Log-Periodic Antenna Theory: τ/σ Basics, Active Region, Design Steps & Compliance for EMC and Broadband Links

Introduction

You need one antenna to cover many bands—without constantly swapping hardware or compromising test repeatability. A log-periodic dipole array (LPDA) does exactly that: stable patterns across a wide frequency span, predictable input impedance, and lab-friendly calibration behavior. In this practical guide I explain τ (tau), σ (sigma), apex angle, and the active region, show how to size and model an LPDA, and map the theory to EMC labs, spectrum monitoring, and product teams. You’ll also get a procurement checklist, compliance notes (FCC/ETSI), and a few ready-to-ship accessories so you can move from theory to deployment with confidence.

1) What an LPDA Is—and Why Engineers Still Choose It

An LPDA is a multi-element directional antenna whose element lengths and spacings follow a logarithmic progression. Adjacent dipoles are fed with alternating phase via a cross-over line, which produces a broadly frequency-independent impedance and radiation pattern. In practice, only a subset of elements radiates efficiently at any given frequency; this sliding subset is the active region and it moves along the boom as frequency changes. That’s the key to wideband performance with controlled directivity.

Compared with narrowband options like Yagi-Uda arrays, LPDAs trade peak gain for band coverage and pattern stability, which is exactly why you see them in EMC labs, test ranges, broadband monitoring, and wideband links.

2) The Theory You Actually Need (Carrel, τ/σ, Apex, Active Region)

2.1 Carrel’s Framework in Plain English

Carrel’s classic 1961 analysis provided the engineering recipe still used today: pick band edges, choose τ (element length ratio) and σ (spacing factor), set apex angle, then compute element count and boom length to achieve the target bandwidth and pattern. He also formalized the active region, relating array size to useful bandwidth and directivity. If you remember one name in LPDA design, make it Carrel.

What the parameters do (high level):

Why it matters: choose τ too small and you need many elements; too large and impedance/pattern ripple grows. σ too small increases coupling and pattern distortions; too large and the array loses the mutual coupling it needs for stable directivity. Carrel’s charts and procedure keep you in safe territory.

2.2 The “Frequency-Independent” Promise—With Real-World Limits

LPDAs are often described as frequency-independent over a design band: input impedance, gain, and pattern repeat roughly with the logarithm of frequency. In practice, SWR ripple and some gain swing are expected; mechanical constraints (boom length, element diameter, cross-over line) set the practical edges. The active region concept explains why patterns remain usable across decades of frequency.

2.3 What Actually Radiates? (Active Region Intuition)

At each frequency, elements whose electrical length is near half-wavelength dominate; upstream (longer) and downstream (shorter) elements mostly store and guide energy. As frequency increases, the active region slides toward the shorter elements. If your layout or housing interferes with that region, your pattern suffers—so leave keep-out space where the active region lives at your most critical frequencies.

3) Design & Modeling: From Spec to Geometry

3.1 Step-by-Step Sizing (Engineer-to-Engineer)

1) Lock the band: determine the required band ratio and center.

2) Pick a starting τ and σ: common starting points are τ ≈ 0.8–0.9, σ ≈ 0.05–0.1 for rod-element LPDAs; then adjust for size and performance targets.

3) Choose apex angle to fit mechanical constraints while keeping acceptable coupling.

4) Compute element count & boom length using Carrel’s relations; budget space for cross-over feed and a current-balun at the input.

5) Simulate (FEM/FDTD): sweep SWR, realized gain, and HPBW across the band; verify active region coverage and confirm H/V plane pattern stability.

6) Prototype & measure: correlate chamber results to simulation; if you’re an EMC lab, obtain a traceable calibration for your production articles.

3.2 Printed/Planar LPDAs (When PCB Wins)

Printed LPDAs (PLPDAs) trade a bit of peak gain for integration, repeatability, and compactness. The open literature shows PLPDAs spanning sub-GHz to multi-GHz with useful realized gain and stable patterns; techniques include tapered microstrip feeds, trapezoidal/triangular elements, and fractalized/Koch edges for miniaturization. Use these when enclosure constraints, reproducibility, or a single-board test antenna are priorities.

3.3 Feed & Balance

The classic LPDA uses a two-wire balanced line that crosses over between elements to alternate phase. Even with a nominally 50-Ω input, add a 1:1 current balun at the feed to suppress common-mode currents on the coax—essential for clean patterns and lab measurements. (You’ll see this practice in many lab-grade antennas and NIST papers on calibration setups.)

4) How an LPDA Compares (So You Pick the Right Tool)

4.1 Side-by-Side Reality Check

Yagis win fixed-channel gain; horns/dishes win ultimate gain; LPDAs win consistency over wide frequency. That’s why LPDAs anchor EMC chambers, broadband test sites, and spectrum-monitoring kits.

4.2 Decision Helper (5 Quick Questions)

1) Is your band ratio > 3:1 and you need one antenna? → LPDA.

2) Is the link single-channel and very long? → Yagi or dish beats LPDA.

3) Is space tight and you need a flat back? → Printed LPDA.

4) Do you need traceable calibration for EMC? → LPDA with NIST-traceable certificate.

5) Will the antenna be used outdoors with long coax runs? → Plan for low-loss cable and waterproof N-type interfaces (see BOM below).

5) Measurement, Calibration & Uncertainty (What Labs Expect)

• 

  5.1 What “Traceable” Looks Like

  A proper EMC/measurement LPDA ships (or can be shipped) with a traceable calibration: method description (OATS or anechoic chamber), antenna factor vs frequency, uncertainty budget, and traceability chain back to NIST/NBS standards. For example, NBS Technical Note 1309 and later NIST Technical Note 2187 describe procedures and uncertainty methods for calibrating antennas used as standards. Use these documents as your reference when auditing a lab or vendor.

