This page is part of the N0NJY General Class self-study course for Technician operators upgrading to General.
G8 covers how information is encoded onto radio signals and what that means in practice. Understanding the different modulation modes — not just their names but how they behave under real operating conditions — is what allows you to choose the right tool for each situation. Understanding digital modes in operational depth prepares you to use them effectively for everyday contacts, emergency traffic, and weak-signal work. Understanding bandwidth tells you why mode selection affects your neighbors on the band as much as it affects you.
A modulation mode is the method by which information is encoded onto a radio carrier wave. The three fundamental analog modulation methods — AM, SSB, and FM — each have distinct characteristics that make them suited for different applications. New operators sometimes treat mode selection as arbitrary, but it is not. Each mode has physics that determines where it works well and where it fails.
In AM, the amplitude (strength) of the carrier varies in proportion to the audio signal. When no audio is present, the carrier is still transmitted at full power. The full AM signal consists of a carrier plus two sidebands — one above and one below the carrier frequency. Each sideband contains a complete copy of the audio information.
AM occupies twice the bandwidth of SSB for the same audio content, and two-thirds of the transmitted power is in the carrier, which carries no information. From a power efficiency standpoint, AM is wasteful. However, AM has one significant advantage: it is easy to detect. A simple envelope detector (a diode and a capacitor) recovers the audio from an AM signal. No synchronized local oscillator is required. This is why AM became the standard for broadcasting — billions of simple, inexpensive receivers could receive it without precision frequency synthesis.
AM is still used on the amateur bands, particularly on 160 and 75 meters where AM nets maintain the older tradition. AM also appears on 10 meters, and some operators use it deliberately for its broad, natural-sounding audio quality compared to SSB. The FCC designator for double-sideband AM with full carrier is A3E.
SSB is derived from AM by removing the carrier and one sideband. What remains is a single sideband containing all the audio information at a fraction of the bandwidth and power of full AM. The carrier is suppressed (typically 40 dB or more below the sideband level) and only one sideband is transmitted.
The efficiency advantage is significant: a 100-watt SSB transmitter delivers all 100 watts to the information-carrying sideband. A 100-watt AM transmitter delivers only 33 watts to each sideband (67% of power wasted in the carrier and suppressed sideband). This is why SSB is the dominant mode for HF voice — it delivers the maximum audio power to the receiver with the minimum bandwidth and power consumption.
The cost is complexity at the receiver. Because the carrier has been removed, the receiver must reinsert it at precisely the right frequency (within a few tens of Hz) for the audio to sound correct. This requires the product detector and BFO described in G7. A slightly mistuned BFO produces the characteristic "Donald Duck" sound of an improperly tuned SSB signal. The FCC designator for SSB with suppressed carrier is J3E.
As covered in G2, the convention for sideband selection is upper sideband (USB) at 14 MHz and above, lower sideband (LSB) below 14 MHz. This is not a legal requirement but a universal operating convention.
In FM, the carrier frequency varies in proportion to the audio signal amplitude. The carrier amplitude remains constant. FM has two properties that make it ideal for VHF and UHF voice:
Capture effect: When two FM signals are received on the same frequency, the stronger one captures the receiver almost completely, suppressing the weaker one. In practice, a signal only 6 dB (four times) stronger will capture the receiver, making the weaker station nearly inaudible. This is why FM repeaters work so well — when you are in range, your signal dominates and you hear nothing but the desired signal. It also means that FM is poorly suited for HF operation where multiple stations may be near the same frequency at varying signal strengths.
Noise rejection: FM receivers use a limiter stage that clips all amplitude variations before the FM detector. Since most natural and electrical noise is amplitude noise, the limiter removes it. Below the capture threshold, FM sounds noisy. Above it, FM sounds essentially noise-free. This threshold effect produces the characteristic clean/noisy transition (the "quieting threshold") that FM operators recognize.
FM on HF is limited to specific applications. On 10 meters, a small FM activity exists around 29.600 MHz. FM is standard for VHF/UHF voice operation and repeaters. The FCC designator is F3E.
