25.5 Machine Condition Monitoring

Back

With the understanding that condition-based maintenance may not be appropriate in all situations, let us

say that some preliminary analysis has been carried out and a decision made to apply machine condition

monitoring and fault diagnostics in a selected part of a plant or on a specific machine. The following is a

list of potential advantages that should be realized:

* Increased machine availability and reliability

* Improved operating efficiency

* Improved risk management (less downtime)

* Reduced maintenance costs (better planning)

* Reduced spare parts inventories

* Improved safety

* Improved knowledge of the machine condition (safe short-term overloading of machine possible)

* Extended operational life of the machine

* Improved customer relations (less planned/unplanned downtime)

* Elimination of chronic failures (root cause analysis and redesign)

* Reduction of postoverhaul failures due to improperly performed maintenance or reassembly

There are, of course, also some disadvantages that must be weighed in the decision to use machine

condition monitoring and fault diagnostics. These disadvantages are listed below:

* Monitoring equipment costs (usually significant).

* Operational costs (running the program).

* Skilled personnel needed.

* Strong management commitment needed.

* A significant run-in time to collect machine histories and trends is usually needed.

* Reduced costs are usually harder to sell to management as benefits when compared with increased

profits.

The ultimate goal of machine condition monitoring and fault diagnostics is to get useful information

on the condition of equipment to the people who need it in a timely manner. The people who need this

information include operators, maintenance engineers and technicians, managers, vendors, and

suppliers. These groups will need different information at different times. The task of the person or

group in charge of condition monitoring and diagnostics must ensure that useful data is collected, that

data is changed into information in a form required by and useful to others, and that the information is

provided to the people who need it when they need it. Further general reading can be found in these

references: Mitchell (1981), Lyon (1987), Mobley (1990), Rao (1996), and Moubray (1997).

The focus of this chapter will be on vibration-based data, but there are several different types of data

that can be useful for assessing machine condition and these should not be ignored. These include

physical parameters related to lubrication analysis (oil/grease quality, contamination), wear particle

monitoring and analysis, force, sound, temperature, output (machine performance), product quality,

odor, and visual inspections. All of these factors may contribute to a complete picture of machine

integrity. The types of information that can be gleaned from the data include existing condition, trends,

expected time to failure at a given load, type of fault existing or developing, and the type of fault that

caused failure.

The specific tasks which must be carried out to complete a successful machine condition

monitoring and fault diagnostics program include detection, diagnosis, prognosis, postmortem, and

25-8 Vibration and Shock Handbook

© 2005 by Taylor & Francis Group, LLC

prescription. Detection requires data gathering, comparison to standards, comparison to limits set

in-plant for specific equipment, and trending over time. Diagnosis involves recognizing the types of

fault developing (different fault types may be more or less serious and require different action) and

determining the severity of given faults once detected and diagnosed. Prognosis, which is a

very challenging task, involves estimating (forecasting) the expected time to failure, trending the

condition of the equipment being monitored, and planning the appropriate maintenance timing.

Postmortem is the investigation of root-cause failure analysis, and usually involves some research-type

investigation in the laboratory and/or in the field, as well as modeling of the system. Prescription is

an activity that is dictated by the information collected and may be applied at any stage of the

condition monitoring and diagnostic work. It may involve recommendations for altering the

operating conditions, altering the monitoring strategy (frequency, type), or redesigning the process

or equipment.

The tasks listed above have relatively crisp definitions, but there is still considerable room for

adjustment within any condition monitoring and diagnostic program. There are always questions,

concerning such things as how much data to collect and how much time to spend on data analysis,

that need to be considered before the final program is put in place. As mentioned above, things such as

equipment class, size, importance within the process, replacement cost, availability, and safety need

to be carefully considered. Different pieces of equipment or processes may require different monitoring

strategies.

25.5.1 Periodic Monitoring

Periodic monitoring involves intermittent data gathering and analysis with portable, removable

monitoring equipment. On occasion, permanent monitoring hardware may be used for this type of

monitoring strategy, but data is only collected at specific times. This type of monitoring is usually applied

to noncritical equipment where failure modes are well known (historically dependable equipment).

Trending of condition and severity level checks are the main focus, with problems triggering more

rigorous investigations.

25.5.2 Continuous Monitoring

Constant or very frequent data collection and analysis is referred to as continuous monitoring.

Permanently installed monitoring systems are typically used, with samples and analysis of data done

automatically. This type of monitoring is carried out on critical equipment (expensive to replace,

with downtime and lost production also being expensive). Changes in condition trigger more detailed

investigation or possibly an automatic shutdown of the equipment.

* Potential advantages of machine condition monitoring include increased machine

availability and reliability, improved efficiency, reduced costs, extended operational life,

and improved safety.

* Some of the disadvantages of condition monitoring include monitoring equipment costs,

operational costs, and training costs.

* The ultimate goal of machine condition monitoring and fault diagnostics is to get

useful information on the condition of equipment to the people who need it, in a timely

manner.

* The specific tasks which must be carried out to complete a successful machine condition

monitoring and fault diagnostics program include detection, diagnosis, prognosis,

postmortem, and prescription.

Machine Condition Monitoring and Fault Diagnostics 25-9

© 2005 by Taylor & Francis Group, LLC

Навигация