Explainers · 2026-07-12 · Patreon guide

Patreon for amateur radio creators: HF propagation and ionospheric layers, half-wave dipole gain and Yagi-Uda arrays, transmission line VSWR and impedance matching, superheterodyne receiver noise figure and MDS, FT8 GFSK LDPC and WSPR digital modes, power amplifier IMD and IP3 Part 97 spurious limits, CW telegraphy and DXCC contesting, and the Apple Tax

Amateur radio Patreons retain patrons because the YouTube video shows the antenna going up and the contact being logged but not the physics: not why a half-wave dipole has exactly 73.1 Ω of radiation resistance and not 50 Ω, not why FT8 decodes reliable messages 24 dB below the noise floor when SSB voice requires 10 dB above it, and not why the intermodulation products of two tones through a power amplifier appear at 2f⊂1;−f⊂2; and 2f⊂2;−f⊂1; rather than at sum and difference frequencies. The patron who understands the MUF formula, the Friis noise chain, and the LDPC belief-propagation decoder does not find that depth anywhere else; canceling the subscription means losing the technical layer they came for.

HF propagation and ionospheric layers: MUF, skip distance, and grey-line

D, E, F1, and F2 layers: electron density and the plasma frequency

HF radio propagation between 3 MHz and 30 MHz uses the ionosphere — the ionized region of the upper atmosphere extending from approximately 60 km to 1,000 km altitude — to refract (bend) signals back toward Earth. The ionosphere is stratified into distinct layers with characteristic electron density profiles driven by solar ultraviolet and X-ray ionization of atmospheric gases. Each layer has a characteristic behavior that determines which HF frequencies it reflects at what time of day and what phase of the solar cycle.

The D layer (60–90 km altitude) forms only during daylight hours when solar UV at wavelengths below 121.6 nm ionizes nitric oxide (NO) molecules, which have an unusually low ionization potential. Electron density in the D layer is relatively low, but it is the dominant source of absorption rather than reflection for HF signals: radio energy excites free electrons, which collide with neutral gas molecules (collision frequency is high at these altitudes due to the dense atmosphere) and convert the signal energy to heat. The D layer absorbs frequencies below approximately 10 MHz during mid-day solar maximum conditions, attenuating 40-meter (7 MHz) and 80-meter (3.5 MHz) paths severely during the day. The D layer disappears at night, which is why 40-meter paths to Europe from North America work well after local sunset — the absorbing layer has recombined.

The E layer (90–130 km altitude) reflects signals in the 3–15 MHz range during daytime conditions. Sporadic-E (Es) is a distinct, unpredictable phenomenon: intense patches of high electron density form in the E layer (mechanism not fully understood but correlated with meteor ablation and wind shear) and can reflect signals up to 50–100 MHz — producing sudden openings on the 6-meter (50 MHz), 4-meter, and even 2-meter (144 MHz) VHF bands over distances of 800–2,500 km. Sporadic-E is the primary propagation mechanism for the VHF bands during summer months in temperate latitudes.

The F layer (150–400 km altitude) is the most important for long-distance HF communication. During daytime it splits into F1 (150–220 km) and F2 (220–400 km) sub-layers; at night the lower-density F1 layer recombines and only the F2 layer remains. The F2 layer electron density varies with the 11-year solar cycle (sunspot number), reaching peak densities near solar maximum that support frequencies as high as 30 MHz (or occasionally higher) for transoceanic paths. The plasma frequency is the key quantity: a radio wave at frequency f is reflected by the ionosphere if and only if f is below the plasma frequency f_p = 9√N Hz, where N is the electron density in electrons/m³. At a typical solar-maximum F2 peak electron density of 10¹² electrons/m³, f_p = 9 × 10⁶ Hz = 9 MHz. This is the critical frequency f_c for vertical incidence.

