Decibels Demystified: The Only Math That Really Pays Off

Why the hobby speaks in decibels

Every specification that matters in amateur radio is written in decibels. Antenna gain, coax loss, filter rejection, receiver sensitivity, intermodulation performance, and the difference between a signal you can copy and one you cannot are all quoted as decibel figures. Operators who never get comfortable with the unit end up buying equipment on faith and troubleshooting by intuition. Operators who spend one evening on it can look at a specification sheet and know immediately whether the product is worth the money.

The reason radio uses decibels rather than plain ratios is that the quantities involved span an absurd range. A receiver might resolve a signal of a fraction of a microvolt while the transmitter beside it produces a hundred volts. Writing those numbers in the same table requires either a great many zeros or a logarithmic scale, and the logarithmic scale has a second advantage: gains and losses along a signal path simply add up instead of multiplying.

The three numbers to memorize

Three decibels is a factor of two in power. Ten decibels is a factor of ten. Six decibels is a factor of four, which is just two threes added together. That is the entire toolkit for practical work, because every other figure you meet can be assembled from those three.

Twenty decibels is ten plus ten, so a factor of a hundred. Thirty decibels is a factor of a thousand. Thirteen decibels is ten plus three, so a factor of twenty. Seventeen decibels is twenty minus three, so a factor of fifty. Negative numbers mean division: minus three decibels is half the power, minus twenty is one hundredth. With those combinations you can convert almost anything you will encounter without touching a calculator, and being able to do it in your head while looking at a product page is genuinely useful.

Power ratios versus voltage ratios

Here is the trap that catches nearly everyone once. The decibel is defined for power ratios as ten times the logarithm of the ratio. For voltage or current ratios, the multiplier is twenty rather than ten, because power is proportional to the square of voltage. This means six decibels is four times the power but only twice the voltage, and twenty decibels is a hundred times the power but only ten times the voltage.

In practice, assume power unless the specification explicitly says otherwise, because almost everything in radio is quoted in power terms. The place voltage decibels appear most often is in receiver sensitivity figures expressed in dBuV, and in oscilloscope or signal-generator documentation. If a number seems off by a factor of two somewhere, check which definition is in play before you assume the equipment is broken.

dBm, dBW, dBd, dBi: the reference matters

A bare decibel figure is a ratio and says nothing absolute. Attach a reference and it becomes a measurement. dBm means decibels relative to one milliwatt, which is the standard currency of RF engineering: zero dBm is one milliwatt, thirty dBm is one watt, and fifty dBm is a hundred watts. dBW is the same idea referenced to one watt, so a hundred watts is twenty dBW.

For antennas, dBi means gain relative to an isotropic radiator, a theoretical point that radiates equally in all directions. dBd means gain relative to a half-wave dipole, which already has 2.15 dBi of gain by virtue of its pattern. The gap between the two references is therefore about 2.15 decibels, and marketing departments are extremely aware of which number looks better. When comparing two antennas, convert both to the same reference before believing either figure. An antenna advertised at 9 dBi and one advertised at 7 dBd are the same antenna.

Building a link budget in your head

Because decibels add, you can trace a whole signal path with arithmetic a child could do. Start with your transmitter output in dBm. Subtract the coax loss. Subtract connector and choke losses if they are significant. Add the antenna gain. The result is your effective radiated power, and the same path in reverse gives you the receive side.

Work an example. A hundred-watt transmitter is fifty dBm. A hundred feet of average coax on 20 meters costs about one decibel, so forty-nine dBm reaches the antenna. A modest three-element beam adds seven dBd, giving fifty-six dBm of effective radiated power, which is about four hundred watts in the favored direction. Now change one variable: use cheap RG-58 on 10 meters instead, where the same run costs three decibels, and you have thrown away half your power before it reaches the antenna. That is the calculation that convinces people to buy better coax, and it takes fifteen seconds.

S-units and the S9 lie

Signal reports use S-units, where the convention is that one S-unit equals six decibels and S9 corresponds to fifty microvolts at the receiver input on HF. If that convention were followed by manufacturers, S-meter readings would be comparable between radios. It is not followed. Real S-meters are frequently non-linear, often compressed at the low end, and vary between models by ten decibels or more at the same input level.

The practical consequences are worth internalizing. A report of S7 from one station and S9 from another tells you almost nothing about the difference in received signal. A change of one S-unit on your own radio, however, is a meaningful relative measurement, because the same meter is making both readings. Use your S-meter for before-and-after comparisons on your own station, and use your ears for absolute judgments about whether a signal is workable. When you want a real number, use the signal-to-noise report that digital modes provide, which is measured rather than estimated.

Where decibels change decisions

Three decibels is the threshold where a change becomes reliably audible on a weak signal, and it is the standard bandwidth reference for filters. Six decibels of extra loss halves your effective range in free space, which is why VHF and UHF operators care so much about feedline quality. Ten decibels is the difference between an unreadable signal and a comfortable conversation, which is why a modest beam antenna transforms a station more than any amount of extra transmit power.

That last point deserves emphasis, because it is where the arithmetic overturns intuition. Going from a hundred watts to two hundred watts is three decibels, a barely perceptible improvement, and costs real money. Replacing a low dipole with the same dipole ten feet higher, or adding a small beam, can deliver six to ten decibels in the direction you care about and improves reception by the same amount, which the amplifier does not do at all. The decibel is the tool that makes that comparison obvious, and that is why it is worth an evening of your time.