A Solution to Years of High Vibration

Problems in Three Reinjection Compressor

Trains Running at 33 MPa Discharge
Pressure

Presenter/ Marcio Felipe

 Authors
Jongsoo Kim
Senior Principal Engineer, Waukesha Bearings Corporation
Bachelor of Science Mechanical Engineering, 1985 Busan National University
Master of Science Mechanical Engineering, 1987 KAIST (Korea Advanced Institute of Science and Technology)
Doctorate of Philosophy Mechanical Engineering, 1991 KAIST (Korea Advanced Institute of Science and Technology)

Marcio Felipe dos Santos

Senior Maintenance Engineer, Major South American Oil Company
Bachelor of Science Mechanical Engineering, 1979 Universidade Federal of Rio de Janeiro

Barry J. Blair
Chief Engineer, Waukesha Bearings Corporation
Bachelor of Science Mechanical Engineering, 1990 University of Virginia
Master of Science Mechanical Engineering, 1990 University of Virginia
Master of Product Design and Development, 2015 Northwestern University
 Abstract
Over a 13 year span, a major South American oil companys
maintenance department fought high vibrations in three gas
reinjection compressor trains. To reduce the chances of
machine trips, technicians field balanced the compressors
every year and replaced worn point contact pivot tilt pad
journal (TPJ) bearings and O-ring squeeze film dampers
(SFDs) with new ones yearly. The downtime from
implementing these preventative measures and from actual
trips in the trains resulted in a loss of capacity of 1% a year
and additional flaring of the gas.
After a thorough analysis of the compressors and inspection of damaged
components, it was determined that the reoccurring problems would be solved by
installing optimized Flexure Pivot tilt pad journal bearings with Integral Squeeze
Film Damper (ISFD) technology into the compressors.
In 2013 the reinjection compressors were placed back into service with Flexure Pivot
TPJ bearings with ISFD technology. Since then the reinjection compressors have
exhibited lower vibration levels that do not grow over time, have had ZERO trips,
and have not required field balancing for continuous operation. Overall efficiency
has increased by approximately 1%.
 Contents
Background
Problem Statement
Modeling and Analysis
Upgraded Bearing and Damper
Design
Optimization
Summary
Lessons Learned
 Unit Background Information
3 gas reinjection compressor trains operating since 2000
Each compressor train has two casings, driven by GE LM2500
First casing (LP) experienced vibration issues
Second casing (HP) did not have vibration issues
Discharge pressure: 33 MPa
Rated speed: 11,456 rpm Vibration Issue

OEM bearing information

114.3 mm bore x 50.8 mm long TPJ
5-pad, load on pad
60% offset
0.6 MPa (86.5 psi) unit load
O-ring damper
 Problem Statement
Vibration trend of LP compressor with OEM
Rotor vibration levels (1X) increased over time bearings over 5 month span
Downtime due to high vibrations resulted in 100 -
about 1% loss in production time and additional

Vibration pk-pk
flaring gas 80 -

(microns)
Field balanced every year (sometime twice year) 60 - __ Compressor DE X
Installation of new OEM bearings every year 40 -
__ Compressor DE Y
__ Compressor NDE X
20 -
0- __ Compressor NDE Y
0- | |
24-Jul-2012 18-Nov-2012
76.2 - Vibration with worn pivots
61.0 - Vibration with new OEM bearings of the
Vibration pk-pk

original design
45.7 -
(microns)

30.5 - __ Compressor DE X
__ Compressor DE Y
15.2 -
__ Compressor NDE X
0 -| | | | |
__ Compressor NDE Y
01-Nov-2011 19-Nov-2011 6-Dec-2011 24-Dec-2011 10-Jan-2012
 Root Cause
Severe pivot wear in TPJ bearings Pivot wear

Bearing clearance increased by 63+ Wear mark on pad Wear mark on housing
microns
O-ring damper performance changed
Damper film eccentricity ratio change
Damper Axial
over time (bottoming out) Pressure Profile

