vi1603 spring r2 final

8/15/2019 VI1603 Spring R2 Final

1/28

Vibrations

Spring 2016 | Volume 33, Number

Dedicated to the dissemination of practical information on evaluating machinery behavior and condition.

www.vi-institute.org

Turbocharger Rolling-ElementBearing Observed During Failure

By Maryon J. Williams, Jr., Ph.D., P.E.


2/28

INTRO TO MACHINERY VIBRATION (IMV)

April 18-21, 2016

Knoxville, TennesseeAugust 15-18, 2016

Oak Brook, Illinois

June 14-17 2016

Asheville, North Carolina

October 4-7, 2016

Salem, Massachusetts

BASIC MACHINERY VIBRATION (BMV)

July 25-29, 2016

Knoxville, TennesseeSeptember 19-23, 2016

San Diego, California

Nov. 28 — Dec. 2, 2016

New Orleans, Louisiana

2016 TRAINING SCHEDULE

Practical Application.

Expert Knowledge.

Real-World Solutions.

Vibration Institute training courses strike the perfect balance among theory, principles,

techniques, case histories, and practical knowledge to help you be a better analyst.

Attend a Vibration Institute training course and take your career and organization to the

next level.

2016 Training Course Schedule and Locations *

MACHINERY VIBRATION ANALYSIS (MVA)

September 19-23, 2016

San Diego, California

Nov. 28 — Dec. 2, 2016


ADVANCED VIBRATION CONTROL (AVC)

April 18-21, 2016

Houston, Texas

ADVANCED VIBRATION ANALYSIS (AVA)

Nov. 29 — Dec. 2, 2016


BALANCING OF ROTATING MACHINERY (BRM) ROTOR DYNAMICS AND MODELING (RDM)

July 25-29, 2016

Knoxville, Tennessee

October 3-7, 2016

Salem, Massachusetts

May 9-13, 2016

Knoxville, Tennessee

* Assigned course dates and locations are subject to change.

Visit www.vi-institute.org for the most up-to-date information on the 2016 training schedule.

Remaining Dates


4/28

2 VIBRATIONS SPRING 20

Contents

columns | departments

4 Feature Article Turbocharger Rolling-Element

Bearing Observed During FailureBy Maryon J. (Skip) Williams, Jr., Ph.D., P.E.

8 Case History Analyzing the Start-Up Vibration

of HOT PUMPSBy Joppu Thomas

16 Case History Rub Identification in Hydrogen

Centrifugal CompressorBy Juan C. Ustiola

SPRING 2016Volume 33, Number 1ISSN 1066-8268

3 Letter from the President By David Correlli, VI Board President

22 2016 Newly Certifed Individuals

24 A Fond Farewell By Robin Ginner

Vibrations is published quarterly inspring, summer, fall, and winter bythe Vibration Institute. Statements offact and opinion are the responsibilityof the authors alone and do not implyan opinion on the part of the officersor members of the Vibration Institute.Acceptance of advertising does not implyan endorsement by the Vibration Institute.

© 2016 by the Vibration Institute. All rightsreserved. Materials may not be reproducedor translated without the express writtenpermission of the Vibration Institute.

Technical EditorBarry T. [email protected]

Editor & AdvertisingNicole [email protected]

List of AdvertisersConnection Technology Center Inc.PCB Inc./IMI SensorsRBTS, Inc.Reliability Web

Postmaster

Send address changes to the VibrationInstitute, 2625 Butterfield Road, Suite128N, Oak Brook, IL 60523.

SubscriptionsVibrations is sent quarterly toInstitute members. The subscriptionrate is $75/year for individuals notaffiliated with the Institute.

Reprints and Back IssuesTo order article reprints or request reprintpermission, please send your requestin writing to [email protected] call the Institute at (630) 654-2254.

Vibration Institute

Please send any correspondenceregarding change of address oradvertising to the Vibration Institute.

