DeltaRex: Why the 8 m Coax Section and Choke Matter

Our DeltaRex installation uses an 8 m section of coax between the 4:1 feed unit and a 1:1 current choke. That arrangement is part of the practical system for obtaining reasonable SWR. It should be installed deliberately, rather than treating all coax between the antenna and shack as electrically interchangeable.

A remote tuner is another workable approach. In the arrangement discussed here, the existing 4:1 feed unit stays in place. The tuner moves close to its coax input, with a short connection between them, so matching is performed near the antenna. That removes the need to use the prescribed 8 m section as the matching arrangement, but common-mode current control still needs its own design.

Both approaches deserve an accurate explanation. The 8 m length is an installation prescription for this antenna system. It is not a general law that eight metres of 50 Ω cable improves any antenna's SWR, and a remote tuner is not a substitute for every function of a choke.

The three parts perform different jobs

The current product description identifies a bottom-fed, wide-angle delta structure with tubular sides and a spaced dual-wire top section. Its 4:1 transformation and its separate 1:1 choke must be considered separately. Neither the triangular outline nor a transformation ratio tells us the complete installed current distribution.

The 8 m is measured along the cable from the 4:1 feed unit to the first choke. It is not the vertical height of the antenna and not necessarily the total coax length to the station. Cable continuing beyond that choke still has differential impedance transformation and loss. A finite choke can also allow residual exterior current beyond it.

What the 4:1 transformation actually means

For an ideal impedance transformer connected in the intended step-down direction:

Z_{coax port} = Z_loop/4

A hypothetical 200 + j120 Ω loop load would therefore become 50 + j30 Ω, before cable transformation. The resistance has reached 50 Ω, but the reactance remains. A hypothetical 40 Ω resistive load would become 10 Ω; the same ratio would make the 50 Ω mismatch larger in that case. A fixed transformer scales the load. It does not search for a match.

Those values are arithmetic examples, not DeltaRex measurements. Real feed units also have finite magnetizing impedance, leakage, capacitance and loss. Their stress depends on actual voltage and current, frequency, load and duty cycle. The nominal ratio alone establishes none of those limits.

Inside the coax: impedance changes with length

Take the load at the antenna end of a uniform cable as Z_L. For a lossless line of characteristic impedance Z_c, length ℓ and phase constant β:

Z_in = Z_c [Z_L + jZ_c tan(βℓ)] / [Z_c + jZ_L tan(βℓ)]

In this equation, Z_L is the impedance looking into the 4:1 unit with the installed antenna attached. It is not the untransformed loop impedance. The electrical length is βℓ = 2πfℓ/(c VF), using the cable's internal differential-mode velocity factor.

For a real 50 Ω characteristic impedance, the reflection coefficient and SWR are:

Γ_L = (Z_L − 50)/(Z_L + 50)
Γ_in = Γ_L exp(−j2βℓ)
SWR = (1 + |Γ|)/(1 − |Γ|)

Changing the length rotates the reflection coefficient without changing its magnitude. The impedance can move through very different resistive and reactive values while the 50 Ω SWR remains constant. The scikit-rf transmission-line example demonstrates the lossless invariant and the reduction in input reflection when attenuation is introduced.

That distinction matters to tuners. A real tuner has finite component values, switching combinations and voltage/current limits. Two loads with the same SWR can occupy quite different places in its matching range. Moving the impedance to a more manageable value can help it find a match or reduce stress, without reducing the untuned line's SWR.

Eight metres has no single electrical length

The following calculation uses an intentionally fixed hypothetical load of 100 + j50 Ω at the cable's antenna end. The line is lossless and exactly 50 Ω. Holding the load constant isolates the cable effect; a real multiband antenna's load would also change with frequency.

Even the nearly real 19.1 Ω result is not a 50 Ω match. Removing reactance is insufficient. These examples also explain why substituting a different cable type can change the load presented to the tuner despite keeping the tape-measure length unchanged.

