Michael-Howard-2
Michael-Howard-2
Michael-Howard-2
Michael-Howard-2
Michael-Howard-2

How can sensing technologies help to diagnose machine health?

Aug. 11, 2016
Sensors provide early indication of equipment degradation.

We asked a panel of 15 industry experts: What impact do sensing technologies have on the ability to diagnose machine problems?

Paula Hollywood, ARC Advisory Group: In the case of diagnosing machine health, monitoring sensors, specifically, vibration and/or temperature sensors have no impact on the machine’s ability to perform its function. In this case, the purpose of the sensors is to monitor the machine’s internal components to provide an early indication of degradation in machine performance. The data from the monitoring sensors provides owner/operators with indications of machine health well in advance of failure. The broad availability of smarter, lower-cost sensors, wireless connectivity and big-data processing tools make it cheaper and easier to collect performance data and monitor equipment health parameters.

Paula Hollywood is senior analyst at ARC Advisory Group.

Michael Howard, DMDII: Sensors can monitor machines in real time and give the machine the ability to notify an operator of an error as soon as there is a problem or, in some cases, even before the problem occurs. Spindle vibration sensors or temperature sensors generate data that can be analyzed to determine if the machine is operating in a normal state or if some operator intervention is required. Sensors can also help to reduce the amount of machine downtime required for routine maintenance, as an operator will have some advance insight into which parts of a machine require more attention or may require replacement or repair.

Michael Howard is project engineer at Digital Manufacturing and Design Innovation Institute (DMDII) in Chicago, Illinois.

Adrian Messer, UE Systems: Ultrasound technology is very capable of finding early-stage bearing failures in rotating equipment. Because we can identify these failures early, it allows for proper planning of repairs, thus decreasing repair costs and avoiding costly downtime due to unplanned outages.

Adrian Messer is manager of U.S. operations at UE Systems.

Jill Oertel, Sick: Intelligent sensors are often self-diagnosing through health output monitoring, which allows machine problems to be easily diagnosed. Via remote connectivity, a sensor can report on various aspects of its condition, from build-ups of dirt and grease that can interfere with performance to operational inconsistencies that may increasingly become worse and require component replacement. Alignment changes resulting from wear or impacts with objects can also be reported.

Also read: Can I monitor and program using the same remote-access hardware or software?

For example, safety laser scanners that safeguard workers from process hazards can be remotely checked and adjusted so maintenance employees can address conditions such as dirty or obscured screens, infringed scanner fields and conditions where a reset may be required, such as the need for device replacement. Signals can be tied back to a PLC via a serial port on the front of the scanner. The scanner can also be connected to a safety controller, and then the data is transmitted via a gateway, for example.

Some intelligent sensors offer the flexibility to alert owners to the need for maintenance for a specified condition, such as minimal vs. excessive dirt build-up. Often, owners can adjust the sensor's parameters to compensate for these conditions, keeping the system operational long enough to coincide with a more convenient, less costly and disruptive maintenance and downtime schedule. And, with the advantage of stored settings, the sensors can easily be reset in conjunction with an overhaul or other major maintenance effort, or to redeploy a machine for a new task.

The health output monitoring abilities allow sensors to diagnose machine problems, allowing longer machine operational time and less unscheduled maintenance.

Jill Oertel is national product manager at Sick.

Helge Hornis, Pepperl+Fuchs: There will be significant changes as new sensing technologies make use of measuring data internally and communicating to the outside through IO-Link externally. Here is an example: Take a retro-reflective photoelectric sensor frequently used in material handling systems such as conveyers. It works by shining light on a reflector and detecting the returned signal. In the past, the sensor determined that it saw the reflector by evaluating the amount of light returned.

Since an object that passes through—let’s say a cardboard box—is much less reflective than the reflector, the amount of light returned is low and the object is detected. But what if we are detecting packing that is metalized and quite shiny and reflective? What about a silver metallic car body or an item made of chrome? Still, a reflector should return more light, but the situation is a bit more challenging. Make this even tougher by having stickers on the box that are reflectors? Modern, marketing-driven designs may demand that even this kind of packaging is securely detected. This is where measuring comes in. Take, for instance, a sensor with pulse ranging technology. In this case, we are not just evaluating the returned signal, but the distance to the reflector. With this technology, the sensor can reliably determine that the reflective material is passing through, but the reflection is not the same intensity and not at the same distance as the reflector. Now add IO-Link and the ability to transmit the real distance to the object or the reflector.

In this case a diagnostics system can continuously evaluate those values and know when the reflector is losing alignment or is getting dirty. Looking at the strength of the returned reflector signal, one can determine when it is time to clean the reflector or lens of the sensor. Looking at the distance to the reflector makes it possible to identify mechanical changes. In either case, predictive maintenance is possible.

Helge Hornis, Ph.D., is technology director—factory automation at Pepperl+Fuchs.

Jason Tranter, Mobius Institute: Rule 1 is that you need vibration sensors that provide accurate, repeatable measurements across the frequency range of interest. Rule 2 is that you need those sensors to be mounted as close to the bearings as possible and, ideally,, in positions that capture the three axes of vibration motion. Rule 3 is that the senses must be able to withstand the environment.

Therefore, the costs of those sensors, including the installation, mounting and wiring costs, are critically important. If they are too expensive, then fewer sensors will be mounted on the machine. Technologies that reduce cost, utilizing wireless communication, for example, increase the likelihood that early and accurate diagnosis can be made because an adequate number of sensors can be installed on rotating machinery. And, if it is possible to incorporate additional data, such as temperatures, speed, and operating states, the ability to perform the diagnosis with more complicated machinery will be enhanced.

