Eliminating oil whip–induced vibration after a steam turbine retrofit

Craig C. Jennings, Exelon Power

Pages: 1 2 3 4

Bad vibes

The old HP turbine had had a history of unstable behavior before the retrofit and had experienced sporadic subsynchronous vibration. However, pre-outage detailed vibration data were not available for precise analysis because the unit was not equipped with the instrumentation package found on late-model turbines.

The rotor bearing system stability of the HP turbine with the original “partial center slot type” bearings (Figure 4) had been analyzed before the outage and was found to be satisfactory with the heavier new rotor. The option of making dimensional changes in the center slot to improve the stability margin was dismissed because of the relatively elevated drain temperature of the HP bearings (180F for an oil supply of 104F).


4. Recycled bearings. The HP turbine’s lower-half bearings with a partial center slot were analyzed as part of the retrofit turbine design and were found to be adequate for the new, heavier rotor. All bearings were rebabbitted during the project. Courtesy: MPS

Vibration data collected on November 19 (before the trip) still showed sporadic subsynchronous vibration, which now appeared during the acceleration ramp-up at approximately 2,840 rpm and disappeared at approximately 3,490 rpm (Figure 5). This sporadic behavior continued during several subsequent operational tests when loading the unit to baseload operation. The only common thread was the unpredictable timing of the vibration.


5. Unexpected spike. Subsynchronous vibration during a start-up with the new HP turbine is shown inside the red circle. Source: MPS

The team determined that the high synchronous vibration recorded on November 20 was caused by a severe rubbing condition between the shaft and the bearings experienced during a normal shutdown, causing the unit to trip on high vibration. The internal rubbing was quickly identified by a large and rapid increase in the synchronous vibration component with large changes in phase angle (Figure 6). At the moment of the trip, the subsynchronous component was negligible, but it suddenly increased during the coast-down that followed the trip (Figure 7). Overall (nonfiltered) vibration during coast-down after the instability was triggered approached 20 mils (Figure 8).


6. Rubbed the wrong way. A large synchronous vibration excitation in the HP turbine was induced by a rubbing condition. The polar plot shown is for bearing No. 2. Source: MPS