Metrix OX9302-BUS, Руководство пользователя

Страница: 23 / 143

Functional Description
23

Rectangle

Hamming

Hanning

Blackman

Flat top
Before calculating the FFT, the oscilloscope weights the signal to be analyzed by a
window that acts as a bandpass filter. The choice of type of window is essential to
distinguish the different spikes of a signal and make accurate measurements.
Time representation
of the signal to be analyzed
Weighting window
Weighted signal
Frequency representation of
the signal calculated by FFT
The total duration of the study interval results in a convolution in the frequency
domain of the signal with a function sinx/x.
This convolution modifies the graphic representation of the FFT because of the
characteristic lateral lobes of the sinx/x function (unless the study interval contains
an integral number of periods).
Five weighting windows are proposed: the menus appear directly upon selection of
the FFT menu.
Effects of undersampling on the frequency representation:
If the sampling frequency is too low (less than the twice the cutoff frequency of the
signal to be measured), the high-frequency components are undersampled and are
aliased (frequency-shifted) in the graphic representation of the FFT.
The "Autoset" function is active. It serves to avoid the above phenomenon and adapt
the horizontal scale: the representation is easier to read.
The "Zoom" function is active. The zoom affects the graphic representation of the
FFT but does not change the conditions of acquisition (TB + depth).
Type of window
Width of the main lobe
at 3dB down (bin)
Max. amplitude of the
secondary lobe (dB)
rectangular 0.88 -13
Hamming 1.30 -31
Hanning 1.44 -43
Blackman 1.64 -58
Flat top
3.72 -93