There are three main types of maintenance food and industrial processors encounter–preventive, predictive and corrective.
Preventive maintenance is planned and then performed on equipment in advance to help ensure failures, and ultimately, unplanned downtime, won’t occur while also mitigating the chance of breakdowns. Examples include regular cleaning, lubrication, inspections, parts replacements and adjustments.
Corrective maintenance, on the other hand, is performed when the equipment fails. It’s unplanned and therefore can potentially result in downtime that can be harmful to the company’s bottom line. Examples include when a bearing fails, a seal breaks or a motor goes out. Corrective is probably the most costly forms of maintenance.
Predictive maintenance is almost like a mix of those two–it involves recognizing when a part is going to fail and planning downtime to fix it. Examples using data, such as rising noise levels, to predict when a bearing is about to fail.
At EnSight Solutions, our team is taking a proactive approach to helping our customers mitigate unplanned downtime by making predictive maintenance a bigger possibility.
One straightforward, cost-effective way of doing that is with a sensor.
How Sensors Work
Sensors are really a great option for predictive maintenance because they can measure a wide variety of variables. Here are some of the most common, along with how they work:
- Heat – Many parts, such as bearings and internal jackets, heat up during use. Companies can use heat sensors to determine what the normal operating temperature of those pieces are during regular operation. In regard to bearings, the sensor can register if the temperature is approaching critical levels, at which point the bearing could encounter a catastrophic failure. For internal jackets, the sensor can measure how often the machine is heating up and cooling down (which results in the metal jacket expanding and contracting) to help determine how long until the jacket cracks and fails. In both instances, the company can order new parts ahead of time and plan the downtime instead of reacting to a machine failing.
- Vibration/Noise – Noise is a biproduct of vibration (the more a piece vibrates, the louder it operates), so a company can really measure either to get the same result. With a simple vibration or decibel sensor, a company can establish a normal operating range (just like it can with a heat sensor). Then, when the vibrations or noise surpasses a certain level, the company knows it’s likely time to order a replacement part.
- Moisture – A moisture sensor is a great way to predict the impending failure of a seal by measuring any water leakage. One way original equipment manufacturers (OEMs) can measure leakage is by installing a chamber below the seal and then installing a sensor that can detect the presence of liquid in the chamber.
Alerts
Whichever sensor would best serve a company’s particular application, the main point is for them to alert team members when the readout goes beyond the normal range. This can be done via the programmable logic controller (PLC). The sensor and PLC can be programmed to alert team members at selected intervals.
Sensors don’t have to be factory-installed by the OEM, either. They can be retrofitted to any machine with a PLC.
No PLC? No problem. Those can also be retrofitted to existing machines.
The EnSight team is continuing to integrate ways that OEMs can become more plugged into the sensor alert system to help companies with their predictive maintenance. If a company allowed the OEM access to the sensor data, the OEM could potentially be alerted when a part was nearing failure, and then preemptively procure and send a replacement part to the customer. This access would also help OEMs study their machines in the field and gain valuable data regarding their uses, life cycles and more.
Considerations
As mentioned previously, sensors are relatively inexpensive, readily available and easy to install. But, before a company goes out and purchases a sensor from a third-party vendor, they should consider who they want to own that data and how they want it used.
It is in the company’s best interest to look to their equipment’s OEM for that support. The OEM is going to be the entity that knows its machines best, along with the function of each part. It is best equipped to understand the data properly in regard to how the machine works, and therefore recommend predictive maintenance at the proper intervals.
Interested in Using Sensors for Predictive Maintenance?
The EnSight Solutions team is currently researching sensors and how we can use them to help improve machine efficiency while reducing downtime. We believe that with sensors, we can better work with our customers to improve their overall operations, while also improving our own machines.
Interested in learning more? Contact EnSight Solutions now to speak to a sales specialist. Click here to locate and contact your sales specialist.