-Pole (~4%), since the added phasing coil is itself frequency-sensitive — detuning away from resonance hurts both the match and the phasing relationship that gives the gain. Real bandwidth is even more build-dependent here than the rest of this page, given the empirically-tuned coil — use this for intuition, and an actual antenna analyzer for the real answer.
Scroll sideways to see the full 1 MHz–5 GHz ruler — hash marks every 5 MHz (1 MHz–1 GHz) and every 100 MHz (1 GHz–5 GHz).
Same illustrative model, zoomed to a linear scale around your design frequency.
Scroll sideways to see the full ±50 MHz window — hash marks every 1 MHz, labeled every 5 MHz.
Standard wire dipole strung parallel to the ground.
| Measurement | Feet & Inches | Inches | Metric |
|---|---|---|---|
| (A) Single Leg Length | - | - | - |
| (B) Total Wire Length | - | - | - |
Tip: Always cut your wire about 6 to 12 inches longer than calculated on each leg. It is much easier to fold wire back on itself or trim it to tune for lowest SWR than to add wire back on!
Half-wave dipole mounted vertically (omnidirectional).
| Measurement | Feet & Inches | Inches | Metric |
|---|---|---|---|
| (A) Single Leg Length | - | - | - |
| (B) Total Wire Length | - | - | - |
Tip: The electrical length of a vertical dipole is exactly the same as horizontal. However, to prevent pattern distortion, route your coax straight out horizontally for at least 1/4 wavelength before dropping it down!
This is an illustrative model, not a simulation or measurement — there's no EM solver behind this page. It approximates the design as a single resonant circuit using a commonly-cited 2:1-SWR bandwidth (~10% for a standard half-wave dipole, ~5% for full-wave, both wider guesses than a J-Pole since thin wire tends to be lower-Q than pipe). For the standard 1/2λ dipole, it also shows narrower (~5%) peaks at the 3rd, 5th, and 7th harmonics — a center-fed half-wave's feedpoint favors odd multiples of its design frequency too (the same reason a 40m dipole is commonly also usable on 15m). The full-λ design doesn't get this treatment; its harmonic behavior follows a different, less-established pattern. Real bandwidth depends on wire gauge, height, and surroundings — use this for intuition, and an actual antenna analyzer for the real answer.
Scroll sideways to see the full 1 MHz–5 GHz ruler — hash marks every 5 MHz (1 MHz–1 GHz) and every 100 MHz (1 GHz–5 GHz).
Same illustrative model, zoomed to a linear scale around your design frequency so you can actually see the shape of the notch (the log-scale chart above compresses it to a sliver).
Scroll sideways to see the full ±50 MHz window — hash marks every 1 MHz, labeled every 5 MHz.
What you're building: a single wire, roughly a half-wavelength long, fed right at one end through an impedance-matching transformer (an "unun," short for unbalanced-to-unbalanced) instead of in the middle like a dipole. It's a popular design because only one end needs support, it packs down small, and — cut for the right band — it'll usually load up reasonably on several higher bands too without retuning.
The transformer's job is to step the antenna's very high end-fed impedance down toward the 50 ohms your coax and radio expect. A true resonant half-wave, fed right at its end, typically presents somewhere around 2000–4500 ohms depending on band, height, and what's nearby — which is why 49:1 became the de facto standard (49 × 50 = 2450 ohms, comfortably within that range). Some commercial and kit transformers instead use 64:1 (3200 ohms) or similar, better suited if your actual end impedance runs higher. Lower ratios like 9:1 or 4:1 are usually a different animal entirely — built for random-wire, non-resonant multi-band operation through a tuner rather than a true resonant EFHW.
Enter whatever ratio is printed on (or wound into) your transformer to see what impedance it's actually designed to match:
If you're winding your own rather than buying a kit, the math behind it is simple: impedance ratio = (turns ratio)2. For 49:1, that means a 7:1 turns ratio (7² = 49) — the antenna-side winding needs to have 7 times as many turns as the coax-side winding.
The standard, widely-used winding for a 49:1 EFHW unun is an autotransformer: one continuous length of enameled magnet wire wound onto a ferrite toroid, with a tap brought out partway through:
Core: an FT140-43 ferrite toroid (about 1.4" OD) is plenty for QRP use up to roughly 100W; step up to an FT240-43 (about 2.4" OD) to handle a few hundred watts. Wire: enameled magnet wire, #14–18 AWG — heavier gauge runs cooler at higher power.
That 3-turn-primary convention is specifically standard practice for higher ratios like 49:1 or 64:1. Lower ratios (9:1, 4:1) are usually a different transformer style entirely (a bifilar-wound transmission-line transformer) rather than just scaling this same design down — follow a proven published design for those instead of extending this math to them.
