Big Data Analytics / Software / Remote Monitoring

The secret to reliability liftoff: How to safely check machinery while it is running

The more work you can do from the perimeter of the equipment without going into lockout and shutting down the equipment, the better.

By John Bernet, Fluke Corp.

Many highly productive plants run 24 hours a day, covering three shifts, and the only planned downtime is over part of the weekend. The larger manufacturers deploy a tiered approach in which reliability personnel monitor critical machinery to predict potential failures or efficiency loss. They want to know of possible problems ahead of time so they can fill out work orders for the maintenance team, who can then make appropriate repairs when machines are planned to be offline.

That works for the big guys, such as auto and heavy equipment manufacturers, but what about plants where the maintenance team is also the reliability team and a skeleton crew is chasing machinery events just to stay ahead of the next disastrous failure?

The key is to find ways to check machinery while it is running, maintain safe practices while doing so, and then accurately document the results. The more work you can do from the perimeter of the equipment without going into lockout and shutting down the equipment down (thereby losing manufacturing throughput), the better.


Since the first humans fashioned tools out of stone, people have used innovation and technology to ease burdens and increase productivity. Now, with the ubiquity of mobile phones and wirelessly connected tools, more workers who are charged with maintaining equipment can at least partly change their orientation toward periodic predictive tasks to check assets for issues – before problems occur.

Wirelessly connected test and measurement tools that are tied together by a cloud-based storage system and software platform can help create liftoff for a reliability program without a massive investment in capital, labor, or IT infrastructure. Starting with even two tools, such as a wirelessly enabled thermal image and a vibration meter, you can create a starter reliability system that can be implemented immediately and then scaled up over time.

With a thermal imager, malfunctioning components will appear warmer than others, while failed components will be cooler or cold by comparison. Periodic inspection can reveal over- and under-heating conditions with switches, circuit breakers, motors, gear boxes, pumps, panels, and other machinery.

For machines with rotating parts, the shaking, shimmying, and thumping of loose, misaligned, or unbalanced parts can be detected with a vibration meter. Indeed, recent developments in vibration sensors, data acquisition, and analysis technologies are making vibration analysis less expensive and more widely available.

For each of these tools, data is logged by date and equipment and can easily be shared to involve more team members and even manufacturer reps or engineers in the process.

And because the thermal imager and the vibration meter can be safely used while equipment is running, they can provide an early warning system to issues.

Routine checks and analysis

A thermal imager provides a safe and easy way to monitor the surface temperature of electrical or electromechanical components. Of more importance, advances in infrared technology mean that you can now detect variance in temperature inch by inch. When charted over time, these instant readings can be used to identify and predict maintenance and repair needs, reducing the potential for failure.

With a vibration meter, you have the capability to measure overall vibration in addition to maintaining a database of real machine values to provide the user with an answer. To best execute inspections, establish a measurement route through the plant. Following the same routine during each inspection can provide time and data advantages as each piece of equipment is viewed, images and readings are captured, and anomalies are recorded. The associated software provides the mechanism to record, track, and recall readings as needed for reference.
Consider how wirelessly connected thermal imagers and vibration meters would enable systematic data collection along the line.

The goals:

  1. Create baselines of critical equipment during normal operation, and then
  2. Continue to capture the same measurements over time so when failure occurs, the data can be used to determine failure mode patterns.

As time and budget allow, power quality analyzers, digital multimeters, handheld oscilloscopes, resistance testers, and more can be added to the system.

Variables to keep in mind when starting such a reliability program:

Optimal conditions

With thermal imaging, ambient temperature, wind conditions and time of day all will affect a scan. Cooler temps and calm winds are optimal but should be balanced by equipment load. The best view into the health of equipment is often at the point of maximum load.  Scheduling thermal inspections during winter and early spring will make the process easier and the results clearer. Subsequent repairs can be completed before the high-demand summer season when maintenance is more difficult.

Vibration analysis, on the other hand, uses a piezoelectric accelerometer to measure vibration from a dynamic machine train. You aren’t measuring the vibration at the source of vibration (from the rotating shaft) but instead are measuring from the machine's bearing housing. This means that you are really measuring the response of the machine’s structure to the vibration from the rotating shaft inside, the components on the shaft, the bearings, the covers, and the foundation. There are many random vibrations mixed in with the vibrations from the rotating shaft. Even the repeatable vibration from the rotating shaft has many variables – resonances, speed and load, location, sensor mounting, environment, operational, noise, excitation, and other machine influences. 

To reduce random vibration, noise, and variables

  • Make sure the machine is at the same speed and load each time a measurement is taken
  • Make sure the machine is running at the same operating conditions
  • Make sure the same machines in the area are running at the same operating conditions.

Data analysis

The goal with both thermal imaging and vibration in a reliability program is to trend the data, capturing regular results over time – and catching irregular results quickly. By establishing a baseline reading on equipment that operates at high temperatures, a range can be established that represents a normal state for that particular asset. Any deviation from that range gives a clear indication of change and the potential for component failure. A similar method can be used for vibration readings with the caveat that exact repeatability is limited.

The real benefit of conducting periodic thermal and vibration inspections is achieved in making repairs during routine maintenance or planned downtime. By starting small and creating a workable program that has room to grow, you can quickly demonstrate to plant managers the value of a reliability program in keeping plant machinery operating as designed and maximizing uptime.