Mode selection should be driven by the operating conditions and purpose:
| Situation | Best Mode | Why |
|---|---|---|
| HF DX contact, marginal conditions | SSB (or FT8 for very weak) | SSB efficiency; FT8 for below-noise conditions |
| HF ragchew, good conditions | SSB or AM | SSB efficiency; AM for audio quality preference |
| VHF/UHF local voice | FM | Capture effect; noise rejection; repeater compatibility |
| Emergency regional comms | SSB (NVIS) or digital | HF NVIS coverage; digital for formal traffic |
| Weak signal DX, below noise | FT8, JT65 | Can decode 20+ dB below noise floor |
| Keyboard QSO, HF | PSK31, JS8Call | Narrow bandwidth; conversational; readable in noise |
| Contest operation, HF | SSB (phone) or CW or RTTY | Rapid exchanges; established contest modes |
| Formal message traffic | Winlink (VARA/Pactor) or SSB phone net | Reliable store-and-forward; or voice net procedure |
Digital modes encode information in ways that allow computers and DSP processors to recover it from signals that human ears cannot decode — sometimes from signals buried deep in noise. Understanding how each major digital mode works, what it is used for, and what its limitations are allows you to choose and use them effectively.
FT8 (Franke-Taylor design, 8-FSK modulation) was released in 2017 by Joe Taylor K1JT and Steve Franke K9AN as part of the WSJT-X software suite. It has become the dominant HF digital mode worldwide because it works under conditions that defeat every other mode.
How it works: FT8 uses eight audio tones (8-FSK — 8-tone frequency shift keying) spaced 6.25 Hz apart, transmitted in 15-second intervals synchronized to UTC. The transmitted signal is 50 Hz wide. Each 15-second frame carries a complete structured message (callsigns and signal report, or locator grid). The receiver collects the entire 15-second transmission before attempting to decode it, using sophisticated error-correction algorithms. This coherent detection across the full frame is what gives FT8 its extraordinary weak-signal performance — it can decode signals 20 to 24 dB below the noise floor, which means it can decode signals you literally cannot hear at all.
Practical use: FT8 operates on standard dial frequencies on each HF band (14.074 MHz on 20 meters, 7.074 MHz on 40 meters, etc.). You set your transceiver to the dial frequency in USB mode and let WSJT-X control the audio frequency within the passband. CAT control is strongly recommended (it handles PTT and frequency automatically). The waterfall display in WSJT-X shows all active stations within the 3 kHz passband simultaneously. You click a callsign to respond; WSJT-X handles the structured exchange automatically.
Limitations: FT8 is not conversational. The exchange is fixed — callsigns, signal report (in dB), and grid square. There is no room for free-form text in a standard FT8 contact. For actual conversation, see JS8Call below. FT8 is also not suitable for formal message traffic — use Winlink for that.
FT4 uses the same technology as FT8 but with 7.5-second cycles instead of 15-second cycles, making contacts roughly twice as fast. FT4 is less sensitive than FT8 (trades sensitivity for speed) and is used primarily in digital contests where speed matters more than ultimate weak-signal capability.
JT65 was developed by Joe Taylor K1JT originally for EME (Earth-Moon-Earth, or moonbounce) communication — bouncing signals off the Moon. The round-trip delay to the Moon is 2.5 seconds, and signals are attenuated by roughly 250 to 260 dB. JT65 achieves this through very slow transmissions (60-second cycles) and powerful error correction. On HF, JT65 was the predecessor to FT8 for weak-signal work. JT9 is a narrower, more efficient variant optimized for the crowded HF bands. Both are now largely superseded by FT8 for general HF weak-signal work but remain in use for EME and specific applications.
PSK31 (Phase Shift Keying, 31 baud) was developed by Peter Martinez G3PLX in 1998 and remains a popular mode for real-time keyboard conversation on HF. It uses binary phase shift keying at 31.25 baud, producing a signal approximately 31 Hz wide — narrower than any other practical voice or data mode. The narrow bandwidth makes PSK31 efficient on crowded bands and gives it excellent weak-signal performance for its data rate.
Practical use: PSK31 is conversational. You type at the keyboard and the other station reads your text in real time. Response time is limited only by your typing speed. On 20 meters, the PSK31 calling frequency is 14.070 MHz. On 40 meters, 7.070 MHz. The most common software is fldigi, which runs on Linux, Windows, and Mac. PSK31 is one of the best modes for casual digital conversation and for demonstrating digital modes to new operators because the exchange is immediately understandable.
Higher-speed variants (PSK63, PSK125) use wider bandwidth for faster data rates when conditions permit.
RTTY (Radio TeleTYpe) is the oldest digital mode still in widespread amateur use, dating to the 1940s in its original mechanical teletype form. Modern RTTY uses audio frequency shift keying (AFSK) through a sound card, or direct FSK keying of the transmitter. The standard amateur RTTY parameters are 45.45 baud (baud rate) and 170 Hz shift between mark and space tones, producing an occupied bandwidth of approximately 250 to 300 Hz.
