Monetize your knowledge: PM and PdM for electric motors

Vibration and temperature are key factors to watch in IoT-enabled condition monitoring.

By Thomas Schardt and Pranesh Rao, Nidec Motor Corp., and Chris Diak, Motion Industries

1 of 2 < 1 | 2 View on one page

Maintaining operation of a production line can be a lot like maintaining the health of a person who is predisposed to heart disease. For the best possible outcome, you must know the warning signs of trouble; you need to monitor that person continuously; and you must be prepared to act quickly should conditions change.

To understand why, consider what happened recently at the manufacturing plant of one of our company’s key suppliers.

This plant operates several production lines, two of which have been equipped with a monitoring system that remotely monitors the lines’ condition. While a heart monitor might use sensors that check for irregular heartbeats, this system uses sensors to closely watch motor vibration.

Shortly after the installation of this monitoring system, one production line’s vibration levels began exceeding the operating parameters that had been established for the motor. Analysis of the vibration data indicated that a motor bearing on that line was on the verge of failure. Further review suggested that the bearing could function for a few more days – enough time to secure a replacement part that could be installed during a controlled shutdown.

Three days later, that’s precisely what the plant did. Production was moved to another line while the bearing was replaced at a fraction of the cost that would have been incurred had the equipment failed unexpectedly. Think of a heart patient whose heart monitor readings suggest an adjustment to his or her blood pressure medication.

Consider now what might have happened had the production line not been continuously monitored. In that case, the motor would have failed and production would have come to an immediate halt while a maintenance team worked to identify the problem.

In addition to diagnostic and repair costs, the company would have faced rush charges, lost production, and possible overtime wages – not to mention the potential customer dissatisfaction brought on by delayed product shipment. The cost of an undetected bearing failure could have easily been 20 times the amount of preventive action. To complete our medical analogy, it would be the difference between a change in medication and emergency open heart surgery.

Positive outcomes, by contrast, result from parties that have the foresight to monitor very simple things: heartbeats and motor vibration levels.

Excess vibration is a warning sign for many potential equipment problems. The vibration signature of a specific piece of machinery, in fact, provides more information about the machine’s mechanical condition than any other factor. But it is not the only thing to monitor. Comparing equipment temperature, noise profiles, and a host of other technology-specific factors can also provide additional insights into equipment condition.

All of this equipment monitoring is at the core of a condition-based (predictive) maintenance program.

Striking a balance between PM and PdM

Many plants today still rely on service contractors that provide regularly scheduled preventive maintenance on critical equipment. This approach – a giant step past the traditional, reactive “we’ll fix it when it breaks” approach – involves following manufacturers’ maintenance guidelines to reduce unscheduled equipment failure and unplanned downtime. Preventive maintenance has a long history of improving safety and bottom-line results.

To go to the next level – condition-based, or predictive maintenance (PdM) – industrial companies and their maintenance, repair, and overhaul (MRO) service contractors will need software-based tools that enable them to access the vast amounts of operational and equipment health data manufacturers already own but are not fully utilizing. The efficiencies possible with a true predictive maintenance approach will depend on the software available to sift through millions of data points to identify those that align with actual equipment condition and performance.

PdM has been improved by technological advances in sensors and in connectivity and communications tools for streaming live data, and declining prices for these have made them more accessible. Special analytical software developed for the industrial internet of things (IIoT) has added needed capabilities, too. Without data analytics, defining an optimal maintenance strategy for critical equipment is like a tightrope walk between operations and maintenance costs on the one hand and the potential cost of an unexpected or unscheduled equipment failure on the other.

Servicing machinery is most cost-effective if it’s done when it is needed (within certain limits). If equipment repairs are addressed too early, a company will spend more on replacement parts and upkeep than it would if maintenance schedules were determined based solely on equipment condition. If it waits too long, on the other hand, deterioration can progress too far, resulting in higher costs to bring equipment performance back to acceptable levels.

1 of 2 < 1 | 2 View on one page
Show Comments
Hide Comments

Join the discussion

We welcome your thoughtful comments.
All comments will display your user name.

Want to participate in the discussion?

Register for free

Log in for complete access.


No one has commented on this page yet.

RSS feed for comments on this page | RSS feed for all comments