Vibration Institute2625 Butterfield RoadSuite 128NOak Brook, IL 60523(630) 654-2254


5/28


6/28


feature article

Skip Williams is Senior Engineer with Condition Analyzing Cor

ration (www.cacvibe.com). He has conducted vibration surv

and has analyzed the data from ships for over 25 years. He c

tributes to ISO/TC 108/SC 2/ WG2 “Vibration of Ships” and to ISO/

8/SC 8/WG12 “Ship Vibration.” He was project leader for ANSI/A

S2.28-2009 “Guide for the Measurement and Evaluation of BroadbVibration of Surface Ship Auxiliary Machinery.”

ABSTRACT

This article shows the failure of a rolling-element bearing in real ti

on the turbocharger of a ship’s Diesel engine. The failure was ant

pated because the bearing had failed previously when the turbochar

reached a certain speed. Instrumentation was prepared to take conti

ous samples of the data before, during, and after the failure. While

time from initial detection of a bearing fault until failure of the bear

can typically take months, the failure reported here occurred in m

seconds. The vibration data at failure indicated a bearing cage prob

which was later confirmed with debris in the bearing cage.

BACKGROUND

A unique situation was provided when it was known that a bear

on the main Diesel engine turbocharger of an oil tanker had failed

previous occasions. The propulsion system consisted of a medium sp

8-cylinder engine with a reduction gear to the propeller of ratio 5:05

(Figure 1). The propeller was controllable pitch, meaning the thrust

ward or aft was controlled by the pitch of the propeller, not by the sp

and direction of the propeller. Also attached to the gear was a PTO g

erator with a step-up ratio of 2.022:1. The normal engine speed wa

constant 595 rpm. The expected forcing frequencies from this mach

are shown in Table 1.

The turbocharger bearings were proprietary anti-friction roller tylubricated with oil pumps operating on the same shaft as the turbine

compressor and mounted on the outside of the compressor and turb

wheels. The compressor bearing was the thrust bearing and consiste

11 balls in two rows. The turbine bearing had 11 rollers in one row. T

bearings were mounted on damped spring assemblies on the radial

axial sides meant to absorb a degree of imbalance in the rotor. Aft

By Maryon J. (Skip) Williams, Jr., Ph.D., P.E.Certied Vibration Specialist IV

Condition Analyzing Corporation

Eatontown, NJ

Turbocharger

Rolling-Element BearingObserved During Failure

Figure 1: Layout of engine, step-up gear to generator/motor

Figure 2: - Waterfall plot during failure


7/28

feature article

SPRING 2016 VIBRATIONS

failure of the turbine side lube oil pump, a replacement rotor was bal-

anced and installed. On two subsequent occasions the turbocharger

bearings had also failed when the rotation rate reached 13,400 rpm.

TURBOCHARGER BEARING FAILURE

The vessel was run at lower load than normal with a propeller

pitch of 55% until the end of the voyage. The greatest vibration

was found in the axial direction with radial vibration low. Measure-

ments were set up on the turbocharger with an axially-mounted gen-eral-purpose accelerometer. Initial readings were taken at Fmax of

5,000 Hz with minimal high-frequency activity. Sampling was then

taken with Fmax of 500 Hz and 800 lines with aterfall slices every

9.6 seconds. With these settings up to 16 minutes could be saved in

a single waterfall. Toward the end of the voyage the propeller pitch

was gradually increased to 73% over a time period of half an hour.

During the increased loading, the turbocharger rpm increased from

10,000 rpm to 13,400 rpm when the turbocharger failed (Figure 2).

Table 1: Speeds & Forcing Frequencies

Component Rotation rate (rpm) Harmonics Comment

Engine 595 2X, 4X, 6X-engine 8 cylinders

Turbocharger 8,000-18,000 11X 11 full compressor vanes

Step Up Gears 595/1203 91X-engine Teeth 91/184 (2.022:1)Reduction Gears 595/118 19X, 76X -engine Teeth 19/96 (5.042:1)

Propeller 118 4X 4 blades

fig.

3

fig.

4

fig.

5

fig.

6

fig.

7

fig.

8

fig.