RTTY uses the Baudot (ITA2) character set rather than ASCII, which limits it to uppercase letters, digits, and a limited set of punctuation. RTTY remains popular in HF contests (RTTY contests are among the largest in amateur radio) and for operators who prefer its retro character. The software fldigi handles RTTY along with PSK31 and many other modes.
JS8Call was developed by Jordan Sherer KN4CRD as a conversational adaptation of FT8. It retains the weak-signal capability of FT8's modulation scheme but allows free-form text messages rather than the fixed structured exchange. JS8Call also supports store-and-forward messaging, relay through other stations, and an informal network for passing text traffic. It is increasingly popular in emergency communications because it combines FT8's ability to work in poor conditions with the free-form messaging needed for actual emergency traffic.
Winlink is a system for sending and receiving email-format messages over amateur radio, using a network of internet-connected gateway stations worldwide. A Winlink station connects to a gateway using one of several HF (or VHF/UHF) protocols, sends its outgoing messages, and receives queued incoming messages. The messages are then delivered to or from conventional email addresses.
Winlink supports several connecting protocols:
Winlink in emergency communications: Winlink is used extensively in ARES, RACES, and served agency communications precisely because it can pass formal formatted messages (ICS forms, FEMA forms, NTS formal message format) over HF when internet and telephone infrastructure are unavailable. A Winlink station can send a message to an emergency coordination center email address over HF, have it delivered to the served agency's email system through an internet gateway, and receive responses back — all without any functioning internet at the disaster site.
APRS (Automatic Packet Reporting System) was developed by Bob Bruninga WB4APR and uses the AX.25 packet radio protocol to transmit position, weather, and status information in real time. APRS packets are transmitted on a standard national frequency (144.390 MHz in North America) and are received by digipeaters (digital repeaters) that re-transmit them to increase coverage, and by iGates (internet gateways) that forward them to the APRS-IS (APRS Internet Service) network, making them visible worldwide on sites like aprs.fi.
Common APRS uses:
APRS on HF: A secondary APRS frequency exists at 10.1515 MHz (30 meters) for HF APRS, which provides much longer range for mobile stations beyond VHF coverage. HF APRS uses 300 baud rather than 1200 baud to account for HF channel characteristics.
AX.25 is the data link layer protocol used for amateur packet radio, APRS, and Winlink VHF connections. It is derived from the X.25 protocol used in early commercial data networks, adapted for amateur radio with callsign-based addressing. Standard VHF packet operates at 1200 baud AFSK (Bell 202 tones at 1200 and 2200 Hz). Higher-speed 9600 baud packet uses direct FSK modulation (bypassing the audio chain) and requires a radio with a direct discriminator output and a modulator that bypasses the audio limiter.
AX.25 packets include a source callsign, destination callsign, up to eight digipeater callsigns for relay routing, a control field, a PID (protocol identifier), and information data. The header information is transmitted in plain text, satisfying the FCC requirement that the meaning of amateur transmissions must be publicly accessible — the protocol is fully documented and openly available.
Bandwidth is the range of frequencies occupied by a transmitted signal. Every mode has a characteristic bandwidth determined by its modulation method and data rate. The FCC requires that transmissions be no wider than necessary for the information being transmitted (47 CFR §97.307). Beyond the legal requirement, bandwidth matters because the HF bands are shared. A wide signal takes up space that other operators cannot use.
The occupied bandwidth of a signal is the frequency range containing a specified percentage of the total transmitted power. The FCC typically measures occupied bandwidth as the range containing 99% of the total power. A signal that is nominally 2.4 kHz wide for SSB may have spectral components extending slightly beyond that if the modulator is overdriven — this is the splatter that adjacent operators complain about.
