Page 172 - Intro Predictive Maintenance

Vibration Monitoring and Analysis 163

installed or after the first scheduled maintenance once the program is established.
These data sets can be used as a reference or comparison data set for all future mea-
surements; however, such data sets must represent the normal operating condition of
each machine-train. Three criteria are critical to the proper use of baseline compar-
isons: reset after maintenance, proper identification, and process envelope.

Reset After Maintenance. The baseline data set must be updated each time the
machine is repaired, rebuilt, or major maintenance is performed. Even when best prac-
tices are used, machinery cannot be restored to as-new condition when major main-
tenance is performed. Therefore, a new baseline or reference data set must be
established following these events.

Proper Identification. Each reference or baseline data set must be clearly and com-
pletely identified. Most vibration-monitoring systems permit adding a label or unique
identifier to any user-selected data set. This capability should be used to clearly iden-
tify each baseline data set. In addition, the data-set label should include all informa-
tion that defines the data set. For example, any rework or repairs made to the machine
should be identified. If a new baseline data set is selected after replacing a rotating
element, this information should be included in the descriptive label.

Process Envelope. Because variations in process variables, such as load, have a direct
effect on the vibration energy and the resulting signature generated by a machine-
train, the actual operating envelope for each baseline data set must also be clearly
identified. If this step is omitted, direct comparison of other data to the baseline will
be meaningless. The label feature in most vibration-monitoring systems permits
tagging the baseline data set with this additional information.

Known Machine Condition
Most microprocessor-based analyzers permit direct comparison to two machine-trains
or components. The form of direct comparison, called cross-machine comparison, can
be used to identify some types of failure modes.

When using this type of comparative analysis, the analyst compares the vibration
energy and profile from a suspect machine to that of a machine with a known oper-
ating condition. For example, the suspect machine can be compared to the baseline
reference taken from a similar machine within the plant. Or, a machine profile with a
known defect, such as a defective gear, can be used as a reference to determine if the
suspect machine has a similar profile and, therefore, a similar problem.

Industrial Reference Data
One form of comparative analysis is direct comparison of the acquired data to indus-
trial standards or reference values. The International Standards Organization (ISO)
established the vibration-severity standards presented in Table 7–2. These data are
applicable for comparison with filtered narrowband data taken from machine-trains