9

Figure 3: T-9 minutes

Figure 4: T-9.6 seconds

Figure 5: Failure

Figure 6: T+9.6 seconds

Figure 7: T+19.2 seconds

Figure 8: T+2 minutes (Maximum vibration at rotation rate)

Figure 9: T+3 minutes


8/28


feature article

The failure was preceded with no indication of abnormal p

rameters on the turbocharger, turbocharger lube oil pump, engi

or gear. What was observed during the load increase was a d

crease in the vibration at rotation rate from 0.09 to 0.03 inch p

second peak, then a slight increase to 0.05 ips peak. From owaterfall slice to the next, an increase in 1X-rotation rate to 0.

ips suddenly appeared, but more importantly a new frequency

0.47X-rotation rate also appeared (Figure 6). In other words, t

time from detection of an abnormal spectrum until the failure w

observed was less than 10 seconds or approximately 2000 rot

tions. The vibration at 0.47X-rotation rate continued for only o

more waterfall slice (Figure 7) and was not seen again. Vibrati

also appeared at the upper rotation-rate sideband at 1.47X-ro

tion rate. Over the next 2 minutes, vibration at 1X-rotation r

increased to 0.73 ips peak (Figure 8). A time sample just after t

failure showed spikes at intervals of 9.4 milliseconds correspon

ing to the vibration at 0.47X-rotation rate (Figure 10).

After the failure the vibration at rotation rate decreased as t

pitch of the propeller was reduced to limit the vibration while t

vessel proceeded into port. The turbocharger never locked

and continued to function until the vessel got to the dock.

When the vessel reached port, the turbocharger was disassem

bled. . The bearing spring assemblies in the bearing were fou

broken as shown in Figure 11, Figure 12 and Figure 13 with

spring clip lodged in a roller of the turbine bearing.

ANALYSIS

The vibration at 0.47X-rotation rate in the regular spectra is t

frequency expected for a Fundamental Train Frequency beari

defect (FTF). The clip lodged in the bearing roller is further e

dence of the failure of the bearing cage.

Vibration on a turbocharger normally consists of three comp

Figure 10: Time sample of axial acceleration from one waterfall

slice to the next before and after the failure. Pulses correspond to

0.47X-rotation rate or FTF.

Figure 11: Turbine bearing (left) and compressor bearing (right)

Figure 12: Turbine bearing with loose spring clip Figure 13: View of installed turbine bearing with spring clips in

rollers. Oil pump also shown.


9/28

feature article


nents: (1) vibration from the engine

at low-frequencies, (2) vibration at

turbocharger rotation rate, and (3)

vibration at compressor vane-pass

frequencies. Vibration from theengine is constant with load and is

sometimes associated with cracking

of the turbocharger foundation. Vi-

bration at turbocharger rotation rate

is associated with imbalance in the

rotor or wear in the bearings. The

manufacture’s recommended alarm

level for ltered rotation rate vibra-

tion was 0.17 ips peak. The alarm

levels for turbocharger rotation rate

used during the this study were 0.27

ips peak for the 1st alarm (alert)

and 0.40 ips peak for the 2nd alarm

(fault). Sometimes vibration at 2X,

3X, or 4X-rotation rate indicates

additional wear, but was not seen in

this case. Vibration from the engine

was normal, and no lube oil failure

occurred this time. Vibration at

compressor vane-pass frequencies

of 11X and 22X-rotation rate also

were not seen in this case.

Possible mechanisms for a bear-

ing failure in a turbocharger:

1. Thrust bearing overload2. Excessive vibration

transmitted from the engine

3. Lube oil failure4. Imbalance in the rotor5. Wrong thrust bearing,

incorrect installation,

excessive axial clearance

The loading on the thrust bear-

ing decreased in the axial direction

with increasing speed (Figure 14

and Figure 15). The decrease in

the axial vibration with increasing

load was due to compression of theaxial load spring with increasing

axial thrust. The turbocharger is an

axial turbine and a radial compres-

sor meaning the exhaust gas pass-

es through the turbine in the axial

direction and the intake air passes

through the compressor in the ra-

dial direction (Figure 16). A major

component of the axial thrust is the

pull of the compressor wheel away

from the turbine.

