Why We Love Doublets—and When a G5RV Becomes a House of Cards
Why We Love Doublets—and When a G5RV Becomes a House of Cards
The doublet makes its bargain openly: give it balanced feedline, a capable tuner and sensible routing. A G5RV can also work well, but its matching section, coax and installation must cooperate on every band you expect to use.
I love the open-wire-fed doublet because it does not pretend to be a 50 Ω antenna on every band. Its mismatch is visible, its feedline is chosen to tolerate that mismatch efficiently, and the tuner is deliberately part of the system. The G5RV is clever too—but its promise is easier to misunderstand because the balanced section and following coax form a frequency-dependent transformation whose success changes from band to band.
Mark’s video and the line he read: Mark, K3ZD—Ham Florida Man—built this discussion around an RF.Guru article written by Joeri/ON6URE. He quotes the G5RV as “a misunderstood trickster, more trouble than it’s worth, prone to crumble like a house of cards or a poorly made apple cake.” He then contrasts that with Joeri’s “honest” doublet: a multiband system that asks plainly for height, balance and a good tuner.
The metaphor stays because it is the article’s voice. The engineering verdict is more exact: a G5RV is not doomed, and a doublet is not automatically perfect. The complete installed feed system decides.
The Doublet’s Honest Bargain
A doublet is a centre-fed wire whose length is chosen for the site and intended bands rather than for a 50 Ω feedpoint on every band. A balanced transmission line carries the resulting impedance to a balanced matching network—or to a suitably rated transition and tuner.
The attraction is not that balanced line makes mismatch disappear. Standing waves still create voltage and current maxima, the line still has finite conductor and dielectric loss, and the tuner still sees a transformed complex load. The attraction is that a well-built air-dielectric line can begin with very low matched attenuation, so its additional loss under severe mismatch can remain modest.
The doublet says: “I’ll work across many bands—but give me height, give me balance and give me a good tuner.”
That honesty comes with obligations. Keep the two-wire line clear of metal and route it symmetrically. Choose its electrical length so the tuner does not encounter impossible voltage, current or impedance extremes on an important band. If an unbalanced tuner and current balun are used, that balun must survive the actual differential voltage, current, common-mode voltage and complex impedance at its location. A genuinely balanced tuner can be a cleaner interface, but it too has finite range and loss.
What Louis Varney Actually Designed
Louis Varney, G5RV, presented his antenna in 1958 as an effective multiband aerial—not as a 20-metre-only antenna. The familiar full-size version uses a roughly 31 m horizontal span and a defined length of balanced line before a transition to coax. Twenty metres is the basic design band: the radiator is about three half-waves long and the matching section produces a convenient transformation there.
On other bands the same structure remains a multiband doublet with a frequency-dependent matching section. Varney expected an antenna tuning unit where the transmitter did not see a suitable load. That distinction matters. The G5RV is not a magic no-tuner antenna, but neither is it an accidental single-band antenna being abused everywhere else.
Historical boundary: “designed around 20 m” is accurate. “Originally intended for 20 m only” is not. The original proposition was multiband operation, with especially convenient behaviour on the design band and matching required elsewhere.
The Matching Section Is a Transmission Line
The balanced section below the G5RV radiator is commonly called a stub, but it is better understood as a transmission-line matching section. Its input impedance follows the antenna terminal impedance, line characteristic impedance, electrical length and loss:
Zin = Z0(ZL + jZ0 tan βl)/(Z0 + jZL tan βl)
When the two conductors carry equal-and-opposite differential current and remain close relative to wavelength, their far fields largely cancel. The section is then principally a feeder and impedance transformer, not a deliberate vertical radiator. Asymmetry, nearby metal, unequal coupling, a poor transition or common-mode current can make the installed line radiate, but that is a system-balance result—not the intended operating principle.
Where the House of Cards Can Appear
The precarious part is not the name G5RV. It is the junction between a strongly mismatched balanced section and a finite length of coax. The coax sees whatever impedance the balanced section presents on that band. If the resulting SWR is high, coax attenuation rises above its matched-loss value. The penalty may be small with short, low-loss coax at a lower frequency, or substantial with a long or lossy run at a higher frequency.
A tuner in the shack can make the transmitter see 50 Ω. It does not remove standing waves or recover heat already lost in the coax and matching components. It also cannot guarantee that the coax exterior is free of common-mode current.
