Condition monitoring for lubricated machinery

Condition monitoring measures selected indicators over time so developing machine or lubrication problems can be found before functional failure.

Useful methods include vibration, temperature, ultrasound, lubricant analysis, visual inspection and process-performance data. The correct combination depends on the asset, likely failure modes and the time available to intervene.

A change in one measurement does not prove a lubrication fault. Compare trends, operating conditions and related evidence, then use the equipment manufacturer's limits and a competent diagnosis before changing the lubricant, interval or machine.

What is condition monitoring?

Condition monitoring is the periodic or continuous measurement and assessment of parameters associated with operating machinery. Its purpose is to support operation and maintenance decisions, not simply collect readings.

Detect a meaningful change from normal condition
Identify the likely fault or degradation mechanism
Estimate severity and urgency
Plan corrective work at a suitable time
Confirm whether the intervention restored normal behaviour

Which monitoring methods are useful for lubrication?

Each method sees different aspects of machine condition. Combining complementary evidence reduces the risk of a false conclusion.

Vibration for rotating faults, bearing defects, looseness and alignment-related patterns
Temperature for changing friction, load, cooling and lubrication conditions
Ultrasound or acoustic methods for friction, impacts, leaks and some lubrication decisions
Oil or grease analysis for fluid condition, contamination and wear evidence
Visual checks for leakage, level, colour, residue, damaged seals and delivery faults
Process data such as pressure, flow, power, speed and product rate

How do you design a condition-monitoring route?

Start with asset criticality and credible failure modes. Select a technique only where a change is detectable early enough to support an action.

List the asset, function and consequence of failure.
Identify the likely failure modes and their warning signs.
Choose suitable measurement methods and locations.
Define machine state, load and speed for comparable readings.
Establish baseline readings from known acceptable operation.
Set review frequency and provisional alert criteria.
Assign responsibility for collection, analysis and action.
Review findings, false alarms, missed faults and programme value.

Why are measurement conditions important?

Vibration and temperature can change with speed, load, product, cleaning cycle and ambient conditions. A route should define where, how and under what operating state the reading is taken.

Use the same sensor orientation and measurement point where possible. Record deviations rather than comparing unlike readings as if they were equivalent.

How should alerts be handled?

An alert should start a defined response rather than create an automatic diagnosis. Check the measurement, compare related indicators and assess operational risk.

Validate the reading and instrument setup
Check operating conditions and recent maintenance
Compare with history and similar assets
Use a second method where appropriate
Escalate according to severity and consequence
Create and complete a corrective work order
Measure again after the action

What records should be kept?

Asset, component and measurement point
Instrument, sensor direction and method
Date, speed, load and operating state
Measured value and relevant spectrum or image
Baseline, alert level and trend
Analyst interpretation and confidence
Recommended action, owner and deadline
Post-maintenance verification result

Records should identify the asset and lubrication point clearly enough for another competent person to repeat the check. Preserve abnormal findings, actions and follow-up results as well as routine readings.

What are the common mistakes?

Collecting data without linking it to failure modes or decisions
Comparing readings taken at different loads or locations
Changing sensors or settings without documenting the change
Using one technique to diagnose every problem
Setting generic alarms without asset context
Leaving alerts unassigned or unclosed
Failing to confirm the effect of corrective work

Frequently asked questions

Is condition monitoring the same as predictive maintenance?

Condition monitoring provides evidence about asset condition. Predictive or condition-based maintenance uses that evidence, together with risk and planning information, to decide when and how to act.

Does every asset need continuous sensors?

No. Continuous monitoring is most useful where failure can develop quickly, access is difficult or consequences are high. Periodic routes and operator checks suit many other assets.

Which method detects lubrication problems first?

There is no universal answer. Asset design, speed, load, fault type and measurement quality determine which signal changes first. Ultrasound, vibration, temperature and lubricant analysis often provide complementary evidence.

Can monitoring replace preventive lubrication tasks?

Not automatically. Use monitoring to improve decisions and verify outcomes, but retain required OEM tasks until a controlled review supports a change.

Summary

Condition monitoring should be designed around asset risk and detectable failure modes. Repeatable measurements, suitable baselines, complementary techniques and an owned response process turn data into useful maintenance decisions.

Monitoring creates value when a reliable observation leads to a proportionate action and the result of that action is checked.

Need help monitoring lubrication performance?

Explore Activate's monitoring and performance verification and lubrication services, or contact the Activate team to discuss an application.