MUF, LUF, and skip distance: the geometry of oblique incidence

Signals transmitted at low elevation angles travel obliquely through the ionospheric layer rather than vertically, which allows them to be reflected at frequencies significantly above the critical frequency. The Maximum Usable Frequency (MUF) for a specific path geometry is: MUF = f_c / sin(θ) = f_c × sec(ζ), where θ is the elevation angle of the signal at the ionospheric reflection point and ζ is the corresponding zenith angle (ζ = 90° − θ). For a typical long-distance path with a reflection geometry at θ = 15° elevation, MUF ≈ 3.9 × f_c. If the critical frequency is 9 MHz, the MUF for that path geometry is approximately 35 MHz — comfortably supporting 10-meter (28 MHz) communication. A frequency above the MUF penetrates the ionosphere and escapes to space; a frequency at or below the MUF is reflected.

The Lowest Usable Frequency (LUF) is the lower bound on usable frequencies for a path, set by D-layer absorption during daytime. Below the LUF, absorption in the D layer reduces the signal level below the noise floor of the receiver, even though the ionosphere does reflect the signal. The LUF and MUF define a usable frequency window for each path at each time of day; propagation programs such as VOACAP, PROPLAB, and ITURHFPROP integrate these values over the diurnal and solar cycle variation of the ionosphere to produce predicted circuit reliability as a function of frequency, month, and solar flux index.

Skip distance is the minimum ground range at which a sky-wave signal at a given frequency is received. Below the skip distance, the signal has a path geometry that would require the signal to penetrate the ionosphere (the elevation angle is too high, so MUF falls below the operating frequency) and the signal is not reflected back to Earth in that zone. The skip distance for a path is: d_skip = 2h × cot(θ_max), where h is the reflecting layer height and θ_max is the maximum elevation angle at which the MUF still equals or exceeds the operating frequency. Increasing operating frequency shortens the skip distance (the signal reaches the required reflection angle at a larger ground range); working near the MUF on a given frequency band produces the longest reliable DX paths with the least absorption. Grey-line propagation occurs at the sunrise and sunset terminator: the D layer on the illuminated side has not yet fully formed (at sunrise) or has just recombined (at sunset), while the F layer remains ionized from recent sunlight exposure; this 15–20-minute window creates enhanced long-path propagation on 40 and 80 meters and is exploited by DX-chasers for the highest DXCC entity totals.

Half-wave dipole antenna physics: radiation resistance, gain, and Yagi-Uda arrays

Resonant length and radiation resistance

The half-wave dipole is the reference antenna for all amateur radio gain calculations because its radiation resistance and pattern are analytically calculable. Physical resonant length: L = 142.5 / f(MHz) meters in practical applications, which is approximately 5% shorter than the theoretical free-space half-wavelength (150 / f(MHz) meters) due to the end effect (capacitive loading at the wire tips that effectively lengthens the electrical antenna without changing the physical length) and conductor diameter effects. The equivalent in feet: L(ft) = 468 / f(MHz). At 14.200 MHz (20-meter phone): L = 142.5 / 14.2 ≈ 10.04 m (32.96 ft) per element, total dipole 20.08 m.

Radiation resistance: A transmitting antenna can be modeled as an impedance at the feedpoint. The feedpoint impedance has a resistive component (radiation resistance R_rad + loss resistance R_loss) and a reactive component (inductive or capacitive). At resonance the reactive component is zero, leaving only the resistive part. For a perfect lossless half-wave dipole in free space, R_rad = 73.1 Ω. This is not an approximation — it is a theoretical result from integrating the radiation pattern of the sinusoidal current distribution on the dipole. The 73.1 Ω is purely determined by the geometry of the current distribution; it is not a design parameter. Any real antenna also has loss resistance R_loss from conductor resistance, dielectric loss, and ground proximity; the antenna efficiency η = R_rad / (R_rad + R_loss). For a properly installed dipole using 12-gauge copper wire above 10 m height, R_loss is typically below 1 Ω, giving η > 98%.

