MANUAL

1. Fluid-Film Bearing Damage
Detection Based on Vibration Data
John J. Yu, PhD, ASME Fellow
Technical Leader
Machinery Diagnostic Services
Bently Nevada
Nicolas Péton
Global Director
Machinery Diagnostic Services
Bently Nevada

2. T U R B O M A C H I N E R Y & P U M P S Y M P O S I A
Presenter Bios
Dr. John Yu joined Bently Rotor Dynamics Research Corporation in 1998, followed by
General Electric - Bently Nevada in 2002. He has performed not only rotor dynamic
research but also machinery vibration diagnostics for customers worldwide, and is now
Technical Leader of Machinery Diagnostic Services at Bently Nevada. He has over 50
technical papers in peer-reviewed journals and conference proceedings. He holds a PhD
in Mechanical Engineering from University of Alberta, and is an ASME Fellow. He
currently serves as an Advisory Committee member to the Asia Turbomachinery &
Pump Symposium.
Nicolas Peton joined GE in 2006 in the Machinery Diagnostic Services group. Previously
he worked for two different manufacturers (Alstom Steam turbine and Cryostar
expander/compressor) where he was in charge of on-site of the startup activities
worldwide. He also worked as an operation and maintenance engineer in the chemical
industry (PPG industry, USA) and as Free Lance for startup activities worldwide. He has
been also a mechanical/acoustical research engineer in research institutes (Technion,
Haifa and TU Berlin). He’s currently Global Director for the Machinery Diagnostic
Services. He’s also a member of the Pump committee. He has a Diplome d’ingénieur
from the Université de Technologie de Compiègne, France.

3. T U R B O M A C H I N E R Y & P U M P S Y M P O S I A
Abstract
This case occurred during startup after outage on a steam turbine generator.
Vibration reached over full scale of 20 mil pp at generator drive end bearing
and therefore tripped the unit. The major vibration component that tripped
the unit was 0.5X sub-synchronous at a level of over 20 mil pp. 1X vibration
excursions existed at constant speed before the trip event. Abnormal shaft
centerline positions were observed. Shaft bow reached 10 mil pp at low
speeds during coast-down. Bearing metal temperature reading was invalid.
After an in-depth data review, diagnostic conclusions and recommendations
were made, followed by corrective actions. Inspection and findings confirmed
bearing damage and rubs. If the vibration issue had simply be treated as rub,
followed by re-start without opening the casing, further catastrophic damages
would have occurred.

4. T U R B O M A C H I N E R Y & P U M P S Y M P O S I A
Outline
1. Introduction
2. Problem Statement
3. Data Review
4. Conclusions and Recommendations
5. Inspection and Findings
6. Resolution and Final Vibration Results
7. Lessons Learned

5. T U R B O M A C H I N E R Y & P U M P S Y M P O S I A
1. Introduction
• Cross-compound steam turbine generator unit with HP turbine speed of 3600
rpm, LP turbine speed of 1800 rpm, and rated output of 750 MW.
• Vibration was monitored by proximity probes at each bearing, plus seismic
sensors. Data collection was performed to support its startup after outage.
HP Turbine
(3600 rpm)
LP Turbine
(1800 rpm)

6. T U R B O M A C H I N E R Y & P U M P S Y M P O S I A
2. Problem Statement
• During startup, when the unit was operating at a constant warmup
speed of 1800 rpm, the vibration reached 20 mil pp at the HP
generator drive end bearing (Brg#5) and therefore tripped the unit.
• Its root-cause needed to be determined as soon as possible to
avoid any delay in unit startup.

7. T U R B O M A C H I N E R Y & P U M P S Y M P O S I A
3.1 Data Review – Direct, 1X, ½X, and 2X trend plots
Vibration excursion
mainly in ½X tripped
the unit.
Brg#5X
Speed Direct 20.02 mil pp 1793 rpm
1X 7.180 mil pp ∠155°
½X 20.41 mil pp ∠299°
2X 4.323 mil pp ∠213°

8. T U R B O M A C H I N E R Y & P U M P S Y M P O S I A
3.2 Data Review – Waveform-compensated orbits
Bearing #5 orbits at 1800 rpm from low to high vibration
Out of full scale
(20 mil pp)

9. T U R B O M A C H I N E R Y & P U M P S Y M P O S I A
3.2 Data Review – Orbit/timebase plot
Out of full scale
(20 mil pp)
Keyphasor® dots
locked (exact ½ X)
Brg#5

10. T U R B O M A C H I N E R Y & P U M P S Y M P O S I A
3.2 Data Review – Orbit plot
Changes in 1X vibration vector. Is it a rub?
Brg#5X Brg#5X

11. T U R B O M A C H I N E R Y & P U M P S Y M P O S I A
3.3 Data Review – Waterfall plot
1X
1
2
X
3
2
X
Abnormal ½ X and its multiples during shutdown
Brg#5

12. T U R B O M A C H I N E R Y & P U M P S Y M P O S I A
½X maintained until 1631 rpm during coast-down
3.4 Data Review – Direct, 1X, and ½X bode plots
991 rpm as resonance or critical speed during coast-down
Brg#5X Brg#5Y
Brg#6X Brg#6Y
½X
1X
Direct