Jason Tranter is founder and managing director of Mobius Institute.

Weishung Liu, Fluke Industrial Group: Sensing technologies can play an important role in diagnosing problems, but further inspection with other technologies is often required.

Tools capture electrical measurements to aid in troubleshooting. By baselining and trending data, we can begin to understand where and when problems occur. We can also correlate electrical measurements with thermography and vibration, for example.

Weishung Liu is product planner at Fluke Industrial Group.

Brett Burger, National Instruments: If you can’t measure it, you can’t fix it. Sometimes the bottleneck in the ability to measure is technology, and sometimes its lack of practicality or business models. Either way, it is imperative for machine diagnostics to get pertinent information, not just data, to the subject matter experts who know how to diagnose machines.

Brett Burger is principal marketing engineer at National Instruments.

Joe Van Dyke, Azima DLI: Sensing technologies have a huge impact on the ability to diagnose machines. Especially impactful are the latest advances that put more and more intelligence right at the machine where, combined with sensor fusion, they can produce accurate diagnostics and send actionable information upstream.

This kind of refined data is far more valuable than raw readings. The ability to be flexible and allow for saving and forwarding the raw data should not be overlooked, however, because the raw data on demand may be required for advanced diagnostics. Sensing technology that reduces power consumption, harnesses or scavenges power from the environment, reduces sensor size and allows remote connectivity continues to enhance our ability to instrument industrial machinery and improve condition-monitoring effectiveness.

Joe Van Dyke is vice president of operations at Azima DLI.

Bob Drexel, ifm efector: The majority of sensors have moved to being microcontroller-based. This has pushed sensors to functionality that has not been possible in the past. Sensors today are combining various sensing values, such as pressure, temperature, acceleration and electromagnetic magnetic field. This combining of units of measure aids in increasing the accuracy of the targeted purpose and also opens the possibility to use these different measurements for extended purposes. This includes things like flow sensors that can give BTU-usage information or pressure sensors that also track vibration patterns.

Coupled with a microcontroller, this allows for digital communication that can send multiple values that will provide operational condition status information to optimize machine performance, longevity, operation condition and granular intelligence about specific aspects of the manufactured components and completed assemblies. Therefore, sensing technologies with dual purpose will continue to aid in the evaluation of the total cost of ownership, as well as flexible business practices.

Bob Drexel is product manager, process sensors, at ifm efector.

Mitesh Patel, TCS: Sensing provides the ability to detect and quantify existence or changes of certain phenomena. Diagnostics rely on this very foundational aspect of any measurement or control system. With the advent of a new sensing technology or type of sensor, a huge number of applications and opportunities to diagnose and improve existing equipment is unlocked. Our skin is a fantastic sensor. While it protects us, it senses various phenomena in our environment. If it becomes possible to have a similar sensor for every machine around us, it will unlock tremendous potential to diagnose machine problems and provide numerous new features.

Mitesh Patel heads Internet of Things for the manufacturing industry solution unit at TCS.

Barry Po, NGrain: In the past, machine maintenance was a costly, labor-intensive activity. By enabling maintainers to anticipate and respond to machine issues before they happen, sensing technologies today are enabling maintenance activities in the enterprise to be done much more quickly and at a much lower operational cost.

Barry Po, Ph.D is senior director, product and business development, at NGrain.

Stew Thompson, CAS Data Loggers: Today’s sensors have the accuracy, sampling speed and connectivity to provide real-time data to users working in fault diagnostics applications. When manufacturers and other businesses are able to catch instances of machine wear and damage as they happen, they realize savings, not only on further equipment damage, but also on costly process delays and shutdowns otherwise caused by failures.

Stew Thompson is technical writer at CAS Data Loggers.

Tim Senkbeil, Belden: Sensors are key in diagnosing machine problems. Common in most machines, position sensors are used to guide items through the process, whether that’s material handling such as conveyor systems, assembly lines, packaging, manufacturing process control or others. Position sensors allow the system to identify jams or other faults that prevent the item from completing the process. Other types of sensors also provide useful feedback for locating a fault within a machine.

Smart sensors or actuators—those that provide additional digital information about the function and health of the sensor itself—can further assist with troubleshooting by eliminating the sensor as the source of the fault. Smart sensors or actuators also improve uptime by providing engineers with health information that can be used to perform preventive maintenance before a component failure causes a system downtime incident.

Tim Senkbeil is product line manager, Industrial Connectivity Division, at Belden.

Tom Edwards, Opto 22: Diagnosing machine problems quickly and accurately has a direct effect on a machine’s operation and uptime, or, in other words, its value as an organizational asset. Sensor data, in conjunction with testing and observation, is indispensable in diagnosing machine problems. Obtaining detailed operational information—both immediate and historical—on a malfunctioning machine can mean the difference between, for example, a technician making an approximate diagnosis and removing the machine for service versus using additional data to make a detailed diagnosis that indicates an on-site repair is possible.

Tom Edwards is senior applications engineer at Opto 22.

Homepage image courtesy of renjith krishnan at FreeDigitalPhotos.net

About the Author

Mike Bacidore | Editor in Chief

Mike Bacidore is chief editor of Control Design and has been an integral part of the Endeavor Business Media editorial team since 2007. Previously, he was editorial director at Hughes Communications and a portfolio manager of the human resources and labor law areas at Wolters Kluwer. Bacidore holds a BA from the University of Illinois and an MBA from Lake Forest Graduate School of Management. He is an award-winning columnist, earning multiple regional and national awards from the American Society of Business Publication Editors. He may be reached at [email protected]