05 Bearings

Rolling Element Bearings
• Objectives
– Identify fundamental frequencies
– Evaluate failure progression
– Determine severity
– Define some characteristics
– Examine case histories
CSI CONFIDENTIAL
• Characteristics
– Harmonics of non-synchronous peaks
– Broad-band energy humps
– Time domain - impacting
– Early stages - spectrum
• low amplitude
• high frequency
CSI CONFIDENTIAL
• With experience, an analyst can find :
– Raceways (rings)
– Rolling elements (balls / rollers)
– Cage (fundamental train)
– Looseness in housing
– Excessive internal clearance
– Spinning on shaft
– Misaligned bearing
– Lack of lubrication
CSI CONFIDENTIAL
• Causes of failures
– 43% improper lubrication (over and under)
– 27% mounting
– 21% other errors, e.g., application, manufacturing
defects, excessive vibration (before and after
installation
– 9% normal life wear out
• Note : sources report that about 10% of all bearings already
have defects on them before installation.
CSI CONFIDENTIAL
What has more effect on the life of the bearing,
speed or load?
CSI CONFIDENTIAL
CSI CONFIDENTIAL
• Identifying Frequencies w/ spectrum

– Non-synchronous peaks w/ harmonics
– BPFI is often between 4x and 16x TS
– BPFO often between 2x and 10x TS
– Look for peaks with TS sidebands
– View spectrum in acceleration
CSI CONFIDENTIAL
• How long will it last?
• History and current condition of machine
– Size and # of defects
– Which component? (roller/cage)
– Geometry
– Rate of progression
• Why is it failing?
– Internal clearance
– Loss of lubricant
– Vibration - internal / external
CSI CONFIDENTIAL
• How long has it been in service?
– When was it installed ?
– When did the defects appear?
• What is the speed?
– 3600 CPM can fail quickly
– 300 CPM may go on for several months
• Experience
– Similar machines
– Similar operating conditions
– Similar failures
• caution : no two identical cases exist
CSI CONFIDENTIAL
• Normal order of failure :
– Races, ball / roller, cage
– Inner races suffer at much lower amplitudes
– Early faults generate predicted frequencies, and
harmonics, frequently for only one race
– Multiple harmonics may mean multiple defect sites or
larger defects
– Appearance of defect frequencies from other brg.
components mean progressive damage
– Rotating race frequencies are modulated by shaft speed,
appear as sidebands
CSI CONFIDENTIAL
– # of sidebands increases as damage progresses

– Disappearance of peaks and / or broadband
energy indicates geometry changes
– Inadequate lubrication can result in accelerated
failure
CSI CONFIDENTIAL
6-11
Analysis Parameter Sets;
A general guideline for a range of 65xTS
CSI CONFIDENTIAL
Alarm Limit Sets
Note : amplitudes usually good for Rpm's above 1K. Modify based on
experience and machine type.
Slow speed machines have failed below .01 IPS.
CSI CONFIDENTIAL
Roller Defect
CSI CONFIDENTIAL
Inner Ring Defect Rolling Element Bearings
Outer Ring Defect Rolling Element Bearings
Evidence of high sub-synchronous energy (.89 orders). Increasing….
BRG1 - TENTER ZONE EXH. FAN MOTOR
MOTOR BPFI-M2H MOTOR INBOARD HORIZONTAL
Max Amp
.13
Plot
Scale
PK Velocity in In/Sec
0.08
08-NOV-90
04-OCT-90
0
04-SEP-90
26-JUL-90
31-MAY-90
09:53:38
RPM= 1750.
31-MAY-90
Freq: 25.96
0 400 800 1200 1600 2000
Ordr: .890
Frequency in Hz Sp 1: .138
CSI CONFIDENTIAL
0.12
Route Spectrum
08-NOV-90 09:59
PK = .1756
0.10
LOAD = 100.0
RPM = 1774.
RPS = 29.57
0.08
0.06
0.04
0.02
0
Freq: 26.01
0 400 800 1200 1600 2000
Ordr: .879
Frequency in Hz Spec: .09154
Label: INNER RACE DEFECT-SEE WVFRM MOD.
Priority: I
Dominate frequency is .89 orders. 2x Belt frequency

