Vertex-Fed Delta Loop or V-Dipole for Inter-European Paths?
Vertex-Fed Delta Loop or V-Dipole for Inter-European Paths?
A point-down, vertex-fed delta loop and a V-dipole can both support regional contacts from 20 through 6 metres. Neither geometry owns “short skip.” Their usefulness follows the three-dimensional installed pattern, polarization toward the required bearing and elevation, feed-system loss, common-mode control and the ionospheric mode available at that time.
The comparison that matters is not “triangle beats V” or the reverse. It is whether the installed antenna puts enough accepted power into the elevation, azimuth and polarization that the current path can use. I start with the route and propagation mode, then compare the complete antenna systems at the same accepted power.
The central rule: shape, height, ground and current distribution work together. A different outline can change the pattern, but a convenient SWR or a promising antenna name cannot identify the useful field toward an inter-European path.
“Short Skip” Is Not One Distance or One Elevation Angle
Regional European contacts can arrive by regular E- or F-region propagation, sporadic E, multiple ionospheric modes and, especially toward 6 metres, mechanisms that do not fit one simple HF sky-wave sketch. Frequency availability, time, season, solar and geomagnetic conditions, path direction and ionospheric structure all matter.
A flat-Earth triangle using one assumed virtual height can illustrate that a shorter one-hop ground distance generally asks for a steeper launch angle. It is not a prediction. Earth curvature, refraction, ionospheric gradients, critical frequency, absorption and mode selection change the result. Recommendation ITU-R P.533 treats HF circuit performance as a path and time problem; Recommendation ITU-R P.534 treats sporadic-E field strength statistically rather than as one fixed reflection height.
Define the path before choosing the antenna. A station seeking 300 km daytime coverage on 20 metres, 1,000 km F-region contacts on 15 metres and summer sporadic-E contacts on 6 metres does not have one common elevation target.
Compare Full Three-Dimensional Patterns
An azimuth cut at one elevation and an elevation cut at one bearing are slices through a three-dimensional pattern. Either can hide lobes and nulls that decide the contact. For each band, compare realised gain over the range of bearings and elevation angles that the intended paths require.
Use realised gain when feed mismatch is part of the comparison, or accepted-power gain when the matching systems are being evaluated separately. State which one is shown. A normalized pattern reveals shape but hides absolute loss; an unnormalized field comparison can reveal the complete system only when source and accepted power are controlled.
Height in wavelengths is important because the direct and ground-reflected fields combine differently as height changes. It is not the only lever. Ground conductivity and permittivity, slope, nearby structures, the feedline and the antenna's own current distribution can move or fill nulls and change the elevation of useful lobes.
The Vertex-Fed Delta Loop
Consider a roughly full-wave triangular loop in a vertical plane, pointing down and fed at the lower vertex. That describes the geometry without assigning it a fixed impedance or pattern.
Current magnitude and phase vary around the loop. The sloping sides and upper horizontal span contribute different field components. Consequently, polarization is direction- and elevation-dependent; it is not accurate to label the complete radiation as purely vertical or purely horizontal. Moving the feed vertex, changing the triangle proportions or bringing one side closer to ground changes the current distribution.
A single vertical-plane loop is not automatically omnidirectional. It often has strong broadside regions, but the actual azimuth pattern depends on shape, electrical perimeter, height, feed location, ground and surrounding conductors. At low height, ground interaction may soften some nulls while deepening or moving others.
The full-wave label does not fix the feed impedance. Conductor diameter, exact perimeter, triangle angles, height profile, ground, feedpoint location and environmental coupling all matter. Measure R+jX in the installed position before selecting the matching network.
The V-Dipole
A V-dipole is a centre-fed dipole whose legs meet at an included angle rather than forming a straight line. The V reduces horizontal span and can change terminal impedance, polarization and the azimuth pattern. A nominal 120-degree angle is a practical starting geometry, not a universal 50 Ω recipe.
The feedpoint may be high while the leg ends are much lower, so one number called “antenna height” is incomplete. Record the centre height, end heights, included angle, orientation and surrounding terrain. These variables determine the installed current distribution and ground coupling.