The 73.1 Ω feedpoint impedance is not 50 Ω, which is why a direct connection to 50 Ω coaxial feedline creates a mismatch. The VSWR of a 73.1 Ω load on a 50 Ω line: Γ = (73.1 − 50) / (73.1 + 50) = 23.1 / 123.1 ≈ 0.188; VSWR = (1 + 0.188) / (1 − 0.188) ≈ 1.46:1. This modest mismatch is acceptable for most transceivers (which tolerate up to 2:1 VSWR at full power) and produces only |Γ|² ≈ 3.5% reflected power loss — 0.16 dB. A 1/4-wave matching transformer or a feedline choke balun is more important for suppressing common-mode current on the feedline sheath (which would make the feedline radiate and distort the pattern) than for correcting the modest 73-to-50 Ω mismatch.

Yagi-Uda directional arrays: directors, reflector, and gain

A Yagi-Uda array adds parasitic elements to the driven dipole to concentrate radiation in one direction. The reflector element is positioned λ/4 (quarter-wavelength, approximately 5.1 m at 14.2 MHz) behind the driven element and is cut approximately 5% longer than the driven element. The reflector induces a current that, by the combination of its position and length, radiates with a phase that adds constructively with the driven element in the forward direction and destructively in the rearward direction. The director elements are cut approximately 4% shorter than the driven element and are positioned λ/4 in front of the driven element (and λ/4 in front of each other for multi-director arrays); they also carry induced currents that reinforce forward radiation.

Gain of Yagi arrays: a 2-element Yagi (reflector + driven element, no directors) produces approximately 3–4 dBd; a 3-element Yagi (reflector + driven + one director) approximately 7–8 dBd; a 5-element Yagi approximately 10–11 dBd; a 7-element Yagi approximately 12–13 dBd. The gain per additional element decreases as the array grows because the useful aperture area is being filled with diminishing spatial resolution. A useful rule: gain (in linear, not dB) scales approximately as the physical boom length in wavelengths for optimized designs — G ≈ 0.4 × (boom_length / λ) + 0.4 for boom lengths from 0.1λ to 4λ. Front-to-back ratio (F/B): a well-optimized 3-element Yagi achieves 20–25 dB F/B; a 5-element design optimized for F/B rather than maximum gain can achieve 35–40 dB F/B. The E-plane 3 dB beamwidth narrows with gain: a 3-element Yagi has approximately 65°; a 5-element approximately 52°; a 7-element approximately 42°. Stacking two identical Yagis vertically with a separation of 0.75λ and phasing them in phase produces approximately 2.7–3.0 dB gain increase over a single Yagi, narrows the vertical beam, and lowers the take-off angle — beneficial for working long-distance ionospheric paths at low elevation angles.

Transmission line theory: characteristic impedance, VSWR, and matching networks

Characteristic impedance and the reflection coefficient

A transmission line is characterized by its distributed inductance L (henrys per meter) and capacitance C (farads per meter) per unit length. The characteristic impedance Z₀ = √(L/C) Ω, independent of line length. For coaxial cable: Z₀ = (138/√ε_r) × log₁₀(D/d) Ω, where D is the inner diameter of the outer conductor, d is the outer diameter of the inner conductor, and ε_r is the relative dielectric constant of the insulating material. For RG-213 coax: D = 7.24 mm, d = 2.26 mm, ε_r ≈ 2.3 (polyethylene), giving Z₀ = (138/1.517) × log₁₀(3.20) ≈ 91.0 × 0.505 ≈ 45.9 Ω — close enough to 50 Ω to be specified as a 50 Ω cable within manufacturing tolerance. LMR-400 has Z₀ = 50 Ω with lower loss (0.68 dB/30.5 m at 100 MHz vs 2.1 dB for RG-213). The velocity factor VF = 1/√ε_r: for polyethylene ε_r = 2.3, VF = 0.659; for foam polyethylene ε_r ≈ 1.5, VF = 0.82.