Significant amount of non-synchronous energy. What could it be?
CSI CONFIDENTIAL
CSI CONFIDENTIAL
6-15
8
Waveform Display
08-NOV-90 09:59
6
RMS = 1.54
LOAD = 100.0
RPM = 1774.
Acceleration in G-s
4
RPS = 29.57
2 PK(+) = 5.89
PK(-) = 5.59
CRESTF= 3.81
0
-2
-4
-6
-8
0 30 60 90 120 150 180 210 240
Time in mSecs
Label: INNER RACE DEFECT-SEE WVFRM MOD.
Priority: I
Impacting is evident. Pattern typical of balls passing over a defect.

Improper belt tension caused the inboard bearing to fail.
CSI CONFIDENTIAL
MOTOR BPFI-M2V MOTOR INBOARD VERTICAL
0.08
Route Spectrum
G G G 08-NOV-90 09:59
0.07
PK = .1454
LOAD = 100.0
0.06 RPM = 1794.
RPS = 29.90
0.05 >FAF 6309

G=BPFI
0.04
0.03
0.02
0.01
0
Freq: 147.86
0 400 800 1200 1600 2000
Ordr: 4.945
Label: BPFI:LOTS OF MOD.-SDBNDS & WVFRM
Priority: 1
MOTOR BPFI-M2V MOTOR INBOARD VERTICAL
10
Waveform Display
08-NOV-90 09:59:38
8
RMS = 1.54
6 LOAD = 100.0
RPM = 1750.
RPS = 29.17
4
Acceleration in G-s
PK(+) = 8.92
2 PK(-) = 7.61
CRESTF= 5.78
0
-2
-4
-6
-8
-10
0 30 60 90 120 150 180 210 240
Time in mSecs
Label: BPFI:LOTS OF MOD.-SDBNDS & WVFRM
Priority: 1
Belt driven agitator. Dominant NON-synch. Peak at @ 5x.
BRG1 - PRIMARY C.S. REJECT AGITATOR
AGIT BRG2 -A1H AGITATOR BRG. #1 - HORIZONTAL
Max Amp
.17
Brg. defect? Blade pass? 29.49 Hz is motor TS.
Plot
Scale
0.20
19-FEB-91
17-DEC-90
Probably sheave run-out.
21-NOV-90
21-NOV-90
11:11:10
RPM= 351.0
31-OCT-90
Freq: 29.49
0 50 100 150 200 250 300 350 400
Ordr: 5.041
CSI CONFIDENTIAL
0.24
Route Spectrum
19-FEB-91 11:48
0.21 Broadband energy. only 2 peaks. exceed .1 I.p.s.
Overall exceeds .7 I.p.s. Why is overall so high? PK = .7351
LOAD = 100.0
0.18 Energy in broadband “humps”. RPM = 351.
RPS = 5.85
Probably outer race defect.
0.15
0.12
0.09
0.06
0.03
0
0 50 100 150 200 250 300 350 400
Frequency in Hz
Label: BROADBAND/SIGNIFICANT CHANGE:BRG
Priority: 1
CSI CONFIDENTIAL
8
Waveform Display
Waveform shows significant impacting. 19-FEB-91 11:48
6
@ 10 G swing. RMS = 1.40
LOAD = 100.0
RPM = 351.
4 RPS = 5.85
Acceleration in G-s
PK(+) = 5.53
PK(-) = 4.85
2 CRESTF= 3.95
-2
-4
-6
0 0.2 0.4 0.6 0.8 1.0 1.2
Time in Seconds
Label: BROADBAND/SIGNIFICANT CHANGE:BRG
Priority: 1
CSI CONFIDENTIAL
BRG1 - CHEM. PLANT SLUDGE PUMP BRG.
PUMP D-17 -P1H PUMP BEARING 1 HORIZONTAL
0.14
Route Spectrum
11-MAY-90 13:22:40
0.12
TS harmonics ! Some exceed amp. of 1xTS. What is the fault? PK = .1514
LOAD = 100.0
Severe looseness ? Note the low amplitude. RPM = 1181.
RPS = 19.68
0.10
0.08
0.06
0.04
0.02
0
Freq: 19.68
0 400 800 1200 1600
Ordr: 1.000
CSI CONFIDENTIAL
Max Amp 11-SEP-90

