3-24
VIBXPER
T II 05.2012
Options
Note
s
ample
frequency
(t
imewaveform
):
The sample frequency for the time
signal defines the upper frequency.
m
easurement
time
(t
imewaveform
):
The length of the time window de-
pends on the sample frequency. The maximum measurement time
is 640 s.
n
umBer
of
lines
(s
pectrum
, c
epstrum
):
Number of lines displayed.
Together with the 'upper frequency', this parameter specifies the
resolution of the spectrum (
∆
f).
w
indow
(s
pectrum
, c
epstrum
):
The measurement of periodic signals
generates time data sets with periodic gaps as a result of the finite
observation period. Thus, the frequency lines appear wider in the
corresponding spectrum. The window functions suppress these
'side lobes'.
Window features:
Hanning: Standard for FFT. Frequency precise, but amplitude er-
rors (<15%); for the analysis of continuous processes with hi
gh
frequency resolution and lowest possible break-off effect.
Rectangular: Frequency errors, but amplitude is precise; suitable
for the analysis of individual pulses, particularly if the pulse is at
the beginning of the time window.
Kaiser: As with Hanning, but lower amplitude errors (<12%)
Flat-Top: Lower frequency accuracy than Hanning, amplitude more
precise. For precision amplitude analysis of continuous processes.
Hamming: Frequency errors as in Rectangular; amplitude errors
(18%) less than in Rectangular, but higher than in Hanning.
Bartlett: Next to Rectangular, this is the simplest window (trian-
gle); lower amplitude and frequency accuracy.
Blackman: As for Hanning, but with higher frequency accuracy;
amplitude errors as in Kaiser (12%); next to Kaiser, the second
best alternative for nearly all applications.
a
verage
type
/
no
.
of
measurements
(
spectrum
, t
imewaveform
, c
epstrum
)
Without: During the measurement each measurement is displayed.
The 'Number of averages" specifies how many single measure-
ments are carried out. The last measurement is evaluated.
Linear: The individual spectra are added and divided by the num-
ber of averages. This averaging type emphasizes recurring station-
ary sections in the spectrum.
Peak-hold: Highest measurement value is displayed. This type of
averaging emphasizes peaks that occur stochastically.
Exponential: Individual measurements are exponentially averaged
(the last spectrum is given the highest weighting).
Time-synchronous: The individual measurements are averaged on
a RPM synchronous basis. Reference transducer required.
o
verlap
(
spectrum
, c
epstrum
):
See section A1, P. 3-21.
m
easurement
range
:
See section A1, P. 3-22.
t
rigger
type
/ l
evel
/ s
tart
(
s
pectrum
, c
epstrum
, t
imewaveform
)
P. 3-22.
Summary of Contents for VIBXPERT II
Page 175: ......
