14.3 Calibrated Signal Generators, Grid Dip Meters, and Component Testing

14.3 Calibrated Signal Generators, Grid Dip Meters, and Component Testing

Diagnosing, aligning, and evaluating radio equipment requires two fundamental diagnostic capabilities: the ability to inject a known, highly calibrated RF signal into a receiver system, and the ability to determine the resonant frequency of unpowered passive components and circuits without applying destructive transmitter power. These roles are fulfilled respectively by the calibrated laboratory RF signal generator and the dip oscillator (historically known as the grid dip meter, or GDO). Mastering these instruments is essential for advanced amateur radio maintenance, filter alignment, and component measurement.

---

Calibrated RF Signal Generators

A calibrated RF signal generator supplies a test signal with specified frequency, amplitude and spectral purity for controlled injection into equipment. Its usable frequency/level range, output matching, phase noise, harmonics and calibration accuracy must suit the measurement.

                 Example Laboratory Signal-Generator Output Stage

 Synthesiser --> [Variable level / amplifier / filter] --+--> [Step attenuator] --> 50 ohm output
                            ^                           |
                            |                           v
                            +------ [ALC comparator / detector]
                                          ^
                                     Level reference

        The detector samples before the large output attenuation.

Key Architectural Features

1. Synthesised Frequency Engine: DDS and/or PLL synthesis referenced to a stable quartz oscillator provides precise frequency setting. TCXO/OCXO options, frequency accuracy, ageing and tuning resolution are specified separately; a sub-Hertz tuning step does not itself guarantee a particular frequency accuracy.
2. Automatic Level Control (ALC): A closed-loop leveling circuit monitors RF output voltage ahead of the attenuator with a precision wideband detector, adjusting PIN diode modulators to maintain constant output amplitude across the entire frequency range.
3. Precision $50\ \Omega$ Attenuator: Switched attenuation combines with the ALC level range to obtain low output levels. Step sizes and total range depend on the model. Shielding and sound connections are essential because leakage bypassing the attenuator can dominate a weak test signal; verify the specified level accuracy and leakage performance for sensitivity work.

Power and Voltage Relationships in a $50\ \Omega$ System

Signal generator output is calibrated in two equivalent units: decibels relative to one milliwatt ($\text{dBm}$) and root-mean-square microvolts ($\mu\text{V}$) across a standard $50\ \Omega$ resistive termination.

$$P_{\text{Watts}} = \frac{V_{\text{RMS}}^2}{R} = \frac{V_{\text{RMS}}^2}{50}$$

Converting milliwatts to decibels:

$$P_{\text{dBm}} = 10 \log_{10}\left(\frac{P_{\text{Watts}}}{10^{-3}\text{ W}}\right) = 10 \log_{10}\left(\frac{V_{\text{RMS}}^2 / 50}{10^{-3}}\right)$$

Conversely, given power in $\text{dBm}$, the equivalent RMS voltage across $50\ \Omega$ is:

$$V_{\text{RMS}} = \sqrt{50 \times 10^{\left(\frac{P_{\text{dBm}} - 30}{10}\right)}}$$

      Reference Calibration Milestones (50 Ω System)

      Output Level (dBm)     RMS Voltage       Significance
      ------------------     -----------       ------------
      +13 dBm                1.0 V             Example generator output level (20 mW)
        0 dBm                224 mV (0.224 V)  Laboratory 1.0 mW standard
      -73 dBm                50.1 µV           HF S9 Standard (IARU Region 1, < 30 MHz)
      -93 dBm                5.01 µV           VHF/UHF S9 Standard (> 30 MHz)
     -107 dBm                1.0 µV            Common reference for receiver sensitivity
     -127 dBm                0.1 µV            Example low-level test signal
     -140 dBm                0.0224 µV         Example very weak test signal

---

Receiver Testing Applications

                             Receiver Sensitivity / MDS Testing

     [ Calibrated Signal ] =====( Double-Shielded Coax )=====> [ Receiver Under Test ]
     [ Generator (50 Ω)  ]                                     [ (CW Mode, 500 Hz BW) ]
                                                                         |
                                                                    Audio Output
                                                                         |
                                                                         v
                                                               [ True-RMS AC Voltmeter ]
                                                                 (or Audio Power Meter)

1. Minimum Discernible Signal (MDS) and Noise Floor Testing

The fundamental metric defining a receiver's absolute sensitivity is its Minimum Discernible Signal (MDS), also designated the receiver's noise floor ($P_{\text{MDS}}$).