Height, top-wire shape, soil, nearby conductors, matching-section routing, velocity factor, coax length and tuner range all change the result. Rain, foliage and contamination can affect both G5RV systems and open-wire doublets; neither architecture receives immunity from weather. The practical question is whether those changes remain inside the matching, loss and stress envelope of the installed system.
The real failure mode: copying a named antenna while changing its wire length, matching-section type, velocity factor, routing and coax—then expecting somebody else’s band table to survive unchanged.
A Fair Doublet–G5RV Comparison
| Question | Open-wire-fed doublet | G5RV system |
|---|---|---|
| What is deliberate? | A conveniently sized radiator, low-loss balanced feeder and tuner as one multiband system | A defined radiator plus balanced matching section, usually followed by coax and a tuner where required |
| Where can mismatch loss accumulate? | Balanced line, tuner and any balun or transition | Balanced section, coax, tuner and common-mode-control components |
| What commonly limits a band? | Tuner range, extreme transformed impedance, component voltage/current or line routing | Coax loss under SWR, tuner range, transition balance, component stress or changed geometry |
| Can it work efficiently? | Yes, when the line and matching network are low loss and correctly rated | Yes, when the band-specific transformed load, coax run and tuner remain within an acceptable envelope |
| Does low SWR prove success? | No—measure loss, balance, stress and installed radiation | No—loss can improve the apparent match, and shack SWR hides line loss |
A comparison should therefore declare the exact radiator dimensions and shape, height and surroundings, balanced-line type and length, coax type and length, tuner reference plane, choke or balun, frequency, power, duty cycle and weather state. Measure complex impedance at the transition and tuner, estimate or measure line loss, check common-mode current and monitor component temperature under power.
Variants Are Different Designs, Not Siblings with One Verdict
The ZS6BKW and other G5RV-derived antennas deliberately change the top-wire and matching-section lengths to move the favourable impedance regions. That can improve the set of bands served in one installation while making another band less convenient. Treat each variant as its own transmission-line design. Do not transfer one model’s matching, loss or tuner claims to another merely because the silhouettes look similar.
This is also why the apple-cake joke needs an engineering footnote. Some cakes stand perfectly well because their ingredients and proportions are known. Trouble begins when the dimensions, line type and coax are substituted freely while the original performance promise is left untouched.
Choose the Architecture That Fits the Site
Choose the open-wire-fed doublet when you can route balanced line cleanly to a capable matching network and want one low-loss feeder to tolerate large band-to-band impedance changes. It rewards builders who are willing to treat the tuner, line length, clearance and balance as part of the antenna.
Choose a G5RV or a documented derivative when its full geometry fits the site, the matching section can hang and route correctly, and the resulting coax loss and tuner loads have been checked on the bands that matter. A short coax run and suitable tuner can make it a useful multiband antenna. A compromised replica with arbitrary line and long lossy coax can become exactly the house of cards Mark describes.
Bottom line: the doublet wins on transparency. The G5RV earns its place when the complete installation proves it. Names do not radiate; current distribution does. A 1:1 SWR does not preserve watts; the feed system does.
Sources and technical context
Mini-FAQ
- Was the G5RV originally a 20-metre-only antenna? No. Varney described a multiband aerial. Twenty metres is its basic design band, while other bands normally require the complete feed system and tuner to be considered.
- Does the G5RV matching section intentionally radiate? Not in the balanced differential mode. It acts principally as a feeder and impedance transformer; common-mode current or asymmetry can make the installed line radiate.
- Is a G5RV always lossy? No. Loss depends on band, transformed impedance, balanced section, coax type and length, tuner and installation. High SWR on coax raises loss, but the size of that penalty must be calculated or measured.
- Is open-wire line lossless under mismatch? No. It can have exceptionally low attenuation, but conductor, dielectric, radiation and transition losses remain, and voltage/current maxima still stress components.
- Does a doublet always beat a G5RV? No. A clean doublet installation is often easier to analyse and can reduce feedline loss, while a well-installed G5RV can be entirely effective on selected bands.
- What should be measured before choosing? Complex impedance at useful reference planes, estimated or measured line loss, tuner range, common-mode current, component temperature and repeatability after environmental changes.