My Jeep Grand Cherokee is a smart vehicle, and it’s comfortable, too. The oil-change predictive maintenance alert has saved me a bunch of dough. A lot of cars have this feature, and some mechanics say that you should rely on this condition-monitoring function.
Of course they would. Oil changes are a great money-making exercise, especially when you use synthetic products like the Jeep does. More than 100 bucks a turn relates to your wallet the same way production numbers relate to a spreadsheet.
The car takes many things into account—speeds driven, stops and starts, engine rpm and distance traveled —while figuring out how to reduce the oil life percentage, so it takes most dynamic issues into account.
Equipment condition monitoring can take many forms. Vibration, temperatures, dynamic modeling and current trends are a few of the measurements that could be used in determining when a piece of equipment can fail.
That is the function of condition monitoring—reducing or in fact removing unplanned downtime due to equipment failure.
I remember a service call I went on at a company that wound small coils for solenoids. I had designed and written the PLC code for the winders, and there were 30 or so of them. The only parameter we used to determine if things were going awry was cycle time. The average cycle time was recorded every week to see which machines may need some attention.
The machine in question just stopped working, and the company really didn’t have a maintenance department as such, and this was back in the 1980s when the PLC was always the cause of the problem.
I asked the operator what was going on when the machine stopped. She related a story to me that rang so true: “I heard a screech and then nothing,” she said. I asked her if she had heard any other noises or unexpected behaviors, and she had been hearing a bit of a whine for a few days.
Condition monitoring, which includes acoustic measurements, would have caught this anomaly before the unscheduled downtime happened.
With the advent of edge computing and IIoT, condition monitoring can be done very easily, be cost-effective and provide invaluable insight into the productive process model, which in the end can improve the bottom line.
One of the byproducts of inefficient operations due to failing devices and processes is wasted resources such as energy. Why give a process more energy than it needs?
Colgate-Palmolive has embarked on an IIoT plan for monitoring its compressed air systems. Air leakage overworks the compressors, and the quality of the air leaking into the immediate environment could be in question.
An added benefit of this leak-detection-monitoring approach is the potential optimization of the air system in general. Why have 120 psi in that system when 80 psi will do? The compressor works less, saving dollars and improving OEE.
The flow sensor it is using for IIoT measures flow, pressure, velocity, volume and energy. It is feature-packed, I would say.
The sensor has Ethernet built-in—web server and message queue telemetry transport (MQTT), as well—so it can be integrated into any control-system architecture.
Condition monitoring is also a data-collection exercise. A smart factory would and could use this data to have processes respond by providing an agile response to a process condition.
Actionable information is all the rage, and IIoT condition monitoring can provide that for you.
The condition response is the analysis of the data collected. This can range from a generated text message to an executed function in the control system.
This closes the loop on the monitoring data. If you are using acoustics to detect bearing wear or vibration analysis, the response to the monitoring is pretty simple—create a work order for replacement at a convenient time.
Other detections such as low oil levels would be managed by the control system, but viscosity changes can create multiple changes in the process, such as changes to pressure/volume and flow. The response to this event could be an increased introduction of fresh oil and an increased rate based on this condition.
There are a ton of machines and processes out there that are older than dirt, and have obsolete control systems. The Organization for Machine Automation and Control (OMAC) was created because of the lack of an Ethernet module for a CNC machine. Open control was born.
But with the advent of IIoT, industrial Ethernet and MQTT/cloud options, very smart sensors can be implemented into any machine or process since they are edge- and present-day enabled.
Colgate-Palmolive and its efforts in trying to reduce its carbon footprint is fascinating. IIoT will be an integral part of any condition monitoring and response system. What can you do to improve your overall process if you really thought about it? The options are endless.