The manufacturer reported that

an incorrect balance procedure had

been used to balance the rotor. Thebalance condition is specic for

each turbocharger by the manufac-

turer, but ISO quality grade 2.5 is

a close approximation. When the

correct procedure was used and the

rotor re-installed, no further prob-

lems occurred.

Still, the absence of signicant

vibration at 1X-rotation rate before

the failure indicated that balance

was not the primary cause. It is

possible that one of the other fac-

tors (wrong bearing, installation

or clearance) might have been in-

advertently corrected during the

re-commissioning repair.

CONCLUSIONS

As vibration analysts we have

several tools available to us to di-

agnose bearing failures, including

high-frequency algorithms. Often

these techniques show a degree of

bearing failure, and the question is,

“When is the bearing going to fail”?

The same question arises when one

decides how often to conduct vibra-

tion tests: should the frequency be

monthly, weekly, or continuously

with automatic shutdown of the

machine?

Bearing failure rates can happen

in months, weeks, days, hours, min-

utes or even seconds depending on

the cause, machine loading, and ma-

chine speed. The example here was

one where the high machine speed

(over 10,000 rpm) clearly played arole. This is one example when a

failure could not be predicted using

traditional vibration analysis tech-

niques even 10 seconds beforehand.

While the conditions of this failure

are unique due to the circumstances

of the installation, there is no one

way to predict the moment of bear-

ing failure with certainty.

Figure 14: Time course of failure: 30-minute period (10 sec intervals)

Figure 15: Zoom on failure event (3-minute period).

Note only two readings at 0.47X-rotation rate

Figure 16: Turbocharge with axial flow turbine and

radial flow compressor


10/28


case history

Analyzing the Start-UpVibration of HOT PUMPS

INTRODUCTION

The subject of this case history was a hot service pump of center-hung

configuration with a multiple volute casing. The pump was directly

driven by a 740 KW induction motor with a rated capacity of 2,800

m3/hr with fluid specific gravity of 0.6. The pumping temperature was 280

C (540 F).

1ST START-UP ATTEMPT

The commissioning team attempted a start-up of the pump many times

However, the pump tripped on high vibration each time. High vibration waexperienced at both the pump drive-end and non-drive-end bearings. The

values exceeded 150 microns (6 mills) at both ends. The recorded run time

was barely a few seconds. The pump had a fluid film bearing (plain sleeve

installed at the drive-end (DE) and a fluid film bearing/antifriction bearing

combination installed on the thrust end (ODE). Since the run time of the

pump was only a few seconds, it was decided to analyze the coast-down

transient data to help determine the cause of high vibration. Due to the

rapid acceleration of the pump during startup, start up transient amplitude

and phase data were not considered meaningful. Therefore coast-down data

By Joppu Thomas ABSTRACT

One line monitoring and diagnostic systems

are widely employed by petrochemical plants

and refineries worldwide for monitoring critical

machinery. These systems provide comprehen-

sive diagnostic capabilities which can be used to

reduce machine downtime and provide consid-

erable savings in maintenance costs. This case

history shows how the diagnostic capabilities

of an online monitoring system were utilized by

a leading refinery in the Middle East to solve a

vibration problem experienced with a 740 KW

Hot Pump. Hot pumps are centrifugal pumps em-

ployed in services with temperatures above 150

C (300 F); warm-up of these pumps are required

prior to start-up. These pumps must be handled

carefully during start-up to ensure normal levels

of vibration and to prevent unwanted trips.


11/28


case history

was recorded and analyzed. Figure 2 below

shows the vibration trend plots for the pump

DE and NDE bearings.