Bearing O-ring
Top view Bottom view
Shell

End Oil Inlet

Seals O-ring groove
Tilt Pad
Casing Fretting damage on bearing OD
 Rotordynamic Model

H14586 Petrobras D-R 7 stage compressor

Supported by FPJs

Division wall
400

200
20 25 30 35 40 45 50 55
Shaft Radius, mm

1 5 10 15 60 65 68

69 70
-200

-400
0 400 800 1200 1600 2000 2400
Axial Location, mm
Baseline Mode Shapes with OEM Bearings
Small log dec for baseline MODEL (no SFD and no aero cross-
coupling stiffness from seals and impellers)

400 Cmin (minimum f=4400.1 cpm

clearance) d=.2065 logd
N=11456 rpm
Shaft Radius, mm

Cmin (minimum
200 clearance)

-200

-400
0 400 800 1200 1600 2000 2400
Axial Location, mm
 Stability with OEM Bearings
Level I stability predicts that the rotor is unstable without SFD

Cmin (minimum
400 clearance) f=4427.8 cpm
d=-.4406 logd
Shaft Radius, mm

N=11456 rpm
200

-200

-400
0 400 800 1200 1600 2000 2400
Axial Location, mm
 Division Wall Seal Contribution
0.6 Existing Brg (Cmax)
Existing Brg (Cmin)
0.4
Estimated aero cross
0.2 coupled stiffness

Log Dec
0.0
QA With DW (sw
-0.2 da ratio=0.5)

-0.4

Insignificant -0.6 Without Division

Wall (DW) seal
improvement to stability -0.8
0.0E+00 1.0E+07 2.0E+07 3.0E+07
with division wall hole Applied Cross Coupled Stiffness, Q N/m
pattern seal
For stability, SFD required
 O-ring Squeeze Film Damper (SFD)
Damper radial clearance (c) = 0.110
mm
Damper radius (R) = 95.25 mm
Effective damper length (L) =37.85 1.25
mm OEM bearing with SFD
1.00
Stability is very sensitive to damper OEM bearing without SFD

Log dec of 1st mode

eccentricity ratio () 0.75

Added O-ring stiffness 0.50

With Applied Cross
0.25 Unstable
Coupled K=2.13E+07 N/m
Damper stiffness and damping coefficients 0.00
(without O-ring stiffness) -0.25
K=1.94E+07 N/m, C=1.88E+05 Ns/m at =0.25 -0.50
K=6.06E+07 N/m, C=2.62E+05 Ns/m at =0.50 -0.75
K=1.70E+09 N/m, C=2.06E+06 Ns/m at =0.90 0.0 0.2 0.4 0.6 0.8 1.0
O-ring SFD eccentricity ratio
 Pivot Wear Effects on Synchronous Vibration
Pivot wear increased operating bearing
clearance and reduced preload, resulting in
increased synchronous vibrations
A bearing with a SFD can make the rotor less
sensitive to pivot wear than a bearing without
a SFD
Pivot wear effect (without SFD) Pivot wear effect (with SFD)
0.08 0.08
No pivot wear
0.07 0.012 mm pivot wear 0.07
SFD:K=4.375E+07 N/m, C=2.625E+05 N-s/m
0.025 mm pivot wear

Response, mm p-p
Response, mm p-p

0.06 0.038 mm pivot wear 0.06

RatedCOS

0.051 mm pivot wear

MaxCOS
MinCOS

0.05 0.05

0.04 0.04

0.03 0.03
No pivot wear
0.013 mm pivot wear

RatedCOS
MaxCOS
MinCOS
0.02 0.02
0.025 mm pivot wear
0.038 mm pivot wear
0.01 0.01
0.051 mm pivot wear

0 0
0 2000 4000 6000 8000 10000 12000 14000 0 2000 4000 6000 8000 10000 12000 14000

