Rethinking Pump Maintenance in a Digital World

Manufacturers are ready to unlock a “hidden plant” if they can get their assets to operate optimally. 

Pump maintenance is often left to technicians, and that can turn into a problem. While techs know how to keep a pump rotating, they don’t always have the training to pick up on the early warning signs of trouble. 

With 30 years of experience, Randy Riddell, has knowhow to spot the first signs of trouble before they become a problem. As site reliability manager at Essity Professional Hygiene North America’s Barton, Ala., paper mill, Riddell has experience with everything from hands-on pump systems diagnosis to manual and wireless preventive maintenance systems. He has also helped design the examinations for the Hydraulic Institute’s Pump System Assessment Profession certification.

In this interview, he reflects on some of what he has learned along the way. 

You have been working in pulp and paper plants for 30 years. How did you start in the industry?
In high school I got a job as a hydraulic tech at Geartek, a local hydraulic pump company. I did everything, from building and repairing to rebuilding and testing hydraulic pumps, motors, and cylinders. I liked it, so I decided to major in mechanical engineering at Mississippi State University. My hands-on experience with pumps helped me get an internship at a paper mill when I was in college, and that internship led to offers from several pulp-and-paper companies when I graduated. They operated huge fleets of pumps, and if you’re a pump person, it was a dream job. 


How did you wind up on the maintenance side?
I started out in project engineering, where I really learned what my professors had been trying to teach me in class. But after a few years I switched to maintenance. After project engineering, it was like working backwards. I started with an existing system but had to apply the fundamentals I learned to diagnose problems and find solutions. I found it more challenging, and I liked that. No two pumps are ever the same and it’s never boring. 

You have been reliability manager Essity’s Barton paper mill since 2014, but your role goes beyond that, doesn’t it?
It does. As the site reliability manager I get involved in nearly any equipment or maintenance related activity. I do have specialty areas such as pumps and systems that I take special interest in. I also oversee our repair programs, day-to-day maintenance, reliability programs, outage execution, and predictive maintenance. I’m also heavily involved in critical failure analysis. Maintenance professionals wear many hats in today’s plant organizations. 

I also consult on pump systems in other situations in our plant and with other plants. I’ve spent a lot of years in pump system reliability and for many problems it takes a deeper look to solve the problems. From design for reliability on capital projects to determining the scope for correcting existing issues on some of our worst pump systems, a pump system specialist can provide a tremendous return for a plant that shows up in the lifecycle of the asset.

I’ll give you an example, one where vibration analysis and pump system assessment combine to solve an issue. During normal vibration analysis on a pump we found vibration had increased about 300% around vane pass and discovered looseness in the acceleration waveform. Most continuous monitoring systems looking at an overall vibration would have called for maintenance to be completed on the pump by changing rotating assembly. However, after assessing the pump system an issue was found on the VFD that was causing the issue. This just shows how important the analysis piece is for determining causes and corrective actions. 

With so much emphasis on running lean, are companies devoting enough resources to reliability?
Pump systems in pulp and paper are such a huge piece of the operation and their impact on maintenance budgets and energy usage make them prime targets for reliability improvement. But I also think that a lot of companies in our industry and in other industries do not realize the maintenance, reliability, and energy impact of pumps and other equipment in their plants. This is really a great opportunity, and managements that address it could unleash the hidden plant inside their factories.

One way to discover that hidden plant is to use predictive maintenance to reduce maintenance costs and catch problems before they lead to failures and work stoppages. Is this something you are doing?
To say that is a hot topic today is an understatement, even though predictive maintenance technology has been around for decades. Implementing predictive maintenance involves a learning curve and many companies are just starting to move up that curve. Just collecting data, establishing baselines, and trending can help identify Stage 4 failure. But the advanced analysis needed to discover Stage 1 failure and other pre-failure conditions will require more tools, experience, and software power. Companies can’t go from having no predictive maintenance program to Industry 4.0 and skip all the steps in between. Learning the application of vibration analysis on equipment requires an understanding of the equipment itself and mastery of the predictive technology. Just applying the technology will not produce the best results. 

