RF Power in dB: Build an Honest HF Station Budget
RF Power in dB: Build an Honest HF Station Budget
Watts describe power at a defined point. Decibels compare two powers or levels. A useful station analysis needs both—and it must keep reference planes, antenna-gain references, modulation and time averaging straight.
“Think in dB” is good advice for comparing station changes, but “stop thinking in watts” goes too far. A dB budget can show that 1,500 W is only 1.76 dB above 1,000 W. Watts, volts, amperes, waveform and time still determine amplifier limits, heating, arcing and RF exposure.
Comparison boundary: a transmitter-power change produces the same dB change at the receiver only while the transmitter, feed system, antennas, propagation path, polarization, receiver bandwidth, noise and all nonlinear or protective behaviour remain unchanged. An HF path is time-varying; a correct arithmetic ratio is not a guaranteed signal report or contact outcome.
dB Is a Ratio; dBW and dBm Are Referenced Levels
Recommendation ITU-R V.574-5 defines the decibel from a power ratio:
power ratio in dB = 10 log10(P2 / P1)
field or voltage ratio in dB = 20 log10(X2 / X1) only when the quantities have the required power relationship and the impedances are handled consistently.
A bare value such as 3 dB is a ratio. A level in dBW is referenced to 1 W; a level in dBm is referenced to 1 mW. Consequently, 0 dBW is 1 W, 30 dBm is 1 W, and the numerical difference between dBm and dBW is 30 dB.
| Power | Level | Change from 1 W |
|---|---|---|
| 1 W | 0 dBW = 30 dBm | 0 dB |
| 10 W | 10 dBW = 40 dBm | +10.00 dB |
| 100 W | 20 dBW = 50 dBm | +20.00 dB |
| 500 W | 26.99 dBW = 56.99 dBm | +26.99 dB |
| 1,000 W | 30 dBW = 60 dBm | +30.00 dB |
| 1,500 W | 31.76 dBW = 61.76 dBm | +31.76 dB |
The rounded mental rules are useful: 2× power is +3.01 dB, 4× is +6.02 dB, 10× is +10 dB and 100× is +20 dB. A 3 dB power increase is a factor of about two; it is not a doubling of electric-field amplitude. Under unchanged far-field conditions, doubling power raises field amplitude by √2, about 1.414.
What the Familiar Power Steps Really Buy
| Change | Exact power-ratio change | Nominal 6 dB/S-unit equivalent |
|---|---|---|
| 1 W → 10 W | +10.00 dB | 1.67 S-units |
| 10 W → 100 W | +10.00 dB | 1.67 S-units |
| 100 W → 500 W | +6.99 dB | 1.16 S-units |
| 500 W → 1,000 W | +3.01 dB | 0.50 S-unit |
| 1,000 W → 1,500 W | +1.76 dB | 0.29 S-unit |
| 100 W → 1,500 W | +11.76 dB | 1.96 S-units |
The arithmetic assumes the powers are comparable—for example, the same definition of peak-envelope, mean or carrier power at the same reference plane. It does not say that 1,500 W is lawful on a particular band or that the receiver will display two more S-points.
An S-unit is a nominal calibration, not propagation physics
The current IARU Region 1 VHF+ Handbook retains the technical recommendation of 6 dB per S-unit. For bands below 30 MHz it associates S9 with −73 dBm of available CW power at the receiver input and specifies detector dynamics. The handbook also observes that commercial implementation is difficult. Receiver AGC law, bandwidth, mode, detector, preamplifier, calibration and display segmentation can all make the front-panel indication depart from that nominal scale.
Even a perfectly calibrated meter reports input level, not readability. Readability depends on signal-to-noise-plus-interference ratio, bandwidth, modulation, fading and the operator or decoder. Treat “two S-units” as a translation of 11.76/6, not as a promised on-air report.
