Skip to the main content

“It tore through the tower”: How AI and the IIoT can help end cooling tower maintenance nightmares

By Grayson Morris

See LinkedIn profile
In this article

Find out how modern technology is helping traditional condition monitoring techniques become round-the-clock warriors in the fight against cooling tower failures.

Day dawns, promising sweltering temperatures. A sudden gust of wind shrieks past a cooling tower. The resulting stress, though minor, is the final straw for one of the fan’s overworked blades. The blade shears off. The vibration switch that’s supposed to shut down the fan at times like these doesn’t; moisture has crept in and killed it. So the fan keeps turning, until the resulting imbalance rips the rest of it from its support, taking the motor and the gearbox with it. All three parts plummet toward the sump basin, taking walkways, piping and a 30-foot section of fill with them. And stopping production at the plant the tower serves. On hot summer days like this, there’s no redundancy to spare.

For cooling tower maintenance and reliability engineers, this is the very real stuff of nightmare.

Not every cooling tower failure is as dramatic as this one—which is inspired by a failure that actually happened in the 1990s—but it serves to illustrate why it’s so important to closely monitor the health of cooling tower components. Periodic manual measurements aren’t enough; maintenance staff need continuous insight in order to spot developing faults before they become failures.

Modern technology to the rescue

Fortunately, most tried and trusted condition monitoring techniques are rising to the task, thanks to artificial intelligence (AI) and the industrial internet of things (IIoT). Permanently installed sensors can now collect vast amounts of high-quality data, which machine learning algorithms can process in near real time. Thanks to this immense scale in both data and processing power, AI-based condition monitoring systems can detect minute changes that signal the start of developing issues, often months in advance.

Sidebar: A new name for an old tool

This AI-powered pattern extraction goes by many names: big data, business intelligence, data mining and more. When it’s used to monitor the health of industrial equipment and extract insights that predict when machine failure will occur, it’s often called predictive analytics. But that’s just a new name for an old tool: it’s still condition-based monitoring—just faster, more accurate and far more scalable than traditional techniques that rely on periodic manual measurements and human analysis. Welcome to Industry 4.0!

So which Industry 4.0 condition monitoring system should I use?

That depends on what equipment you want to monitor. Some things haven’t changed: for example, oil analysis still can’t monitor components that don’t use lubrication, no matter how much AI you throw at it. You also need to know what types of faults you want to spot early. Vibration is still a champion for early bearing failure, while MCSA is the winner when it comes to electrical failures inside the motor. And finally, you need to consider the environment where the sensors will be mounted.

So you see, each technique has its strengths and weaknesses. And that means the best choice of all is to use the right technique for each situation. There’s a lot to consider in planning the best condition monitoring strategy for your cooling towers, but don’t despair: we’ve written a comparison guide to help you find your way. It’s a detailed guide to five of the most reliable condition monitoring techniques for pumps, fans, motors, gears, bearings and more. We hope it helps you put a permanent stop to your cooling tower nightmares!

The condition monitoring comparison e-book for cooling towers

Learn how current, voltage, oil, sound, vibration and heat can give you early insight into developing cooling tower failures—and which techniques work best in which situations to keep critical cooling tower equipment 100% up and running.

This e-book covers:

  • the anatomy of a cooling tower
  • condition monitoring in cooling towers
  • examples of when different techniques will detect different failures
  • energy and performance insights

Fill in the form to download the e-book.


72 See all resources

Extend pump lifespan and save energy with SAM4’s new real-time pump curve

We wanted to help our clients keep their pumps operating at their best efficiency points in the first place. So we’ve developed a tool to do just that. Meet SAM4’s new real-time pump performance dashboard!

  • Blogs
Read article

How SAM4 detects developing faults in industrial pumps

In this report, we explain exactly how SAM4 detects common faults in industrial pumps using actual results from anonymized SAM4 data, to help engineers evaluate SAM4’s value for their own operations.

  • Technical documents
  • White papers
Read article

Condition monitoring in the water and wastewater industries

Learn how SAM4 reduces unplanned downtime in critical water and wastewater processes.

  • Blogs
Read article