| Mode | Approx. Occupied Bandwidth | Notes |
|---|---|---|
| CW (at 25 WPM) | 50–100 Hz | Keying speed and waveform shaping determine bandwidth |
| PSK31 | 31 Hz | The narrowest practical digital mode |
| FT8 | 50 Hz | Fixed; software controls frequency precisely |
| FT4 | 90 Hz | Wider than FT8 due to faster symbol rate |
| RTTY (170 Hz shift) | 250–300 Hz | Mark and space tones plus sidebands |
| PSK63 | 63 Hz | Twice the speed of PSK31, twice the bandwidth |
| SSB voice (normal) | 2.4–2.8 kHz | Audio bandwidth determines RF bandwidth |
| AM voice (double sideband) | 6 kHz | Two full audio sidebands plus carrier |
| NFM (narrow FM, VHF) | 10–15 kHz | 5 kHz deviation × 2 + audio sidebands |
| WFM (broadcast FM) | 150–200 kHz | 75 kHz deviation; stereo; pilot tone |
On a crowded band, adjacent channel spacing must accommodate the bandwidth of the signals being used. On 20-meter SSB, contacts are typically separated by 3 kHz to avoid mutual interference. On a digital segment using PSK31 or FT8, signals can be packed much more closely because they occupy far less bandwidth. This is why digital sub-bands can support many more simultaneous contacts than phone sub-bands of the same size.
When a transmitter is overdriven and produces splatter — distortion products extending beyond the nominal signal bandwidth — the effective occupied bandwidth grows dramatically. A badly overdriven SSB station that nominally occupies 2.4 kHz may actually occupy 10 kHz or more with its intermodulation products. This is why ALC adjustment (covered in G4) is not just a matter of audio quality — it is a matter of being a considerate band occupant.
Spread spectrum is a modulation technique in which the transmitted signal intentionally occupies a bandwidth much wider than the minimum necessary to transmit the information. This seems counterintuitive — why use more bandwidth than you need? — but spread spectrum provides two advantages: resistance to interference and jamming, and the ability for multiple users to share the same frequency band simultaneously.
In a spread spectrum system, the transmitted signal is spread across a wide bandwidth using a pseudo-random spreading code known to both transmitter and receiver. The receiver uses the same code to "despread" the signal, recovering the original data. To an observer without the spreading code, the signal appears as low-level noise spread across a wide bandwidth — it is not recognizable as a discrete signal on a conventional spectrum analyzer.
Direct sequence spread spectrum (DSSS): The data stream is multiplied by a high-rate pseudo-random binary sequence (the spreading code), spreading the signal across a bandwidth determined by the chip rate of the code. GPS signals use DSSS. 802.11b Wi-Fi uses DSSS.
Frequency hopping spread spectrum (FHSS): The transmitted frequency hops rapidly through a sequence of frequencies according to a pseudo-random pattern known to both stations. Each dwell time at one frequency is brief; the total bandwidth covered by all the hops is the spread spectrum bandwidth. Bluetooth uses FHSS.
Spread spectrum is authorized in the amateur service under specific conditions defined in 47 CFR §97.311:
The prohibition below 222 MHz is the most important exam point. Spread spectrum is not permitted on HF or on the 6-meter, 2-meter, or 70-centimeter (144 MHz) bands. It is permitted on 222 MHz and higher.
In practice, spread spectrum operation by amateur stations is relatively rare. The main amateur spread spectrum activity is experimental, using software-defined radio platforms to explore the techniques. The more practical relevance for most operators is understanding that the Wi-Fi, Bluetooth, and cellular signals that increasingly pollute the spectrum near amateur radio installations are all spread spectrum systems.
The FCC uses a standardized system of emission type designators in license documents and regulations. New operators often confuse these designators with mode names. The designator is a technical description of the signal; the mode name is what operators call it in practice.
| Designator | What It Means | Common Name |
|---|---|---|
| A1A | Amplitude modulation, on-off keying, telegraphy (no audio) | CW (Morse code) |
| A2A | Amplitude modulation, on-off keying with audio tone | Tone-modulated CW |
| A3E | Amplitude modulation, double sideband, full carrier, telephony | AM phone |
| F1B | Frequency modulation, FSK, telegraphy | RTTY (direct FSK) |
| F2B | Frequency modulation, audio-tone modulated FSK, telegraphy | AFSK RTTY |
| F3E | Frequency modulation, analog telephony | FM phone |
| G3E | Phase modulation, analog telephony | PM phone |
| J3E | Single sideband, suppressed carrier, telephony | SSB phone |
| J2B | Single sideband, suppressed carrier, FSK | SSB RTTY (used in some systems) |
The structure of the designator is: first character = type of modulation of main carrier (A=AM, F=FM, G=PM, J=SSB suppressed carrier); second character = nature of signal modulating the carrier (1=digital no audio, 2=digital with audio, 3=analog); third character = type of information (A=telegraphy/CW, B=telegraphy/printing/RTTY, E=telephony/voice).
A transmitter ideally produces only the desired signal at the desired frequency. In practice, non-linearities in the amplifier stages produce unwanted emissions — spurious emissions — at other frequencies. The FCC requires that these be suppressed to specified levels.