Build the Budget Between Named Reference Planes
A directional link budget can be written as:
Prx(dBW) = Ptx(dBW) − Ltx(dB) + Gtx(dBi) − Lpath(dB) + Grx(dBi) − Lrx(dB)
Every term needs a definition. State where transmitter power is measured; whether line, tuner and matching loss are measured under the installed load; whether antenna gain is dBi or dBd and for which elevation, azimuth and polarization; and whether receive losses and gain are referenced to available power. HF path loss is not fixed: the in-force ITU-R P.533-14 predicts quantities such as median field strength, available receiver power and signal-to-noise ratio for specified HF circuits, while ITU-R P.372-17 treats atmospheric, man-made and galactic radio noise.
A controlled comparison
These two hypothetical budgets preserve the useful example from the source article:
| Term | Station A | Station B |
|---|---|---|
| Conducted transmitter power | 100 W = 20.00 dBW | 500 W = 26.99 dBW |
| Loss before antenna input | −1.00 dB | −4.00 dB |
| Antenna gain in the path direction | +5.00 dBi | −2.00 dBi |
| Directional EIRP term | 24.00 dBW ≈ 251 W | 20.99 dBW ≈ 126 W |
Under those assumptions, Station A is 3.01 dB stronger in that direction even though its transmitter power is lower. This proves the arithmetic of the stated example, not that 100 W stations generally beat 500 W stations. The gain values must come from the complete installed antennas. Ordinary antenna gain already reflects antenna efficiency; do not subtract the same antenna loss again. If a realised-gain figure already includes port mismatch, do not also subtract that mismatch.
Loss Is Also a Ratio—and It Needs an Operating Point
| Net power loss | Power delivered | Power dissipated or diverted |
|---|---|---|
| 1 dB | 79.43% | 20.57% |
| 2 dB | 63.10% | 36.90% |
| 3 dB | 50.12% | 49.88% |
| 6 dB | 25.12% | 74.88% |
| 10 dB | 10.00% | 90.00% |
At 1,500 W net input, a genuine 1 dB dissipative loss would leave about 1,191 W and account for about 309 W elsewhere. That is why dB is convenient for the budget while watts remain essential for thermal design. Do not apply a matched-line catalogue attenuation blindly under high SWR: frequency, line length, load magnitude and phase, conductor/dielectric heating, tuner state and the chosen incident-versus-net-power convention all matter.
ERP and EIRP Use Different Antenna References
Current 47 CFR §2.1, reproducing ITU Radio Regulations terminology, defines ERP in a given direction from gain relative to a half-wave dipole and EIRP from gain relative to an isotropic antenna. For the same system and direction, the ideal reference conversion is:
EIRP(dBW) = power at antenna input(dBW) + gain(dBi)
ERP(dBW) = power at antenna input(dBW) + gain(dBd)
EIRP(dBW) ≈ ERP(dBW) + 2.15 dB
Loss before the antenna input can be subtracted when starting from transmitter power. Never combine dBd with an EIRP label or dBi with an ERP label. For a path calculation, use gain in the path direction. Where a rule limits EIRP or ERP, follow that rule's definition—often the maximum direction and specified operating conditions—not a convenient take-off angle.
1,500 W Is an Example, Not a Worldwide Permission
Power limits depend on administration, licence class, band, location, emission and the specified power quantity. The current Belgian BIPT amateur frequency and technical-characteristics table varies limits by certificate and band and explicitly uses ERP or EIRP on some allocations. In the United States, current 47 CFR §97.313 sets a 1.5 kW PEP ceiling but also requires the minimum power necessary and specifies lower limits in several cases.
Those are jurisdiction-specific examples, not a substitute for the operator's current authorisation. Peak-envelope power, mean power and carrier power are different quantities. A legal PEP value cannot be inserted unchanged into an average-heating or time-averaged exposure calculation.
A dB Budget Does Not Prove RF-Exposure Compliance
From 100 W to 1,500 W, power rises by a factor of 15, or 11.76 dB. With unchanged far-field geometry, field amplitude rises by √15 ≈ 3.87. HF operators are often close enough to antennas, feedlines and conductors that a simple far-field EIRP-over-distance shortcut is not adequate.
The ICNIRP 2020 RF guidelines treat 100 kHz–30 MHz personal exposure as near-field for reference-level assessment and require both electric- and magnetic-field conditions to be considered. They also specify temporal and spatial averaging and separate whole-body, local and induced-current restrictions. Applicable national rules may differ. In the United States, 47 CFR §97.13(c) separately requires amateur stations to ensure RF-exposure compliance.
