Interesting Rotor Dynamics Observations on Oil Whirl and Whip

SpectraQuest Inc.
8205 Hermitage Road Richmond, VA 23228 (804)261-3300 www.spectraquest.com April, 2006

Abstract: In this study, the effects of load on oil whirl and whip were studied by using
a rotor dynamics simulator with fluid film bearings. Rotor displacement during machine run up and coast down under different loading conditions were measured, analyzed and presented. Besides the oil whirl and whip introduced in traditional textbooks, harmonics of oil whip are observed. Some other vibration components associated with oil whip are also observed.

1. Introduction
Oil whirl is a common problem with journal bearings used on machines equipped with pressure lubrication systems operating at relatively high speeds. If the shaft is moved off center due to load, eccentricity, or imbalance, then the clearance on one side of the bearing will be greater than that on the other side, as shown in Fig. 1.

Figure 1 Shaft off-center in Journal Bearing As the lubricant rotates at less than 50% of shaft speed, it must squeeze through the narrow area where the shaft is closest to the bearing. The average speed of the lubricant increases inside the gap and slows down when it leaves the gap. Such a speeding up and slowing down process creates turbulence on both sides of the gap, and a vortex develops in the high-pressure lubricant zone.
1

The shaft that rides on the oil vortex performs much like a surfboard that rides the surface of a wave. The so-called oil whirl, whose frequency is somewhat less than half of the shaft rotational speed, causes instability. The oil whirl stays proportional to the shaft frequency and drops out when shaft RPM drops below the instability threshold. The oil whirl frequency approaches the first critical speed of the shaft as the shaft exceeds more than two times its first critical speed, creating a resonant condition called oil whip. The oil whip frequency remains constant at the first critical speed of the shaft and drops out when shaft frequency drops below two times its first critical speed. Both the phenomena can be severe and result in a penetration of the lubrication film. When this happens, the shaft impacts against the bearing and serious damage may happen. In this study, a series of tests were carried out on a rotor dynamic simulator with fluid film journal bearings to observe the oil whirl and whip phenomena.

2. Experimental Setup
The test is conducted on the SpectraQuest rotor dynamic simulator which is illustrated in Fig. 2. The rotor dynamic simulator system is comprised of two major subsystems: the simulator and the lubrication oil pump assembly.

(a) side view

(b) front view Figure 2. Rotor Dynamic Simulator

In the simulator, the rotor shaft is supported by fluid film journal bearings. The length of the rotor shaft between the two bearings is 28 inches and the shaft diameter is 0.5 inches. The clearance between the journal and bearing is 2/1000 inches. The lubrication oil used is mineral oil with ISO 13 viscosity. The pressure of the oil is 5 psi when it leaves the tank. The coupling between the motor and rotor shaft is helical flexible coupling. Proximity probes are mounted on the journal bearing housings to measure the relative displacement of the shaft with respect to the bearing housing. Two proximity probes are mounted on each bearing housing in the vertical and horizontal directions so they can provide shaft orbit graph. In order to reduce the interference between these two probes, their positions should have an offset as illustrated in Fig. 3.

Figure 3. Mounting of Proximity Probes Three types of disks were used in the test as shown in Fig. 4. The small disk possesses a smaller diameter and weighs 767 grams. The thin and large disks possess the same diameter. However, the large disk possess larger thickness than the thin one. The thin disk weighs 741 grams and the large disk weighs 1404 grams.

Figure 4. Disks Used in Test

3. Experimental Procedure
Tests were conducted under the following load conditions. 1. no disks installed on rotor shaft 2. one small disk installed 3. two small disks installed 4. one thin disk installed 5. one large disk installed 6. two large disks installed For each load condition, the disk(s) was installed at the center of the shaft and the simulator ran up to over two times the first critical speed, stayed at the maximum speed
3

for a while and then coasted down. For each test, rotor shaft displacements were measured and processed using VibraQuest, a data acquisition and vibration analysis system.

4. Experimental Results
The waterfall plots of the displacement data for different load conditions are generated using SpectraQuest VibraQuest software package and presented in this section. 4.1 Shaft without Disk The data are illustrated using waterfall plot in Fig. 5. In Fig. 5, it can be noticed that the oil whirl appears after a threshold speed is reached. The locking of the oil whip can be identified clearly. A careful inspection of Fig. 5 identifies some interesting vibration components. They are parallel with the 1X ridge and appear with the occurrence of oil whip. The reason for these components is not investigated in this study.

Figure 5 Waterfall Plot for Shaft without Disk (linear scale) 4.2 One Small Disk Installed Figure 6 illustrates the data for run-up and coast-down with one small disk installed on the rotor shaft. The oil whirl and whip can be distinguished easily in Fig. 6. Figure 7 presents the same data in dB scale. In Fig 7, one interesting observation is that some oil whip harmonics can be identified. Figure 8 presents the data in intensity plot in dB scale. Sidebands around oil whip and its harmonics can be observed in Fig. 8.

Figure 6. Waterfall Plot for Shaft with One Small Disk (linear scale)

Figure 7. Waterfall Plot for Shaft with One Small Disk (dB scale)

Figure 8. Intensity Plot for Shaft with One Small Disk (dB scale) 4.3 Two Small Disks Installed Figure 9 illustrates the data for run-up and coast-down with two small disks installed on the rotor shaft. In Fig. 9, the oil whirl is not as obvious as those shown in the previous figures.

Figure 9. Waterfall Plot for Shaft with Two Small Disks (linear scale) 4.4 One Thin Disk Installed Figure 10 illustrates the data for run-up and coast down with one thin disk installed on the rotor shaft. The oil whirl and whip can be distinguished easily in Fig 10. Figure 11 presents the same data in dB scale. Like the unidentified components in Fig. 5, in Fig 11, besides the oil whip harmonics, there are some unidentified vibration components with constant frequencies.
6

Figure 10. Waterfall Plot for Shaft with One thin Disk (linear scale)

Figure 11. Waterfall Plot for Shaft with One thin Disk (dB scale) 4.5 One Large Disk Installed Figure 12 illustrates the data for run-up and coast down with one large disk installed on the rotor shaft. Oil whip can be identified easily. The oil whirl appears during the coast down.

Figure 12. Waterfall Plot for Shaft with One Large Disk (linear scale) 4.6 Two Large Disks Installed Figure 13 illustrates the data for run-up and coast-down with two large disks installed on the rotor shaft. There is no oil whirl or whip observed.

Figure 13. Waterfall Plot for Shaft with Two Large Disks

5. Summary
In this rotor dynamics study, the effects of load on oil whirl and whip instability were studied by using a SpectraQuest rotor dynamic simulator. Rotor displacements during machine run up and coast down under different loading conditions were measured, analyzed and presented. Oil whirl and especially oil whip, appear in five out of the six tests conducted. Moreover,
8

harmonics of oil whip are observed. These harmonics can be distinguished easily in the dB scale waterfall plot. Some other vibration components associated with oil whip were also observed. The reason for the whip harmonics and other oil whip associated vibration components is needs to be investigated in the future study.