Using your ratio entered above, here's a starting point for turns and roughly how much wire to buy:
Why it works on more than one band: a half-wave wire also resonates reasonably well at its odd harmonics — cut for 40m, it'll typically also load on 20m, 15m, and sometimes 10m. Don't expect those harmonic bands to be quite as clean a match as the design frequency; that's exactly what the external tuner is for.
Off-Center Fed Dipole (OCFD)
Ideal for 80m fundamental (e.g., 3.550 MHz). Offers broad bandwidth for 80/40/20/15/10m.
| Measurement | Feet & Inches | Inches | Metric |
|---|---|---|---|
| (A) Short Leg | - | - | - |
| (B) Long Leg | - | - | - |
| Total Wire Length | - | - | - |
Setup Note: An OCFD requires a 4:1 or 6:1 Current Balun at the feedpoint to match the roughly 200-300 ohm impedance back to 50 ohms for your coax.
Illustrative model, not a simulation — uses a ~12% assumed 2:1-SWR bandwidth at the design frequency, plus narrower (~6%) peaks at the 3rd, 5th, and 7th harmonics, matching the real, well-documented multiband behavior of a half-wave OCFD. Verify with an analyzer.
Scroll sideways to see the full 1 MHz–5 GHz ruler — hash marks every 5 MHz (1 MHz–1 GHz) and every 100 MHz (1 GHz–5 GHz).
Scroll sideways to see the full ±50 MHz window — hash marks every 1 MHz, labeled every 5 MHz.
End-Fed Half-Wave (EFHW)
Ideal for 40m fundamental (e.g., 7.150 MHz). Easily rigged as a sloper for 40/20/15/10m.
| Measurement | Feet & Inches | Inches | Metric |
|---|---|---|---|
| (A) Main Radiator | - | - | - |
| (B) Counterpoise (0.05λ) | - | - | - |
Setup Note: An EFHW requires a 49:1 Unun. While the coax shield often acts as a counterpoise, cutting a dedicated short counterpoise wire (B) attached to the ground lug of the Unun helps keep common mode current out of your shack.
Illustrative model, not a simulation — uses a ~8% assumed 2:1-SWR bandwidth at the design frequency, plus narrower (~4%) peaks at the 3rd, 5th, and 7th harmonics, matching the real, well-documented multiband behavior of a half-wave EFHW. Verify with an analyzer.
Scroll sideways to see the full 1 MHz–5 GHz ruler — hash marks every 5 MHz (1 MHz–1 GHz) and every 100 MHz (1 GHz–5 GHz).
Scroll sideways to see the full ±50 MHz window — hash marks every 1 MHz, labeled every 5 MHz.
Every antenna on this page is built around one core idea: resonance. Radio waves travel at the speed of light, and a wave's length (its wavelength) is tied to its frequency — the higher the frequency, the shorter the wave. In the units hams use every day, wavelength in feet works out to roughly 984 divided by frequency in MHz. A "half-wave" antenna is built to roughly half that length; a "quarter-wave" antenna to roughly a quarter. Building a piece of wire or pipe to one of these fractional lengths lets standing waves of voltage and current naturally form along it, which is what makes it efficient at launching (or receiving) radio energy at that frequency — and considerably less efficient at frequencies it wasn't cut for.
None of the formulas here use the pure "free space" wavelength number, and that's deliberate. A real wire or pipe isn't infinitely thin and isn't floating in a vacuum — it has some thickness, and its ends act a little like a small capacitor, which effectively makes it "look" electrically longer than it physically is. To compensate, a resonant antenna's actual physical length ends up a bit shorter than the textbook free-space number — typically around 4–5% shorter for thin wire. That's why the classic dipole formula uses 468 (not the free-space value of 492) divided by frequency in MHz to get feet, and why every calculator on this page bakes in a similar practical correction rather than the raw physics number.
This same effect is bigger the fatter the conductor gets, which is why the J-Pole and HF Dipole calculators let you pick your actual pipe size or wire gauge: a 1" copper pipe needs to be cut noticeably shorter than 1/2" pipe for the same frequency, while different wire gauges on a dipole barely move the needle (thin wire is thin wire, electrically). Picking your real size applies a small, clearly-labeled correction on top of the base formula — it's still an approximation, so the SWR-and-trim advice above applies even more once you stray from the baseline size.
Once an antenna is built, the way you check whether it's actually resonant where you think it is — and whether it presents a good match to your radio — is with SWR (Standing Wave Ratio). Your coax and radio are designed around a 50-ohm impedance; if the antenna's impedance at the feedpoint doesn't equal that, some of the power heading up the cable reflects back down it instead of radiating. The ratio between the power going up and the power bouncing back is the SWR. A perfect match reads 1:1; most radios are happy anywhere under about 2:1; higher than that wastes power, can shorten your range, and on some radios triggers automatic power reduction (or worse, in extreme, sustained cases). An SWR meter or antenna analyzer connected between the antenna and the radio is how you actually verify this, and it always has the final word over any calculator — that's why every section on this page recommends cutting a little long and trimming down while watching your SWR, rather than trusting a cut-once calculated number blindly.
