2.3 Receiver Performance: Sensitivity, Selectivity, and AGC

2.3 Receiver Performance: Sensitivity, Selectivity, and AGC

Beyond Standard level: Third-order intercept point, numeric shape factors and noise-figure cascading are Advanced-level material. At Standard you need to recognise the terms sensitivity, selectivity, dynamic range and blocking, and describe what each one means.

Evaluating the quality of an amateur radio receiver requires examining specific performance metrics that determine how well the radio can pick up extremely weak signals in distant communications (DXing) and cope with extremely strong adjacent signals during high-density contests. The three critical benchmarks in receiver design are sensitivity, selectivity, and dynamic range.

---

1. Receiver Sensitivity and Thermal Noise Floor

Sensitivity is the measure of a receiver's ability to pick up weak radio signals. It is defined quantitatively as the minimum RF input signal voltage (expressed in microvolts, $\mu\text{V}$) or power (in $\text{dBm}$) applied to the antenna terminal that produces a specified Signal-to-Noise Ratio (SNR) or SINAD (Signal-plus-Noise-and-Distortion to Noise-and-Distortion ratio) at the audio output.

• Standard SSB/CW Benchmark: $0.15, \mu\text{V}$ to $0.25, \mu\text{V}$ for a $10\text{ dB } (S+N)/N$ ratio.
• Standard FM Benchmark: $0.25, \mu\text{V}$ to $0.50, \mu\text{V}$ for a $12\text{ dB SINAD}$ ratio.

Thermal Noise Floor (Johnson-Nyquist Noise)

The ultimate physical limit to receiver sensitivity is thermal noise generated by random electron motion in resistive components. The thermal noise power available in a given bandwidth is calculated using the thermal noise formula:

$P_N = k \cdot T \cdot B$

Where:

• $k = 1.38 \times 10^{-23}\text{ J/K}$ (Boltzmann's constant)
• $T = 290\text{ K}$ (Standard room temperature in Kelvin)
• $B = \text{Receiver noise bandwidth in Hertz (Hz)}$

Expressed in decibels relative to one milliwatt ($\text{dBm}$) at room temperature ($290\text{ K}$):

$P_N (\text{dBm}) = -174\text{ dBm} + 10 \log_{10}(B)$

Noise Floor Calculations for Standard Bandwidths

• CW Bandwidth ($500\text{ Hz}$): $P_N = -174 + 10 \log_{10}(500) = -174 + 27 = -147\text{ dBm}$
• SSB Bandwidth ($2400\text{ Hz}$): $P_N = -174 + 10 \log_{10}(2400) = -174 + 33.8 = -140.2\text{ dBm}$

Notice that narrowing the receiver IF bandwidth reduces total noise power, significantly improving weak-signal readability!

Noise Figure (NF) and Noise Factor ($F$)

No real-world receiver is perfectly noiseless. Active components (transistors, mixers) add internal noise to the signal as it passes through the stages. Noise Factor ($F$) is the ratio of input SNR to output SNR:

$F = \frac{\text{SNR}_{\text{in}}}{\text{SNR}_{\text{out}}}$

Noise Figure (NF) is simply the noise factor expressed in decibels: $\text{NF (dB)} = 10 \log_{10}(F)$

According to Friis' Formula for Noise, the noise figure of the very first stage (the RF preamplifier) dominates the overall noise figure of the entire receiver system. Adding a low-noise preamplifier at the antenna terminals improves overall receiver sensitivity on VHF/UHF bands where thermal noise dominates.

---

2. Receiver Selectivity and Bandwidth Standards

Selectivity is the receiver's ability to differentiate between a desired signal and unwanted interfering signals operating on closely adjacent frequencies. Selectivity is established primarily by the bandpass filters located in the Intermediate Frequency (IF) amplifier section.

Standard Bandwidth Requirements by Operating Mode

Filter Skirt Selectivity and Shape Factor

An ideal filter would possess a perfectly rectangular passband ("brick-wall filter"), passing all frequencies inside the band with zero attenuation and instantly blocking all frequencies outside. Real-world physical filters have sloped sides known as filter skirts.

