18
Oct
Vibration data were collected on the motor and fan bearing housings (Figure 1) during a routine monthly data collection. Failure of the machine would result in serious production losses - 407o of the refinery capacity per day. Overall vibration amplitudes were higher than normal at several points on the motor in the horizontal direction and the axial direction of the fan non-drive end bearing, typically 2.2 and2.9 mrr,ls (0.09 and 0.11 IPS) rms respectively (Figures 2 and3).


Figure 1. Fan. [manufacturer: Aeromeccania Stranich ; 980 RPM; 160 kw;capacity: 185,292 kg/h flue gas; motor bearings: 6319C3; fan bearings: DE 22224 EKC3, NDE 22219 EK-C3 (non-locating)1.
Figure 2. Vibration Trend of the Motor.
Figure 3. Axiat Vibration Tfend of Fan Non-drive End.


The data did not show the exact location of the problem or its severity. A decision was made to increase vibration monitoring to once per week.

A review of data collected over the next four weeks indicated a fufiher increase in overall axial vibration levels of the fan non-drive end bearing - about 4.25 mrn/s (0.17IPS) rms.In addition, there was an increase in broad-band activity in the specrrum (Figure 4).
Figure 4. Fan Non-Drive End, Axial Direction.
Figure 5. Strip Chart from the Non-drive End Bearing.


Figure 6. Loose Lock Nut and Washer.
Additional vibration data were taken from the fan nondrive end bearing in the axial direction using strip chart processing (240 sec). Figure 5 shows the significant fluctuation in overall vibration levels - ranging from 2 to 7.8 mm/s (0.08 to 0.31 IPS) rms.

The vibration spectrum recorded from this position indicated early signs of mechanical looseness. The recommendation was to stop the equipment as
soon as possible and examine the condition of the non-drive end bearins.

Examination of the fan non-drive end bearing highlighted several problems. The lock washer had loosened from the lock nut and caused the lock nut to loosen from the taper sleeve (Figure 6).

The inner race of the bearing had slipped over the tapered sleeve; the tapered sleeve then slipped over the shaft (Figures '7,8,9, and 10).

One millimeter of the shaft outer diameter was removed, and the taper sleeve was replaced by one with a smaller inner diameter and the same outer diameter.

Figure 7. Taper Sleeve Slipped Under Inner Race and Over the Fan Shaft.
Figure 8. Outer Race of Taper Sleeve.
A new bearing of the same size was installed. The process unit was brought on line at 100Vo capacity after 48 hours. According to the SKF book Bearing Failures and Their Causes smearing may occur on the external surfaces of heavily loaded bearings.

In this case, the smearing may have resulted from movement of the bearing ring relative to the taper sleeve as well as the taper sleeve relative to the shaft. Smearing can be avoided by selecting heavier interference fits.

Figure 9. Inner Race of Bearing.
Figure 10. Outer Surface of Shaft.
Article by

Ashraf Abdel-Rahim Ali

MIDOM-MIDOR Refinery

Alexandria, Egypt

Vibration Vol 23 No 2 June 2007 , Vibration Institute, USA

© 2007 Vibration Institute, USA
BACK