YO Yagi Optimizer — Ham Radio Antenna Design Guide
YO (Yagi Optimizer) is a Windows program written by Brian Beezley K6STI that remains one of the most practical tools for designing multi-element Yagi antennas. Unlike general-purpose NEC2 engines that require you to specify every element position manually, YO is purpose-built for Yagi design — it takes your target specifications (gain, F/B, SWR, number of elements, boom length) and optimises the element dimensions automatically using a genetic algorithm. The results are Yagi designs that genuinely perform as predicted and export cleanly to NEC2 format for further analysis.
YO occupies a specific and useful niche in the antenna design toolkit. General-purpose NEC2 tools like MMANA-GAL and 4NEC2 can model and optimise Yagi antennas, but they require you to specify the initial geometry and understand what parameter adjustments move the design toward your target. YO reverses this — you specify what you want the Yagi to achieve and let the optimiser find the geometry that achieves it. This makes YO particularly valuable for exploring the limits of what a given number of elements can achieve over a given boom length.
When YO excels
Designing a Yagi from scratch to a performance specification. Exploring gain vs. boom-length trade-offs. Finding designs that maximise F/B ratio for a given element count. Generating starting-point designs for further refinement in MMANA or 4NEC2. Verifying published designs against independent optimisation.
YO's limitations
Designs Yagis only — not loops, dipoles, phased arrays, or other antenna types. Uses a simplified NEC2 implementation without full Sommerfeld-Norton ground modelling. Boom correction is approximate. For final design verification, export to MMANA-GAL or 4NEC2 with accurate ground modelling.
Alternatives
4NEC2 with its optimiser handles Yagis and all other NEC2 geometries. MMANA-GAL's optimiser is well-suited to wire antennas. NEC-Win Pro adds stepped-diameter correction. For a first Yagi design, YO's Yagi-specific interface is the fastest path to a working design.
YO is available as a free download from several mirror sites — search for "K6STI YO Yagi Optimizer download" or "YO6.0.exe". The program is a standalone Windows executable requiring no installation — copy the .exe file to a folder and run it. On modern 64-bit Windows systems, YO runs correctly without any compatibility settings as it is a 16-bit or early 32-bit application that Windows handles through WOW64. On Linux, it runs under Wine with the same simple approach: place the exe in a folder and run with wine yo.exe.
YO does not require internet access, does not install registry entries, and leaves no footprint beyond its own folder. The program includes a built-in help file (press F1) that documents every parameter and the optimisation algorithm. Read the help file before your first optimisation run — understanding what YO's merit function is actually optimising prevents misinterpreting the results.
YO's Input Parameters — What You SpecifyYO requires you to specify the following design inputs before optimisation. Each parameter constrains the search space and influences which designs the optimiser can find:
The constraints — boom length, element diameter, SWR limit, and frequency range — define the feasible design space. YO's optimiser then searches within that space for the combination of element lengths and spacings that maximises the chosen objective (gain, F/B, or a weighted combination). The key design trade-offs that these parameters capture are:
- More elements on the same boom: More elements allow higher gain but require closer spacing, which reduces bandwidth and increases the difficulty of achieving good SWR across the full frequency range
- Longer boom with the same number of elements: Wider element spacing improves bandwidth and makes matching easier but may not maximise gain per unit boom length
- Larger element diameter: Thicker elements broaden the SWR bandwidth and reduce sensitivity to element length errors, but increase wind loading and cost
- Tighter SWR limit: A tighter SWR constraint (say, 1.2:1 instead of 1.5:1) forces the optimiser to sacrifice some gain in exchange for better impedance matching across the band
Yagi Design Target Estimator
Estimates realistic performance targets for a Yagi of a given element count and boom length, based on established DL6WU / NEC2-optimised design data. Use these as YO input targets.
Launch YO and select File → New to start a fresh design. Enter the design frequency, number of elements, and element diameter. For a uniform-diameter Yagi (all elements the same OD), enter the diameter once. For a stepped-diameter design, YO requires the stepped-diameter correction — enter the equivalent uniform diameter using the electrical diameter calculation, or use a post-optimisation correction step.
