A LDPE company in the US avoided costs of 250 M for line cleanings, labor and lost production. They made use of deep diagnosis by the PROGNOST®-NT system. Both Hyper / Secondary compressors (commissioned 1976) have been equipped with a monitoring system. In 2005, the customer decided to replace the existing system with the latest PROGNOST®-NT technology .
Old monitoring system
This system monitored process pressures and temperatures, lube oil pressures and temperatures, main bearing temperature, plunger proximity (x/y). In 2002, the customer added impact monitors to the system.
Both machines have a failure history:
- Main and connecting rod bearing failures (resulted in crankshaft damage)
- Main crosshead and auxiliary drive yoke slipper failures
- Discharge valve spring failures
Some of these failures resulted expensive in line cleaning procedures.
Change of monitoring strategy
In 2005, the PROGNOST®-NT was installed. Existing measuring loops have been integrated and new loops added. This resulted in the following configuration:
- Crosshead vibration
- Cylinder vibration
- Plunger temperatures (sliding contact TC’s)
- Integration of plunger proximity probes
- Integration of main bearing temperatures
- Integration of relevant DCS process values
First results of new monitoring strategy
Within the first months, several damages occurred and where precisely detected by the early failure module of the system.
- April 4, Discharge Valve Failure, machine 2
- June 1, Discharge Valve Failure, machine 2
- July 5, Discharge Valve Failure, machine 1
- July 7, Discharge Valve Failure, machine 2
- Aug 22, Discharge Valve Failure, machine 2
- Sep 4, Packing Failure, machine 1
- Sep 5, Discharge Valve Failure, machine 2
Benefits and Conclusion
PROGNOST Systems supported the customer in the early detection of various valve failures within the first six weeks of system operation. This allowed early intervention before more serious valve damages could send valve pieces downstream. All in all 250 M dollars comparable to two line cleanings (plus labor and lost production) have been saved.
Cylinder vibration data stopped the guesswork in diagnosing, and is illustrated by the collected data.