Predictive Maintenance System Curbs Pump Downtime

Sensor-based system sends data to the cloud, where servers identify potential problems.

Every year, countless manufacturers struggle with unplanned downtime after a pump or blower unexpectedly fails. In the worst cases, the pump or blower can take down a whole process line, leading to major revenue losses.

Now, however, a new sensor-based, predictive maintenance and analytics system offers hope for such scenarios. PumpSense -- which combines sensors, software, and wireless communications -- promises to tell factory floor managers when a pump or blower is deteriorating and, if so, why. The product is designed for integration by OEM pump manufacturers.

“Our platform monitors 24/7, and it doesn’t just tell you if your bearing is getting bad,” noted Biplab Pal, chief technology officer and founder of Prophecy Sensorlytics, LLC. “It finds the root cause of your problem and tells you why the bearing is getting bad.”

 

PumpSense displays information on an app, which displays the condition of bearings, valves, filters, belts and oil. (Source: Prophecy Sensorlytics, LLC)

 

Although diagnostics systems for pumps and blowers have existed previously, Prophecy Analytics claims its new system is a departure from the past in several key ways: It’s simple, inexpensive, wireless, always on, and more deeply analytical than predecessors.  

The key to the system is its use of cloud-based servers to analyze the 24-hour, seven-day-a-week stream of data from the pumps and blowers. It uses a MEMS sensor and vacuum sensor mounted on the pump to measure such variables as vibration, temperature, pressure and vacuum. It then “talks” via Bluetooth wireless to a hub, which gathers and converts the data, and sends it along to an Internet-based router. From there, data goes to cloud-based servers that run predictive analytics software and then send the results back to an app on a tablet or smartphone. The handheld device then displays vital information, such as filter and oil status, bearing conditions, vibration data, and vacuum performance, among other parameters. Color-coded data windows -- green, yellow, and red -- alert users to the urgency of the information.

The use of wireless is important because it simplifies set-up and reduces the cost of the system, Prophecy Sensorlytics says. In the past, manufacturers sometimes used Ethernet cables to make connections, but such systems could be challenging near crowded, dirty process lines, Pal said. To solve the problem, Prophecy Sensorlytics created its own proprietary middleware to enable wireless connections to “self-repair” in the event of temporary disconnection. That way, the system is always on, and always streaming data, Pal said.

Moreover, the system’s analytics eliminate the need for outside experts to make sense of the data. “In the past, you would have needed expertise on staff, or you would have had to consult with a third party to get any value out of your data,” noted Jim Zinski, president and chief operating officer of Prophecy Sensorlytics. “With this product, the analytics are done in the cloud and summarized on a dashboard, which also lets you view the underlying trends.”

The company is also rolling out a pair of similar products called MachineSense and ElectroSense for industrial applications. MachineSense monitors the condition of bearings, valves, belts, gearboxes and stators on industrial machinery. And ElectroSense monitors electrical parameters on lines for electric motors and components.

The goal for all the products is to help manufacturers promote uptime, Pal said. That way, time and revenue aren’t lost, and factory floor interruptions are minimized. “You should always be able to know if something catastrophic is coming, and stop the unplanned downtime,” Pal said.  

Senior technical editor Chuck Murray has been writing about technology for 33 years. He joined Design News in 1987, and has covered electronics, automation, fluid power, and autos.

 

 

Comments (0)

Please log in or register to post comments.
By submitting this form, you accept the Mollom privacy policy.
  • Oldest First
  • Newest First
Loading Comments...