Condition Monitoring Adam Adgar School of Computing and Technology.

Condition Monitoring

Adam AdgarSchool of Computing and Technology

Why is it needed?

► Improve industrial plant performance► Provide early warning of potential

failure► Continuous drive for improved

efficiencies► Reducing unexpected plant downtime► Removing need for routine maintenance

How does it improve things?

►Maintenance performed on a 'just-in-time' principle rather than on a routine basis Reduction in replacement-parts holding

► Reduces or eliminates unscheduled shut-down of plant or equipment lead to additional costs, usually out of all

proportion to the cost of the repairs subsequently found to be necessary

What is it?

► Dynamic systems (electrical, hydraulic, mechanical or thermal) possesses a normal characteristic 'signature' when operating in the desired state

► Changes in this signature may indicate the onset of a failure

► Small differences between normal and abnormal signatures may be hidden by 'noise' in the system

► Modern transducers and associated signal-analysis techniques can now discriminate between truly random variations and significant trends which, with knowledge of the system parameters and normal characteristics, can be used to predict time to failure

How is it done?

► Need to identify changes in the condition of a machine that will indicate some potential failure

► Physical characteristics are identified that collectively indicate the current condition of the machine

► Each of these characteristics is measured, analysed, and recorded so that trends can be recognised

► Over a period of time, the progress of these trends represents the deterioration of machine condition and can be used to determine maintenance actions.

Where is it done?

► Originally developed to anticipate failure in high-speed gearboxes in the aerospace industry

► Now being applied to an ever-growing range of industries

► If process variables such as vibration, speed, pressure, temperature, voltage, current etc. can be measured, then it is likely that a process signature can be identified which can lead in turn to the application of Condition Monitoring techniques

Example

►Different types of monitoring systems are used according to the criticality and condition of the individual asset

►Data: Body Temperature 99°F►Diagnosis: Appendicitis► Prognosis: Bad► Action: Appendectomy

tomorrow

Technologies Applied

1. Vibration analysis2. Thermographic analysis3. Oil analysis4. Ultrasonic analysis5. Motor current signature analysis

(MCSA)6. Performance trending/visual

observations

1. Vibration Analysis► Involves measuring the forces created from vibrations.► 50 to 70% of rotating component failures are

attributable to misalignment► Non-intrusive.► Foundation of rotating machinery predictive

maintenance.► Limited analysis on reciprocating machinery now

possible.

2. Thermographic Analysis► Involves the use of infrared scanners to detect

differences in surface temperatures.► Based on the principle that machinery failures will be

preceded by changes in temperature.► Foundation of electrical component predictive

maintenance► Can be used to analyze patterns of heat loss or gain► Non-intrusive► Non-contact

3. Oil Analysis

► Identifies the condition of the hydraulic fluid or lubricant

Viscosity analysis Contamination analysis Acid/Base number Oil additive analysis

► Identifies the conditionof the machine.

Wear particle analysis

4. Ultrasonic Analysis

► Measures the sonic sound patterns of machines being monitored

► Airborne ultrasounds (20 kHz - 100kHz) Cannot penetrate solid surfaces Can penetrate cracks Radiate in a straight line

therefore the source is easy to locate

Travel a short distance

5. Motor Current Signature Analysis

► Detects mechanical and electrical problems in motor driven rotating equipment.

► Based on the concept that an electrical motor driving a mechanical load acts as an efficient transducer.

Motor senses mechanical load conditions and translates variations in these conditions into variations in electrical current conditions.

6. Performance Trending

►Monitoring the performance of a machine through the trending of process data. Pressure Temperature Flow Amperage

Visual Observations

►Oldest and most common technique► Involves looking, listening, touching► Virtually all equipment

Benefits / Summary

►Opportunity to organising avoidance strategies to minimise lost time and unexpected costs

► Provides information on presence of a fault condition indication of fault type diagnosis of the cause of a fault level of severity of fault prediction of time-to-failure