PUMP COAST-DOWN &

SLOW ROLL DATA

Following the first start up attempt, coast-

down and slow roll data from the pump was

examined. A slow roll vibration level of 125

microns (5 mils) and 163 microns (6.4 mills)

was observed at the drive end (DE) and non-

drive-end (NDE) of the pump respectively

at a speed ≤ 10% of operating speed (~ 300

rpm). In addition, this radial vibration at the

drive and non-drive-end was observed to be

in-phase. Both orthogonal set of probes (X

& Y) installed on the pump bearings indi-

cated high amplitudes. Slow roll data on

rotating equipment is typically collected at

speeds < 10% of the operating speed of the

machine. At these slow speeds, the dynam-

ic forces due to problems such as unbalance

are very low and can be reasonably dis-

counted; the “vibration” seen at these very

slow speeds can be due to mechanical or

electrical run-out, a physical bow or bend in

the rotor, or other problems that are not al-

ways easy to determine without this unique

data. Prior tests at these hot pumps had rou-

tinely measured slow-roll “vibration” at or

near 10 microns (1/2 mill).

BEARING INSPECTION

Owing to the high vibration noticed, the

contractor decided to open the pump bear-

ings for a quick inspection. The drive-end

and non-drive-end bearing clearances

were checked and found to be within rec-

ommended tolerances. The wear marks on

the bearings also appeared nor-

mal. It was advised to the contractor to do a

manual slow roll check on the pump rotor

after assembly of the bearings and check for

run-out. Typically the slow roll vectors are

around 5 microns (0.2 mills). This is mostlydue to mechanical/electrical run-out of the

probe target surface and some small residual

bow of the rotor. However the values mea-

sured during manual rotation were much

larger around 26 microns (1 mill).

EXPLORING THE REASON

FOR VIBRATION

Many hot pumps employed in refineries all

around the world experience high levels of

vibration during start-up. However, many of

these cases go unnoticed due to absence of

shaft proximity probe monitoring systems.

A set point multiplier is a useful feature pro-

vided for machinery with proximity probe

based monitoring systems which exhibit

Certifcation SurveillanceIn order to protect the Institute members who have justly earned certication a

a Vibration Analyst, the Vibration Institute wants to pursue individuals who falsi

Institute certication in any manner.

If you are aware of any instance in which you believe an individual is falsifying hi

her certication status, please call or contact the Vibration Institute immediately.

If you are aware of any individual who you believe is violating basic ethics, pleas

contact the Institute as soon as possible. Failure to do so degrades the reputatio

of your certication and the Vibration Institute.

(630) 654-2254

[email protected]


12/28


case history

high levels of vibration during transi

condition. TRIP/SET POINT MULTIP

ERS enable the ALARM & TRIP values

be temporarily elevated until steady s

conditions are achieved in these machin

Since most machines are designed to sand operate below set levels of vibrati

caution must be exercised before appli

tion of a multiplier. Arriving at a pro

value and duration for a set-point multi

er involves making a database of vibrat

levels and their duration during multi

start-stops of machine. Pumps emplo

in hot service commonly experience st

up vibration due to uneven expansion of

casing during warm-up. In extreme ca

poor warm-up can lead to shaft bows. V

ious warm-up piping arrangements are

forward by pump manufacturers.

API 686 gives guidelines for warm

piping arrangements. Typical warm-up

rangements for a top-suction, top discha

pump (center-hung conguration) norma

consists of two routes of entry for the wa

up uid - through the discharge check va

bypass and through the warm-up line at

bottom of the pump casing. The entry ro

through the discharge check valve byp

introduces uid into the discharge volute

the pump. The introduction of uid thro

the bottom of the pump casing ensure

uniform bottom-up heating of the cas

and rotor.

Due to the high start-up vibration ex

rienced at the pump in question, an inv

tigation of the warm-up piping system

rangement was carried out by the ren

engineering team. It was noted that at a

pumps, the discharge check valve by-p

route was absent thus heating was carr

out solely through the bottom casing dra

These “bottom warm-up only” pumps c

sistently experienced lower levels of vibtion during start-up.

The investigation also discovered a p

ticular hot pump which experienced h

vibration during start-up until a discha

drain line was introduced. Introduction

the discharge drain line was intended a

temporary measure until repairs to the ch

valve by-pass arrangement could be ma

The drain line from the discharge was

Figure 2: Trend of pump vibration during start-up and coastdown?