.23
Plot
Scale
0.24
16-AUG-90
0
13-JUL-90
08-JUN-90
11-MAY-90
0 400 800 1200 1600
Frequency in Hz
Note the pattern through Aug. In Sept. it looks like the machine
healed itself ! Take a closer look at the data.
CSI CONFIDENTIAL
0.04
Route Spectrum
11-SEP-90 10:58:30
Now check the broadband energy.
PK = .1193
Low full scale range, high overall. LOAD = 100.0
0.03 Excessive clearances caused the looseness. RPM = 1168.
RPS = 19.47
What other data …….?
0.02
0.01
0
0 400 800 1200 1600
Frequency in Hz
Label: FINAL STAGE BRG.-VERY BROADBAND!
Priority: 1
CSI CONFIDENTIAL
3
Waveform Display
High energy, severe impacting, irregular pattern. 11-SEP-90 10:58:30
2 RMS = .4992
LOAD = 100.0
RPM = 1168.
RPS = 19.47
1
Acceleration in G-s
PK(+) = 2.20
PK(-) = 2.52
CRESTF= 5.05
0
-1
-2
-3
0 60 120 180 240 300

Time in mSecs
Label: FINAL STAGE BRG.-VERY BROADBAND!
Priority: 1
CSI CONFIDENTIAL
BRG1 - FILM TRIM TAKEAWAY BLOWER
BLOWER BRG-F2H FAN OUTBOARD HORIZONTAL
Max Amp
.16
Plot
Scale
23-JUL-90
0.16
20-JUL-90
0
19-JUL-90
23-MAY-90
27-APR-90
0 200 400 600 800 1000
Frequency in Hz
Significant change in high frequency range from April to late July.
Pattern resembles looseness. Are these peaks harmonics of TS.
CSI CONFIDENTIAL
0.20
Route Spectrum
D D D D 23-JUL-90 07:03:50
PK = .3165
0.16 LOAD = 100.0
RPM = 914.
RPS = 15.24
>FAF 214K
0.12 D=BPFO : 62.53
0.08
0.04
0
Freq: 54.51
0 200 400 600 800 1000
Ordr: 3.578
This blower has an outer race defect. Data shows the fault freq. overlays.
Label: BEFORE REPLACING BAD BEARING
Don’t line up perfectly. What could cause this? Use pattern recognition.
Waveform data confirms our suspicion.
CSI CONFIDENTIAL
10
Waveform Display
23-JUL-90 07:03:50
8
RMS = 1.98
6 LOAD = 100.0
RPM = 914.
RPS = 15.24
Acceleration in G-s
4
PK(+) = 7.61
2 PK(-) = 7.40
CRESTF= 3.84
0
-2
-4
-6
-8
-10
0 50 100 150 200 250 300 350 400
Time in mSecs
Label: BEFORE REPLACING BAD BEARING
Extremely high levels of energy, impacting.

Classic bearing failure. Note the pattern.
CSI CONFIDENTIAL
Multi-spectral plot displays harmonics of TS. (Looseness) High freq.
activity appears on this roller bearing in June of 89.
BPFI - PRESS ROLL BRG.- PAPER MACH.
PRESS ROLL-FR8 FELT ROLL #8 (ON ARM FRM 3PB)
Max Amp
.24
Plot
Scale 19-JUN-89
0.30
11-APR-89
0
09-JAN-89
19-SEP-88
19-JUN-89
11:33:32
FPM= 1404.
02-OCT-87
Freq: 6.167
0 100 200 300 400 500 600
Ordr: 1.000
CSI CONFIDENTIAL
6-26
0.35
Analyze Spectrum
19-JUN-89 11:33:32
0.30
Typical pattern of cracked inner race. PK = .3042
LOAD = 100.0
Mounds of peaks represent 1X FPM = 1403.
RPM = 370.
0.25
modulation of BPFI. harmonics.
0.20
0.15
0.10
0.05
0
Freq: 12.53
0 100 200 300 400 500 600
Ordr: 2.032
Label: BPFI W/1xRPM SDBND--ACCEL;1600
Priority: B
CSI CONFIDENTIAL
2.5
Waveform Display
2.0 19-JUN-89 11:33
Acceleration in G-s
RMS = .4601
1.5
LOAD = 100.0
FPM = 1460.
1.0 RPM = 385.
0.5 PK(+) = 1.80

