The Power of Water
This is a great example that shows how powerful water can be in destroying a bearing, after only 1 week, and also highlights that when you perform vibration analysis the normal Velocity data should never be forgotten about.
This case history comes from a great friend of mine Matthew Plant linkedin.com/in/mathew-plant-46000456 .
Matt collected the vibration data and performed the analysis and recommendations on his findings to his client.
The asset is an Automotive Dynamometer test system in an altitude test facility, the bearing supports the dynamometer rolls. The unladen (no vehicle) rolls shaft weighs 3 tonnes and speed is variable from 0 to 720 rpm (0-250kmh). There is a SKF 22228CCKW33 installed at both shaft ends, the bearing in question however is location end, all radial loads are within spec.
This forms part of a routine maintenance condition based monitoring program. The client reported activation of the facility water sprinkler systems and a service inspection was scheduled to ensure no asset was damaged due to the water sprinkler activation.
All the data is the survey before the incident and the survey after the incident. The data was collected one week after the incident due to de- contamination works.
On analysis of the vibration data the following points were noted;
Figure 1 is the overall Velocity trend. The overall Velocity increased from 0.137 mm/s RMS to 0.602 mm/s RMS. Even still low this was an increase of 440%
Figure 2 compares the before and after incident Velocity spectra’s. This clearing indicates a change in the bearing condition after the incident. The top green plot is after the incident on site.
Figure 3 is the Velocity spectrum and this show activity that is dominated by the bearing outer raceway defect frequency.
Figure 4 is the PeakVue Max Peak trend from before the incident at 2.019g’s and after the incident at 5.623 g’s
Figure 5 compares the PeakVue spectra’s from before (Blue plot) and after the incident (Green plot).
What can be seen is a 3.566 Order and harmonics. This 3.599 Order is the fundamental defect frequency for the SKF 22228CCKW33 installed. You can also note that 2XBSF is the highest frequency.
Figure 6 compares the before (Blue plot) and after (Green plot) of the raw Acceleration time waveform. This also indicated a high increase in the acceleration impactive data (note crest >5).
Figure 7 compares the before (Blue plot) and after (Green plot) Acceleration spectra’s. This also shows an increase in the friction and impactive levels.
Vibration Analysis Summary and Recommendations:
Due to the high increase in all vibration parameters and defect frequencies evident for the bearing outer raceway and rolling elements it was advised to replace the bearing.
What was the ‘Alarm bell’ for the analysts was the Velocity data.
Obviously water ingress was the instigator in the corrosion; however it was noted that the SKF SNL housings should withstand wash down. Further inspections pointed to the cap lifting eye being absent allowing water to enter the enclosure through the W33 lubrication groove.
On inspection the damage due to the bearing after running one week after the water incident was highly evident.
The new bearing was then installed using the hydraulic nut drive up method.
A reliable plant is a safe plant
…..an environmentally sound plant
….. a profitable plant
……a cost-effective plant