NOTIFICATION: The Technology Channels is now an automated compilation of industry press releases.
Click here to download complimentary copies of Fastmarkets RISI’s pulp and paper newsletters.


Don’t wait for failure Part II

Read so far

Don’t wait for failure Part II

August 13, 2010 - 01:34

BRUSSELS, Aug. 13, 2010 (RISI) -Pulp and paper mills are under ever-increasing pressure tomeet shareholder expectations of increased profitability through improved plant efficiency. Machineryproblemscan make it harder to meet this challenge, not only by affecting product quality but also by causing mill downtime.In orderto minimise this downtime, maintenance must be carefully scheduled so that it occurs prior to a machine failure - but not so often as to incur unnecessary maintenance costs.

In part II we explore the real world benefits of continuous online vibration measurement. Part I can be readhere

A good example of continuous vibration monitoring can be found at a large pulp and paper mill in Europe. Here, three large fans provide air to a recovery boiler that produces steam for power generation within the mill. All three fans are required for the boiler to work correctly, and any failure requires the boiler to be shut down for repair. With the boiler offline, the loss of steam and power forces the entire plant to halt production.

It takes a full day to shut down and restart the boiler or to switch to a backup boiler, so even a relatively quick repair forces production to stop for at least a day. The fans are therefore critical to the smooth running of the plant, and it is essential that they are carefully monitored to maintain continuous production.

Historically the main problem with the fans has been the bearings, specifically the bearing cages. Damage to the bearing cages leads to a bearing failure affecting the housing, shaft, and possibly the coupling. There is even the chance that a motor could be damaged beyond repair. A bearing takes roughly two hours to fix while a major repair to the shaft, coupling, or motor could take between 12 and 18 hours. To minimise downtime, most parts are kept in stock as they can take days or weeks to deliver.

The mills' maintenance team are dedicated to identifying potential problems with the fans well in advance so they can schedule planned maintenance. However, because deterioration of the bearing cage condition produces little heat or vibration, it is difficult to identify the problem using traditional observations. Changes to its state are subtle and abnormal readings fluctuate considerably so they can be missed when testing on a periodic basis.

Previously the maintenance team collected and analysed vibration and temperature data from the boiler fans every two weeks using portable data analysers. The latest results were then compared with previous results to spot any trends and possible faults. When substantial levels of vibration were identified, the maintenance team would shut down the fans. Unfortunately by this point, considerable damage may have already occurred.

In addition, despite the periodic testing, some failures were still going undetected. When examining the historical information the team found that there was no change in the temperature of the oil at any time before a cage failure, although an oil probe showed high ferrous levels. A higher than normal vibration level was recorded shortly before the previous cage failure, but a second reading taken the very next day showed perfectly normal levels.

The maintenance team realised that to gain a true picture of the fan condition, vibration readings needed to be taken on a continuous basis. There was also a need for advanced analysis technology to identify and flag any minor changes that would indicate potential deterioration of the cages.

To address this problem, they decided to enhance the monitoring equipment by upgrading to an online predictive intelligence system. This would collect more data and then use it effectively to identify problems and help predict potential faults. They chose the CSI 6500 Machinery HealthTMMonitor and AMS Suite predictive maintenance software, which is Emerson's primary online vibration monitoring solution supporting predictive maintenance strategies.

This technology works in conjunction with the existing process monitoring equipment to provide a complete picture of the health of the fans. Overall vibration levels are displayed so the operator can immediately see what impact any process changes have on the health of the machine. Detailed vibration data from the CSI 6500 is fed to Emerson's AMS Suite predictive maintenance software application, enabling the maintenance personnel to easily monitor the detailed health of the fans.

The CSI 6500 uses Emerson's PeakVue digital technology to immediately detect stress waves for an early indicator or bearing wear. This technology employs a unique method of processing the stress waves to enable accurate, trendable diagnostics for roller bearings and gearboxes.

The CSI 6500 also indicates the severity of any fault, enabling technicians to call for immediate maintenance, if necessary. The technology not only takes readings, but also identifies when there are any changes. The technology can spot defects, even in a cage, and can also identify which bearing has the problem.

Since this new online vibration monitoring equipment was installed, it has identified a number of potential problems including a failing fan bearing cage, failing outboard motor bearing, and an unbalanced fan. These problems have been identified much earlier than when periodic testing was used. As a result, it has been possible to schedule maintenance, reduce the length of the shutdowns, and avoid any unscheduled downtime. In the three years since the solution was installed, the system has helped prevent one significant machinery failure a year that would not have been spotted using traditional techniques, saving an estimated €720,000 in lost production.

By identifying any problems with the cages or bearings at an early stage, the mill has been able to schedule repairs and avoid shutting the plant unnecessarily. This has increased the production throughput as well as lowering overall maintenance costs.