A quarter-wave line gives Z_in = Z_c²/Z_L. A 50 Ω quarter-wave line transforms 100 Ω into 25 Ω: both are 2:1 SWR relative to 50 Ω. A purpose-designed quarter-wave transformer uses a suitable characteristic impedance as well as a suitable electrical length. The phrase “quarter-wave coax” alone does not establish a match.

Outside the coax: the choke defines a current boundary

At HF, the differential transmission current on the centre conductor is paired with the opposite current on the shield's inner surface. The shield's exterior can carry another current relative to the surroundings. That exterior path is affected by the antenna, ground, mast, house, cable routing and choke.

Roy Lewallen, W7EL, explains these separate currents and the action of a current balun. His analysis also shows why nearby structures and asymmetric connections matter. The practical consequence here is that a common-mode choke acts on an external current path that the ordinary two-conductor transmission-line equation does not describe.

Placing the choke about 8 m from the feed unit leaves that exterior conductor between the loop system and a substantial common-mode impedance. If that conductor is excited, its input impedance and current distribution can influence the antenna-system load and radiation. Moving the choke changes the available path. It can therefore change the measured SWR through a change in the installed load, not just through differential cable transformation.

The choke is finite: Z_CM(f) = R_CM(f) + jX_CM(f). It is neither a perfect break nor an instruction that every point before it must carry strong current and every point after it must carry none. Coupling can excite other sections, and the installed suppression depends on the surrounding impedances.

I would not assign a percentage of DeltaRex radiation to the coax, or claim that exterior current accounts for a specified SWR improvement, without measurements or a validated model of that installation. The 8 m/choke arrangement is the practical prescription. Separating its physical mechanisms requires more evidence than a single sweep at the radio.

The coax datasheet's velocity factor applies to the internal mode. The exterior field occupies air, jacket, nearby structures and ground. Its propagation and resonance cannot be calculated by automatically applying the internal VF. Nor is eight metres a universal common-mode quarter wavelength across several HF bands.

Reasonable SWR must have a reference plane

An SWR result should identify the frequency, installation, cable type and length, choke position, measuring point and tuner state. “At the antenna” could mean the loop terminals or the coax side of the 4:1 unit; these are different ports. A measurement after the shack tuner describes yet another interface.

When the tuner is bypassed, coax length can change the impedance, loss and exterior-current boundary seen by the instrument. When the tuner is active, it can produce a low transmitter-side SWR while substantial standing waves remain between tuner and antenna.

Loss adds a further complication. With propagation constant γ = α + jβ:

Γ_in = Γ_L exp(−2γℓ)

For the usual approximately real 50 Ω cable model, a one-way matched attenuation of A dB reduces the magnitude of the returned reflection by 10^−A/10. The returning wave has travelled through the cable twice. A lower indicated SWR can therefore coexist with less useful power delivered to the antenna. Cable attenuation should be measured or calculated separately; it should not be credited as an antenna improvement.

My use of “reasonable SWR” describes a practical installation and matching objective. It does not mean that every band, height, roof and tuner will produce the same trace or a guaranteed 1:1 result. The relevant outcome is a stable match within component limits, acceptable losses and useful radiation.

The remote-tuner alternative retains the 4:1 feed unit

For the remote arrangement, the signal path is:

Radio → station feeder → remote tuner near antenna → short RF connection → existing 4:1 feed unit → DeltaRex

The conventional arrangement is:

Radio / shack tuner → station feeder → 1:1 current choke → prescribed 8 m coax → 4:1 feed unit → DeltaRex

These show the matching order. In the remote arrangement, common-mode suppression must also be provided on the appropriate coax and, where necessary, power/control paths. The exact choke positions depend on the tuner's RF connections, chassis and supply arrangement. The tuner is not drawn as if it automatically performs that job.

