QSB and Fading on HF: Why Signals Rise and Fall and What to Do

You are in the middle of a QSO on 20 meters when the other station slides from a solid S9 down to the noise, comes back a minute later and then drops again. That rhythmic rise and fall is QSB fading, named after the Q code that asks whether your signals are fading, and it is one of the most common experiences on the HF bands. It is not a fault in your radio or a problem with your antenna. It is a natural result of the way short-wave signals travel through a moving ionosphere. This guide explains what causes fading, how to recognize the different kinds, and what you can do at the receiving end to make the best of it.

What QSB Means

QSB is one of the standard Q codes used on the air: sent as a question, QSB? means are my signals fading, and sent as a statement, QSB means your signals are fading. Our guide to Q codes lists the others that you will hear. In everyday use, hams say there is a lot of QSB when a signal changes strength over seconds or minutes. The term covers slow changes that last a minute or more and fast flutter that changes many times per second, although the physical causes of the two are different.

The key to understanding fading is that a short-wave signal rarely arrives by one path. The Wikipedia article on skywave propagation explains that at any distance sky waves will fade, because the ionospheric layer that reflects the signal is not fixed, but undulates like the surface of the ocean. It adds that even more serious fading can occur when signals arrive via two or more paths. That second statement is the root of most of what you hear.

The Main Causes

Multipath interference

A signal may reach your antenna by one hop and by two hops, by an E-layer reflection and an F-layer reflection, or by a refraction that splits into two rays. Each path has a different length, so the signals arrive with a different phase. When they are in phase, they add and the signal is strong; when they are out of phase, they cancel and the signal fades. Because the layer moves, the path lengths change by fractions of a wavelength, and the signal swings between the two states. The wavelength at 14 MHz is about 21 meters, so a change of only a few meters in the path length is enough to turn a peak into a null. This is the most common cause of QSB on long paths.

Polarization changes

When a radio wave passes through the ionosphere in the presence of the Earth’s magnetic field, its polarization can rotate. A signal that leaves your antenna horizontally polarized may arrive with a different polarization, and the strength that your antenna picks up depends on the match. As the ionization changes, the polarization changes with it, and the received level swings. This is one reason why receiving with both a horizontal and a vertical antenna can help, and why our article on vertical versus horizontal polarization notes that the polarization of the received signal is often mixed after a skywave path.

Focusing and absorption

The ionosphere does not only reflect; it also absorbs and sometimes focuses signals. A change in the absorption in the D layer, for example at sunrise or sunset, brings a slow rise or fall over minutes. Irregularities in the F layer can focus the signal and produce a short peak. Solar flares can cause sudden changes, as described in our guide to the HF radio blackout, and such a change is not periodic like multipath fading.

Types of Fading Compared

Selective Fading and Why Voice Sounds Strange

Fading is not always uniform across the whole signal. Because the cancellation depends on the phase, and the phase depends on the frequency, one part of the signal may be in a null while another part is strong. This is called selective fading. On an AM broadcast signal, you can hear it as a distorted, hollow sound as the carrier fades and the sidebands remain. On SSB it produces a signal that is intelligible but metallic, with certain letters lost. A wide signal is affected more than a narrow one, since the width spans more of the interfering pattern. That is one reason why CW and narrow digital modes tend to be less disturbed by selective fading than a broad voice transmission.

Flutter and Polar Paths

When a path crosses the polar region or the auroral zone, the signal often takes on a rapid, raspy flutter, sometimes called polar flutter or auroral flutter. The reflecting medium is disturbed and moving quickly, and the signal is scattered in many directions. The flutter can sound like a hoarse buzz on a CW signal or make voice hard to understand even when the level is high. It is more likely during geomagnetic disturbances, which you can watch through the K index described in our guide to solar flux, A index and K index. When flutter is present, trying a different path, such as the long path or a different band, may be better than changing your receiver.

What You Can Do at the Receiver

You cannot prevent fading, but the receiver can make it easier to follow. Several settings and techniques are worth knowing.

• Use the AGC properly. A fast attack with a slow release helps the receiver follow the fade without pumping on every noise burst. A very fast release raises the noise in the nulls. Our article on AGC attack, decay and hang time shows how to set it.
• Narrow the bandwidth wisely. A narrow filter reduces noise in the nulls and the effect of selective fading, but too narrow a filter can make an unstable signal harder to copy. Our guide to receive filter bandwidth covers the trade-off.
• Try another antenna. The nulls at one antenna are not the same as at another. Switching between a dipole and a vertical, or between two beam headings, often rescues a signal that has just disappeared.
• Use diversity reception. Two receivers on two antennas of different polarization, with the audio from both mixed in stereo, can follow a fade better than one. Some radios offer a dual receiver for this purpose; with SDR software the audio of two antennas can be combined.
• Use synchronous detection. Some receivers can lock to the carrier of an AM signal, which reduces the distortion of selective fading by choosing one sideband.

