Why a 9:1 Long-Wire Works… Why It’s Less Efficient than the EFHW/OCF

Key takeaway: A 9:1 end-fed “random wire” can absolutely work and can cover many HF bands with one wire. But for the same physical space, it is usually less efficient than a resonant or near-resonant antenna system such as an EFHW, OCF dipole, or RF.Guru EFOC. The difference is not because the wire refuses to radiate; it is because the complete system includes transformer loss, tuner loss, counterpoise loss, common-mode current, and sometimes high-SWR coax loss.

In this article, “9:1 random wire” means a non-resonant end-fed wire used with a 9:1 UNUN and, normally, a tuner. It is not the same antenna system as a resonant EFHW.

This article explains where the losses come from, why a 9:1 random-wire antenna can vary so much from band to band, and why the lengths 51 m, 37 m, and 22 m are sensible starting points when using a modern ferrite 9:1 UNUN.

Why a 9:1 Random Wire Works

A 9:1 UNUN transforms the feed-point impedance by a 9:1 impedance ratio. In practice, that often moves the impedance closer to a range that coax, a tuner, or a radio’s internal ATU can handle. When the wire length is chosen well, the impedance seen at the UNUN is not too extreme on most HF bands.

That is the real reason these antennas are popular: they are simple, flexible, easy to deploy, and can provide useful multiband coverage from a single wire.

However, “easy to tune” does not automatically mean “maximum efficiency.” A low SWR at the radio only means the radio sees an acceptable load. It does not prove that most of the RF power is being radiated by the wire.

Where the Efficiency Is Lost

The wire itself is usually not the main problem. A reasonably long and elevated conductor can radiate. The losses normally come from the parts around the wire:

• the 9:1 UNUN, especially when it is forced to work into difficult impedances
• the tuner, especially when it must match very high, very low, or highly reactive loads
• the ground, radial, or counterpoise path
• the coax, if the tuner is in the shack and the feedline is operating at high SWR
• common-mode current, when the coax shield becomes part of the antenna unintentionally

Loss inside the 9:1 UNUN

A modern 9:1 UNUN is normally built on one or more ferrite cores, not a powdered-iron T-200-2 core. Its loss is not a fixed number. It depends on ferrite mix, core size, number of cores, winding layout, load impedance, frequency, power level, and duty cycle.

That is why generic loss tables for “a 9:1 UNUN” can be misleading. A good ferrite 9:1 may be quite efficient over much of HF when it is working into a reasonable impedance. The same transformer can become lossy or hot when it is forced to transform a very reactive, very high, or very low impedance.

• Good impedance range: transformer loss can be modest, especially on the mid-HF bands
• Difficult impedance range: loss and heating can rise quickly
• Low bands: transformer loading, tuner range, counterpoise quality, and wire length all become more critical
• High power or digital modes: even moderate loss can create significant core heating

So the important point is not “a 9:1 always loses X dB.” The real point is that a 9:1 random-wire system often asks the transformer to work across a very wide impedance range, and that makes installation quality and transformer design much more important.

Additional tuner losses

A random wire can present very different impedances from band to band. On some bands the tuner sees a comfortable load. On others it may see a very high resistance, a very low resistance, or a large reactive component.

When the tuner is working hard, practical losses can become noticeable. Loss can be low with an efficient external tuner and a moderate load, or much higher with a small internal tuner, high circulating current, high power, or high-duty-cycle digital modes.

A tuner can make the radio happy, but it cannot recover power already lost in the transformer, counterpoise, coax, or tuner components.

Ground and counterpoise losses

A 9:1 end-fed wire needs a return path. If you do not provide one deliberately, the antenna system will find one anyway.

• Ground rod only: simple, but often lossy because RF current flows through soil
• Coax shield as counterpoise: common, but can create feedline radiation, noise pickup, and RF in the shack
• Radials or elevated counterpoise: usually cleaner, more predictable, and more efficient

This is one of the biggest differences between a quick emergency random wire and a well-installed station antenna. The same wire and same UNUN can perform very differently depending on the return path.