13/28

SPRING 2016 VIBRATIONS 1

case history

lized to introduce warm-up uid into the casing bot-

tom. Following the introduction of the discharge

drain line, this pump experienced lower levels of vi-

bration during start-up.

The investigation concluded that the hot pump

warm-up piping arrangements played a signicantrole in the start-up vibration levels experienced at

the pumps. Subsequently, it was agreed to perform

only bottom-up warm-ups of the pumps. Also it was

agreed that a controlled warm-up was required with

a maximum casing top to bottom temperature differ-

ential not to exceed 50 C (122 F). On the basis of all

the above incidents, the engineering team concluded

that heating from the pump top to bottom is an infe-

rior practice as it results in an excessive temperature

differential within the pump casing (top to bottom).

This temperature differential creates a very large

“humping” or “bowing” tendency since the pump ex-

pands more at the top than the bottom. As a result,

the bearing brackets & housings might shift down at

both ends of the pump (center-hung arrangement)

and the rotor will bow along with them. However, it

is worth mentioning that a few hot pumps of similar

arrangement from a different manufacturer did not

show a signicant effect on start-up vibration, irre-

spective of their warm-up routing – the exact cause

for this was not known.

TRIP/SET POINT MULTIPLER

With the reason for excessive start-up vibration iden-

tified, it was necessary to estimate and apply a suit-

able TRIP or SETPOINT MULTIPLER to the ma-

chines. The criterion for arriving at the trip multiplier

duration and set points are based on machine histo-

ry. A carefully compiled data base of the vibration

data for a machine during multiple start-ups is very

valuable information. The SET POINT multiplier

duration should be sufficient to overcome start-up

temperature transients and stabilization. A bowed

rotor or unevenly heated casing on these machines

can endure anywhere between 45 to 180 seconds be-

fore the vibration levels stabilize below alarm levels.

Depending on the machine, sometimes it is advisable

to apply a multiplier to a specific vibration channel,based on prior history where the other channels ex-

hibit lower levels of vibration.

CONCLUSION

The pump was finally put in service with the modi-

fied warm-up arrangement and exhibited acceptable

vibration levels. The operation team modified the

warm-up procedures and piping per the findings of

the investigation to ensure a bottom-up heating of


14/28


15/28

NNECT TO CONFIDENCE 7939 Rae Boulevard • Victor, NY 1CONNECTION TECHNOLOGY CENTE

VIBRAT ION ANALYSIS HARD

HAZARDOUS AREA

RATED SENSORS IN STOCK, SAME DAY SHIPPING

US & Canada: 1-800-999-5290International: +1-585-924-5900

Biaxial & Triaxial Accelerometers Dual Output: Temperature & Acceleration

Unconditional Lifetime Warranty!

If any CTC Vibration Analysis Hardware product should

ever fail, we will repair or replace it at no charge.


16/28


the pump. The TRIP MULTIPLER settingwas also judiciously applied. A multipli-

er of 2 times trip value (2X) for a duration

of 120 Seconds was applied which ensured

temperature stabilization of the pump. Sub-

sequent bearing inspection was also done to

physically verify that the applied multiplier

did not exceed the bearing clearance values.

Insufficient warm-up or fast warm-ups can

lead to high vibration in the pump during

start-up due to a thermally induced bow ofthe rotor. A thermally induced bow can also

occur due to a temperature differential in the

mechanical seal area. Some other factors

which can induce a thermal bow are:

1. Length of time pump remainedstationary (not rotating)

in a hot condition.

2. Casing differential temperatures.

REFERENCES

1. Fundamentals of Rotating

Machinery Diagnostics by Dona

E. Bently and Charles T Hatch.

2. Rotordynamics by Agnieszka Muszynska.

3. Design and Opeartion of pumps f

Hot standby services by Charles.C

.Heald and David.G.Penry.