Harmonics: Integer multiples of the transmit frequency. A 14 MHz signal produces harmonics at 28 MHz, 42 MHz, 56 MHz, etc. These harmonics must be suppressed to at least 43 dB below the fundamental (for transmitters over 5 watts). A low-pass filter at the transmitter output is the standard solution, as covered in G7.
Intermodulation distortion (IMD): When two signals are present in a non-linear device, they produce sum and difference products at additional frequencies. In an overdriven SSB transmitter, the two-tone products of the voice audio produce IMD components at frequencies adjacent to the desired signal — this is what operators hear as splatter. The solution is proper ALC adjustment and microphone gain setting.
Parasitic oscillations: Unintended oscillations at frequencies unrelated to the input signal, caused by stray inductance and capacitance in amplifier stages. Parasitic suppressors (small resistors or ferrite beads in the amplifier circuit) prevent parasitic oscillations.
The G8 subelement covers signals and emissions as tested in the 2023–2027 FCC General Class question pool. All pool questions are covered below.
Q1 (G8A01) — How is an FSK signal generated?
Q2 (G8A02) — What is the name of the process that changes the phase angle of an RF wave to convey information?
Q3 (G8A03) — What is the name of the process that changes the instantaneous frequency deviation of an RF wave to convey information?
Q4 (G8A04) — What emission is produced by a reactance modulator connected to a transmitter RF amplifier stage?
Q5 (G8A05) — What type of modulation varies only the instantaneous amplitude of the transmitted signal?
Q6 (G8A06) — Which of the following is characteristic of QPSK31?
Q7 (G8A07) — Which of the following phone emissions uses the narrowest bandwidth?
Q8 (G8A08) — Which of the following is an advantage of using single sideband, as compared to other analog voice modes?
Q9 (G8A09) — Which of the following HF digital modes is the narrowest in bandwidth?
Q10 (G8A10) — What is the name of the modulation mode where the signal amplitude is reduced to a very low level to encode data bits, then restored to its normal level?
Q11 (G8A11) — What is the modulation envelope in an AM signal?
Q12 (G8B01) — What is the term for the process of combining two signals in a non-linear device to produce additional signals at the sum and difference frequencies?
Q13 (G8B02) — If a 1 MHz signal is mixed with a 3.785 MHz signal, what is the frequency of the resulting difference frequency?
Q14 (G8B03) — If a 14.35 MHz signal is mixed with a 455 kHz signal, what is the resulting lower frequency?
Q15 (G8B04) — If a 14.35 MHz signal is mixed with a 455 kHz signal, what is the resulting upper frequency?
Q16 (G8B05) — What is the approximate bandwidth of a PACTOR 3 signal at maximum data rate?
Q17 (G8B06) — What is the total bandwidth of an FM phone transmission having a 5 kHz deviation and a 3 kHz modulating frequency?
Q18 (G8B07) — What is the frequency deviation for a 12.21-MHz reactance-modulated oscillator in a 5-kHz deviation, 146.52 MHz FM phone transmitter?
Q19 (G8B08) — Why is it important to know the duty cycle of the mode you are using when calculating PEP or average power?
Q20 (G8B09) — What is the relationship between the PEP and the average power of a CW signal?
Q21 (G8B10) — What is the frequency deviation of an FM phone transmission having a maximum frequency swing of plus or minus 7.5 kHz?
Q22 (G8B11) — What is the difference between the frequencies of two signals that produce a 1000 Hz audio beat note when received by a CW receiver?
Q23 (G8C01) — On what frequencies are spread spectrum transmissions authorized?
Q24 (G8C02) — What describes a spread spectrum communication system?
Q25 (G8C03) — What is the term for the reduction in receiver sensitivity caused by a strong signal near the desired frequency?
Q26 (G8C04) — What signal is used to convey position data in the APRS network?
Q27 (G8C05) — What is the purpose of the automatic position reporting system (APRS)?
Q28 (G8C06) — What is the most common data rate used for APRS packets?
Q29 (G8C07) — Which of the following digital modes uses variable-length coding for data compression?
Q30 (G8C08) — In what segment of the 20-meter band is PSK31 most commonly found?
Q31 (G8C09) — How does the receiver in a PACTOR system acquire the incoming signal?
Q32 (G8C10) — What is Winlink?
Q33 (G8C11) — What is the difference between direct-sequence spread spectrum (DSSS) and frequency-hopping spread spectrum (FHSS)?