Testing Procedure:

1. Terminate the receiver antenna input with a calibrated $50\ \Omega$ signal generator via a high-isolation, double-shielded coaxial cable.
2. Set the receiver to CW mode with Automatic Gain Control (AGC) disabled (or set to manual RF gain) and a standard CW intermediate-frequency filter (typically $B = 500\text{ Hz}$). Connect a true-RMS audio voltmeter or audio power meter across the loudspeaker terminals.
3. Switch the generator RF output off while maintaining the specified $50\ \Omega$ source termination, and record baseline audio voltage $V_{\text{noise}}$. It includes source-termination thermal noise and receiver-added noise. If an attenuated signal is used instead, verify that it is well below the expected MDS; $-145\text{ dBm}$ is not automatically negligible.
4. Turn on the generator, tuned to the receiver passband center (e.g., $700\text{ Hz}$ audio beat tone). Increase generator output until the audio voltmeter reading increases by exactly $3\text{ dB}$ (a voltage increase by a factor of $\sqrt{2} \approx 1.414$):

$\frac{V_{\text{signal+noise}}}{V_{\text{noise}}} = \sqrt{2} \implies \frac{P_{\text{signal+noise}}}{P_{\text{noise}}} = 2.0 \quad (+3\text{ dB})$

1. The generator output level in $\text{dBm}$ producing this $3\text{ dB}$ rise is the MDS. In modern high-grade HF receivers, MDS is typically $-138\text{ dBm}$ to $-142\text{ dBm}$ in a $500\text{ Hz}$ bandwidth.

Theoretical Noise Floor Relationship:

For a matched source at the standard reference temperature of about $290\text{ K}$, the input-referred receiver noise floor in equivalent noise bandwidth $B$ is approximately:

$P_n = -174\text{ dBm/Hz} + 10 \log_{10}(B) + \text{NF}$

where $B$ is in hertz and $\text{NF}$ is the receiver noise figure in decibels. A receiver with $\text{NF}=0\text{ dB}$ adds no noise beyond the source thermal noise. Assuming $B=500\text{ Hz}$:

$P_n = -174 + 10 \log_{10}(500) = -174 + 27 = -147\text{ dBm}$

A practical receiver with an $8\text{ dB}$ noise figure exhibits an $\text{MDS} = -147 + 8 = -139\text{ dBm}$.

2. S-Meter Calibration Standards

The S-meter on a communications receiver monitors Automatic Gain Control (AGC) voltage to indicate incoming signal strength. Historically, uncalibrated consumer receivers used arbitrary S-meter scales. To establish scientific uniformity, the International Amateur Radio Union (IARU) Region 1 adopted a definitive technical recommendation:

               IARU HF S-Meter Signal Level Scale (50 Ω System)

     S-Unit:     S1     S2     S3     S4     S5     S6     S7     S8     S9   S9+20dB
     dBm:      -121   -115   -109   -103    -97    -91    -85    -79    -73     -53
     µV (RMS):  0.20   0.40   0.79   1.58   3.16   6.31   12.6   25.1   50.1   501.0
                 <------------------ 6 dB per S-Unit ------------------>

Using a calibrated signal generator, an amateur tests receiver S-meter tracking by injecting $-73\text{ dBm}$ to verify $S9$, $-79\text{ dBm}$ for $S8$, and so forth. In practice, many commercial transceivers exhibit poor S-meter linearity, compressing lower S-units to $2-3\text{ dB}$ per unit while expanding upper units.

3. Receiver Selectivity and Filter Passband Profiling

With AGC disabled and the receiver in its linear range, establish an on-frequency audio reference, then step the generator off-frequency while increasing its level to restore that output. This gives attenuation versus offset. A common receiver shape-factor convention compares the $-60\text{ dB}$ and $-6\text{ dB}$ bandwidths; half-power bandwidth instead uses about $-3\text{ dB}$:

$\text{Shape Factor} = \frac{\text{Bandwidth at } -60\text{ dB}}{\text{Bandwidth at } -6\text{ dB}}$

An ideal rectangular response has shape factor $1$. For example, $3.84\text{ kHz}$ at $-60\text{ dB}$ divided by $2.40\text{ kHz}$ at $-6\text{ dB}$ gives $1.6$. A value closer to one means a sharper transition when compared at the same attenuation levels.