The steepness of a filter's skirts is defined by its Shape Factor, calculated as the ratio of the bandwidth at $-60\text{ dB}$ attenuation to the bandwidth at $-6\text{ dB}$ attenuation:

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

• An ideal theoretical filter has a Shape Factor of 1.0.
• High-grade commercial crystal filters achieve shape factors between 1.5 and 2.0.
• Modern Digital Signal Processing (DSP) sharp linear-phase filters can achieve shape factors approaching 1.1 to 1.2, offering extraordinary adjacent-channel rejection.

---

3. Dynamic Range, Intermodulation, and Desensitisation

While sensitivity handles weak signals and selectivity handles nearby signals, Dynamic Range determines how well a receiver can process extremely weak signals in the presence of extremely strong signals elsewhere on the band without suffering distortion.

Intermodulation Distortion (IMD)

When two or more strong, unwanted signals ($f_1$ and $f_2$) pass into a non-linear receiver front-end or mixer stage, they mix together to produce Intermodulation Distortion (IMD) products. The most troublesome are the 3rd-Order IMD products because they fall extremely close to the original frequencies:

$f_{\text{IMD3}} = 2f_1 - f_2 \quad \text{and} \quad 2f_2 - f_1$

If $f_1 = 7.100\text{ MHz}$ and $f_2 = 7.105\text{ MHz}$, the 3rd-order product lands at $2(7.100) - 7.105 = 7.095\text{ MHz}$. If a weak desired station is operating on 7.095 MHz, it will be completely obliterated by this phantom signal generated inside the receiver!

Third-Order Intercept Point (IP3)

The Third-Order Intercept Point (IP3 or TOI) is a theoretical figure of merit used to quantify strong-signal handling. It represents the hypothetical input power level (in $\text{dBm}$) at which the power of the 3rd-order IMD product would equal the fundamental signal power. Higher positive IP3 values (+15 dBm to +30 dBm) indicate superior resistance to intermodulation distortion.

Desensitisation (Blocking)

Desensitisation (or receiver blocking) occurs when an exceptionally strong off-frequency signal (such as a nearby commercial broadcasting station or local ham transmitter) enters the receiver front-end. Even if the strong signal is outside the IF passband, its high power level overloads active transistors, driving them out of their linear region into saturation. This drastically reduces the gain for all signals, causing weak desired signals to instantly disappear or drop in volume.

---

4. Automatic Gain Control (AGC) Operation

Signals arriving over ionospheric skywave paths constantly fluctuate in amplitude due to fading (QSB). If receiver gain remained fixed, audio volume would jump wildly between whisper-soft and ear-splitting levels.

Automatic Gain Control (AGC) solves this by automatically adjusting the gain of the IF and RF amplifier stages in inverse proportion to incoming signal strength.

AGC Feedback Mechanism

1. A portion of the amplified IF signal is tapped off after the crystal filter and fed to an AGC detector diode.
2. The AGC detector rectifies the AC signal into a negative DC control voltage proportional to RF signal level.
3. This DC control voltage is fed back to the bias networks of the RF and IF amplifier stages.
4. As incoming signal strength increases, the DC voltage reduces active device gain, holding the audio output level virtually constant.

AGC Time Constant Settings

Receivers feature selectable AGC decay time constants tailored to different operating modes:

• FAST AGC (Decay time ~20 to 100 ms): Quick recovery time. Ideal for Morse code (CW) or fast data digital modes (FT8), allowing the receiver to recover sensitivity in the spaces between code elements.
• SLOW AGC (Decay time ~1 to 3 seconds): Slow recovery time. Essential for Single Sideband (SSB) voice operation. Prevents the receiver gain from surging upward during brief pauses between words, which would cause loud background atmospheric noise "pumping."
• OFF / MANUAL AGC: Disables automatic control. The operator manually adjusts the RF Gain control. Used in heavy contest environments or weak-signal EME (Earth-Moon-Earth) work to prevent strong nearby stations from pumping the AGC circuit.