Measurement Computing eZ-PostView rev.2.0 - Manual (Page 140)

G-2, Glossary 957397 eZ-Analyst
Crosstalk
An undesired transfer of signals between system components or channels. Crosstalk often
causes signal interference, more commonly referred to as noise.
dB
Decibels. A logarithmic unit of amplitude for noise or vibration. While there are several
definitions to dB, in our application we are using dB to express the ratio of the magnitudes of
two quantities equal to 20 times the common logarithm of the ratio.
dB = 20 log (x/dBEUref)
In this equation, “x” refers to the value of the voltage being measured and “dBEUref”
refers to the reference voltage.
Decay %
Decay %, also referred to as Response Decay Percent, is used when an Exponential Window is
applied to the FFT Response Channel. It is the weighting factor present at the last block value.
For example, a decay of 5% means that the weighting factor at the beginning of block is
100% and that the weighting factor of the last block is 5%. In other words, the weighting
factor dropped (decayed) to 5% over the length of block.
Digital
A digital signal is one of discrete value, in contrast to a varying signal. Digital data is
represented by combinations of binary digits (0s and 1s).
Dynamic Stiffness
Force/Displacement. A transfer type that is the reciprocal function of Receptance. Dynamic
Stiffness is derived by taking the inverse of the magnitude, and multiplying the phase angle
by “-1.”
EU
Engineering Unit.
Excitation
Some transducers [e.g. strain gages, thermistors, and resistance temperature detectors
(RTDs)] require a known voltage or current input in order for the sensor to operate. This
known input is called the Excitation.
Exponential Weighting Window
An exponential weighting window is equal to 1.0 at the beginning of the block and decays
exponentially to a smaller value at the end of the block (see Decay % ). Exponential is used
only with transient data that is captured with pre-trigger to assure that the initial values in all
data channels are very close to zero. Exponential can be used with all transient excitation
methods in order to force the signals to decay close to zero, even if the block length is not
sufficient to capture all of the naturally occurring response. If the data decays naturally to a
low amplitude [within the block], so that leakage is not significant, exponential windowing can
improve the signal-to-noise ratio by giving reduced weight to the very low-amplitude data at
the end of the block.
Fast Fourier Transform (FFT)
An efficient technique for calculating the frequency of components in a time-domain waveform
based on digitized voltage measurements. The result is a display of amplitude versus
frequency and phase versus frequency.
Fast Fourier Transform (FFT) Reference Window is applied to the output of a transducer
to avoid collecting extraneous signals caused by an excitation device, such as an impulse
hammer. Possible selections are: Response, Rectangular, and Cosine Taper.
Fast Fourier Transform (FFT) Response Window
The FFT (Response) window is a time-domain, weighting window. A response window is
usually applied to data to reduce FFT leakage errors. FFT theory assumes that the signal
being analyzed is periodic in the data acquisition block. When this is not the case, energy
from a signal at a specified frequency can leak into nearby spectral bins causing spectral
amplitude inaccuracies. Applying a windowing function controls, but doesn’t completely
eliminate, the error by multiplying each data frame by a suitable time-domain weighting
window. This calculation reduces the amplitude/magnitude of the data near the ends of each
data frame prior to performing the FFT and forces the data to be nearly periodic in the
window, thus reducing leakage errors. Response window options include Flat-Top, Blackman-
Harris, and Hanning.