John Portune’s Window-Line Argument Fails at the Foundation

Why the reasoning in A Novel Approach to Using Window Line overreaches the data

John Portune’s 2018 QST article describes a real experiment: 450 Ω window line placed inside polyethylene foam pipe insulation, then tested in contact with concrete, wet soil, and metal roofing. The weak point is not that he measured something. The weak point is what he tries to make that measurement prove. His data may support a narrow practical observation about added loss in one specific setup. It does not support the broader claim that foam-covered window line can now be routed much like coax in the real world.

Spacing Is Not Shielding

This is the first place where the argument breaks. Coax works near metal because its differential fields are largely confined inside the cable. Open two-wire line is different. Its fields are not confined in the same way, which is why routing, nearby objects, and environmental symmetry still matter.

A foam sleeve can absolutely reduce direct contact, abrasion, and some immediate coupling by increasing distance. That part is real. But it does not create an electromagnetic shield around the line. It adds spacing, not containment. That difference is not cosmetic. It is the whole point.

That is also why classic ladder-line guidance never treated distance from metal and ground as superstition. Those warnings exist because nearby conductive or lossy materials can change the line’s electrical behavior, disturb balance, and in some installations push the feed line into becoming part of the antenna system rather than a neutral transfer path.

The Measurements Are Narrower Than the Claims

Portune’s measurement window is much narrower than the confidence level of his conclusion. In his own description, the key loss curves are examined from 18 to 22 MHz, with the argument centered near the first half-wave resonance around 19.8 MHz. That is a useful way to isolate a real-loss number for that test line. But it does not answer the larger installation question amateurs actually care about.

It does not tell you whether the line remains well balanced in a real station. It does not tell you whether common-mode current is being excited. It does not tell you whether the feed line is starting to radiate. It does not tell you whether the antenna pattern or noise pickup changes. And it does not prove that the same behavior holds across multiple HF bands where current and voltage maxima appear at different places along the line.

In other words, “modest added attenuation in this test” is not the same thing as “routing no longer matters.”

The Measurement Technique Cannot Isolate Ladder-Line Loss

Portune’s headline loss test is not a direct measurement of the ladder line alone, but of a composite fixture consisting of a 50 Ω VNA, a 9:1 Guanella current balun, the 450 Ω window line, and an open-circuit far end. Because the loss is inferred from the reflected signal after a round trip, the result necessarily includes not only the line itself, but also balun loss, launch geometry, mismatch effects, and any differential-to-common-mode conversion introduced by nearby concrete, wet soil, or metal surfaces. The line is absolutely part of the measurement, since the open termination creates a full reflection and a standing-wave pattern along the feed line, but the method cannot cleanly separate true differential-mode line loss from balun imbalance or fixture-related error. That is the real weakness of the technique: it may show that one specific balun-plus-line setup looked acceptably low-loss in a small-signal VNA test, but it does not prove that foam-covered window line itself behaves like coax in real installations.

Loss Alone Is Not the Whole System Story

This is the second place where the reasoning overreaches. Open line over or near ground is not just an isolated ideal two-conductor line. In real installations it becomes part of a larger multiconductor environment. Once that happens, you are no longer dealing only with differential-mode line loss. You also have to care about imbalance, environmental asymmetry, common-mode excitation, and the possibility that the feed system itself starts influencing the antenna.

That is why a small change in measured attenuation does not settle the bigger argument. A feed line can show modest added loss and still behave badly in other ways that matter just as much at the station: altered pattern, shifted current distribution, extra received noise, increased local coupling, RF in the shack, or feedline radiation.

Real Installations Are Messier Than the Test Geometry

Another problem is the jump from a controlled test geometry to much broader real-world reassurance. A short run placed on a patio roof, on concrete, or in foam on the ground is not the same as a line routed past gutters, siding, a metal window frame, house wiring, rebar, conduit, foil insulation, or the metal skin of a shed or RV.

That is where symmetry starts to matter as much as average spacing. It is not enough to say both conductors are “inside foam.” If one side of the line couples more strongly to nearby structure than the other, the line can become unbalanced even though the mechanical spacing looks acceptable. That was not fully tested, yet the article’s tone encourages readers to generalize into exactly those more complex situations.

The 450 Ω Label Does Not Mean the System Is 450 Ω

A second foundation-level weakness appears in the matching discussion. The article uses a 9:1 Guanella current balun for measurement and then slides too easily from “450-ohm line” to a practical 50-to-450 transformation mindset. But the 450 Ω printed on window line is its characteristic impedance, not the fixed impedance that the complete antenna system will present at the shack end.

The actual impedance seen at the source end depends on the antenna load, line length, frequency, and standing-wave condition. That is basic transmission-line behavior. So a 9:1 balun is not a universal answer merely because the feed line says 450 Ω on the jacket. A 450-ohm feed line is a transmission line, not a 450-ohm resistor.

This matters because it is easy for readers to come away with the wrong mental model: line label equals system impedance, so 9:1 must be inherently appropriate. That is not how real feed systems behave.

A Matched Feed Point Does Not Automatically Mean a Clean System

The article also states, in essence, that the antenna does not care how it is fed as long as the feed point is matched. That is only safely true in the ideal case where the feed system remains balanced and non-radiating. Once the line couples to nearby metal or ground and common-mode current appears, the feed system itself can influence radiation pattern, current distribution, local noise pickup, and the total behavior of the antenna system.

A tuner finding a happy match is therefore not proof that the feed line is behaving well. It only proves that the transmitter sees an acceptable impedance at that point. Those are not the same thing.

What the Article Can Fairly Claim

There is a fair and useful version of Portune’s conclusion. It would be something like this: short sections of foam-protected 450 Ω window line may sometimes work better in adverse routing situations than many amateurs expect, at least in terms of added loss in one tested setup. That is a reasonable practical observation.

But that is a much smaller claim than suggesting the old cautions about routing open line near metal, on the ground, or through hostile installation environments were largely wrong. They were not wrong. They were incomplete simplifications of a real electromagnetic problem.

The Bottom Line

The base assumption in the article is wrong: increasing spacing with foam is not the same as shielding an open line. Because that assumption is wrong, the broader argument built on top of it does not hold. The measurements may still be interesting. The workaround may still be useful. But it does not rewrite the rules of open-wire feedline behavior.

At best, Portune demonstrated a limited practical exception. He did not overturn the reason ladder line is normally kept away from metal, away from the ground, and routed with balance in mind.

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.