Yet there is still a big move now to sensor every piece of equipment to get continuous feedback on operations. Does that make sense today?
To some degree, yes. I see only a few critical applications where it is really financially feasible to execute, at the moment. That said, wireless sensors for predictive maintenance are maybe 20 years old, but in the last five years or so the technology has made some real advances. We’ve run trials on several different systems. Most of them are developing but there are several, like Pumpsmart from ITT Goulds, that have been around for many years. Everyone seems to be fighting for market share. I think some of the pump OEMs, have a very good handle on what they can do and are very advanced. 

Still, there are gaps. We ran a trial of a continuous monitoring systems in parallel with our system. Our system found four defects and avoided those failures. The trial system never even alarmed. It just didn’t have the ability to do the type data collection and analysis that our system had. Of course, we also have a great analyst, Gary Holden. He’s a Level 3 certified vibration analyst and he’s very good at what he does. Combine that with the equipment/system we use and it’s just hard to beat. 

So, tell me about how you do predictive maintenance?
We use an Emerson hand-held data 2140 collector/analyzer and MH software to do more advanced analytics. The most important thing we do is vibration analysis, though we also gather other data, such as oil analysis. We have a set route that ensures a technician analyzes each piece of equipment once every month. Being able to identify Stage 1 failure with our technology and experience allows us to be very effective for most failure modes within a 30-day P-F interval [the time between potential and functional failure]. 

Do you see yourself going to wireless sensing in the future?
At the corporate level, Essity is partnering with Augry to develop a company-wide condition monitoring program. We’re putting a lot of resources into it, and many plants have gotten off to a good start on their predictive maintenance journey. Our journey has us at a much advanced level, even though we are only doing continuous monitoring on some satellite systems. When we execute here we will have to figure out how to merge with our current program, which is operating at a very high level, so we can keep progressing as AI technology continues to advance. 

Bottom line, what will it take to make predictive maintenance work?
Somebody has to own it and keep it working. That ranges from keeping accelerometers working to updating database with new information such as a different bearing was installed. Someone has to have software upgrades completed and calibrate the equipment. Predictive maintenance calls must also be validated to confirm the analysis and learn from it. This is what makes a good program great. Sometimes it seems like the more automation we get, the more experts it takes to keep it working well. Predictive maintenance can be extremely complicated if you want it done right and desire excellence as the outcome. 

Pumps are getting smarter. Are we moving toward more automated operations?
We’re definitely moving in that direction, especially with the use of smart variable frequency drives. This gives us more data. For example, a smart VFD can give us flow feedback without having to put a flow meter on the pump. These are allowing us to simplify the layout of our plants by eliminating some of the traditional control loop valve strategies. But it’s not one size fits all. You really have to look at each situation, but smart VFDs are certainly a future reality.

Let’s talk about budgeting. When you choose pumps, are your decisions based on lifecycle cost analysis? Is there any pushback from accounting on this?
They are absolutely based on lifecycle costs. On the maintenance side, this is relatively easy because accountants typically focus on capital projects rather than pump replacements. I also use a lifecycle approach when I collaborate on projects, Operating cost and maintenance are the two largest pump cost of ownership pieces. When you get these two right on pump and system selection you end up usually getting the reliability piece right. 

A final question: You were involved in creating the assessment test for Hydraulic Institute’s Pump System Assessment Professional certification. Did you learn anything from the experience?
I was pulled onto that committee by Mike Pemberton, who had worked for ITT Goulds, because he wanted a team member with my practical pump maintenance and reliability background. Everyone on the committee had different backgrounds and I learned about some pump applications that I had never encountered before. This is what makes my job fun, you never quit learning and everyday is a new adventure. We all had different pump experience, so I’m sure I also posed problems some of the others had never seen. The goal was to create a test that let professionals show their broad, deep background in pump system assessment. I can see pump system professionals being needed in a big way in the future. Our process plants are full of complicated pump system issues.

Subscribe to the Pump Systems Matter Newsletter through our Contact Us form. 

SUBSCRIBE TODAY

Get the latest pump industry news, insights, and analysis delivered to your inbox.

UPCOMING EVENTS