When a transmission line of impedance Z₀ is terminated with a load Z_L, the reflection coefficient at the load is: Γ = (Z_L − Z₀) / (Z_L + Z₀), a complex number with magnitude |Γ| between 0 (matched, no reflection) and 1 (open or short circuit, total reflection). The fraction of forward power reflected is |Γ|². The VSWR (Voltage Standing Wave Ratio) is: VSWR = (1 + |Γ|) / (1 − |Γ|), ranging from 1:1 (matched) to ∞:1 (total reflection). Practical interpretation: VSWR 1.5:1 ⇒ |Γ| = 0.20 ⇒ reflected power = 4%; VSWR 2:1 ⇒ |Γ| = 0.333 ⇒ reflected power = 11.1%; VSWR 3:1 ⇒ |Γ| = 0.50 ⇒ reflected power = 25%. In a lossless transmission line, VSWR has the same value everywhere on the line; in a lossy line, VSWR decreases toward the transmitter because the reflected wave is attenuated on its return trip.

A quarter-wave matching transformer transforms impedance: a λ/4 section of transmission line with characteristic impedance Z_T transforms a load impedance Z_L to an input impedance Z_in = Z_T² / Z_L. To match a 73.1 Ω dipole to a 50 Ω feedline: Z_T = √(50 × 73.1) = √3655 ≈ 60.5 Ω; a 60 Ω λ/4 transformer (constructed from RG-6 cable at 75 Ω, two sections in parallel give 37.5 Ω — not ideal; commercial 60 Ω Heliax achieves this directly). A single-stub matching network uses a short-circuit or open-circuit stub of calculated length at a calculated position on the feedline to cancel the reactive component of the load impedance and transform the resistive component; the required position and stub length are read from a Smith chart. The Smith chart is a graphical representation of the complex reflection coefficient Γ plane with constant-resistance and constant-reactance circles overlaid; moving along a transmission line appears as rotation around the center of the Smith chart.

Superheterodyne receiver design: noise figure, MDS, and dynamic range

IF conversion and image rejection

A superheterodyne (superhet) receiver converts the desired signal frequency f_s to a fixed intermediate frequency (IF) f_IF by mixing f_s with a local oscillator (LO) at frequency f_LO. The mixing product at |f_s − f_LO| = f_IF is selected by the IF filter; all other frequencies are rejected. Common IF frequencies in HF receivers: 455 kHz (classic superhets), 8.215 MHz, 9.0 MHz, 10.7 MHz, or 70.0 MHz in high-performance SDR-based designs. The IF filter bandwidth determines the receiver's selectivity: a 2.4 kHz SSB filter resolves adjacent channels 3 kHz apart; a 500 Hz CW filter enables copying CW through a QRM pile-up.

The image frequency problem: a mixer is a non-linear device that produces output at |f_s − f_LO| for any input; a signal at f_image = f_LO + f_IF (if f_LO is set below f_s) also produces an output at f_IF and is indistinguishable from the desired signal. The image rejection ratio (IRR) in dB measures the suppression of the image frequency signal. A preselector filter (bandpass filter before the mixer, covering the desired band) suppresses the image if the IF is high enough that the image frequency falls outside the preselector passband; at a 455 kHz IF on 14 MHz, the image is at 14.000 + 0.910 = 14.910 MHz — within the same 20-meter band, giving essentially no image rejection. High-IF designs (10.7 MHz or higher) place the image 21.4 MHz above the signal on 10.7 MHz receivers, well outside the 20-meter band. Modern SDR receivers use quadrature (I/Q) demodulation to inherently reject the image by 30–40 dB without a preselector.

Noise figure, Friis chain, and minimum detectable signal

The noise figure (NF) of an amplifier or receiver measures the additional noise the device contributes relative to the thermal noise floor: NF = 10×log₁₀(F) dB, where F is the noise factor F = (S/N)_in / (S/N)_out (the ratio of SNR at the input to SNR at the output; for a noiseless device F = 1, NF = 0 dB). A preamplifier with NF = 3 dB has F = 2; it adds noise equal to one thermal noise floor power (kTB) in addition to the amplified input noise.

