By Danny Horvat, E73M / N4EXA โ antenna design engineer and founder of MyAntennas.com.
Since 2015, we have shipped tens of thousands of EFHW antennas and transformers to hams in over 130 countries, and along the way, I have answered the same questions thousands of times: Does it need a counterpoise? How high? Why is my SWR different from the chart? Where does the choke go? This page is the complete answer โ everything I know about end-fed half-wave antennas in one place, based on decades of professional and amateur antenna work, not internet folklore.
What Is an EFHW Antenna?
An end-fed half-wave (EFHW) antenna is exactly what the name says: a wire one half-wavelength long at its lowest operating frequency, fed at the end instead of the center. Because a half-wave wire fed at its end presents a very high impedance โ on the order of 2,000โ3,000 ohms โ it requires an impedance transformer (typically 49:1) to match the 50-ohm coaxial cable from your radio.
The beauty of the design is in the harmonics. A wire that is a half-wave on 80m is a full wave on 40m, two waves on 20m, and so on. Each of these harmonic resonances also presents high impedance at the wire’s end, so a single wire and a single transformer give you multiple bands with no traps, no radials, and no tuner on most bands.
This is not a new idea. Josef Fuchs patented the end-fed concept in 1927, and in QST (July 1928) he noted that the single-wire connection “has the advantage of working without ground or counterpoise.” Nearly a century later, the physics has not changed โ only the quality of the transformers has.
The Transformer: Where the Engineering Lives
The wire is simple. The transformer is where a good EFHW is won or lost. Our transformers use a 49:1 ratio wound on ferrite cores sized for the power rating โ from single-core 100W LP units for POTA up to triple 3-inch-core 2K-Plus units with roughly 0.5 dB insertion loss for full legal limit operation.
A few facts about these transformers that surprise people:
A multimeter will always show a dead short. The primary and secondary windings share a common connection at the SO-239 shield, so DC continuity between any two points on the box is normal. It is how every RF transformer of this type works โ it is not a defect.
They essentially do not fail in normal use. The ferrite cores in our 1K-Plus and 2K-Plus models would need to reach 300ยฐC to be damaged, and the plastic enclosure would melt long before that. Across tens of thousands of units sold since 2015, we have never seen a transformer fail except from shipping damage or a fall onto concrete. When SWR suddenly changes, the cause is almost always a cable, connector, or lightning arrestor โ more on that below.
Technically, an EFHW transformer is a UNUN, not a BALUN. It transforms one unbalanced impedance (the coax) to another unbalanced impedance (the end-fed wire). If you see our matching units labeled as UNUNs, that is why.
The Counterpoise Question, Settled
No. An EFHW antenna does not need a counterpoise. Fuchs said so in 1928, and my own measurements confirm it today.
The antenna finds its path to low impedance through the coaxial cable shield, which is grounded at the entrance to your home per NEC-810. That is all the “other half” the system needs. The popular internet advice to hang a 0.05-wavelength wire off the transformer’s ground lug does nothing โ I have demonstrated on video that SWR is identical with and without it: watch the measurement here.
The confusion comes from misusing the word. The ARRL Handbook defines a counterpoise as a wire or group of wires mounted close to ground, insulated from it, forming a low-impedance, high-capacitance path to ground. A single pigtail hanging in the air is neither low-impedance nor high-capacitance โ it is not a counterpoise, it is a decoration.
More about counterpoise in detail is at this link: About Counterpoise
Installation: Configurations That Work
One of the most underrated features of an EFHW is flexibility. Because it is fed at the end, you need to get coax to only one end of the wire, and the wire itself can bend. All of the following configurations work:
Horizontal (flat top). The classic setup between two supports. Simple and effective for regional and DX work, depending on height.
Inverted V. One center support, ends sloping down. Keep the apex angle at 90 degrees or more.
Inverted L โ my personal favorite. The transformer mounts low (as low as 3 feet), the wire runs vertically 20โ30 feet, then horizontally to the far support. The vertical section gives you low-angle radiation for DX. This is the configuration on my own station, and my DXCC results with it speak for themselves โ see my QRZ page. Keep the vertical wire a few feet away from the supporting tree trunk or pole, and get the horizontal portion as high as you can.
Sloper. Good for DX; keep the low end at least 10 feet up.
Vertical. Works, and works without radials.
Inverted U and zig-zag. For tight lots. Vertical sides of an inverted U should be no more than half the length of the horizontal section. Ninety-degree bends are fine; the wire can follow a fence line or zig-zag through trees.
Height and Clearance
Minimum recommended heights: 20 feet at all points along the wire for the EFHW-8010 and EFHW-7510. The EFHW-4010 is far less height-sensitive thanks to its high-impedance feed point and performs well anywhere from 10 to 50 feet, which is exactly why it is the go-to for stealth and HOA installations.