A shallow V may remain predominantly horizontally polarized in many directions; a steeper V can add stronger vertical components. The azimuth pattern can have broader or partially filled nulls compared with a straight dipole, but it is not automatically circular or omnidirectional.
As with the loop, operation on bands where the conductor is several wavelengths long creates additional lobes and nulls. That does not inherently make the antenna inefficient. It makes coverage more directional and more sensitive to geometry.
Polarization Is Part of the Pattern, Not a Permanent Label
The transmitted field at any direction can be resolved into orthogonal polarization components. A vertical-plane loop can launch both components, and a V-dipole can too. Their ratio varies with bearing and elevation.
The ionosphere can rotate and mix polarization, and multipath can deliver several components at the receiver. That does not make launch polarization irrelevant. Polarization mismatch, fading and receiving-antenna response can still affect the link at a particular moment.
Compare co-polar and cross-polar realised gain toward the paths of interest rather than assigning each antenna one polarization word. If diversity matters, measure the correlation and received SNR of the actual channels.
Height Changes the Two Geometries Differently
Raising a V-dipole centre while leaving its ends fixed changes the leg angle and average height. Raising the top support of a point-down delta loop while leaving the lower vertex fixed changes its shape and average height. Those are not pure translations, so their pattern and impedance changes cannot be attributed to height alone.
For a controlled comparison, either scale and move the complete geometry or state exactly which points moved. On every band, record all conductor coordinates relative to local ground. A 6 m support is about 0.3 wavelength on 20 metres but roughly one wavelength on 6 metres; the same installation therefore produces very different ground interference and lobing across the range.
Do not turn fractions such as 0.25λ or 0.5λ into guaranteed regional-coverage angles. They are useful model inputs. The installed three-dimensional result, over the actual ground and terrain, is the evidence.
Matching and Loss Can Reverse an Apparent Winner
The two antennas may present different complex loads. One may connect conveniently to 50 Ω coax while the other needs an impedance transformer or tuner. That difference is part of the complete system, but a lower SWR is not the same as higher radiated power.
Measure or bound feedline, transformer, tuner and conductor loss. A tuner at the station can protect the transmitter while high SWR remains on the upstream coax. A transformer must be characterised with representative complex loads, not only a nominal resistor. Voltage, current, temperature, waveform and duty cycle set the operating boundary.
For an A/B comparison, equal transmitter output is not enough if the feed systems accept or dissipate different power. Compare at equal accepted power, and state whether loss before the antenna terminals is included.
Both Antennas Need a Controlled Feedline Boundary
An ideal symmetric loop or dipole is a balanced load. Coax is an unbalanced transmission line. Installation asymmetry, unequal coupling to ground, support hardware and feedline routing can drive current on the coax exterior and distort both impedance and pattern.
A current balun or common-mode choke can be appropriate when its differential transfer, complex common-mode impedance and powered limits suit the installed load. Its name does not prove balance. Measure exterior current on both sides and over every band being compared.
Route the feedline as part of the controlled geometry. Taking it away from the feedpoint in a direction that minimizes immediate coupling can help, but “right angles” are not a universal cure when the wire shape, support and field vary. The current measurement decides.
A Closed Loop Is Not Automatically Quieter
Receive noise at the radio is the sum of external noise accepted through the antenna pattern and polarization, losses that change antenna temperature, common-mode pickup on the feedline, receiver noise and local coupling. A closed conductor does not guarantee a quieter receiver.
A loop can be quieter than a V-dipole in one installation because its nulls, polarization or placement reject a local source, or because its feed system carries less common-mode current. Reverse the orientation or noise-source position and the ranking can reverse. Compare signal-to-noise ratio, not S-meter noise alone.
