UNCLASSIFIED
TM-TOOL-006
RF NOISE BRIDGE — IMPEDANCE MEASUREMENT
Wheatstone Bridge, 0.5–60 MHz, R + jX Measurement, Null Detector
Prepared by: Mervyn Martin, KO6NNH  •  Merced, California  •  26 May 2026
Amateur Radio / Electronics — Not for commercial use

Chapter 1 — Introduction and Scope

This manual covers the construction and operation of an RF noise bridge for antenna impedance measurement from 0.5 to 60 MHz. The bridge measures complex impedance Z = R + jX by balancing a calibrated reference arm against the unknown antenna. Null detection uses either an external communications receiver, a built-in audio amplifier, or an AD8307 logarithmic detector with digital readout. Accuracy: R ±5Ω (0–200Ω range), X ±10Ω (−200 to +200Ω range).

Chapter 2 — Theory of Operation

2-1 Bridge Circuit

A standard 4-arm Wheatstone bridge topology adapted for RF. Two fixed 51Ω arms establish the bridge reference. One variable arm (R_cal + jX_cal) is the calibrated reference. The unknown arm (Z_ant) is the antenna under test. The null detector sees zero voltage when the bridge is balanced:

Z_ant = R_cal + jX_cal   (at null)

A 1:1:1 trifilar-wound transformer (T1) on a BN-43-202 core drives the two bridge legs from the noise source. Trifilar winding ensures <1° phase imbalance between the two drive ports across 0.5–60 MHz.

2-2 Noise Source

Two designs are provided:

  • NGZ (Zener): 5.1V zener diode biased into avalanche breakdown produces wideband white noise. Output amplified by a MAR-6 MMIC (+20 dB, DC–2 GHz) to approximately −25 dBm into 50Ω.
  • NGT (Transistor): Reverse-biased NPN transistor base-emitter junction noise; higher output level and more uniform spectral density than zener. Amplified by ERA-3SM MMIC.

2-3 Null Detection Modes

ModeHardwareSensitivityNotes
A — External RXCommunications receiver/SDR<−60 dBmBest; frequency-selective
B — Audio amplifierERA-3SM MMIC + headphones−50 dBmStandalone; no RX needed
C — AD8307Log detector + CYD display−70 dBmDigital null indication

Chapter 3 — Equipment and Materials

ComponentValue / PartQuantity
Bridge transformer T1Trifilar, BN-43-202, 6 turns #26 AWG1
Fixed bridge resistors R1, R251Ω 1% metal film, 1/4W2
Variable resistance R_cal0–200Ω wirewound pot (non-inductive)1
Variable reactance X_cal0–100 pF variable cap + 0–50µH roller inductor1 each
Noise source MMICMAR-6SM or ERA-3SM1
Zener D1BZX55C5V1, 5.1V 500mW1
RF bypass capacitors100 pF NP04
RF choke RFC110µH SRF >50 MHz1
ConnectorsBNC female: noise out, DET, ANT3
Power9V battery (80 mA typical)1

Chapter 4 — Construction

4-1 Transformer Winding

  1. Wind three identical windings of 6 turns #26 AWG simultaneously on a BN-43-202 binocular core. Keep the three windings in the same rotational direction (color-code the wire starts).
  2. Check winding: with an ohmmeter, verify all three windings are equal resistance (±0.1Ω) and have no shorts between windings.
  3. Test leakage inductance: connect one winding to NanoVNA port 1 with port 2 shorted to the other two windings. Leakage inductance should be <0.1µH at 10 MHz.

4-2 Calibration Dials

Mark the R_cal potentiometer dial in 10Ω steps from 0 to 200Ω using a resistance meter. Mark the X_cal capacitor dial in terms of −jX at 10 MHz using the formula X_C = 1/(2πfC). Mark the inductor in terms of +jX at 10 MHz using X_L = 2πfL. Actual X depends on operating frequency; provide a reactance chart (see Appendix A) for other frequencies.

Chapter 5 — Operating Procedures

5-1 Measuring Antenna Impedance

  1. Connect antenna to ANT BNC. Connect communications receiver (tuned to a clear frequency in the antenna's operating range) to DET BNC.
  2. Enable noise source (power switch ON). Receiver S-meter should show S3–S7 noise level. If no noise: check battery, MMIC bias.
  3. Set R_cal to 50Ω (center scale) and X_cal to zero (cap at minimum, inductor at minimum).
  4. Adjust R_cal for minimum S-meter reading. Note: the null may be sharp; sweep slowly.
  5. If null is imperfect at all R_cal values, the antenna has significant reactance. Adjust X_cal (capacitive for a high-SWR dipole that is too long; inductive for one that is too short) until the null is satisfactorily deep (≥10 dB below noise floor).
  6. Read R and X from the calibration dials. Reported impedance: Z_ant = R_cal + jX_cal at the measurement frequency.

Chapter 6 — Calibration

  1. Connect a precision 50Ω non-inductive resistor to ANT port. Set R_cal = 50Ω, X_cal = 0. Verify null at ≥20 dB below noise floor. If null is <20 dB: re-check bridge transformer balance.
  2. Connect a known 100Ω resistor. Set R_cal = 100Ω, X_cal = 0. Verify null is achieved at that setting.
  3. Connect a 47 pF NP0 capacitor. At 10 MHz: X_C = 1/(2π×10e6×47e-12) = −338Ω. Set R_cal = 0Ω (resistor is pure reactance), X_cal to −338Ω (capacitive). Verify null.
  4. Record calibration date and reference components used in log.

Chapter 7 — Verification and Acceptance

  1. Measure a known 50Ω dummy load: R must read 50 ± 3Ω, X must read 0 ± 10Ω.
  2. Measure a known antenna of known resonant frequency (verified by NanoVNA). At resonance, X should read 0 ± 15Ω; R should agree with NanoVNA feedpoint resistance within 10%.
  3. Log: date, reference impedances measured, null depth achieved, measured vs. expected values, operator.

Appendix A — Reactance vs. Frequency Chart

Component1 MHz7 MHz14 MHz28 MHz
47 pF cap−3386Ω−484Ω−242Ω−121Ω
100 pF cap−1592Ω−227Ω−114Ω−57Ω
1 µH inductor+6.3Ω+44Ω+88Ω+176Ω
10 µH inductor+63Ω+440Ω+879Ω+1759Ω

Appendix B — Worked Example

Measuring a 40m dipole at 7.100 MHz. Null achieved at R_cal = 72Ω, X_cal = +18Ω (inductive). This means the dipole is slightly too long (excess inductive reactance). To resonate it: shorten each element by:

ΔL = λ × (X / (2 × R)) × k   (approximate)
ΔL = 42.3m × (18 / 144) × 0.97 = 2.5 cm per side