Rotor Speed, rpm Rotor Speed, rpm

 Bearing Upgrades (Journal Bearing)
1. Flexure Pivot Tilt Pad Journal
Bearing
No pivot wear
Integral pivot
Maintains bearing clearance Conventional
Conventional TiltJournal
Tilt (CT) Pad Pad Flexure Pi
Bearing (TP, Rocker
Journal Bearing (Rocker Back) Jour
High pivot stiffness Back)
No pivot stiffness effect on bearing
dynamic coefficients
Tight control of clearance and preload
Electrical Discharge Machining (EDM)
No pad flutter
Conventional Tilt (CT) Pad Journal Flexure Pivot
Flexure Pivot Tilt
(FP) Tilt Pad
Pad
Bearing (TP, Rocker Back) Journal Bearing
Journal Bearing
 Bearing Upgrades (Damper)
2. Integral Squeeze Film Damper
(ISFD) Damper Film S-spring
Accurate stiffness control by
eliminating O ring support
No change in stiffness and
damping over time
Designed to counter static load
Optimized damping 0.11 mm O-ring
0.35 mm
Original Damper
Less cavitation Clearance
ISFD Clearance

Conventional SFD ISFD

 Optimization of ISFD
Optimized stiffness
and damping are
4.375 E+07 N/m
(250,000 lb/in) and
2.625E+05 N-s/m
(1500 lb-s/in)
Additionally, the ISFD
is designed to center
the Flexure Pivot TPJ
under gravity load by
countering the static
deflection
 Redesigned Bearing (Drop-in)

Optimized

Original Design Optimized Design

Conventional TPJ with SFD Flexure Pivot TPJ with ISFD Technology
5-pad, Load On Pad 4-pad, Load Between Pad
Shaft diameter Shaft diameter
114.300 +0/-0.013 mm 114.300 +0/-0.013 mm
Bearing bore Bearing bore
114.414 +0.025/-0 mm 114.427 +0.025/-0 mm
Clearance range Clearance
0.124/0.156 mm range 0.124/0.156 mm
Preload range 0.293/0.501 Preload range 0.230/0.273
L/D 0.444 L/D 0.500
Pad arc 60 Pad arc 72
Pivot Offset 60% Pivot offset 55%
 1st Mode Shapes: Original and Upgraded Bearings
Applied destabilizing cross-coupling 400 OEM bearing w/out O-ring SFD f=4427.8 cpm
stiffness of 2.13E+07 N/m to mid span d=-.4406 logd

Shaft Radius, mm
N=11456 rpm
200
Without SFD, unstable
0
Both O-ring SFD and ISFD can make the
rotor stable -200

No change in stiffness and damping of -400

ISFD over time 0 400 800 1200 1600 2000 240
Axial Location, mm

400 Upgraded bearing (Flexure f=3594.5 cpm

OEM bearing w/ O-ring SFD f=3624.7 cpm
d=1.0362 logd 400
Pivot TPJ +ISFD) N=11456 rpm
d=.3504 logd

Shaft Radius, mm
N=11456 rpm
Shaft Radius, mm

200 200

0 0

-200 -200

-400 -400
0 400 800 1200 1600 2000 2400 0 400 800 1200 1600 2000 24
Axial Location, mm Axial Location, mm
No Subsynchronous Vibration (SSV) with Upgrade
Small SSV with O-ring SFD
With OEM bearing
No SSV with ISFD
(Acceptance test)

Upgrade bearing
 Vibration Improvement (Comp A & B)
Compressor A
TC-125227,60
01B VIBRAES e DESLOCAMENTO AXIAL DO COMPRESSOR
25,10 D12 75,50 21,90
19,20 15,50 72,50 22,40
100100- 100 100 100
50,60 85,40 25,60 Compressor A with Upgrade
14,10 \\bsolsiopu\2VI12
22,60
58,60 90,70 12,90 25,30

vibration dropped to

Vibration pk-pk
8080- (4 month span) \\bsolsiopu\2VI12
70 24,00

(microns)
6060-
less than half and 50

4040-
\\bsolsiopu\2VI12
17,10

maintained that
\\bsolsiopu\2VI12
30 42,30

20 20-
level over time
\\bsolsiopu\2ZI12
\\bsolsiopu\2ZI12
0 04/7/2012
- 0 11:29:00,112
24/7/2012|
0 0
21:44:24 16/11/2012 09:43:49 365,00 dias 14/3/2013|16:01:00 8/5/2013 11:56:35 |
4/7/2013 11:29:00,112