The Friis noise formula cascades the noise figures and gains of a receive chain: F_total = F₁ + (F₂ − 1)/G₁ + (F₃ − 1)/(G₁×G₂) + …, where F_n and G_n are the noise factor and power gain of the n-th stage. The key insight: the first stage in the chain dominates the total noise figure if it has significant gain (G₁ ≫ 1). This is why a low-noise preamplifier at the antenna (before feedline loss) dramatically improves system sensitivity: if the feedline has 3 dB loss (a factor of 2 in power, equivalent to adding 3 dB of noise figure), placing a preamplifier with G = 20 dB and NF = 2 dB before the feedline gives F_total ≈ F_preamp = 1.58 (NF = 2 dB), whereas without the preamp the feedline loss adds 3 dB to the receiver's noise figure.

The Minimum Detectable Signal (MDS) is the smallest signal power that can be detected with SNR = 1 (0 dB): MDS = kTB × NF = −174 dBm + 10×log₁₀(BW Hz) + NF dB. At BW = 2,400 Hz (SSB) and NF = 10 dB: MDS = −174 + 10×log₁₀(2400) + 10 = −174 + 33.8 + 10 = −130.2 dBm. Practical SSB copy requires SNR ≈ 10 dB, so the copy threshold is at MDS + 10 dB = −120.2 dBm. S-meter convention: S9 = −73 dBm (VHF/UHF) or −93 dBm (HF by IARU standard); S1 = −121 dBm at HF. A signal just above the noise floor is approximately S1 on a calibrated S-meter.

Dynamic range measures the receiver's ability to handle strong signals while copying weak signals simultaneously. The blocking dynamic range is the range between MDS and the signal level that compresses the receiver by 1 dB (P_1dB input compression point). The intermodulation dynamic range (IMDR) is the range between MDS and the two-signal level that produces third-order intermodulation products at MDS level: IMDR (dB) = (2/3) × (IIP3 − MDS), where IIP3 is the input third-order intercept point. A high-performance HF receiver achieves IIP3 > +20 dBm and NF < 10 dB, giving IMDR > 100 dB.

FT8 digital modes: GFSK, LDPC coding, and weak-signal sensitivity

FT8 modulation and protocol structure

FT8 (Franke-Taylor 8-FSK) was released in 2017 as part of the WSJT-X software suite authored by Joe Taylor K1JT (Nobel Physics Laureate 1993) and Steve Franke K9AN. It uses 8-tone Gaussian frequency-shift keying (8-GFSK): eight discrete tone frequencies separated by 6.25 Hz, with Gaussian pre-filtering to limit adjacent-channel interference. The baud rate is 6.25 symbols per second (one symbol per 160 ms); each symbol carries log₂(8) = 3 bits. The total transmission occupies 79 symbols × 160 ms = 12.64 seconds and is transmitted within one 15-second slot (synchronized to the GPS-disciplined UTC clock at 0, 15, 30, and 45 seconds).

Message structure and LDPC encoding: an FT8 message carries 77 bits of information (caller and callee call signs in a compressed format, 4-character Maidenhead grid locator in 15 bits, or signal report in 3 bits for CQ, reply, and acknowledgment messages). A 12-bit CRC is appended to detect decoding errors, giving 89 bits. The 89-bit message is encoded with a rate-1/2 LDPC code specifically designed for FT8, producing a 174-bit codeword. LDPC (Low-Density Parity-Check) codes are linear block codes defined by a sparse binary parity-check matrix H where the codeword satisfies H×x = 0 (mod 2). The FT8 LDPC code has a carefully structured Tanner graph (the bipartite graph representation of H) that allows efficient iterative decoding by the sum-product (belief propagation) algorithm: each variable node passes soft probability estimates to connected check nodes, which compute consistency corrections and pass them back; the iteration converges (typically in 20–50 iterations) to a codeword that satisfies all parity checks or declares a failure. The 174-bit codeword is mapped into 58 symbols at 3 bits per symbol (58 × 3 = 174); 7 synchronization symbols are added at specific positions in the 79-symbol sequence, giving the decoder a known pilot pattern for synchronization and timing.