4. Review of Vibration problem in power station Boiler feed Pumps

David France, Weir pumps,Glasg

case history

Figure 4: Warm-up piping arrangement

Certication as a Vibration Analyst is valid for ve years from the

date of current certication level. After ve years, and in com-

pliance with ISO 18436: Part I, certied Vibration Analysts are

required to recertify. Recertication at the current level of certica-

tion can be achieved in one of two ways:

Renewal. You may provide evidence of continuing education ex-

perience, training and/or technical activity. Points for renewal canbe earned for vibration-related activities including work experi-

ence, professional development, attending industry, association or

chapter meetings, and vibration-related presentations and pub-

lished articles.

Re-examination. You may take the certication exam at the level

you are currently certied. This requires scheduling an examination

and securing a proctor per established Vibration Institute protocol.

Vibration Analysts are certied on the basis of ability to function

at a specied level. The motivation for re-certication is to ensu

that the Vibration Analyst maintains the capability to function at

level certied.

Points toward recertication can be earned in various ways. The

Vibration Institute Certication Committee has approved renewa

requirements as follows:

Category I: 24 points (beginning January 2011)

Category II: 28 points (beginning January 2011)

Category III: 32 points (beginning January 2013)

Category IV: 36 points (beginning January 2014)

Visit www.vi-institute.org and click on Certification to learn more

about earning points for re-certication!

Recertifcation Requirements

Figure 3: Pump BODE for coast-down (Pump DE)


17/28

23rd Annual Seminar & Short Course On

ROTOR DYNAMICS & BEARINGS TECHNOLOGIES Lateral & Torsional Vibration Analysis / Fluid-Film & Rolling-Element Bearings

Gear

Loading

Bearing No. 1

Bearing No. 2

Aerodynamic

Loading

Coupling

Sinusoidal

force input

DISC

Unbalance

Z

Y

X

Body Force

Direction

May 2-5, 2016, PENN STATE Great Valley, Malvern, Pennsylvania, USA

For more information & seminar brochure visit our web site or contact us.

RBTS’ 23 rd ANNUAL SEMINAR

A 4-day seminar and workshop for engineers and technical managers involved in

ROTATING MACHINERY design, operation, maintenance, diagnosis, and

troubleshooting, with emphasis on machinery rotor dynamics, drive train

torsional vibrations, and bearing systems (fluid-film & rolling-element) that

support, guide, and locate the rotating assembly. Detailed coverage of the field

of fluid-film / rolling-element bearings and rotor dynamics, illustrated by the

presentation of case histories and the application of advanced software formodeling, analyses, and troubleshooting real life bearings and vibration problems

encountered in rotating equipment. No previous experience is required.

1st Day

2nd Day

3rd Day

:

:

:

May 2, 2016

May 3, 2016

May 4, 2016

Seminar “FLUID-FILM BEARINGS”

Seminar “ROTOR DYNAMICS”

Seminar “ROLLING-ELEMENT BEARINGS”

3rd Day

WorkshopMay 5, 2016

Advanced Rotating Machinery Dynamics “ARMD™”

software training/demonstration applied to bearings,

bearing systems, rotor dynamics and torsional vibrationmodeling, analyses and interpretation of generated results.

:

Flexible registration for any number of days

4th Day

: WorkshopMay 4, 2016

Prepared & Presented by:

RBTS, Inc. Rotor Bearing Technology & Software, Inc.

1041 West Bridge Street

Phoenixville, Pennsylvania 19460, USA

Tel: 610-415-0412

Fax: 610-415-0413

email: [email protected]

web: rbts.com

Rotating Machinery Dynamics Journal/Thrust Bearings Spherical Roller Bearing


18/28


case history

Rub Identification in HydrogenCentrifugal Compressor

MACHINE DESCRIPTION

The machine to be analyzed is a 5-stage

Hydrogen Centrifugal Compressor

manufactured in 1995 by Nouvo Pig-

none. It is driven by a steam turbine whose

normal operating speed is between 8,000

and 9,500 rpm. The machine is monitored

using X-Y proximity probe pairs on each

fluid film bearing as shown in the instru-

mentation diagram (Figure 1) along with

two axial probes and a Keyphasor.

SYMPTOMS

A sudden increase in the vibration lev

was observed in the DCS and PI softw

(Figure 2). After observing the abnorm

increase in the vibration levels, an imme

ate request was made to the vibration s

cialist to analyze and identify the cause

the sudden change.

