FM vs. SSB: Bandwidth, Emission Choice, and the Sideband Convention

Concept

T7A introduced modulation as the general idea of varying a carrier's amplitude, frequency, or phase to load information onto it, without settling on which property any particular mode actually varies. This lesson gets concrete about two of the amateur world's most common answers to that question: single sideband, SSB, and frequency modulation, FM. Both are voice modes you'll hear constantly on the air, but they arrive at 'carrying a voice' by two genuinely different routes, and each has habits of behavior that make it the right tool for some situations and a poor one for others. T3A through T3C spent real time on how a radio signal actually behaves once it leaves an antenna: fading, multipath, the difference between a strong local path and a marginal long-distance one. That behavior is exactly what decides which of these two modes an operator reaches for, so this lesson is where those two threads, modulation vocabulary and propagation reality, finally meet and produce a genuine operating decision.

Concept

Start from plain amplitude modulation: vary a carrier's strength in step with an audio signal, and the result is a carrier plus two mirror-image copies of that audio, one shifted above the carrier's frequency and one shifted below it, called sidebands because they sit to each side of the carrier. Both sidebands carry the same information, so transmitting both is redundant, and the carrier itself carries no information at all, just a reference tone sitting in the middle. Single sideband radio strips away that redundancy: filter out the carrier and one of the two sidebands entirely, and transmit only the one sideband that's left. Nothing about the recovered audio is lost, since the same information rode on either sideband, but every watt the transmitter produces and every hertz of spectrum used now goes toward the one signal that actually reaches the listener, rather than half of it going to waste on a spare copy and a silent reference tone. SSB, in one sentence, is amplitude modulation with everything but the useful part filtered away before transmission.

Concept

FM takes an entirely different property of the carrier and varies that instead: the carrier's frequency shifts slightly higher and lower in step with the audio, while its amplitude stays constant throughout. That constant-amplitude property matters a great deal in practice. Most of the electrical noise a receiver picks up, ignition noise, motor noise, atmospheric static, shows up primarily as amplitude variation, and a well-designed FM receiver can be built to simply ignore amplitude altogether and pay attention only to the frequency shifting underneath, discarding most of that noise before it ever reaches your ears. FM receivers also tend to lock onto whichever signal arriving on a channel is strongest and largely disregard weaker ones sharing that same frequency, a behavior called capture effect, useful on a busy repeater where you want the strongest, clearest station coming through cleanly rather than a mixed jumble of several at once. But it also means FM doesn't degrade gracefully: below a certain signal strength, capture effect breaks down and the audio doesn't fade gently, it turns to noise fairly abruptly.

Concept

Every mode's bandwidth, meaning how much of the radio spectrum it actually occupies, is set by how much information it has to carry per second, and voice modes and picture modes sit at very different points on that scale. CW, on-off keying of a carrier to send Morse code, carries the least information of anything discussed here: only two states, on or off, changing at a human sending speed, so it fits into the narrowest sliver of spectrum of any mode in common amateur use. SSB voice needs to carry the full range of frequencies present in a human voice, so it occupies a slice of spectrum roughly matching that vocal range, narrow compared to many signals, but meaningfully wider than CW's simple on-off pattern. FM voice, carrying that same vocal information, ends up occupying substantially more spectrum than SSB does for an equivalent voice signal, because FM's design deliberately spreads the signal's energy across a wider swing in frequency; that extra spread is exactly what buys FM its noise-rejecting and capture-effect behavior from the previous block, so the bandwidth cost and the noise immunity are two sides of the same design choice, not two unrelated facts. Fast-scan television goes further still: encoding a full moving picture, redrawn many times a second, carries vastly more information than any voice signal, and its bandwidth reflects that, wider than FM voice by roughly the same kind of proportion that FM voice is wider than CW. Line them up from narrowest to widest and the order runs CW, then SSB voice, then FM voice, then fast-scan TV, tracking the amount of information each one is actually carrying, not any inherent property of amplitude versus frequency modulation on its own.

Concept

SSB comes in two flavors, since either the upper or the lower half of that mirror-image pair can be the one kept, and amateur practice long ago settled on a convention for which one to use where. Below 10 MHz, hams by convention use the lower half; at 10 MHz and above, including every VHF and UHF band where SSB sees use, the convention switches to the upper half. Neither choice is technically superior to the other; a receiver set up to demodulate one can't correctly pull audio out of a transmission using the other, so the entire point of the convention is that everyone operating on a given band uses the same one. Get it backwards and your voice comes out garbled and unintelligible on the other end, not because anything is broken, but because the receiving station's radio is demodulating the wrong half of that pair for where you actually put your signal.

Concept

Bandwidth and signal strength trade against each other in a way that makes SSB the natural choice whenever a contact is marginal. A transmitter only has so much power to give a signal, and spreading that fixed amount of power across a wider slice of spectrum means less of it lands in any given sliver of frequency, which is exactly where a receiver's own noise lives too. Concentrate the same power into SSB's narrower bandwidth instead, and more of that power is packed into the same slice a receiver is listening in, giving a marginal signal a better chance of standing out above the noise floor. That's precisely the situation T3A through T3C described: a distant contact riding a fading, multipath-battered, or barely-there propagation path, right at the edge of readability. FM's constant-amplitude, capture-effect design needs a signal comfortably above a threshold to work its magic at all; below that threshold it doesn't fade gracefully, it falls apart, as the previous blocks already described. SSB has no such cliff: a weak SSB signal gets harder to copy as it weakens, but it degrades gradually rather than suddenly, which is exactly the behavior an operator wants when reaching for a station at the ragged edge of what a path will support.

Concept

Flip the situation around, a strong, reliable, comfortably-above-threshold local signal, the kind a VHF or UHF repeater or a nearby packet link is built around, and FM's tradeoffs stop being drawbacks and start being exactly what the job calls for. A repeater exists to give local stations a strong, consistent path, so the signal reaching it rarely sits anywhere near FM's threshold problem, and capture effect becomes a genuine convenience: it keeps a channel intelligible even with a little interference or a second station briefly overlapping the first, rather than producing an unreadable mix of both. Packet radio rides that same reliability for a different reason: a data connection needs a receiver that reconstructs the sender's tones cleanly and consistently, and FM's noise-rejecting, constant-amplitude behavior gives a packet modem a clean signal to decode as long as the path stays within FM's comfortable range, which a local VHF link generally does. The advantages that matter least on a long, marginal HF contact become the deciding factor once the contact is short-range and solid; the mode choice follows the path itself, not a fixed rule about which mode is simply better.

Concept

Put the whole lesson together and mode choice stops looking like an arbitrary rule to memorize and starts looking like the product of three questions answered in sequence: how much information does this signal need to carry, deciding its inherent bandwidth, narrowest to widest running CW, then SSB voice, then FM voice, then fast-scan TV; is the path to the other station strong and reliable or weak and marginal, deciding whether FM's threshold behavior or SSB's graceful degradation suits it better; and, if SSB is the choice, which half of that sideband pair convention calls for on this particular band. T7A gave this lesson its vocabulary and T3A through T3C gave it the propagation reality that vocabulary gets applied to; T8A is where those two threads produce an actual operating decision. From here the course turns to the non-voice and image modes waiting just past SSB and FM: CW's narrow bandwidth already introduced here, and fast-scan television's wide one, both get a closer look next.