Subscribe and listen to AW’s podcast!
Subscribe and listen to the Automation World Gets Your Questions Answered podcast!
Listen Here

Understanding Equipment Failures and How to Respond

Learn about the high costs of reactive maintenance and how moving to more proactive—and even predictive—maintenance can be achieved without major investments and executed in small, easily incorporated steps.

Transcript

David Greenfield, Automation World

0:13

Welcome to the Automation World get your questions answered podcast where we connect with industry experts to get the answers you need about industrial automation technologies. You can find even more answers by subscribing to automation world at subscribeaw.com. I'm David Greenfield, editor in chief at automation world. And the topic we'll be addressing in this episode is understanding equipment failures and how to respond to them. Joining me to answer this question is Stephen Lacy with Belden, a supplier of industrial automation cables, connectors, networking and cybersecurity technologies. So thanks for joining me, Steven.

 

Stephen Lacy, Belden

0:51

Thank you, David. Nice to be here.

 

David Greenfield, Automation World

0:53

So, Stephen, you know, though different types of equipment will of course fail for different reasons. What do you see as being the major mechanical, electrical or user causes of equipment failure in industrial operations.

 

Stephen Lacy, Belden

1:09

We've seen some of the heavy hitters there are age a lot of automation and electrical equipment is 15, 20, 25 years and aging out fast. We see high absolute temperatures either in a in an enclosure or the production environment itself. Daily seasonal temperature cycles, put mechanical stresses on chips, and dries out capacitors. Dust builds up on enclosure filters and printed circuit boards. And it's slightly conductive, which, which helps accelerate failures. And then finally, vibration. For equipment that stalled on machines, whether it's even slight or high vibration, wears out connector, contact points and solder joints over time for mechanical equipment, motors and servos. The failure modes there we see our bearing and joint lubrication challenges, overloading of motors, misalignment in the X, Y or Z planes of the driven load that's connected to two motors or servos. Off spec materials, many times can cause equipment jams, and then finally, operator mistakes, just as a normal part of doing business and running a production environment.

 

David Greenfield, Automation World

2:39

So given those types of failures you just described, are there certain types of equipment that are more likely to fail from these reasons and others? 

 

Stephen Lacy, Belden

2:48

Yes, the ones we are we're most familiar with the kind of make the Pareto chart of the of the heavy hitters. Again, variable frequency drives, power supplies, most of them being 24 volt. But of course, there's there's other other voltages, non industrial rated network switches are quite common and can cause a lot of grief, non industrial non hardened PCs being used for for HMIs. Ethernet media converters are definitely there, particularly the less expensive ones that have a external power supply that plugs directly onto the wall, for example, protocol converters, they seem to age out and when they start to fail, it's a kind of intermittent present intermittent problems that make them more more challenging to find out which one is giving you the headache. And then finally, AC and DC motors due to bearing bearing issues winding or brush failures.

 

David Greenfield, Automation World

4:00

So with all of those reasons and types of failures that are common that we've been discussing here so far, you know, it's there's still the fact that many manufacturers still rely on reactive maintenance rather than being more proactive. So based on your experience working with end users and industry, can you talk about what the typical cost factors are that are associated with this more generally used reactive approach?

 

Stephen Lacy, Belden

4:27

repair or replacement of the failed components obviously, is the first thing but that's assuming it can be quickly sourced and shipped and we all know that is that can be quite challenging nowadays, there's maintenance labor to disassemble repair and then reassemble the equipment. And depending on where the failed component is, you know, within the machine or the plant area, that may require working at height, which which has an associate of which has an associated risk. There's makeup Production labor which may be required during overtime hours. unplanned maintenance can also have a safety component again depending on on the hazards involved random failures puts a facility into an off normal operational state that the upstream and downstream equipment and the operations team may not be prepared for. So, you definitely want you know, you would like to avoid that if you win when you can. equipment failures can greatly lower the production KPIs of a plant or a facility. And unreliable production equipment can create an atmosphere of uncertainty over whether production goals can be met. So the goal here is to really ensure that the production environment is set up and is robust enough. So meeting production goals can be solely based on personnel performance, and not on machine or equipment issues. So you definitely like to get rid of those. Finally, production delays are always a concern, since they could make your customer try a competitor's product. 

 

David Greenfield, Automation World

6:28

And that's an area you know, you would you'd like to you'd like to stay away from,

for all the obvious reasons. They're absolutely, yeah. So you know, right now, you know, we're talking about reactive versus proactive maintenance. But, you know, across industry right now, there's a lot of interest in predictive maintenance to allow for even greater levels of proactive maintenance. But is it possible without specialized predictive maintenance software to predict when equipment is going to fail?