Enter the maximum boom length you can physically install. Set the SWR limit based on your system — 1.3:1 gives good matching headroom; 1.5:1 allows more gain at the expense of matching. Set the frequency range to span your entire operating range — for the 20 m band, use 14.000–14.350 MHz. Narrower ranges allow higher gain; wider ranges sacrifice gain for bandwidth. YO will find designs that stay within SWR limit across the full specified range.
YO offers several objective options: Maximum Gain, Maximum F/B ratio, and Minimum SWR. For most Yagi designs, Maximum Gain with F/B and SWR as constraints is the most useful starting point. YO's optimiser uses a genetic algorithm (called "Genetic Optimizer" in the menu) that maintains a population of designs and evolves them toward the objective. The classic "steepest descent" option is faster but more likely to find local optima — use the genetic algorithm for a more thorough search.
Click Optimize → Genetic Optimizer. YO displays a real-time plot of the best design found so far, updating as better designs are discovered. The optimisation can run for minutes to hours depending on element count and computer speed. Let it run for at least 10–20 minutes before evaluating the result — early solutions are often locally optimal but not globally optimal. If the merit function (gain in dB) is still improving when you check back, let it continue. If it has plateaued for more than 5 minutes, the search has likely converged.
After optimisation, YO displays the best design's element table (lengths and positions), gain vs. frequency plot, F/B ratio vs. frequency plot, and SWR vs. frequency plot. Verify that SWR stays within your specified limit across the full operating range — YO's constraint enforcement can occasionally produce designs where SWR just barely meets the constraint at one frequency while being higher than specified at another. If so, tighten the SWR limit and re-run.
From File → Export, save the design as a NEC2 format file (.nec). Import this file into MMANA-GAL, 4NEC2, or EZNEC and re-run the simulation with a realistic ground model (σ and εr for your site) and at your actual antenna installation height. This step catches the differences between YO's free-space or simplified ground model and the real-world environment. The NEC2 verification often shows that the boom correction needs adjustment and that the gain is slightly different over real ground — update the design accordingly before building.
YO's output contains all the information needed to build the Yagi. Understanding each output field is essential before moving to construction:
The feed impedance of 22.5 Ω is characteristic of a well-optimised multi-element Yagi — mutual coupling from the parasitic elements pulls the feed point resistance well below 50 Ω. This is not a problem but requires a matching network. Common solutions include a gamma match, beta (hairpin) match, or a 4:1 or 2:1 balun transformer depending on the exact impedance. YO reports the impedance to 50 Ω reference as SWR, showing whether any matching is needed and how much transformation is required.
Boom Effect Correction in YOYO includes a boom correction algorithm to account for the detuning effect of a conductive boom on element resonant frequencies. When elements pass through or mount on a conductive aluminium boom, the boom acts as a short section of transmission line at each element crossing, electrically lengthening each element. YO's correction shortens each element in the model to compensate.
The boom correction magnitude depends on the ratio of boom diameter to element diameter and the element spacing. YO applies its own formula based on the work of DJ9BV. If you are using a boom diameter significantly different from common values, or if elements are isolated from the boom (using insulating plates), you may need to disable or adjust the boom correction. The key rule: elements passing directly through the boom (making electrical contact) require boom correction. Elements mounted on isolated plates above or below the boom require no boom correction — but their effective position on the boom must be adjusted for the distance from the boom centreline to the element centreline.
YO vs. Other Yagi Design Tools| Tool | Optimisation | Yagi-specific features | Ground modelling | Best use |
|---|---|---|---|---|
| YO (K6STI) | Genetic algorithm | Excellent — boom correction, stepped diameter, matching | Free space or simplified | Initial Yagi design from specifications |
| MMANA-GAL | General optimiser | Good — any geometry | MININEC | Refinement and verification of any antenna |
| 4NEC2 | General NEC2 optimiser | Full NEC2 control | Sommerfeld-Norton available | High-accuracy final verification |
| EZNEC Pro | General NEC2/NEC4 | Stepped diameter correction | NEC4 ground models | Professional final design |
| AO (K6STI) | Antenna Optimizer — general | Any antenna type | NEC2 | General purpose follow-on to YO |
This worked example shows the complete YO workflow for a 5-element 20 m Yagi with a 7 m boom, targeting DX operation with good F/B ratio:
Target: 5-element Yagi for 14.000–14.350 MHz, maximum gain with SWR ≤ 1.3:1 from a single matching network, boom ≤ 7 m, element OD = 25 mm aluminium tubing, boom OD = 50 mm. Expected gain from reference table: approximately 12.5 dBi, F/B approximately 25 dB.