PK(-) = 2.44
0 CRESTF= 5.30
-0.5
-1.0
-1.5
-2.0
-2.5
-3.0
0 100 200 300 400 500 600 700

Time in mSecs
Label: BPFI W/1xRPM SDBND--ACCEL;1600
B alls or ro lle rs pas sin g o ve r a crac k in the race caus e a s harp im pac t a nd ringin g pattern
Priority: B
in the w a ve fo rm . T he pattern is quite dis tinc t. T he am p litude s , ho we ver, are o nly

m o derate be c aus e of the s lo w roll s pee d an d the lo w F m ax .
R e m em be r als o th at B P F I de f ec ts of te n pr o ve m or e d iff icult to de te ct be c aus e o f the ir
po o r tr ans m is sio n path to the tr ans d uc er . T a ke ser io usly the ver y pr ese nc e o f B P F I
de f ec ts or patter ns w he n co nduc ting diag nos tic s — r egar dle ss of am plitude .
CSI CONFIDENTIAL
LUBE - LUBE XAMPLE-CENTER HUNG PUMP
NO LUBE -P1V PUMP BEARING 1 VERTICAL
0.07
Route Spectrum
16-APR-91 10:46
0.06
PK = .1364
LOAD = 100.0
RPM = 1745.
O n the plot be lo w , a bo ut 93 H z se par ates o ne pe a k f ro m the ne xt am o ng the gr o up o f RPS = 29.08

0.05
s e ven pe aks be tw ee n 80 0 a nd 16 00 H z. W hat c o uld c ause this d ata? T he s pac ing is
a bo ut 3.2 o r ders — B S F o r B P F O ?
0.04
0.03
0.02
0.01
0
0 400 800 1200 1600
Frequency in Hz
Label: CLASSIC LACK OF LUBE EXAMPLE
CSI CONFIDENTIAL
LUBE - LUBE XAMPLE-CENTER HUNG PUMP
8
Waveform Display
16-APR-91 10:46
6
RMS = 1.89
LOAD = 100.0
4 RPM = 1745.
RPS = 29.08
Acceleration in G-s
2 PK(+) = 6.29
PK(-) = 5.67
CRESTF= 3.32
0
-2
-4
-6
-8
0 60 120 180 240 300

Time in mSecs
Label: CLASSIC LACK OF LUBE EXAMPLE
CSI CONFIDENTIAL
A gro up o f pe a ks ap pe ars in the 80 0- to 1 60 0-H z ran ge se parate d by 8 0 a nd 13 0 H z (in
this case, 93 H z). LUBE - LUBE XAMPLE-CENTER HUNG PUMP
0.07
Route Spectrum
16-APR-91 10:46
0.06
PK = .1364
LOAD = 100.0
RPM = 1745.
0.05 RPS = 29.08
0.04
0.03
0.02
0.01
0
Freq: 1108.1
0 400 800 1200 1600
Ordr: 38.10
Label: CLASSIC LACK OF LUBE EXAMPLE Dfrq: 92.46
CSI CONFIDENTIAL

05 Bearings

Uploaded by

Copyright:

Available Formats

05 Bearings

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

05 Bearings

Uploaded by

Copyright:

Available Formats

Rolling Element Bearings

• Identifying Frequencies w/ spectrum

– # of sidebands increases as damage progresses

Dominate frequency is .89 orders. 2x Belt frequency

Impacting is evident. Pattern typical of balls passing over a defect.

0.05 >FAF 6309

Max Amp 11-SEP-90

0 60 120 180 240 300

Extremely high levels of energy, impacting.

0.5 PK(+) = 1.80

0 100 200 300 400 500 600 700

in the w a ve fo rm . T he pattern is quite dis tinc t. T he am p litude s , ho we ver, are o nly

O n the plot be lo w , a bo ut 93 H z se par ates o ne pe a k f ro m the ne xt am o ng the gr o up o f RPS = 29.08

0 60 120 180 240 300

0.05 RPS = 29.08

You might also like