Yagi-Uda Antenna Design: A Basic Tutorial for Engineers

This section covers practical implementation of impedance matching networks. After determining gamma_s and gamma_l, one converts these to impedances and designs matching networks. At 1.5 GHz, discrete LC components require careful consideration of stray capacitance and component parasitics. Microstrip transmission line matching is often preferred, using characteristic impedance lines (commonly 50 ohms) and stubs. The design includes bias integration considerations: blocking capacitors must be placed to prevent DC from shorting through matching networks. Capacitor values are selected based on impedance criteria (e.g., 1 nF at 1.5 GHz provides -j0.1 ohms, negligible compared to 50 ohms). The solution is verified by checking impedances looking into each port.

When direct connection cannot achieve maximum power transfer, impedance matching networks transform load impedance using lossless L and C components. Common topologies include gamma networks (L-section), pi networks, and T networks. Gamma networks transform impedance when R' > R using XL = Q × R and XC = R' / Q. Pi networks (cascaded gamma networks) provide better frequency selectivity: XC1 = R × Q, XC2 = R' × Q, XL = (R × R') / (Q × (R + R')). T networks provide similar performance. Parallel resonant circuits (LC tanks) can also be used for impedance matching, with transformed impedance R' = n² × R. Higher Q means narrower bandwidth and better frequency selectivity. The design procedure involves: determining required impedance transformation ratio, choosing suitable topology, selecting Q based on bandwidth requirements, calculating component values, and verifying through simulation. Approximation formulas are valid only when Q is high and load current is negligible compared to circulating currents. The Smith chart is a graphical tool for analyzing impedance matching and transmission line problems in microwave engineering.

Quarter-wave transformers use transmission line sections exactly one-quarter wavelength long to transform impedance between two points. They work by creating reflections that cancel out when summed, making the load appear perfectly matched. The characteristic impedance is the geometric mean of the two impedances: Z_transformer = √(Z₁ × Z₂). For matching a 100 ohm loop antenna to 50 ohm coax, use 70.7 ohm coax cut to quarter-wavelength. This provides single-frequency matching and is useful for surplus hardline cables. Feed line matching at the antenna maximizes power transfer with minimal loss, preventing signal bouncing and high voltages/currents. The delta match moves feed points on dipoles to find better impedance matches, originally developed for matching 600 ohm open-wire lines to dipoles with impedances around 70-90 ohms at center. The gamma match is half of a T-match configuration, consisting of an adjustable capacitor and a shorted transmission line stub. It transforms the low impedance of driven elements (around 1000 ohms) to match 50 ohm coaxial cable. The gamma match connects the feed line shield directly to an electrically neutral point on the antenna, providing mechanical stability and preventing insect nests. The hairpin match (beta match) uses a coil across the driven element, combined with the element's inherent capacitance, forming an L-network for impedance transformation. Both gamma and hairpin matches are single-band solutions effective only at one frequency.

The Gamma Match is an impedance matching device that connects unbalanced coaxial cables to balanced dipole antennas, preventing common mode currents (I3) that cause interference. It consists of a coaxial section forming a variable capacitor, with the center conductor inserted into a tube and the outer shield shorted to the dipole. Four key parameters govern design: (1) Dipole length = 142.5 / frequency (MHz), (2) Gamma Match capacitance = 8 picofarads per wavelength, (3) Gamma Match length = 10% of dipole length, (4) Separation = dipole length / 70, and (5) Gamma Match diameter = dipole diameter / 3. The wavelength is λ = 300 / frequency (MHz). For 28.5 MHz, this yields a 5-meter dipole, 82.2 picofarads capacitance, 50 cm Gamma Match length, and 7.14 cm separation. Standard coaxial cables (RG-58, RG-213) provide 110-120 picofarads per meter, so 50-60 cm provides necessary capacitance.

Impedance matching maximizes power transfer and minimizes signal reflection when connecting systems; baluns (balanced-to-unbalanced transformers) enable connections between balanced feedlines (like twin lead with opposing currents) and unbalanced feedlines (like coaxial cable), with common implementations including the Bazooka balun (using a λ/4 metal sleeve), λ/4 coaxial balun, and compact designs that increase outer shield impedance to minimize unwanted current components.

This section covers advanced antenna configurations and stacking techniques for enhanced performance. It begins with cage dipole antennas that replace single wires with multiple parallel conductors, creating larger effective diameters that increase bandwidth and allow full-band coverage on HF bands like 20 meters with SWR below 2:0. The section explains how stacking identical antennas increases gain by approximately 3 dB per doubling of power, with vertically stacked Yagis providing 3 dB more gain than singles and four-element stacks providing 6 dB. Near-Vertical Incident Skywave (NVIS) antennas radiate nearly vertically upward for short-skip communications during daylight hours, particularly valuable in mountainous regions. The section covers log-periodic antennas designed by Harold Beverage for AT&T long-distance service, featuring logarithmic element progression for wideband operation. Finally, it explains the Beverage antenna as a traveling-wave receiving antenna with negative gain but high directionality and noise immunity for weak signal reception on 40, 80, and 160 meter bands.

