# What is the difference between peak-power, average power and true RMS power in RF measurement?

What difference between peak-power, average power and true RMS power in RF measurement? I have read a peak-power measurement can calculate average power measurement also. Does any of these needs signal to be demodulated?

• Demodulation may or may not be required but this has nothing to do with how the power is measured or represented. – Andy aka Jun 26 '14 at 7:30
• Peak voltgae is not the real power.. – user46337 Jun 27 '14 at 2:44

1. If a RADAR transmitter sent out a 10 kilowatt pulse 1 microsecond long every millisecond, its peak power output would be 10 kilowatts, and its average power output would be 10 watts.

2. There is no such thing as "true RMS" power.

3. No.

• Field how did you calculate avg output power as 10 watts? And I have been looking for RF power meters and have seen specifications like 'True RMS value of measured signal irrespective of signal waveform' – sk1 Jun 26 '14 at 9:11
• 1. If 10000 watts is radiated for 1 microsecond out of every 2 microseconds, the average radiated power will be 5000 watts. Similarly, if 10000 watts is radiated for 1 microsecond out of every 1000 microseconds, the average radiated power will be 10 watts. 2. You can have "instantaneous power", "peak power", "continuous power", or "average power", but regardless of what advertising fluff you read, "RMS" applies to voltage and current waveforms, not to power. – EM Fields Jun 26 '14 at 10:51

These are detectors, or aggregation functions during a sweep.

Your analyzer allows you to set up a number of sweep points across the spectrum. The sweep time is divided by the number of sweep points, and this time is integrated into a single point.

• The "max peak" detector uses the highest value during the sweep point interval
• The "average" detector averages the magnitude
• The "RMS" detector averages the power (proportional to magnitude squared).

If you have a fairly coarse grid (few points), you can get visible peaks using the max peak detector, which allows you to find the signal quickly, but the power displayed will be too large, while the RMS and average detectors would report values that are too small:

With a finer grid, all three detectors report values close to the actual power for a CW signal, because now they only differ in how the noise is integrated:

However, for a modulated signal (pulse, 8 us off, 2 us on) you can clearly see the difference (note that I used a long sweep time here):