Sensitivity: FT8 decodes at approximately −24 dB S/N (signal-to-noise ratio in a 2500 Hz reference bandwidth), meaning the signal is 400 times weaker in power than the noise floor in that bandwidth. This extraordinary sensitivity is possible because the 77 information bits are spread over 12.64 seconds of transmission (4.84 ms per information bit equivalent duty cycle), the LDPC code corrects errors in received symbols, and the decoder correlates all received samples against the known structure. Comparing to SSB voice (requires +10 dB S/N for copy) and CW (requires −10 to −15 dB S/N with experienced operators and narrow filters): FT8 gains approximately 34 dB over SSB and 9–14 dB over CW. WSPR uses 4-FSK at 1.4641 baud with 2-minute transmissions (162 symbols) and achieves −28 dB S/N threshold — 4 dB more sensitive than FT8, at the cost of much lower throughput (call sign + 4-char locator + transmit power level only, one message per 2 minutes). Operators run WSPR beacons 24 hours per day to map propagation paths that would otherwise go undetected; a 5-watt WSPR transmitter on a modest antenna is routinely spotted at 15,000 km distances during favorable F-layer conditions.

Tier structure for digital modes educators: Signals ($5–8/month, waterfall image archives, decoded spot logs for major propagation events, Discord), Data Files ($15–25/month, raw IQ recordings from SDR, WSPR spot CSV exports, propagation analysis documents with MUF calculations for featured paths), QSL + Propagation Reports ($35–50/month, patron-specific WSPR beacon operation — transmitting the patron’s call sign grid from the creator’s antenna for one month + full spot-log analysis).

Power amplifier linearity: IMD, IP3, and FCC Part 97 spurious emission limits

Intermodulation distortion and third-order intercept

A power amplifier is not a perfectly linear device. When two tones at frequencies f₁ and f₂ pass through a non-linear amplifier characterized by a Taylor series transfer function y = a₁x + a₂x² + a₃x³ + …, the cubic term a₃x³ produces third-order intermodulation distortion (IMD) products at frequencies 2f₁−f₂ and 2f₂−f₁. These third-order products fall at offsets of (f₂−f₁) above and below the two original tones — within the passband of the transmitter and potentially within adjacent channel allocations. For a two-tone test with tones at 14.200 MHz and 14.201 MHz (f₂−f₁ = 1 kHz), the IMD products fall at 14.199 MHz (2×14.200−14.201) and 14.202 MHz (2×14.201−14.200) — 1 kHz outside each tone, within the SSB phone bandwidth.

The IP3 (third-order intercept point) is the extrapolated power level at which the fundamental and third-order IMD output powers would be equal, if the amplifier remained linear above its operating range. For the output IP3 (OIP3): the fundamental output power increases 1 dB per 1 dB of input increase; the third-order product increases 3 dB per 1 dB of input increase; they would theoretically intersect at OIP3 if extrapolated. The IMD suppression ΔIM (in dB below the carrier) is related to OIP3 by: OIP3 = P_out + ΔIM/2, where P_out is the output power of each tone in dBm and ΔIM is the suppression of the IMD product below the tone level in dB. A standard for amateur SSB operation is IMD ≤ −31 dB (IMD3 31 dB below the PEP output) for acceptable adjacent-channel interference. A commercial-grade amplifier achieves −35 to −40 dB IMD3.

Harmonic suppression and FCC Part 97 spurious emission limits

Harmonics are signals at integer multiples of the carrier frequency (2f₀, 3f₀, 4f₀, …) produced by the non-linearity of the amplifier. A push-pull amplifier topology cancels even-order distortion products (2f₀, 4f₀, …) because the two amplifier halves produce even-order products with equal magnitude but opposite phase that cancel at the output transformer. This is why push-pull amplifiers are preferred for HF power amplifiers: the most problematic harmonic (2nd) and 4th harmonic are suppressed by topology, leaving only odd harmonics (3rd, 5th, …) which require the low-pass output filter to suppress. A 7th-order elliptic low-pass filter following the amplifier provides 70–80 dB of harmonic suppression.