Clearance matters as much as height. Keep the wire at least 3 feet from any metal โ towers, poles, gutters, metal roofing โ and expect some interaction even then. Never let the wire touch a tree limb directly; use rope and an insulator, and stay 2 feet from trunks. And do not lay wire on a roof: asphalt shingles are carbon-based and affect the antenna much like metal. A wire on shingles behaves like a wire lying on the grass.
The Reactive Near Field: Why Your SWR Isn’t Like the Chart
Every antenna has a reactive near field extending about 0.159 wavelengths in all directions. Objects inside that zone โ buildings, towers, vehicles, other antennas, your house wiring โ couple to the antenna and change its feed point impedance, exactly like parasitic elements on a Yagi. On 80m that zone is roughly 41 feet; on 40m about 22 feet; on 20m about 11 feet; on 10m about 5 feet.
This is the single most common reason a customer’s SWR curve differs from ours: our published curves are measured with the antenna in the clear, and most backyards are not. A metal shed 20 feet away will raise SWR on 80m and 40m while leaving 20m and up untouched โ the shed is inside the near field on the low bands and outside it on the high bands. The antenna is not defective; it is measuring your yard.
Tuning and Trimming the Wire
Every antenna leaves here resonant, with roughly an extra foot at the end insulator for tuning to your favorite part of the band. Your installation โ height, bends, nearby objects โ shifts resonance somewhat, so fine-tuning on site is normal.
The method: find your current SWR minimum with an analyzer, then calculate the trim using the half-wave formula, length in feet = 468 รท frequency in MHz. For example, if your antenna is resonant at 3.45 MHz (468 รท 3.45 = 135.7 feet electrical length) and you want 3.55 MHz (468 รท 3.55 = 131.8 feet), trim the difference: about 3.8 feet. Cut, don’t fold โ a folded-back wire behaves unpredictably. And cut conservatively; you can always trim more, but you can’t trim less.
Two rules to remember: shortening the wire moves resonance up on all bands simultaneously, and wire stretch over time moves it down. If your SWR minimum has slowly drifted lower over months of tree movement, the wire has stretched โ trim it back.
SWR, Power, and What the Tuner Actually Does
Perspective on SWR numbers: 1.5:1 reflects 4% of your power, 2:1 reflects 11%, 3:1 reflects 25%. Anything under 2:1 is excellent and safe at full-rated power. Between 2:1 and 3:1, reduce to 250โ500W on high-power models. Above 3:1, keep it to 200W or less. At 100W, any SWR your tuner can match is fine.
Here is what many hams miss: your antenna tuner changes the SWR the radio sees, not the SWR at the antenna. The reflected power still flows through the transformer, and with a kilowatt at 3:1, a quarter of that power is heating the ferrite core. That is why “SWR creeps up during long transmissions” is not a defect โ it is an overheating core telling you to fix the match or reduce power. Never combine high power, a tuner, and SWR above 2:1.
Full details and charts: About SWR.
Common Mode Chokes: Where, When, and Whether
You may not need a CMC at all. If your coax shield is grounded at the entrance to your home per NEC-810 and most of the cable run is on or in the ground, the system is already taken care of โ I run 1500W on my own EFHW-8010-2K+ with no choke, because my cable is grounded at the entrance and lies on the ground.
You do want a CMC if the coax is not grounded outside before entering the house, or if you have RFI symptoms: computer resets, WiFi dropouts, interference on cable TV, or noise coupling from house wiring onto the feedline.
Placement is critical and this is where most installations go wrong: the choke belongs at the entrance of the cable to the home, before the shield contacts any grounding. If you put the choke inside at the radio and then run a ground wire from the radio chassis to an outside rod, you have built a bypass around your own choke โ it does nothing. And skip the single snap-on ferrites; they are ineffective at HF. A proper choke needs multiple passes through ferrite material, which is exactly what our CMC-130-3K is.
While we are near the topic: long ground wires from the shack to an outside rod are not a safety improvement at HF โ at radio frequencies, a long wire is an impedance, not a ground. Ground the coax shield at the home entrance per NEC-810 and keep it simple.
Band Physics: 80m, 75m, and the Truth About 160m
Why One Wire Cannot Cover All of 80/75m
A half-wave at 3.5 MHz is about 134 feet; at 4.0 MHz it is about 117 feet โ a 17-foot difference. No single wire can be resonant across that entire range, which is why any single-wire antenna covers roughly 100โ150 kHz of 80/75m with SWR under 2:1. It is physics, not a product limitation, and it applies to every dipole and EFHW ever built.
This is exactly why we build two versions. The EFHW-8010 puts its low-band SWR minimum near 3.55 MHz for CW and FT8 operators. The EFHW-7510 uses a small PCB with matching capacitors in the center of the wire to shift that minimum to about 3.85 MHz for 75m SSB โ same transformer, same 130-foot length, all other bands identical. If you own an 8010 and migrate to 75m SSB, you only need the 7510 wire assembly, not a new antenna.