Operating Techniques When the Signal Fades

Fading affects how you operate as well. Repeat important information, such as callsigns and report numbers, so that a null does not hide it. In a contest, send the exchange twice if the other station seems unstable. In a net, use short phrases and phonetics for names. In CW, a slightly slower speed gives more time for copying in the peaks. If you are the one who is fading, the other operator may suggest a different frequency or a different time, and, if the fading is deep, changing band by one step is sometimes enough.

Digital modes deal with fading in their own way. FT8 and other modes with forward error correction can decode a signal even when parts of the message are lost, which is why they work well on paths where voice is almost unreadable. The same property makes them useful for checking whether a path exists at all. With a long fade, the decoder may lose a whole transmission, but the next period may be received without a problem. When you see a station that decodes at some periods and not at others, you are watching QSB in the logs.

Fading by Band and Time of Day

The pattern of fading changes with the band and the hour. On the lower bands, 160, 80 and 40 meters, the strongest effects appear around sunrise and sunset, when the D layer forms or vanishes and the paths change quickly. A signal from the other side of the world may rise out of the noise for ten minutes and then disappear as the terminator moves. On 20 meters during the day, long paths with several hops show multipath fading with periods of a few seconds. On 15 and 10 meters, when the band is open by a single strong F2 reflection, signals are often steady and strong, and the sudden onset of fading is itself a sign that the path is marginal and about to close.

Seasonal and solar conditions matter too. Near the maximum of the solar cycle the upper bands support many strong single-hop paths, so there is less fading, while at the minimum the high bands rarely open and the lower bands carry most of the traffic, with the fading that goes with multi-hop paths. Geomagnetic storms add flutter and deep fades at high latitudes. If you live in northern Europe, as many readers of this site do, trans-polar paths to North America and Asia are especially prone to flutter in disturbed periods. For general background on how fading and signal variation arise in every kind of radio link, the Wikipedia article on fading is a useful companion, and it also describes diversity schemes in a broader context.

Fading or Interference? Telling Them Apart

Not every change in level is QSB. A signal that drops when a neighbor switches on a device may be suffering from local noise, as described in our article on finding RF noise, and a signal that disappears and returns in a regular rhythm tied to the other station’s keying may be the result of a power supply or an automatic level control problem at the transmitter. To separate the cases, check whether other stations on the same band fade at the same time. If all of them do, the cause is the path. If only one does, the problem is at the other end, or it lies in a path that is special to that signal. If the fading appears on the noise as well, check the AGC and the noise reduction settings, since an aggressive noise reduction can create a sound that resembles fading.

Seeing and Measuring Fading

If you want to learn the pattern of fading on a path, listen to a beacon or a steady carrier and watch the S meter or the waterfall. A waterfall display shows the multipath as streaks that move across the signal, with the fading appearing at different frequencies across the width of the trace. A spectrum display on an SDR, or a panadapter, is a useful aid because you can see the selective fading directly. A logging tool for the received level, even a simple script that records the S meter value every second, shows whether the fade has a regular period. A regular period of a few seconds usually points to multipath between two stable modes, while an irregular pattern suggests several paths or a moving layer.

The Bottom Line

QSB fading is the normal rise and fall of an HF signal, caused mainly by two or more paths interfering, by changes in polarization and by moving absorption in the ionosphere. It comes as slow fading over minutes, as multipath fading over seconds, as selective fading that distorts wide signals and as fast flutter on polar paths. You cannot cure it, but a sensible AGC setting, a suitable filter bandwidth, a second antenna, repeated information and error-correcting digital modes make signals far easier to copy. Treat fading as information about the path rather than a fault, and try another antenna or band before blaming the equipment.

Frequently Asked Questions

What does QSB mean?

QSB is a Q code meaning that signals are fading. Sent as a question, QSB? asks whether your signals are fading.

What causes QSB on HF?

Mostly interference between signals that arrive by two or more paths, together with changes in polarization and absorption as the ionosphere moves.

What is selective fading?

Fading that affects only part of the frequency range of the signal, so some frequency components fade while others do not. It distorts wide signals such as AM and SSB voice.

Why does my signal flutter on a polar path?

Paths through the auroral and polar regions are scattered by a disturbed and moving ionosphere, which adds a rapid flutter, especially in geomagnetic disturbances.

Does a better antenna stop QSB?

No, but a second antenna of a different polarization or direction can help, because the fades at the two antennas are not the same.

Do digital modes cope better with fading?

Modes with forward error correction, such as FT8, can decode messages from signals that fade, and they often work where voice is hard to read.