A Log-Periodic Dipole Array (LPDA) is a broadband unidirectional antenna consisting of multiple half-wavelength dipoles arranged in a logarithmic progression, with the longest dipole at the back and the shortest at the front; it achieves continuous broadband coverage with a frequency range ratio of approximately 4:1 through a phasing harness that twists 180° between each antenna element, creating directional characteristics similar to a Yagi antenna while maintaining broadbanded operation.

LPDA (Log-Periodic Dipole Array) is a high-gain wideband directional antenna operating from 400 MHz to 8 GHz. Unlike Yagi antennas with single driven elements, LPDA has multiple connected dipole elements where each pair forms a dipole. The design achieves wide bandwidth because each dipole has its own resonant frequency - larger dipoles handle low frequencies while smaller ones handle high frequencies. Typical gain ranges from 6-10 dBi. A design issue is the SMA connector placement at the high-frequency end, where cables interfere with radiation patterns. LPDA is useful for Wi-Fi applications and directional finding.

This lecture explains how Yagi-Uda antennas can be modified to achieve broader bandwidth through quasi-Yagi configurations using high-dielectric substrates, specialized feed networks (including power dividers with λ/2 phase shift lines and CPW feeds), and optimized parasitic element dimensions, sacrificing some gain for bandwidth improvements of 40-48% at X-band; it then introduces log-periodic dipole arrays, which inherently provide ultra-wideband operation (1:10 frequency ratio) by arranging multiple dipoles with geometrically scaled dimensions (length and spacing following a constant ratio τ between 0.7-0.95) and 180° phase shift between adjacent elements, where the design angle α and space factor σ determine the antenna's directional characteristics.

The Log-Periodic Dipole Array (LPDA) provides wide bandwidth unlike Yagi-Uda. All dimensions vary logarithmically with frequency, with the active resonant element shifting toward smaller dipoles as frequency increases. Key parameters are scaling factor τ = R(n+1)/R(n) = L(n+1)/L(n) < 1 and space factor σ = D(n)/L(n). Design curves show constant directivity for various τ and σ values. For 6.5 dB gain at 54-216 MHz, τ≈0.82 and σ≈0.15 are optimal. The design procedure involves finding the longest dipole (0.5λ at lowest frequency), calculating intermediate lengths using τ, and determining element spacing using D(n)=2σL(n). Adding a reflector behind the longest dipole improves low-frequency gain performance.

Construction loads differ fundamentally from design loads due to shorter exposure times and different loading conditions. Two distinct wind load periods govern analysis: the continuously monitored work period (35 mph three-second gust) applies during active work hours, while the construction period (75-80% of code-specified wind speed) applies to all members remaining after work concludes. Roof installation before wall siding is critical as siding creates windscreens increasing differential loading on frames. Critical structural elements include primary frames, roof and wall bracing, strut purlins, heav struts, and flange braces. Flange braces are particularly vital as they control unbraced lengths of columns and roof beams to resist wind-induced buckling. Recommended minimum requirements include at least one row of purlin bracing per bay and brace bays every five roof bays.

Wind loading during construction follows ASCE 37 Chapter 6 procedures with reduction factors: 0.75 for construction under six weeks, and 15% reduction on elements beyond the first three. Open frameworks during construction have significantly higher drag coefficients (1.6-2.0) than enclosed final buildings, resulting in larger design forces requiring temporary measures. Stability analysis involves creating models for each construction sequence using ASCE 37 to determine required loads, considering material staging, snow/ice loading, and matching analysis to actual erection sequences.

Structure during erection may experience less wind force than the final enclosed building, but total base shear due to wind during construction could be similar to or higher than wind on the final enclosed building. The shape factor CF accounts for square-sided members collecting more wind force than projected area indicates. Shielding is limited by design codes—even with multiple floor beams lined up, ASCE only allows 20% reduction on any beam past the third. Erection engineering involves three types of tasks: structure stability (analysis of partially completed structure at each stage), element stability (analysis of major element installation), and staged construction analysis (accumulation of dead load stresses).

ASCE 37 Chapter 6 provides wind load guidance for construction, applying reduction factors since construction periods are shorter than 50-year design lives. For example, a Michigan structure with 115 mph design wind speed might use a 0.8 factor for six-week construction, yielding 92 mph design wind speed. Two methodologies exist: treating each structural element individually (solid freestanding walls/signs approach) versus analyzing the entire structure as a lattice framework. Shielding effects reduce loading on downstream elements—first three elements get 100%, subsequent elements get 85%. The German standard DIN 1055 provides stacking factor equations for multiple parallel elements.

Sag is affected by environmental conditions beyond conductor weight and span length. Ice loading increases effective weight: Wi = density of ice × πT(D + T), where T is ice thickness and D is conductor diameter. Wind force acts horizontally: Ww = wind pressure × (D + 2T). The total weight per unit length is the vector sum: WT = √[(W + Wi)² + Ww²]. These factors are critical in winter and stormy conditions, significantly increasing sag and requiring careful engineering calculations.