Coax losses at high SWR

If the tuner is in the shack, the radio may see a good match, but the coax between the tuner and the UNUN may still be operating with high SWR. That mismatch does not “destroy” power by itself, but it increases the effective loss of real coax because real coax has attenuation.

• Short, low-loss coax: may keep the penalty modest
• Longer coax or thinner coax: can add noticeable loss
• Higher HF bands: usually make coax loss worse
• Remote tuner near the feedpoint: often reduces this problem substantially

This is why many operators describe a 9:1 random-wire system as “several dB down” from a good resonant or near-resonant antenna in the same location.

Real-World Efficiency by Band

Because 9:1 random-wire systems are non-resonant, their efficiency changes strongly from band to band. The same antenna may be surprisingly good on one band and noticeably weaker on another.

For a decent installation with a suitable ferrite UNUN, a short coax run, and a reasonable counterpoise, these are useful practical expectations:

• 40–12 m: often very usable, especially when the wire length avoids extreme impedances
• 80 m: more dependent on wire length, counterpoise quality, tuner loss, and transformer loading
• 160 m: possible with longer wires, but usually a compromise unless the full system is designed for low-band operation

Compared with a good resonant or near-resonant antenna in the same space, a 9:1 random-wire system is often several dB behind on at least some bands. The difference is usually not caused by the wire alone. It comes from the complete system: UNUN, tuner, counterpoise, coax, common-mode current, and installation geometry.

Why 51 m, 37 m, and 22 m Are Smart Choices

Good 9:1 random-wire lengths are not magic. The goal is to avoid the worst feed-point impedance extremes, especially lengths that land close to half-wave or multiple-half-wave conditions on the bands you want to use. Those points can create very high impedance and make the UNUN, tuner, and coax work harder.

These lengths are sensible starting points, not sacred numbers. Final SWR and tuner behavior will depend on height, slope, nearby metal, soil, coax length, counterpoise layout, and the actual UNUN design. Trimming or adding a meter or two can make a difficult band much easier to match.

Why EFHW, OCF, and EFOC Designs Usually Outperform a 9:1 Random Wire

For the same span of wire and the same supports, a resonant or near-resonant antenna system usually has the advantage because less RF power is spent in the matching system and return path.

• Lower matching loss: the transformer or balun is working closer to its intended impedance range
• Less tuner work: many bands can be used directly or with only light matching
• More controlled return current: the antenna structure, not random coax or lossy soil, carries more of the RF current
• More predictable patterns: current distribution is easier to model and repeat
• Stronger low-band performance: especially on 80 m and 160 m, where small losses add up quickly

This is why a 9:1 random wire is excellent as a universal, emergency, portable, or restricted-space antenna, while EFHW, OCF, dipole, and EFOC systems usually make better everyday station antennas when the available space allows them.

Getting the Most Out of a 9:1 Random Wire

• Use a deliberate return path: radials, an elevated counterpoise, or a well-designed ground system will usually beat “whatever the coax happens to do.”
• Keep coax short when possible: especially if the tuner is in the shack.
• Place the tuner near the feedpoint if practical: this reduces high-SWR loss on the coax run.
• Use the choke in the right place: if the coax is acting as the counterpoise, choke near the station entry; if you provide a real counterpoise or radial system, a choke near the UNUN can help keep RF off the feedline.
• Choose a transformer designed for the bands you use: low-band performance depends heavily on ferrite mix, core size, winding method, power rating, and impedance range.
• Derate power on difficult bands: high SWR, digital modes, and lossy matching can heat transformers and tuners quickly.
• Start with sensible lengths: 51 m, 37 m, and 22 m are practical starting points; adjust slightly for your installation.

With these measures, a 9:1 random wire can perform far better than many operators expect. The difference between a poor 9:1 installation and a good one is often not the wire — it is the counterpoise, transformer, tuner position, and coax layout.

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

Questions or experiences to share? Contact RF.Guru here.