---

Dip Oscillators (Grid Dip Meters / Gate Dip Meters)

While signal generators evaluate active receivers, determining the resonant frequency of an unpowered passive circuit—such as an antenna trap, loading coil, parasitic choke, or uninstalled tuned circuit—is the domain of the dip oscillator.

          Dip Oscillator                         Unpowered Test Resonator

       [Tunable LC oscillator]
               |
        [Exposed coil]      <~~ magnetic coupling ~~>   o---[ L ]---o
               |                                       |           |
        [RF strength detector                          +----||-----+
          or grid-current meter]                            C
               |
          Meter dips when the resonator loads the oscillator.

1. Operating Principle: The Physics of the "Dip"

A dip oscillator consists of a calibrated, tunable LC oscillator equipped with a set of interchangeable, plug-in external inductor coils covering discrete bands from $1.5\text{ MHz}$ up to $250\text{ MHz}$ or higher:

1. Indication: A traditional valve grid-dip meter indicates oscillator grid current. Solid-state dip meters can instead rectify sampled RF voltage and amplify the detector output; they do not all measure gate current, and a BJT has no gate-source junction.
2. Coupling: Bringing the exposed oscillator coil near an accessible test coil creates mutual inductive coupling between the oscillator and the unpowered resonator.
3. Dip: At resonance the test circuit absorbs energy and loads the oscillator, reducing its RF amplitude. The instrument is arranged to show that reduction as a downward meter deflection. The common mechanism is resonant loading, regardless of the particular detector circuit.

2. Applications in the Amateur Radio Station

A. Determining Resonant Frequency of Antenna Traps and Elements

Antenna traps isolate wire sections on multiband dipoles and Yagis. To verify the resonant frequency of an unpowered trap without applying transmitter power, hold the dip meter coil parallel to the trap inductor coil, slowly rotate the frequency dial, and note the frequency where the needle dips. This confirms whether the trap is tuned to the intended band (e.g., $14.15\text{ MHz}$ for a 20-meter trap).

B. Measuring Unknown Inductors ($L$)

An accessible inductor can be estimated from its resonance with a known capacitor. Allow for stray capacitance and frequency-calibration error. A closed toroid may need a suitable auxiliary coupling winding or a dip meter with a capacitive probe; placing the normal coil beside it is not a reliable general method.

1. Connect a precision $1\%$ silver-mica or C0G ceramic capacitor of known value (e.g., $C_{\text{ref}} = 100\text{ pF}$) in parallel across the unknown inductor ($L_{\text{unknown}}$).
2. Couple weakly to the accessible resonator and find $f_0$. Keep leads short; the formula uses total capacitance, including relevant stray and probe capacitance.
3. Calculate the unknown inductance using Thomson's resonance formula:

$f_0 = \frac{1}{2\pi \sqrt{L \cdot C_{\text{ref}}}} \implies L_{\text{unknown}} = \frac{1}{4\pi^2 f_0^2 C_{\text{ref}}}$

C. Measuring Unknown Capacitors ($C$)

Conversely, to measure an unknown capacitor, solder it across a known standard inductor ($L_{\text{ref}}$), dip the combination for $f_0$, and solve for capacitance:

$C_{\text{unknown}} = \frac{1}{4\pi^2 f_0^2 L_{\text{ref}}}$

D. Checking Coaxial Transmission Line Stubs

Quarter-wave shorted stubs and half-wave open stubs behave as parallel resonant circuits at their fundamental electrical frequency. Forming a tiny 1-turn loop at the end of the stub and coupling it to the dip meter coil allows rapid determination of the stub's true electrical resonant frequency and velocity factor.

E. Wavemeter (Absorption Frequency Meter) Mode

In absorption wavemeter mode, set the instrument's controls as specified in its manual so the tuned circuit and detector respond to external RF rather than using an oscillator-loading dip.

• RF from a powered source couples to the tuned circuit; the detector produces an upward peak as the dial is tuned through that component.
• Compare peaks with the expected fundamental and harmonic frequencies. A peak alone does not unambiguously distinguish a fundamental, harmonic or parasitic oscillation. Begin well away from a power stage, couple loosely to avoid overload, and do not touch live RF or high-voltage circuitry.