 

Stephen Lacy, Belden

6:52

It is, yes, that's, that's a condition statement, but it's getting better, better and easier all the time. We start by first identifying all the production of production critical components and equipment in the is in the automation, networking sensors and electrical component realm. If, if the facility of the plant does not have drawings or accurate, accurate drawings, then sketches are fine, possibly with photos. Once you know what you have, the next step is start by installing inexpensive wireless sensors. Those are nice because they operate outside of your current you know it and OT networks which are production critical. The sensors available nowadays can measure position, current motion, distance, pressure, velocity, you name it. So you're there's an inexpensive sensor to identify most, excuse me most, most any, most any critical parameter. Next is replaced failed power supplies, with smart monitored supplies. There are some units out with IO link that's built in. And some of the parameters that are included in that link is actually the remaining useful life of the supply, which plays to the to the predictive analytics nature of the exercise. installed network monitoring software to detect when a network switch may be starting to fail or has failed, you'll see spikes in network traffic, that that don't align to anything else and those may be indicative of an impending failure. And you can install wireless motor vibration and temperature sensors on motors with an analytic software package. And I would just note that the solution ideas described can operate either on the existing facility networks or they can be on a separate smaller network dedicated just to the equipment monitoring and analytics.

 

David Greenfield, Automation World

9:10

So considering all those options facing manufacturers, you know when it comes to equipment failure prediction approaches, where do you suggest manufacturers start?

 

Stephen Lacy, Belden

9:21

I would start by quickly estimating the costs associated with downtime for a specific machine or a specific plant area. This is really important since it helps determine the ROI or at least give you give you a ballpark ROI and enables you to quickly get a proof of concept approved internally. If you haven't started you know any industry 4.0 or dated data driven programs start small on one machine or one process area which limits the cost and the risk See, this also allows all stakeholders from top to bottom of the company to engage and see the results of the concept. And then suggest better, faster, cheaper ways of doing of doing following ones. This approach also allows time for the for the facilities personnel to understand the value of the effort and get buy in from them on the approach.

 

David Greenfield, Automation World

So getting back to the cost issue of all this earlier, we were talking about the reactive maintenance costs, but how expensive is it to apply these more predictive approaches that you recommend?

 

Stephen Lacy, Belden

10:37

Right so in industry 4.0, and industrial internet of things by their nature, they're designed not to be expensive, to be able to smart small and to be scalable. The main thing here is to start small—one machine one process. You can start by adding a few new sensors to existing equipment, and adding signals that are already available from your existing equipment, your existing machines, your PLCs and, or and or a DCS if you have one. As the journey continues, you may find value adding additional signals from other existing plant systems, such as the warehouse management system, your integrated enterprise resource planning system, your ERP system, manufacturing execution systems, QA, QC systems, laboratory management systems, etc. To help determine correlations and identify the root cause of the problem. Separately, small plant floor displays can be added so all employees from top to bottom know the root causes of the production loss. And note that everything described above is and typically should be vendor agnostic. So you have lots of choices in the in the marketplace. 

 

David Greenfield, Automation World

12:05

To help put all of this into perspective. Steven, can you share some details around equipment failure issues that Belden is helping its customers with?

 

Stephen Lacy, Belden

12:13

So we're installing wireless vibration and temperature sensors on all our motors and all our manufacturing facilities. Machine learning software is then used for anomaly detection. It analyzes those two variables and then alerts via email or text when they've gone out of when they've exceeded their normal operating range. So that's one area where we're starting with one of the customers we're working with now, they have a problem with polypropylene plastic conveyor belts that frequently fail, the belt failures stop all finished products from leaving the plant and have a significant cost six figures a year. To address that. We designed a custom belt tension and belt speed monitoring assembly, and a small control panel that collects those signals along with other existing VFD signals via an edge device and sends those to the cloud. We're in the cloud, we're building a predictive model of each belts remaining useful life. And the model will automatically adjust over time as the causes of the belt breaks are identified and corrected. So we're really excited by both these use cases. And we're looking forward to doing more. 

 

David Greenfield, Automation World

13:39

Well, very interesting, you know, thank you again for joining me for this podcast, Steven. And thanks, of course to all our listeners. And please keep watching this space for more installments of automation world get your questions answered. And remember you can find us online at automation world.com And subscribe to our print magazine and subscribe aw.com to stay on top of the latest industrial automation technology insights, trends and news

 

Discover New Content
Access Automation World's free educational content library!
Unlock Learning Here
Discover New Content
Test Your Machine Learning Smarts
Take Automation World's machine learning quiz to prove your knowledge!
Take Quiz
Test Your Machine Learning Smarts