A J-pole is really two pieces of pipe doing two very different jobs. The long element (measurement A) does almost all of the actual radiating. It's fed near one end rather than in the center, which gives it a very high impedance right at its base — commonly thousands of ohms, nowhere close to the 50 ohms your coax wants to see. Connecting coax straight to it wouldn't work well at all.
That's the entire reason the short element (measurement B) exists. It's shorted to the long element at the very bottom, which turns the pair into a simple two-wire transmission line that's closed off (shorted) at one end — not unlike a piece of ladder line with the far end jumpered together. Basic transmission-line theory says a shorted line reads as a dead short, 0 ohms, exactly at the short itself, and presents more and more reactance the further you move away from it — on its own, that's not yet a usable 50-ohm match, just an adjustable "electrical spring." The real match happens when that reactance combines with the long element's own impedance above the tap: pick the right tap height (measurement C — typically just an inch or two on 2m/70cm), and the two cancel out into something very close to a plain, real 50 ohms. Tap in there, center conductor to the long element and shield to the short element, both at that same height, and you've built an impedance-matching network out of nothing but the antenna's own geometry — no separate matching components required. That's the trick a J-pole is built around.
A few practical consequences fall directly out of that mechanism:
Mounted vertically, a J-pole radiates equally in every horizontal direction (omnidirectional) with a low takeoff angle, which is exactly the coverage pattern most VHF/UHF FM and repeater work is looking for.
The dipole is the simplest resonant antenna there is: two straight legs, each roughly a quarter-wavelength long, fed where they meet in the middle. Current peaks at that center feedpoint and tapers toward zero at the far tips of each leg; voltage does the opposite, low in the middle and high at the ends. That current-and-voltage relationship is what gives a center-fed half-wave dipole its textbook feedpoint impedance of around 72 ohms in free space — close enough to 50-ohm coax that a direct connection gives a workable match, usually well under 2:1 once it's up in the air and trimmed.
Horizontal versus vertical mounting is purely about orientation, not electrical length — the length formula doesn't change either way. What does change is the radiation pattern and polarization:
One thing worth watching on either orientation: a dipole is a balanced load (symmetric, with no natural "ground" side) fed by unbalanced coax (one conductor is literally the shield). Without something to bridge that mismatch — even something as simple as a few tight loops of the coax taped together near the feedpoint to choke off common-mode current — some RF current will creep onto the outside of your coax shield, which can skew the pattern and make your SWR oddly sensitive to how the feedline is routed. It's a cheap, easy addition that heads off a genuinely common source of "why is my dipole acting weird" problems.
An OCFD is a regular dipole with one deliberate change: the feedpoint is moved off-center, typically splitting the wire around 33% and 67% of its total length instead of an even 50/50. Moving the feed off-center changes the impedance seen there — instead of the tidy ~72 ohms of a center feed, you get something in the 200–300 ohm range that also happens to present a usable match not just at the design frequency, but at several harmonically related bands too (a wire cut for 80m, for example, will typically also work reasonably on 40, 20, 15, and 10m). That multi-band behavior without touching the antenna is the entire appeal of an OCFD.
The tradeoff is that 200–300 ohms is a poor direct match for 50-ohm coax, so an OCFD needs a 4:1 or 6:1 current balun at the feedpoint to transform that impedance down — and, just as importantly, to choke off common-mode current, since this is still fundamentally a balanced antenna being fed by unbalanced coax, same as the plain dipole above.
An EFHW takes the opposite approach from a center- or off-center-fed dipole: it feeds the wire right at one end instead of anywhere near the middle. The very end of a half-wave wire is a voltage maximum and current minimum, which means the impedance there is extremely high — commonly several thousand ohms, nowhere near 50. Matching that requires a step-up transformer, almost always a 49:1 unun (some designs use 64:1) — a specially wound ferrite-core transformer that trades impedance for current, the same way a mechanical gear ratio trades speed for torque.
Because that high-impedance end of the antenna provides very little of a return path on its own, EFHW builds add a short counterpoise wire (roughly 5% of a wavelength, calculated as measurement B on this page) connected to the unun's ground lug. It isn't there to radiate — it's there to give stray common-mode current a short, defined path to follow instead of finding its way onto your coax shield and back into the shack. Like the OCFD, an EFHW cut as a half-wave on one band will also present a usable, if imperfect, match on its odd harmonics, which is why it's a popular "several bands off one wire" antenna — often rigged as a sloper or inverted-L since only one end needs to be up high.