3. Operating Precautions: The Loose Coupling Rule

---

Step-by-Step Worked Component Calculations

Worked Example 1: Measuring Unknown Inductance with a Dip Meter

A builder has an unmarked air-wound RF coil. A precision $C_{\text{ref}}=100\text{ pF}$ capacitor is connected across it with short leads. Weak inductive coupling gives a dip at $f_0=7.15\text{ MHz}$. Assuming stray capacitance is negligible for this estimate, calculate the inductance in $\mu\text{H}$.

Calculation:

$L = \frac{1}{4\pi^2 f_0^2 C_{\text{ref}}}$

1. Compute the denominator terms:

$4\pi^2 \approx 39.4784$

$f_0^2 = (7.15 \times 10^6\text{ Hz})^2 = 5.11225 \times 10^{13}\text{ Hz}^2$

$4\pi^2 f_0^2 = 39.4784 \times (5.11225 \times 10^{13}) \approx 2.01825 \times 10^{15}$

1. Multiply by capacitance $C_{\text{ref}}$:

$\text{Denominator} = (2.01825 \times 10^{15}) \times (100 \times 10^{-12}\text{ F}) = 201{,}825$

1. Calculate inductance $L$:

$L = \frac{1}{201{,}825} \approx 4.955 \times 10^{-6}\text{ H} \approx 4.96\ \mu\text{H}$

Result: The estimated inductance is $4.96\ \mu\text{H}$ under the stated capacitance and frequency assumptions. The result’s accuracy is limited by those measurements and the coupling/loading, not the number of displayed decimal places.

---

Worked Example 2: S-Meter Generator Output Level Determination

A communications receiver S-meter is to be tested on the 20-meter band ($14.2\text{ MHz}$). Assuming the receiver follows the international IARU Region 1 standard ($S9 = -73\text{ dBm}$, $6\text{ dB}$ per S-unit), determine:

1. The generator output level in $\text{dBm}$ required to test an $S5$ reading.
2. The corresponding RF voltage in microvolts ($\mu\text{V}$) across the receiver's $50\ \Omega$ input.

Calculation:

1. Compute the required power in $\text{dBm}$: From $S9$ to $S5$ is a drop of 4 S-units:

$\Delta P = 4 \times 6\text{ dB} = 24\text{ dB}$

$P_{S5} = P_{S9} - \Delta P = -73\text{ dBm} - 24\text{ dB} = -97\text{ dBm}$

1. Compute the equivalent voltage in $\mu\text{V}$: Because each $6\text{ dB}$ step represents a halving of signal voltage:

• At $S9$: $V = 50.1\ \mu\text{V}$
• At $S8$: $V = 25.1\ \mu\text{V}$
• At $S7$: $V = 12.6\ \mu\text{V}$
• At $S6$: $V = 6.31\ \mu\text{V}$
• At $S5$: $V = 3.16\ \mu\text{V}$

Verify using the power formula:

$V = \sqrt{50 \times 10^{\left(\frac{-97 - 30}{10}\right)}} = \sqrt{50 \times 10^{-12.7}} = \sqrt{50 \times 1.995 \times 10^{-13}} = \sqrt{9.976 \times 10^{-12}} \approx 3.16 \times 10^{-6}\text{ V} = 3.16\ \mu\text{V}$

Result: The signal generator must be set to $-97\text{ dBm}$ ($3.16\ \mu\text{V}$) to indicate $S5$.

---

Worked Example 3: Receiver Noise Figure from Measured MDS

An operator testing an amateur VHF receiver on $144.2\text{ MHz}$ with an equivalent noise bandwidth of $500\text{ Hz}$ measures a Minimum Discernible Signal (MDS) of $-136\text{ dBm}$. Calculate the receiver's Noise Figure (NF) in decibels.

Calculation:

Recall the theoretical thermal noise floor equation:

$\text{MDS} = -174\text{ dBm/Hz} + 10 \log_{10}(B) + \text{NF}$

1. Calculate the bandwidth factor for $B = 500\text{ Hz}$:

$10 \log_{10}(500) = 10 \times 2.6990 \approx +27.0\text{ dB}$

1. Substitute into the formula:

$-136\text{ dBm} = -174\text{ dBm} + 27.0\text{ dB} + \text{NF}$

$-136\text{ dBm} = -147.0\text{ dBm} + \text{NF}$

$\text{NF} = -136 - (-147.0) = +11.0\text{ dB}$

Result: The receiver exhibits a Noise Figure of $11\text{ dB}$.

---

Comparison Table: RF Diagnostic and Measurement Instruments