13. T U R B O M A C H I N E R Y & P U M P S Y M P O S I A
3.5 Data Review – Shaft centerline plots
Abnormal shaft
centerline plot at Brg#5
Brg#5 Brg#6
Brg#4
Brg#3

14. T U R B O M A C H I N E R Y & P U M P S Y M P O S I A
4.1 Conclusions and Recommendations
Root-cause of much higher vibration:
• 1X and ½X components ?
➢1X, followed by dominant ½X (20 mil pp),thus tripping the
unit. ½X maintained above 1631 rpm during shutdown
• Shaft bow ?
➢High runout at low speed during shutdown (10 mil pp)
• Change in bearing condition ?
➢Normal shaft centerline plots except at Brg#5
– Yes
– Yes
– Likely at Brg#5

15. T U R B O M A C H I N E R Y & P U M P S Y M P O S I A
4.2 Conclusions and Recommendations
Root-cause of much higher vibration (Cont.):
• Rub?
➢1X vectors changed at constant speed
➢Much higher runout during shutdown, indicative of shaft
bow
➢Distorted orbits at Brg#5
➢Likely due to changes in shaft centerline position
• What caused ½X vibration ?
➢½ X occurred at 1631 to 1800 rpm during shutdown. The
first critical speed is 991 rpm. Thus, the ½ X frequency was
at 815-900 cpm < 991 cpm. Normal-tight or normal loose?
– Yes
– Instability due to excessive clearance

16. T U R B O M A C H I N E R Y & P U M P S Y M P O S I A
4.3 Conclusions and Recommendations
Root-cause of much higher vibration (Cont.):
• Bearing damage?
➢At constant speed of 1800 rpm and during coast-down, shaft
centerline position was far beyond the bearing clearance
boundary at the bottom at Brg#5.
➢Unfortunately, bearing metal temperature reading was
invalid due to the sensor issue.
– Most likely

17. T U R B O M A C H I N E R Y & P U M P S Y M P O S I A
4.3 Conclusions and Recommendations
Conclusions:
• Brg#5 appeared to have been damaged based on
shaft centerline position.
• Rub seemed to occur, causing varying 1X vibration
and distorted orbits, plus significant shaft bow
during shutdown.
• ½X vibration occurred due to oversized bearing
(normal-loose).

18. T U R B O M A C H I N E R Y & P U M P S Y M P O S I A
4.4 Conclusions and Recommendations
Recommendations:
• Open Brg#5 for inspection
• Check rub locations

19. T U R B O M A C H I N E R Y & P U M P S Y M P O S I A
5.1 Inspection and Findings
• The findings indeed
indicated Brg #5 bearing
wiped, mostly at the left
bottom with additional 42
mils clearance due to wear
beyond as-left clearance of
24 mils in vertical direction,
matching the diagnosis.
Obviously the wear was due
to journal rubbing against
the babbitt surface.
Looking from turbine to generator
Babbitt wear at
the left bottom
Fracture due to heat
Babbitt material
transferred to

20. T U R B O M A C H I N E R Y & P U M P S Y M P O S I A
5.2 Inspection and Findings
• It was found that a fine strainer was mistakenly left in place,
causing oil reduction and starvation, and finally wiping the
bearing. Bearing was shipped offsite to be re-spun.
• Rubs occurred on inner and outer oil deflectors at Brg #5 as
well as that at adjacent Brg #4 generator side. All three were
shipped offsite for teeth replacement.
• Brg#5 hydrogen seal casing oil deflector rubbed and
replaced with new one.
• Adjacent Brg #4 turbine side oil deflector clearance and
alignment acceptable.

21. T U R B O M A C H I N E R Y & P U M P S Y M P O S I A
5.3 Inspection and Findings
• No damage was observed on Brg#5 journal surface, and its
dimensions were verified to be acceptable.

22. T U R B O M A C H I N E R Y & P U M P S Y M P O S I A
6.1 Resolution and Final Vibration Results
• The bearing was repaired and re-installed correctly.
• Damaged oil deflectors were replaced with new ones.

23. T U R B O M A C H I N E R Y & P U M P S Y M P O S I A
6.2 Resolution and Final Vibration Results
Normal full-spectrum
waterfall plot at Brg#5
LP vibration transmitted
through foundation, not ½X
Brg#5

24. T U R B O M A C H I N E R Y & P U M P S Y M P O S I A
6.3 Resolution and Final Vibration Results
Normal shaft centerline
plot at Brg#5
Brg#5 Brg#6
Brg#4
Brg#3

25. T U R B O M A C H I N E R Y & P U M P S Y M P O S I A
7. Lessons Learned
• Correct assembly after outage should be ensured to avoid costly repairs. All
important sensors should be ensured working properly.
• The ½X vibration up to 20 mil pp was due to the increase in bearing clearance, not
simultaneously due to a rub against the bearing wall.
• However, rubs did occur, which first damaged bearing babbitt surface and resulted in
a change in shaft centerline position. This change caused the shaft to rub against oil
deflectors. All these rubs were indicated by 1X vibration excursions.
• Shaft centerline is one of the most important plots in machinery diagnostics. It can
be used to diagnose bearing damage correctly when bearing metal temperature
reading is invalid.
• Had the vibration issue been simply treated as rub, followed by re-start without in-
depth data analyses and actions, further catastrophic damages would have
occurred.