Non ROUTE
How to Collect “User” Defined Non ROUTE Data
1,000 Hz
= 2.5 Hz
400 lines
For example, if you choose 400 lines and a frequency range from 0 to 1,000
Hz, the basic resolution of the spectrum will be 1,000 divided by 400 or 2.5 Hz.
This specifies that the frequency axis is divided into 400 segments spaced 2.5
Hz apart.
Window
– The type of window used in the FFT processing. A window function must be
applied to any periodic time record prior to performing an FFT to minimize leakage
errors. The Hanning and
Flattop window functions attenuate to zero both the leading
and trailing edges of the time domain buffer (to prevent leakage error caused by
discontinuities in the time record). Press the right arrow to view the choices:
Hanning
- A dynamic signal analyzer window function that provides better
frequency resolution than Flattop, but with reduced amplitude accuracy.
Useful for machine vibration measurements, general purpose measurements,
and measurements containing random noise.
Flattop
- A dynamic signal analyzer window function, which provides the best
amplitude accuracy for measuring discrete frequency components. Useful for
calibration or machine vibration measurements using displacement probes in
fluid film bearings.
Uniform
– A dynamic signal analyzer window function with uniform weighting
across time. Useful for measuring transients or mechanical response
measurements and in tracking mode.
Averages
– Determines the number of average samples taken for the measurement.
Enter the number of FFT averages to be collected (from 1 to 4096). Four to six
averages are adequate and are normally used for machine monitoring. The higher the
number of averages, the slower the data collection.
Overlap
–
Overlap processing is advantageous when the time required to gather a time
record is much longer than the time needed to calculate an FFT spectrum. In the
Microlog, this occurs at frequencies below 1,000 Hz (60,000 CPM).
For lower frequencies, the amount of overlap can be increased to reduce the time
required to collect a given number of averages. Recognize, however, that the greater
the overlap, the more information shared between averages.
Overlap processing is used to obtain enough new ensemble data for an accurate
average. If the maximum frequency is low and the FFT process time is fast, the average
sum would include a high percent of old data with maximum overlap.
Below 2 kHz, 50% overlap and six averages is a reasonable
ROUTE setup.
•
Enter a value in the text field and press an Enter button.
Type
– Choose from the following average types:
Spectral
– The summation of the magnitude of each spectral line is divided by
the total number of averages (ensemble averaging). This is the most
frequently used method of averaging for routine data collection and analysis.
A - 14
SKF Microlog - GX Series
User Manual
