To determine the RMS value of a stationary waveform, most low-cost (and a few expensive) "RMS" meters simply measure the average of the absolute value of the input, using a full-wave rectifier, then multiply it by 1.111, the ratio of RMS: Average for sine waves. Though valid for undistorted sine waves, this approach leads to substantial errors with random noise, triangular waves, or square waves (but they may be calibrated), and to even worse errors with arbitrary waveforms, for which no calibration is possible. Read full article