P-F Curve brings contributions to Maintenance

The P-F curve is typically used in reliability analysis of monitored assets, based on Reliability Centered Maintenance (RCM) concepts.

Asset failures are not events, but degradation processes. The P-F curve is used to represent the condition of a piece of equipment or component over time, which makes it possible to identify these processes and act proactively to avoid failure, as shown in the figure below.

The horizontal axis (X) of the P-F Curve represents the service or operation time of an asset or asset component.

The vertical axis (Y) represents the resistance to failure or, in some cases, the performance of the asset.

In the P-F curve, the performance or condition of an asset declines over time, leading to functional failure (i.e. loss of function for which it was intended).

Thus, the goal of the P-F Curve is to determine the P-F interval. That is, the interval between Potential Failure and Functional Failure.

To foment the P-F curve, predictive and sensitive maintenance techniques are used to evaluate the machine’s condition, giving greater effectiveness to the repair cost forecasts for each stage of the equipment’s useful life.

Therefore, another practical application of this curve is to justify the importance of these maintenance techniques over the life cycle of an equipment or component.

Thus, there is much value in understanding the P-F curve and the activities associated with it, because the only way to prevent equipment failure is to prioritize actions to mitigate the P.

A potential failure is a condition that indicates whether functional failure is imminent or even occurring.

In other words, the potential failure is the way the failure presents itself in the equipment, and it can be said that Potential Failure is the same thing as Failure Mode.

If a potential failure is detected between point P and point F, actions can be taken to prevent the consequences of the functional failure.

Examples of Potential Failures in pumps, blowers, compressors, and turbines:

• Elevation in the temperature levels of the bearings;
• Elevation in vibration levels;
• Drop in flow rate;
• Drop in pressure;
• Elevation in the noise levels.

Examples of Functional Failures in pumps, blowers, compressors, and turbines:

• Shaft locked or no torque transmission (broken bearing);
• Total loss of pressure and flow;
• Mechanical seal breakage;
• Impeller/rotor does not pressurize the fluid.

For the application of preventive and predictive techniques to be successful, it is necessary to study the interval between the latent failure point, that is, the state from which it is possible to detect some parameters that indicate the proximity of a failure, and the functional failure point itself.

This interval is crucial in determining the appropriate frequency for inspections.

Once it’s desirable to detect a latent failure before the functional failure is realized, then the periodicity of inspections should be less than the P-F interval.

In terms of predictive, as shown in the curves, different approaches can be taken and cross-referencing information from different techniques is essential for optimized diagnosis and prognosis.

Ultrasonic measurements, oil inspection, thermography, and vibration measurements are generally used in the time interval from the occurrence of the potential failure to the functional failure.

Regarding to preventive maintenance, the use of periodic tasks directed by means of a checklist and recognition of a change in the machine’s condition by sensory perception are essential.

This way, system reliability in the P-F region will depend on the good use of the techniques and tasks associated with predictive and preventive maintenance.

Note that significant action can only be performed depending on how quickly a potential failure develops, so it is important to know the machine being monitored and its maintenance history (if any).

Following Figure A, tasks under condition or directed tasks are so called because the items that are inspected are left in service, under condition, so that they continue to develop the desired performance standards.

This is one of the prerogatives of predictive maintenance. Targeted tasks require checking potential failures so that action can be taken to prevent the functional failure or avoid its consequence.

In this case, other processes such as Root Cause Failure Analysis (RCFA) and Failure Mode and Effects Analysis (FMEA) can be used to speed up diagnostics.

In terms of proactive maintenance, early detection of potential failure, therefore within the skillful period on the P-F curve, results in:

• Fewer unplanned maintenance actions;
• Less repair costs;
• Longer asset uptime;
• Greater reliability for components, machines and their processes.

In addition, among the advantages in evaluations, indicators and maintenance reports are:

• Description of the potential failure and measures for maintenance intervention;
• Descriptions of reasons for failure, which implies proactive maintenance based on assertive information;
• Avoids high load conditions (overload) when possible;
• Defines needed changes to avoid additional load.

Therefore, with a clear understanding of the P-F curve, it is possible to determine intervals for preventive and predictive actions, so that they can be taken at the right time: as close as possible to potential failure and as far away as possible from functional failure. This implies great advantages for asset reliability.

If a company can figure out the PF range of its assets, it has already taken the first step to working with zero break rate.

Although activities that prevent failures and extend the P-F range often don’t get the attention they deserve, these efforts probably produce the best return of investment for companies.

With actions based on predictive analysis (vibration, temperature, and contaminant analysis in the lubricant, for example) and correlated with each other, the cause of potential failures can be found.

By doing so, functional failures will be mitigated and the life span of the assets can be continuously extended.

If companies start to follow this understanding in the learning process based on predictive analytics, this will imply in an extension of the asset life cycle.

The figure below schematically represents in the P-F curve how predictive techniques at various levels of accuracy and applicability help increase the assets lifetime.

The DynaPredict is a good solution to be applied in the P-F range, which offers continuous monitoring of vibration and temperature of machines and components and helps detect potential failures and early stages of functional failures, helping to adopt maintenance measures and ensure optimization of the equipment’s life span.