Start genetic optimiser with these constraints. After 20 minutes the optimiser converges on a design with 12.42 dBi gain, 24.8 dB F/B, and feed impedance 22.5 + j3 Ω. SWR to 50 Ω is 2.23:1 — needs a gamma or hairpin match to provide the approximately 2.2:1 impedance step-up.
Export the NEC2 file and import into MMANA-GAL. Run with real ground (σ = 0.005, εr = 13, typical suburban soil) at 12 m boom height. The gain over real ground shows 16.8 dBi peak (including ground reflection) at 12° take-off angle — excellent DX performance. Feed point impedance 21.8 + j5 Ω — close to YO's free-space result.
Use the gamma match calculator (from the antenna impedance guide) with source = 22 Ω, load = 50 Ω, frequency = 14.175 MHz. This gives the gamma rod length, spacing, and series capacitor value needed to transform 22 Ω to 50 Ω. Add these to the build specification alongside the YO-generated element table.
Is YO still useful or are there better tools now?
YO remains genuinely useful because of its Yagi-specific interface and fast genetic algorithm for Yagi design. Its strength is the speed with which an experienced operator can go from a performance specification to a working element table. Modern tools like 4NEC2 are more powerful and general, but their interfaces require more knowledge to drive efficiently for Yagi design. Many experienced Yagi designers use YO for initial design and 4NEC2 for final verification.
How many optimisation iterations does YO need?
A 3-element Yagi typically converges in 2–5 minutes. A 7-element Yagi may take 20–60 minutes on a modern PC. A 15-element Yagi can take several hours. The genetic algorithm makes measurable progress for much of the run time, so longer runs consistently produce better results — let it run as long as you can. The progress plot shows when the merit function has truly plateaued.
Should I trust YO's boom correction?
YO's boom correction is based on DJ9BV's empirical formula and works well for typical boom-to-element diameter ratios (boom diameter 1.5–3× element diameter). For very large booms relative to element diameter (boom OD > 4× element OD) or very small booms (boom OD < element OD), the correction becomes less accurate. Verify boom correction accuracy by comparing YO-designed elements against NEC2 verification with the boom explicitly modelled as additional wire segments.
Can YO design a stepped-diameter Yagi?
YO has limited stepped-diameter correction — it applies the Leeson correction for a two-section element (inner and outer diameters). For more complex stepped-diameter elements (three or more sections, as commonly found in commercial aluminium-tube Yagis), 4NEC2 or EZNEC Pro with their more sophisticated stepped-diameter algorithms provide more accurate results. Use YO for uniform-diameter element design and export to 4NEC2 for stepped-diameter refinement.
What does the YO merit function actually maximise?
YO's genetic algorithm maximises a weighted merit function that combines forward gain and penalises designs that violate the SWR constraint across the specified frequency range, designs that violate the boom length constraint, and designs with very poor F/B ratio (if F/B optimisation is selected). The merit function weights can be adjusted in the advanced settings — increasing the SWR penalty makes the optimiser prioritise matching over raw gain. Understanding this allows you to tune YO's behaviour for different design priorities.
Why does my YO design perform differently from what YO predicts?
YO uses a free-space model without detailed ground interaction. Real antennas over real ground at finite height behave differently — the ground reflection adds gain at low angles (typically +5–8 dBi over free-space gain at optimum height) and the feed point impedance shifts slightly. Additionally, construction tolerances, boom corrections, and element sag all deviate from the ideal model. Export to MMANA-GAL with real ground parameters and re-verify to close the gap between YO prediction and real-world performance.