FCC Part 97 spurious emission limits for amateur stations: for transmitter output power at or above 50 W PEP, spurious emissions (harmonics, IMD products, and any other unintended emissions) must be attenuated at least 50 dB below the PEP carrier level. For output power between 5 W and 50 W: at least 40 dB below PEP. For output below 5 W: at least 30 dB below PEP. A 100 W (50 dBm) transmitter must suppress all harmonics to below 50 dBm − 50 dB = 0 dBm (1 mW) at the antenna port. A 2nd harmonic at 28 MHz from a 14 MHz transmitter at this level would interfere with amateur 10-meter operations; well-designed transceiver and amplifier combinations typically achieve −60 to −70 dBc at harmonics.

Amplifier class efficiency: Class A conducts current over the full 360° of the RF cycle; maximum theoretical efficiency 50%; produces the lowest distortion and is used in low-level driver stages. Class AB conducts over slightly more than 180° of the cycle; efficiency 60–70%; used in HF solid-state final amplifiers (LDMOS transistors are preferred for HF due to high breakdown voltage and low output capacitance). Class B: 180° conduction; efficiency π/4 ≈ 78.5%. Class C: less than 180° conduction; efficiency > 78.5% but with heavy harmonic content — suitable only for constant-envelope modes (FM, CW) where the output filter removes harmonics; cannot be used for SSB, FT8, or any amplitude-varying mode. Vacuum tube amplifiers (3-500Z, 8877) in Class AB tetrode or triode configuration remain popular for high-power HF operation (1500 W legal limit in US amateur radio) due to their inherent high breakdown voltage and tolerance for antenna mismatch.

Tier structure for HF amplifier and RF electronics educators: RF Bench ($8–12/month, amplifier build photo documentation, spectrum analyzer screenshots of harmonic levels, Discord), Measurement Files ($20–35/month, two-tone IMD test result data files, SPICE simulation files for amplifier stage designs, filter coefficient and component values spreadsheets), Design Consultation ($50–80/month capped 4–6, patron submits amplifier schematic and receives documented IMD analysis with specific correction recommendations).

CW telegraphy and DXCC contesting

Morse code mechanics: PARIS standard and Farnsworth method

CW (Continuous Wave) telegraphy is the transmission of Morse code by keying the carrier of a radio transmitter on and off. Morse code uses a combination of short marks (dits, duration 1 unit) and long marks (dahs, duration 3 units) separated by spaces (intra-character: 1 unit; inter-character: 3 units; inter-word: 7 units). The speed in words per minute (WPM) is defined by the PARIS standard: the word “PARIS” (P-A-R-I-S followed by a word space) contains exactly 50 unit elements. Therefore: WPM = 1200 / unit_duration_ms. At 20 WPM: unit duration = 60 ms; dit = 60 ms; dah = 180 ms; inter-character space = 180 ms; inter-word space = 420 ms. At 30 WPM: unit = 40 ms.

CW occupied bandwidth: the keyed carrier produces sidebands at the keying rate and its harmonics. For clean, sharp keying with no shaping: BW ≈ 4 × WPM × 10 Hz (a rough rule; the exact spectrum depends on the rise/fall time of the keying envelope). At 20 WPM: BW ≈ 800 Hz. With a shaped keying envelope (Gaussian rise/fall time of approximately 5 ms), the occupied bandwidth at 20 WPM is reduced to approximately 200–300 Hz — within the 500 Hz filter bandwidth used in CW operation. Excessively fast key rise times (clicks) produce wideband interference across the entire band; most modern CW transceivers apply automatic rise-time shaping.