5.2 Plots You Should Demand

• H- and V-plane patterns at several anchor frequencies across your band
• Gain vs frequency sweeps with averaging method stated
• SWR/return loss across the declared band (e.g., ≤ 2:1 typical targets)
• Polarization purity notes if required by your test method

If a vendor cannot provide these within a week, that’s a red flag for lab work.

5.3 Arrays and Special Geometries

Government and research reports have analyzed LPDA arrays for enhanced directivity or uniform chamber fields. If you’re building a custom field generator (e.g., EMC susceptibility), these are valuable starting points.

6) International Differences You Can’t Ignore (EIRP & Power Back-Off)

• 

  LPDAs are often used with transmitters. Antenna gain changes your EIRP, and EIRP—not just conducted power—is regulated. Two standard reference points:

• United States (FCC Part 15): Many Part 15 device limits assume a reference antenna gain (commonly 6 dBi). If your antenna gain exceeds that, you must reduce conducted power by the excess gain to keep EIRP within the limit. See the current eCFR for the operative text.

• European Union (ETSI EN 300 328, 2.4 GHz example): The “headline” 20 dBm (100 mW) EIRP shows up repeatedly; clear-channel assessment thresholds scale with EIRP. If you fit a higher-gain antenna, your transmitter power must back-off accordingly to keep the same EIRP.

One-line EIRP worksheet:
EIRP (dBm) = TX power (dBm) + Antenna gain (dBi) − Cable/connector loss (dB). Compare the result to the applicable rule; if it’s higher, reduce TX power. Keep the math with your technical file.

7) From Theory to BOM: Cables & Connectors That Don’t Waste dB

A wideband antenna deserves a proper feed system. For outdoor or long-run installs, standardize on low-loss coax and N-type interfaces, and minimize adapters.

• Low-loss field run: LMR-400 N-male ↔ SMA-RP male assembly — stable impedance and lower attenuation on long pulls.
• Enclosure pass-through: N-female bulkhead for LMR-400 — preserves shielding and IP while routing to inside electronics.
• Waterproof connectorization: N plug for LMR-240, crimp — weather-ready terminations.
• Service adapter: Waterproof N-female ↔ N-male adapter — for quick swaps without re-cabling.
• Short bench pigtail (for instruments or tight enclosures): 50 Ω RG174 SMA male ↔ SMA female jumper.

8) Worked Example: Sizing an LPDA for 400–1200 MHz

Spec: 400–1200 MHz band, realized gain ~6–7 dBi, SWR ≤ 2:1, moderate boom length.
Start: choose τ = 0.86, σ = 0.07 → a common mid-range set that balances element count and ripple.
Compute: Carrel’s method yields element lengths/spacing and an apex angle that fit a manageable boom, with ~12–15 elements.
Validate: simulate and check that the active region at 400 MHz sits well clear of the mast and feed hardware; at 1.2 GHz, verify the cross-over line does not disturb the short-element region.
Finalize: add a 1:1 current balun at the input, spec LMR-400 for the 25 m rooftop run, and calculate EIRP with your radio’s conducted power.

9) Printed LPDA Notes (Compact and Calibratable)

Recent open-access papers and theses show 0.5–8 GHz printed LPDAs with ~5–7 dBi realized gain over most of the band and manageable boards (e.g., 420×350 mm). Techniques include bow-tie/trapezoidal elements, tapered microstrip feeds, and Koch-edged dipoles for miniaturization. If you want a repeatable reference antenna for a small chamber—or you need to mount an LPDA on a product—planar is attractive.

10) Vendor Data Pack & Acceptance (What to Ask For)

10.1 Required Documents & Plots

This is the same “evidence bundle” EMC auditors and sophisticated buyers expect.

10.2 Accessory & Cabling Checklist

• Balun model and placement, common-mode control note
• Coax type and run length with loss estimate
• Connector/IP rating and pass-through details
• Grounding & bonding note, especially for OATS work

11) Interactive Quick Test (Answer in 60 Seconds)

1) Band edges known? If not, define (f\text{low})/(f\text{high}) first.
2) Band ratio > 3:1? If yes, LPDA likely beats a Yagi; if no, weigh Yagi’s cost-per-dB.
3) Calibration needed? If yes, ask for traceable AF and uncertainty.
4) Outdoor run > 10 m? If yes, upgrade to LMR-400 and N-type terminations (see internal links).
5) Compliance on TX path? Compute EIRP and apply power back-off per FCC/ETSI rules.

12) FAQs (Schema-Ready)

Q1. What do τ and σ physically control?
A. τ controls how fast elements shrink; σ is the normalized spacing. Together with apex angle, they set bandwidth, element count, and pattern stability.

Q2. What is the “active region” in an LPDA?
A. The subset of elements around half-wavelength at the operating frequency that actually radiates. It slides along the boom as frequency changes.

Q3. How is a printed LPDA different from a rod LPDA?
A. Printed LPDAs integrate on PCB or substrate; they’re more compact, reproducible, and integration-friendly but often have lower gain than full-size rod arrays.

Q4. Can I use an LPDA for transmit as well as receive?
A. Yes—LPDAs are reciprocal—but mind your EIRP for compliance; gain boosts TX EIRP.

Q5. What’s the typical gain of an LPDA?
A. Usually 6–8 dBi for well-designed rod LPDAs; printed designs may be ~4–6 dBi.

13) Welcome Your Inquiry

Whether you’re specifying an EMC test antenna, building a wideband monitoring link, or integrating an LPDA into a product, we can ship traceable-calibrated rod or printed LPDAs and the low-loss cabling to match.
Contact us today for quotes, datasheets, or sample requests:
📧 sales@bafitop.com | 📞 +86-15817341810