Practical Comparison Without a Universal Winner
| Question | Vertex-fed point-down delta loop | V-dipole |
|---|---|---|
| Mechanical footprint | Closed triangular perimeter; lower feed access can be convenient, but the loop needs its full vertical and lateral shape supported | Open two-leg span; often one high centre support plus two end anchors |
| Installed polarization | Direction-dependent mixture from horizontal and sloping/vertical current components | Often predominantly horizontal for a shallow V, with increasing mixed components as the legs steepen |
| Azimuth coverage | Usually has favoured broadside regions and nulls whose depth/location depend on height, shape and ground | Dipole-family broadside regions with nulls modified by V angle, end height and surroundings |
| Elevation coverage | Set by the full current distribution, height profile, ground and terrain | Set by the full current distribution, centre/end heights, ground and terrain |
| Feed impedance | Installation-specific R+jX; do not assume one loop value or fixed transformer | Installation-specific R+jX; the included angle can move it but does not guarantee 50 Ω |
| Receive noise | Potentially favourable if pattern, polarization, placement and common-mode control reject the dominant source | Potentially just as favourable under the same criteria |
| Multiband use | Higher electrical perimeter produces additional lobes, nulls and changing feed impedance | Higher electrical length produces additional lobes, nulls and changing feed impedance |
Build the Test Around the Intended Routes
- Define the paths: list bearings, ground distances, frequencies, seasons and likely regular-layer or sporadic-E modes.
- Document both geometries: record every conductor coordinate, feedpoint, support, ground profile, feedline and nearby structure.
- Model in three dimensions: use measured or bounded ground parameters and inspect realised gain by bearing, elevation and polarization on each band.
- Calibrate the feed systems: record R+jX at declared planes and measure or bound transformer, tuner and feedline loss.
- Map exterior current: verify that neither coax becomes an uncontrolled extra radiator.
- Compare equal accepted power: use rapid A/B/B/A switching or simultaneous receivers with fixed gain, bandwidth and detection.
- Log propagation context: record time, frequency, path, ionospheric indicators and fading so the comparison does not confuse a changing channel with antenna performance.
One day of contacts can reveal a useful operating preference, but it does not establish the complete pattern. Repeat across paths and conditions, restore the baseline after every swap and publish the uncertainty with the result.
Safety Is Part of the Installation Choice
Both antennas can place high RF voltage at conductor ends and matching components. Maintain clearance from people, animals, buildings, vegetation and power conductors; provide mechanical strain relief; prevent access during transmission; and de-energise before adjustment.
Assess RF exposure using the actual power, duty cycle, antenna geometry and accessible locations. A lower feedpoint may simplify adjustment, but it can also make an RF-voltage region easier to touch.
Bottom line: choose the delta loop when its supported geometry and measured pattern fit the target routes and site. Choose the V-dipole when its simpler support and measured pattern do. For inter-European work, the winner can change by band, bearing, height and ionospheric mode—and the feed system can erase a small pattern advantage if its loss or common-mode current is ignored.
Primary technical references
- Recommendation ITU-R P.533 — Method for the prediction of the performance of HF circuits
- Recommendation ITU-R P.534-6 — Method for calculating sporadic-E field strength
- Recommendation ITU-R BS.705-2 — HF transmitting and receiving antenna characteristics and diagrams
- IEEE Std 145-2025 — Standard for Definitions of Terms for Antennas
- Numerical Electromagnetics Code — NEC-2 User's Guide, Part III
- Roy Lewallen, W7EL — Baluns: What They Do and How They Do It
- ICNIRP — Guidelines for limiting exposure to electromagnetic fields, 100 kHz to 300 GHz
Mini-FAQ
- Is a point-down delta loop omnidirectional? No. Its installed azimuth pattern has favoured directions and nulls set by current distribution, loop plane, shape, height, ground and nearby structures.
- Does a delta loop guarantee quieter reception than a V-dipole? No. Compare received SNR with pattern, polarization, placement, feedline loss and common-mode pickup controlled.
- Which antenna is better for inter-European contacts? Neither universally. The useful result depends on band, route, ionospheric mode, installed three-dimensional pattern, loss and common-mode control.
- Does a 120-degree V-dipole always present 50 ohms? No. Included angle, conductor, height, end position, ground and surroundings determine its installed R+jX.
- Do both antennas always need the same choke? No. Characterise the selected current balun or choke for each installed load, frequency and power, then verify exterior-coax current.
- How should they be compared on air? Use equal accepted power, rapid A/B/B/A switching or simultaneous receivers, fixed settings, restored baselines and propagation context across several paths.