The Farnsworth method for CW learning uses full-speed character elements (individual dits and dahs at the target WPM) while extending the inter-character and inter-word spaces. For example: characters at 20 WPM but inter-character spaces at 10 WPM effective overall speed — the student learns to hear the full-speed character as a sound pattern (“dah-dit” for N rather than counting “long-short”) while having time to process it before the next character. This is critical because the brain must recognize characters as audio patterns rather than consciously counting elements; copy speeds above 10 WPM are impossible to count consciously.

DXCC entities, contest exchanges, and LoTW

The DXCC (DX Century Club) award is the amateur radio world’s premier station award, issued by the ARRL. An entity counts for DXCC if it is defined by the ARRL DXCC list, which currently contains 340 current entities (political entities, island groups, territories, and sovereignty regions satisfying the DXCC entity rules). To earn the basic DXCC award, an operator must have confirmed two-way contacts with at least 100 different current entities on any combination of bands and modes. Single-band DXCC awards (One Band award at 100 entities on a single band), single-mode DXCC awards (CW, Phone, Digital), and the 5-Band DXCC (B5DXCC, contacts with 100 entities on each of five HF bands: 80, 40, 20, 15, and 10 meters) are earned separately.

Contest exchanges vary by contest. The CQ World Wide DX Contest (CQ WW, the largest amateur contest, held last full weekends of October for SSB and November for CW): exchange is RST + CQ Zone (the ARRL/ITU zone number 1–40 based on geography; the US mainland is zones 3, 4, and 5). The ARRL Sweepstakes: exchange is serial number + precedence (Q/A/B/U/M/S based on license class and history) + call sign + check (last two digits of the year the call sign was first licensed) + ARRL/RAC section (84 sections within North America). The RST (Readability-Strength-Tone) system: Readability 1–5 (1 = unreadable, 5 = perfectly readable); Signal Strength 1–9 (1 = barely perceptible, 9 = extremely strong); Tone 1–9 for CW only (9 = perfect sinusoidal CW note). The exchange “599” has become conventional in contest contacts regardless of actual signal conditions; an accurate evaluation would more often be “449” or “339” for signals that are being worked through noise.

LoTW (Logbook of The World) is the ARRL’s digital QSL confirmation system. Each contact is logged in ADIF format (Amateur Data Interchange Format), signed with the operator’s private certificate (issued by ARRL after identity verification), uploaded to LoTW, and matched against the other station’s upload. Confirmed matches are creditable for DXCC, Worked All States (WAS), and other ARRL awards. Contest log submissions use CABRILLO format — a plain-text log format with a standardized header block (contest name, call sign, category, claimed score) followed by one QSO record per line (frequency, mode, date, time, call, exchange sent, call, exchange received, transmitter ID).

Tier structure for contesting and DX educators: Log Review ($5–8/month, contest rate analysis graphs, claimed score vs adjudicated score comparisons, Discord with band-opening alerts), Contest Logs ($15–25/month, CABRILLO log files from major contests with annotated QSO-rate graphs, LoTW confirmation statistics, antenna configuration notes for each contest), DX Mentor ($40–60/month capped 5–8, patron submits a recent contest log for documented analysis — rate curves, multiplier strategy, antenna selection decisions — with specific improvement notes).

iOS rates and the Apple Tax for ham radio creators

Amateur radio is a technical hobby with a substantial desktop computing component: software-defined radio (SDR) receiver software runs on a PC or Mac; WSJT-X FT8 and WSPR logging requires a computer connected to the transceiver; antenna modeling (EZNEC, 4NEC2) runs on a PC; contest logging (N1MM+, N3FJP) runs on a PC; LoTW certificate management and upload runs on a PC. This desktop-heavy workflow produces lower iOS rates than typical creator niches, moderating the Apple Tax impact. YouTube ham radio and amateur electronics content: 45–60% iOS (technical tutorial content at the lower end; general ham radio commentary and contest video at the higher end). Instagram ham radio: 62–72% iOS. TikTok ham radio: 68–78% iOS.

Frequently asked questions

Related: Patreon for amateur radio creators guide · Patreon for electronics creators · Patreon for electronic music creators · How the Apple Tax works · All explainers