Soft dB Opus Suite Data Logger Module - Manual (Page 45)

44
This parameter is used only for the Elliptic and Inverse Chebyshev filter
types. This parameter is the desired attenuation in the stop band of the
filter. It must be greater than zero and you must express it in decibels.
All filter parameters can be saved and recalled using these buttons. Note that the filter interface
automatically creates a .flt file when the filter operation is launched with the Apply Filter and Quit
button. This automatically saved .flt file will have the name of the filtered wave file and it will be in
the same folder. This feature is very useful for following the filtering historic of a wave file.
This control allows selecting the group of channels to which the filter will be applied. Note that the
non-filtered channels will be included in the wave file created by the filter function.
This button launches the filtering operating. The original wave file will be used as a source for the
filter and the filtered signal will be saved in a new wave file. Also, a .flt file containing the filter
parameters is automatically created in the same folder that the filtered wave.
9.2.2 Integral x(t)
This function performs a time-integration on the signal of a group of selected channels. Some
modules in the analysis functions tab allow integration in the frequency domain. This function is
useful for analyzing the signal coming from an accelerometer, especially if the speed and position
must be analyzed in the time domain. There are many available techniques for performing time
integration. We select the filtering approach, which has the advantage of being numerically stable.
So, a simple first order or second order low-pass filter is used to perform a simple or double
integration. The user must specify a cut-off frequency for the filter. We suggest using an Fcut(Hz)
parameter that fits with the sensor used. A typical accelerometer sensor has a flat frequency
response that starts at around 20 Hz. The interface of the Integral x(t) function is shown in the next
figure: