What is the difference between average power and peak power measurement?
Average vs Peak Power Measurement
Understanding the distinction between average and peak power is fundamental for RF system design, component selection, and regulatory compliance.
| Parameter | SOLT Cal | TRL Cal | eCal |
|---|---|---|---|
| Accuracy | Good | Excellent | Good-very good |
| Standards Needed | 4 (S,O,L,T) | 3 (T,R,L) | 1 (module) |
| Bandwidth | Broadband | Band-limited | Broadband |
| Setup Time | 5-10 min | 10-20 min | 1-2 min |
| Best For | Coaxial, general | On-wafer, waveguide | Production, speed |
Calibration Procedure
(1) Average power (P_avg): the mean power over a specified time interval T: P_avg = (1/T) × integral(0 to T) P_inst(t) dt. For CW: P_avg = P_peak (constant envelope). For pulsed CW: P_avg = P_peak × tau/PRI = P_peak × duty_cycle, where tau is the pulse width and PRI is the pulse repetition interval. For modulated signals: P_avg depends on the modulation statistics. (2) Peak power (P_peak): the maximum value of the instantaneous power envelope: P_peak = max(P_inst(t)) over the measurement interval. This is the power at the crest of the highest peak. P_peak = P_avg + PAPR (in dB, where PAPR is the peak-to-average power ratio). (3) Peak envelope power (PEP): the average power during one RF cycle at the peak of the modulation envelope. For AM: PEP = P_carrier × (1 + m)^2, where m is the modulation index. For 100% AM: PEP = 4 × P_carrier. For SSB: PEP = P_avg × speech_peak_factor (typically 6-10 dB above average). PEP is the FCC-specified metric for amateur radio transmitter power. (4) Pulse power: the average power during the pulse on-time. For a rectangular pulse: pulse power = P_avg / duty_cycle. For shaped pulses (Gaussian, Hamming): the pulse power depends on the pulse shape; it is the time-averaged power over the pulse duration.
Error Sources
(1) Pulsed radar: a typical radar transmits 10 kW peak, 10 us pulse, 1 ms PRI (1% duty cycle). P_avg = 10000 × 0.01 = 100 W. Average power sensor reads 100 W (correct for thermal analysis and prime power calculation). Peak power sensor reads 10 kW (correct for component sizing and range equation). (2) LTE downlink: P_avg = +43 dBm (20 W) at the base station output. PAPR = 8 dB. P_peak = +51 dBm (126 W). The PA must handle +51 dBm peaks without excessive compression. The duplexer and antenna must handle +51 dBm peak without PIM or arcing. (3) Wi-Fi 6 (802.11ax): P_avg = +20 dBm (100 mW) at the access point output. PAPR = 12 dB for 80 MHz OFDM with 1024-QAM. P_peak = +32 dBm (1.6 W). The PA must be extremely linear (EVM < -35 dB for 1024-QAM). (4) Pulsed Doppler radar: more complex pulse pattern (burst of pulses). P_avg during the burst ≠ overall P_avg. Peak power sensor tracks each individual pulse. Average power sensor reads the overall average (including inter-burst dead time).
- Performance verification: confirm specifications against the application requirements before finalizing the design
- Environmental factors: temperature range, humidity, and vibration affect long-term reliability and parameter drift
- Cost vs. performance: evaluate whether the application demands premium components or standard commercial grades
- Interface compatibility: verify impedance, connector type, and mechanical form factor match the system architecture
- Margin allocation: include sufficient design margin to account for manufacturing tolerances and aging effects
Fixture Considerations
Average power of any signal: thermocouple sensor (waveform-independent, guaranteed accurate). Corrected diode sensor (faster, wider range, but requires waveform correction for high PAPR signals). Peak power: peak power sensor (diode-based with > 100 MHz video BW): captures the envelope of each pulse or modulation peak. Peak and average simultaneously: dual-function sensors (e.g., Keysight U2040XA): can report both peak and average from the same measurement. Or: use two sensors on a directional coupler (one thermocouple for average, one peak diode for peak). Time-gated power: some power meters support time gating (measuring the average power only during a specified time window). Useful for: measuring the power of a specific time slot in a TDMA signal, measuring the power during the preamble or payload of a packet, and measuring the pulse droop of a radar transmitter.
Frequently Asked Questions
What power rating should I use for my cable selection?
Cables and connectors have two power ratings: (1) Average (continuous) power rating: determined by the thermal capability (heat dissipation). Example: LMR-400 cable is rated at 140 W average at 1 GHz. Using it at 100 W average: acceptable. At 200 W: the cable overheats and may be permanently damaged. (2) Peak power rating: determined by the voltage breakdown of the dielectric and connectors. Example: LMR-400 peak power rating: 16 kW at 1 GHz. If the signal has +43 dBm average (20 W) with 12 dB PAPR: P_peak = 316 W. This is well below the 16 kW peak rating. For your cable selection: verify that P_avg < average power rating AND P_peak < peak power rating. Also derate by 50% for altitude (reduced dielectric strength) and 10% for elevated temperature.
How does duty cycle affect power measurement?
The duty cycle is the fraction of time the signal is present: duty_cycle = pulse_width / PRI. Effect on average power: P_avg = P_peak × duty_cycle. A 100 W peak transmitter at 1% duty cycle: P_avg = 1 W. Measurement consideration: a thermocouple or CW-calibrated diode sensor reads the average power directly (1 W in this case). To determine the peak power: either use a peak power sensor, or calculate: P_peak = P_avg / duty_cycle (only valid if the duty cycle is precisely known and the pulse shape is rectangular). For very low duty cycles (< 0.01%): the average power may be below the sensor sensitivity. Use: a higher-sensitivity sensor, or a peak power sensor that directly measures P_peak, or a coupler to sample the pulse and a sensitive receiver (oscilloscope or SA).
Can I measure average power with an oscilloscope?
Yes, with limitations. An oscilloscope can measure the RF voltage waveform and calculate power: P = V^2/(2×Z0) for a sinusoidal CW signal, or P_avg = (1/T) × sum(V[n]^2/Z0)/N for a digitized waveform. Requirements: the oscilloscope bandwidth must be at least 5× the RF frequency (to accurately capture the waveform). For a 1 GHz signal: need a 5 GHz oscilloscope. A calibrated probe with known impedance and insertion loss. Proper termination (50 ohm). Accuracy: typically ±1-3 dB (limited by probe calibration and oscilloscope vertical accuracy). Much less accurate than a dedicated power sensor. When this approach is useful: when both time-domain waveform and power are needed simultaneously (e.g., pulse shape analysis, transient power measurement). For just power measurement: use a power meter.