Table of Contents
Advertisement
Number of Averages
The larger the number of spectra used for averaging, the more any noise
spikes in vibration signals are reduced, and the more accurately true
spectral peaks are represented. However, the larger the number of
averages, the more data needs to be collected, and the longer it will take
to obtain the average spectrum.
Four averages are sufficient in most cases. You can set the number of
averages to one if spectra averaging is NOT required (e.g. with run-up
and coast-down recordings).
Averaging Types
The following averaging types are available:
Linear averaging — This is suitable in most cases. The
•
amplitude value at each frequency of a spectrum is added to
the same frequency of the next spectrum. The sum is then
divided by the number of averages taken.
Exponential averaging — This is occasionally used during
•
manual measurements to give the advantage of a fast update
rate, while maintaining some averaging. The most recent
spectra have more influence on the average than earlier
measurements. Averaging is continuous until you stop the
measurement.
Peak hold — This is useful for measuring the maximum
•
amplitude peak that occurs over a period of time. It is also
useful for measuring resonances during a bump test. Peak hold
does NOT average the amplitude values. It records and displays
the maximum amplitude of each spectral line.
Average Overlap
Overlapping is a means of quickly collecting and displaying data. As
vibration measurements are collected, a percentage of the new data is
combined (overlapped) with each subsequent measurement. The higher
the overlap percentage, the less newly acquired data is needed to
generate a spectrum and the faster the spectrum can be displayed. An
overlap percentage of 50% is ideal in most cases.
Section 5 — Set Measurement Options
119
Advertisement
Table of Contents
