FFT Analyzer Basic FAQ

This is due to the effect of window correction. Normally, there are window-induced components on both sides of any frequency component in the frequency spectrum, but a certain coefficient is applied to correct for this effect when calculating the overall value. The same correction is applied during frequency calculus, which is why this phenomenon occurs.

<Reference example>

[Figure 1]
If there is amplitude at 6.25 Hz, which is next to 0 Hz, then amplitude values will also appear at 0 Hz and 12.5 Hz due to the effect of the time window.
Furthermore, if there is amplitude at 0 Hz, an amplitude value will also appear at 12.5 Hz due to the time window (this occurs due to fluctuations and residual offset in the circuit).
In the case of a Hanning window, the overall value is calculated by dividing the sum of the power values of all frequency data by 1.5 (the coefficient "1.5" here is called the "equivalent noise bandwidth").

[Figure 2]
When integration is performed, the integral value at 0 Hz cannot be defined, so the 0 Hz component is forcibly set to "0". At this time, the leakage at 6.25 Hz is also set to 0, so when divided by 1.5, the overall value becomes smaller than the peak value.
In integration, the influence of low-frequency components becomes significant, and frequency components below the resolution of the FFT analyzer become uncertain. For integration results, we recommend using partial overall, specifying a frequency bandwidth (generally 3 Hz or higher).