Why Back-to-Back EFHW Measurements Keep Fooling People

Updated: December 2025 — Technical follow-up clarifying why “back-to-back EFHW transformer measurements” are structurally misleading, even when performed carefully with VNAs or wattmeters.

If you have ever published anything critical about EFHW transformers, you know what follows: screenshots of VNAs, smooth S21 plots, and the claim that “efficiency is proven” because two 49:1 transformers were measured back-to-back.

This article exists because that test is still widely misunderstood — and because it continues to be used to make claims it cannot support.

A back-to-back test can be executed correctly, repeatably, and with good instruments — and still tell you almost nothing about how an EFHW system behaves on air.

What a Back-to-Back Test Actually Measures

When two identical EFHW transformers are connected back-to-back, their impedance transformations cancel. What remains is a neat, well-behaved 50 Ω → 50 Ω two-port network.

Under those conditions, the test mainly shows:

• Small-signal copper loss in the windings
• Core loss at a very modest flux swing
• Insertion loss of a nearly matched RF component

That is not EFHW operation. That is easy-mode transformer operation.

In other words: the test answers the question “How lossy is this device when used as a nearly matched RF component?” — not “How efficient is this EFHW system on air?”

The EFHW Feedpoint Is Not a Resistor

A real EFHW feedpoint is not 2450 Ω resistive, not stable, and not benign.

Depending on band, height, environment, and routing, the impedance seen at the end of a half-wave wire is:

• High (often several kΩ)
• Strongly reactive
• Highly frequency-dependent

Back-to-back testing silently replaces that ugly, reactive reality with a friendly resistive termination.

The result is lower internal voltage stress, lower circulating current, and dramatically reduced copper and core heating compared to real EFHW use.

Reactive Current Is the Real Enemy — and the Test Removes It

In an EFHW system, much of the current at the transformer primary is reactive current.

That current does not radiate — but it does:

• Heat copper via I²R loss
• Increase core flux swing
• Push ferrite into higher loss regions

A back-to-back fixture eliminates this mechanism almost entirely. The test environment is deliberately constructed so that reactive circulating currents collapse.

This is why “looks cool on the VNA” and “runs cool on the mast” are not the same statement.

The Shunt Capacitor: Match Cosmetic, Not Efficiency Magic

The now-famous 100–120 pF capacitor placed across the 50 Ω side of many EFHW transformers deserves special attention.

Yes — it can improve SWR at certain bands, especially 20–10 m. But that does not mean it reduces losses inside the transformer.

What it actually does:

• Retunes the impedance seen by the radio
• Masks mismatch at the input
• Leaves winding current largely unchanged for a given delivered power

In back-to-back measurements, this capacitor often just makes the already artificial 50 Ω test network look prettier.

A better SWR plot is not evidence of lower copper loss or lower core loss.

The “Good SWR, Bad Efficiency” Trap

There is a more subtle and more dangerous failure mode.

The shunt capacitor can resonate with the transformer’s magnetizing inductance. When that happens:

• The reactive components cancel
• The remaining impedance appears resistive
• Any core loss becomes a real power sink

The SWR can look excellent while real power is being dissipated inside the ferrite.

Back-to-back tests are structurally incapable of warning you about this — because the resonance simply improves the apparent match.

Common-Mode Currents Vanish in the Fixture — But Not on the Mast

An EFHW is not a two-terminal device.

The coax shield, mast, ground coupling, and nearby structures form an invisible third conductor. That is where common-mode current lives — and where many EFHW problems originate.

A symmetric back-to-back setup provides no mechanism to excite that mode.

So the test cannot tell you:

• How much your coax will radiate
• Whether your choke strategy works
• Why the match changes when you move the feedline
• Why RFI appears only at power

“But My VNA Uses Milliwatts” — The Missing Nuance

Small-signal measurements are valid — within limits.

For strictly linear networks, S-parameters scale with power. But EFHW transformers are not strictly linear at operating power:

• Ferrite permeability shifts with flux density
• Core loss is amplitude- and frequency-dependent
• Winding resistance rises with temperature
• Voltage stress and breakdown margins matter

A milliwatt-level VNA sweep cannot certify high-power efficiency or thermal behavior.

Why the 10–17 MHz Region Is Where Things Fall Apart

The trouble bands are not random.

As frequency increases, the EFHW feedpoint impedance drops because multiple half-waves are fed from the same point. The 49:1 ratio becomes increasingly mismatched.

At the same time:

• Leakage inductance becomes more visible
• Parasitic capacitance matters more
• The shunt capacitor resonance often lands in this region

The result is often “nice SWR, ugly internal stress” — exactly the condition back-to-back testing hides.

You Can Simulate This — And It Explains Everything

This entire failure mode can be demonstrated with SPICE-based simulation using non-ideal transformer models.

When you compare:

• A back-to-back 50 Ω → 50 Ω setup
• The same transformer feeding a high-impedance, reactive load

You will see that:

• Input behavior can look benign in both cases
• Internal winding currents and losses do not

The plots make it painfully obvious why one test reassures — and the other reveals.

What Actually Meaningful Evaluation Looks Like

If the goal is to understand real EFHW behavior:

• Use realistic, high-impedance complex loads — not just 50 Ω
• Test at power and observe temperature rise and drift
• Measure common-mode current on the feedline

Anything less is measuring convenience — not performance.

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.