Safety boundary: transmitter watts, directional EIRP, duty factor, modulation envelope, frequency, near-field E and H fields, antenna geometry, accessible conductors and occupancy all matter. This article cannot supply a generic “safe distance.” Increased power also raises voltage, current, heating and fault energy in feedline, matching, switching, choking and grounding hardware.
Common-Mode Control and Receive Noise Are Not Free Gain
Changing a choke or feedline route can alter feedline current, loss, the radiation pattern, local RF voltage, RFI and receive-noise coupling. The result may improve or worsen field strength in a particular direction. It is not defensible to enter “better common-mode control” as an assumed gain term without measuring or modelling the installed system.
More transmit power cannot lower the receiver's own noise floor. Reducing coupled household noise or using a better receive pattern can improve received signal-to-noise ratio, but the improvement must be measured with the same bandwidth, detector and time statistic. Likewise, antenna height or take-off angle is not a fixed number of dB: its value depends on frequency, path, ground, pattern and ionospheric conditions.
A Repeatable Upgrade Method
- Name the objective. More received power in one direction, better SNR, less RFI, lower component temperature and regulatory compliance are different objectives.
- Lock reference planes. Record whether power is measured at the transmitter, after the tuner, at the feedline input or at the antenna port, and whether it is PEP, mean, carrier, incident or net power.
- Use one gain reference. Keep dBi with EIRP and dBd with ERP; state direction and polarization.
- Characterise real losses. Include the installed frequency, load/SWR, line length, tuner state, connectors and common-mode path.
- Measure both ends of the problem. Compare absolute field or received power and noise in a fixed bandwidth over enough repetitions to separate the change from fading.
- Recheck limits. Higher power may change equipment ratings, interference, exposure, access controls and the applicable legal ceiling even when the link-budget arithmetic is perfect.
Engineering conclusion: use watts for absolute power, heating and limits; use dB for ratios and cascaded budgets; use dBW or dBm only with their stated references. A 1.76 dB upgrade can matter at a threshold, but whether it is the best upgrade depends on the full transmit path, receive noise, uncertainty, safety and law.
Primary Sources Checked
- Recommendation ITU-R V.574-5 — Use of the decibel and the neper in telecommunications
- 47 CFR §2.1 — ITU/FCC power, gain, ERP and EIRP definitions
- IARU Region 1 VHF+ Handbook 10.03 — S-meter recommendation
- Recommendation ITU-R P.533-14 — HF-circuit performance prediction
- Recommendation ITU-R P.372-17 — Radio noise
- BIPT — Frequency bands and technical characteristics authorised for radio amateurs
- 47 CFR §97.313 — Amateur transmitter power standards
- ICNIRP 2020 — Guidelines for limiting exposure to electromagnetic fields, 100 kHz–300 GHz
Mini-FAQ
- What is the difference between dB, dBW and dBm? dB expresses a ratio. dBW is a power level referenced to 1 W, and dBm is referenced to 1 mW. Thus 0 dBW and 30 dBm both represent 1 W.
- How much stronger is 1,500 W than 100 W? The power ratio is 15, or 11.76 dB. That change reaches the receiver only if the rest of the linear system and path are unchanged; it is not a guaranteed readability or S-meter change.
- Does doubling transmitter power double my signal? It doubles received signal power under unchanged linear conditions, a 3.01 dB increase. It raises far-field electric-field amplitude by about 1.414, not by two.
- Is one S-unit always 6 dB? Six decibels per S-unit is the IARU nominal recommendation. Real receiver displays can depart from it because of calibration, AGC, bandwidth, mode, detector and display behaviour.
- Are ERP and EIRP interchangeable? No. ERP uses half-wave-dipole gain in dBd; EIRP uses isotropic gain in dBi. For the same system and direction, EIRP is about 2.15 dB higher than ERP.
- Does a correct dB budget prove that my station is legal and RF-safe? No. Legal limits use jurisdiction-, band- and licence-specific power definitions. Exposure also depends on frequency, modulation, duty, averaging, geometry, near fields and access.