160m: Do It Right or Not at All
I regularly get asked whether the EFHW-8010 can be pushed onto 160m with a tuner. Electrically, the 130-foot wire is only a quarter-wave at 1.8 MHz, presenting a feed point impedance below 100 ohms. The 49:1 transformer converts that to around 2 ohms โ your tuner will struggle, and losses are enormous; in some cases 10% of your power reaches the wire while 90% becomes heat. On top of that, a wire cut for 160m produces useful harmonics only on 80m and (barely) 40m โ above the fourth harmonic, nothing lands in an amateur band.
For serious 160m work, use the right tool: an MEF-104 or MEF-107 transformer with roughly 260 feet of wire, at least 50 feet up (0.1 wavelength for reasonable efficiency), and expect the narrow 50โ60 kHz bandwidth that band physics dictates for any single wire.
Choosing Your EFHW
Three questions decide it: how much space, how much power, and which low band.
Limited space, portable, POTA: the EFHW-4010-P โ 63 feet covering 40/20/15/10m, happy at 10โ50 feet of height, and available with thin black wire for stealth installations. The lightweight P version is our most popular POTA antenna.
Full HF coverage, 80m CW/FT8: the EFHW-8010 โ 130 feet, 80 through 10m.
75m SSB operators: the EFHW-7510 โ same footprint as the 8010, SWR minimum moved to 3.85 MHz.
Power versions: LP and P (100W, lightweight), 1K (1kW ICAS, two cores), 2K (2kW, three cores), and 2K-Plus (2kW, three large 3-inch cores, lowest insertion loss at roughly 0.5 dB). Wire matters too: 1kW models use 16 AWG XLPE (~100 lb tensile strength, lighter and more concealable); 2kW models use 14 AWG (~200 lb). Whatever you hang, remember that trees in the wind exert hundreds of pounds of force โ install with generous sag. The wire should hang naturally, never guitar-string tight.
Experimenters: our MEF transformer line (Multiband End Fed) is the same engineering without the wire โ you cut your own half-wave using 468 รท frequency in MHz. The MEF-130 covers 1โ30 MHz including 160m; the MEF-330 covers 3โ30 MHz at higher power; the MEF-104 and MEF-107 are the 160m specialists.
Frequently Asked Questions
Does an EFHW antenna need a counterpoise?
No. The coax shield, grounded at the home entrance per NEC-810, provides the return path. A 0.05-wavelength pigtail changes nothing โ measured and demonstrated on video.
How high does an EFHW need to be?
20 feet minimum at all points for the EFHW-8010/7510. The EFHW-4010 works well from 10 to 50 feet. For 160m antennas, 50 feet minimum.
Can I bend the wire?
Yes. Ninety-degree bends, inverted L, inverted U, and zig-zag runs all work. Keep 3 feet from metal, 2 feet from tree trunks, and never rest the wire on roof shingles.
Why did my SWR suddenly change after months of perfect operation?
In order of likelihood: a failed gas discharge tube in a lightning arrestor, a degraded cable or connector, or a wire stretched from tree movement. In tens of thousands of units sold, we have never seen a transformer fail in normal use โ the cores would need to reach 300ยฐC, and the box would melt first. Bypass the arrestor and test the coax before suspecting anything else.
My multimeter shows a short across the transformer. Is it broken?
No โ that is normal for every RF transformer of this type. The windings share a common connection at the SO-239 shield, so DC continuity is expected. A transformer problem shows up on a VNA test with a 2,000โ3,000 ohm load, not on a multimeter.
Can I use an amplifier with an EFHW?
Yes, within the model’s power rating. Keep SWR under 2:1 at full power; between 2:1 and 3:1, reduce to 250โ500W. Remember, the tuner only matches the radio side โ reflected power still heats the transformer core.
Will an EFHW work in my attic?
I do not recommend it. The wiring in your walls, HVAC ducts, carbon-based shingles, and in some homes foil-backed insulation detunes any antenna, and RFI into your own electronics is nearly guaranteed. For HOA situations, a stealth outdoor EFHW-4010 with thin black wire is dramatically better. If the attic is truly the only option, use a remote tuner and inexpensive wire, and accept the compromise.
What coax should I use?
Up to 200W, RG8X is fine. Up to 1kW, RG8X, RG213, or LMR-240. At 1500W, use LMR-400 or DXE-400Max โ that is what I run at my own station.
About the Author
Danny Horvat, E73M / N4EXA, is an antenna design engineer with professional experience in the design of HF NVIS systems, HF, VHF, UHF, and Microwave antennas up to 5.6GHz, and the founder of MyAntennas.com (EuroXpress Corporation, Zephyrhills, Florida).
Related reading: EFHW Installation Manual (PDF) ยท About SWR ยท About Counterpoise