24-Jul-2012
DR02B VIB.HOR.MANC LA D12
14-Mar-2013 8-Aug-2013
Compressor B
DR02B VIB.VERT.MANC.LA D12
Compressor A with OEM
DR02B VIB.HOR.MANC.LNA D12
DR02B VIB.VERT.MANC.LNA D12
__ Compressor DE X
bearing
vibration also down 100 - __ Compressor DE Y
to below 50 m 80 - __ Compressor NDE X
Vibration pk-pk

from 90 m __ Compressor NDE Y

(microns)

60 -
(original) and kept 40 -
the same level over
time 20 -

0- | | |

2-Jun-2013 12-Dec-2013 15-Oct-2014

Compressor B with OEM Compressor B with Upgrade (10-month span)
bearing
 Vibration Improvement (Comp C)
100 -
Compressor C was Compressor C (vibration trend over 10
80 -
also upgraded with a years)

Vibration pk-pk
__ Compressor DE X

(microns)
60 -
Flexure Pivot TPJ with __ Compressor DE Y
40 - __ Compressor NDE X
ISFD technology __ Compressor NDE Y
20 -
Again, the vibration 0- |
level decreased to
|
8-Mar- 30-Dec-2014
2005
below 30 m with the 100 -
Compressor C with OEM Compressor C with
upgrade and 80 - bearing Upgrade
Vibration pk-pk

maintained that same 60 -

(microns)

level due to no 40 -

change in bearing 20 -
0 -|
clearance and SFD
| |
3-Jun-2013 17-Apr-2014 14-Oct-2014

performance over
time
 Summary
Three reinjection compressor trains
suffered from excessive vibration
over many years
Original configuration: point contact
pivot tilt pad journal bearings with
an O-ring SFD

The root cause was excessive pivotPivot Wear mark on housing Fretting Damage
wear and degradation of the O-ring
SFD
Bearing bore increased
Stiffness and damping changed
over time
 Summary
These compressors were retrofitted with
optimized Flexure Pivot tilt pad journal
bearings with ISFD technology
Operating exceptionally well
Since 2013
Low vibration levels
50% drop pk-pk compared to OEM 2.5
ISFD optimization
(with areo cross coupling stiffness of 1.5E+07 N/m)

bearings 2.0
26.25 MN/m
30.625 MN/m

Do not grow over time

35 MN/m

Log. Dec. of 1st mode

1.5 43.75 MN/m

No field balancing required so far (2 1.0

years)
0.5 C=2.63 E05 N-s/m

0.0

No trips (continuous production) -0.5

0.0E+00 1.0E+05 2.0E+05 3.0E+05 4.0E+05 5.0E+05 6.0E+05

No expensive bearing replacements

Damping coefficient of ISFD, N-s/m

Overall efficiency increased by 1%

 Lessons Learned
Increase in synchronous vibrations over time may be an indication of
bearing clearance increasing from pivot wear and/or change in O-
ring damper performance
Pivot wear may accelerate over time by increasing imbalance due to
deposits on impellers (heavy due machines)
Without eliminating pivot wear, just replacing the worn bearing with
new build of the same design is NOT a long-term solution
Proper bearing and damper selection and optimization can reduce or
eliminate the likelihood of increasing vibrations and pivot wear
Flexure Pivot technology is a proven design to eliminate pivot wear
ISFD technology maintains performance over time (no degradation)
 Feedback and Questions
Case Study: A Solution to Years of High Vibration
Problems in Three Rejection Compressor
Trains Running at 33 MPa Discharge Pressure