I’d like to draw your attention to this issue’s Maintenance & Asset Management Journal. It features an article by Gary Einarson, vibration technologist at the Yara Belle Plaine nitrogen fertiliser plant in Saskatchewan, Canada. In it he describes how the plant suffered from a longstanding problem of valve wear and failure in positive displacement pumps.
Vibration data wasn’t picking up the problem. Over a period of several years, the team sought a way of detecting or monitoring the onset of wear and of anticipating failure.
Eventually they surmised that fluid leaking through the small opening between a worn valve and its seat would emit ultrasound, and they discovered that this could be successfully detected by monitoring spike energy time waveform readings. The ability to detect incipient failure has greatly improved pump reliability.
It’s an excellent example of real-world problem solving out in the field. There must be many more stories out there like it. M&AMJ is always on the lookout for material, and if you feel a project you’ve undertaken or a problem you’ve solved could be turned into an article to provide similarly useful insights to colleagues in industry, please get in touch. You can send a synopsis, in the first instance, to david@maintenanceandengineering.com