A tuner placed beside the 4:1 unit sees approximately the transformed loop-system impedance, plus the effect of the short output connection. It adjusts its network to present a suitable load to the long station feeder. The previously prescribed 8 m section is no longer required for this matching arrangement. This is a different feed configuration, so the previous exterior-current boundary and its pattern contribution cannot be presumed to remain identical.

If the tuner is installed in the shack, or simply at the far end of the same 8 m section, that section remains on the mismatched side of the tuner. The differential transformation and possible additional loss in it remain. Calling a tuner “remote” only helps explain the RF system once its actual position is specified.

The retained 4:1 unit also remains inside the power and loss budget. The tuner must accommodate the impedance presented through it on each intended band, and the output connection and feed unit must withstand the associated RF voltage and current. Successful tuning at low power alone does not establish continuous-duty capability.

What improves, and what still needs checking

With a successful remote match, most of the long feeder can operate near its intended impedance. This can reduce mismatch-related feeder loss and stress compared with putting the same long feeder between the antenna and a shack tuner. The result depends on cable attenuation, the original mismatch and the remote tuner's own loss.

The remote tuner does not change the loop dimensions or remove the effect of ground and roof. It can change installed currents if its chassis, output lead, choke or supply wiring changes the external RF paths. The useful comparison therefore keeps the radiating geometry and current-control arrangement documented.

A remote tuner can be a useful engineering choice without being a realistic choice for every station. Mounting space, access, power, control, weather exposure, matching range and cost all matter. A passive outdoor arrangement with station-side matching remains a legitimate practical solution when its losses and stresses are acceptable.

Installing and checking the conventional arrangement

1. Keep the intended geometry. Record the loop's heights above ground, orientation and nearby conductors. A rooftop relocation changes the installation before any cable adjustment is made.
2. Retain the 4:1 feed unit and measure the prescribed section. Use 8 m of coax along its route to the first 1:1 current choke. Record cable type and actual routing. Do not assume a random tightly wound coil is equivalent to a routed section; winding can add common-mode impedance and coupling.
3. Choose the choke for the operating bands and duty. Its relevant impedance is common-mode impedance versus frequency. Nominal transmitter watts alone do not determine ferrite dissipation or RF voltage stress in this installation.
4. Measure with the tuner bypassed first. Save complex impedance and SWR at the station reference plane. If the feed-unit port is characterized, use calibration or appropriate de-embedding and maintain the intended exterior-current boundary; connecting the analyzer differently can itself alter that boundary.
5. Check matching and loss separately. Confirm the station tuner covers the required operating frequencies. Account for total feeder loss and mismatch rather than judging efficiency from the transmitter-side SWR after tuning.
6. Inspect the outside current path. With a suitable calibrated RF current probe, compare several locations before and after the choke at a controlled power. A current minimum at one point does not prove negligible current everywhere.
7. Verify the intended operating duty. Increase power within all component ratings, watching for match instability and excessive heating. Preserve records so seasonal or weather-related changes can be investigated rather than hidden with a new tune cycle.

For the remote arrangement, repeat the same checks with the tuner in its actual outdoor position and its normal power/control wiring connected. Document the short connection to the retained 4:1 unit and the new choke locations. Tune success, current suppression and thermal margin are separate observations.

My recommendation

For the conventional DeltaRex installation, use the specified 8 m coax section and choke as a defined part of the system. It offers a practical route to manageable matching without requiring an outdoor tuner. Keep its cable route, antenna height and operating conditions in the description of any SWR result.

When a suitable remote tuner is available, place it near the antenna on the coax side of the retained 4:1 feed unit, keep the output connection short, and design the common-mode control around that installation. Then compare the two approaches using accepted power, feeder and tuner losses, exterior currents and the required radiation directions.

The engineering objective is straightforward: deliver useful power to the installed antenna while controlling unintended current paths. A pleasing number at the radio is one useful check within that larger task.

Questions, antenna-factor records or height trials to share? Contact RF.Guru.