VIBRATION ANALYSIS

Rubs are generally defined as contact

Juan C. Ustiola,

Vibration Specialist Category III.

ABSTRACT

The following case history ex-

plains how a rubbing condition

was identified in a centrifugal

compressor by the utilization

of phase, bode and orbit plots.

This information was required

during the troubleshooting pro-

cess to analyze sudden changes

observed in the vibration levels

during compressor operation.

An abnormal increase was no-

ticed in the control room by the

night shift crew. The rotating

equipment group was quickly

dispatched to investigate the

problem.

KEYWORDS:

Rub, rubbing, bow, orbit, phase,

bode, shaft centerline, vibration

analysis.

(Right) Figure 2: PI software showing the sudden

increase in vibration levels at the DE X & Y probes.

Figure 1: System 1 Machinery Diagram


19/28


case history

tween stationary and rotating

components of a machine. A

rub in a rotating machine usu-

ally doesn’t happen by itself; it

is usually a secondary effect of

several malfunctions occurring

at the same time. There are two

types of rubs, as shown in figure

3. The first is an impact rebound

type rub; in this case an impact

occurs, and then the rotor bounc-

es back, and the process is repeat-

ed and repeated, etc. The second

is a friction rub that causes fairly

constant contact between a rotat-

ing and stationary part.

Our case had all the symptoms

of a friction type, where the rub

looks like unbalance. The vibra-

tion characteristics were high

1X vibration, with no major

amplitude in harmonics and no

sub-harmonic frequencies.

There are certain myths about

rubs and the way it is presented.

In our situation (friction type) one

of the most common effects ob-

served in the vibration response

are changes to the 1X vector due

to thermal bowing of the rotor. A

rub under different circumstances

is manifested in different ways.

For example, if operating speedis sufciently above the rst crit-

ical speed (resonance frequency),

then a sub-synchronous frequen-

cy may be present such as 1/2 X,

1/3 X, or 1/4 X.

In practice it has been observed

that vibration due to a rub occur-

ring at 1/2 X might take place

when the running speed is above

two times the rst critical speed

and 1/3 X when the running speed

is above three times the rst crit-

ical speed. In our case this con-

dition did not apply since our op-

erating speed was not far enough

above the rst critical speed. Full

spectrum waterfall indicates the

change in the 1X vibration ampli-

tude (Figure 4).

It is possible to sort rubbing out

from a balance issue by looking at

the 1x amplitude and phase over

time. If a shaft is unbalanced, the

phase angle should not change

over time at a constant speed.

However if the rotor is bowed

due to a rub, the amplitude and

phase angle may shift over time

as shown in gure 5.

In turbomachinery is always

important to trend the 1x val-

ue (Amplitude and Phase) since

the symptoms of many prob-

lems occur at that fre-

quency. When rubbing

occurs, sudden changes

in the phase are expected

at a constant speed; this

was observed in this case

study, where two sudden

steps in phase & ampli-tude were observed within

a period of 13 days. These

changes primarily occur

due to the contact of sta-

tionary & rotating parts;

heating from friction is

generated at certain areas

of the rotor (where the rubs

Figure 3: Types of rubs [1]

Figure 4: Waterfall spectrum showing

increase at 1X Vibration Amplitude.

Figure 5: 1x Amplitude & Phase shift over the time at constant speed.

Figure 6: Rotor in Healthy Condition (June 2013).


20/28

case history

are) which in turn results in a bowed shaft.

Bode plots revealed the change in the critical speed where the

rotor stiffened due to the rub. In some cases it can be higher than

the normal critical speed, since we are modifying the natural fre-

quency.

Several diagnostic plots are useful when diagnosin

rub. Chief among these is the shaft orbit plot (Figure

In some occasions shaft orbit presentations can give qu

dramatic indications of a rub as evidenced by at sp

on portions of the orbit where contact is occurring

restriction of orbital motion. [2]

FINDINGS

The drawing shown in figure 9 highlights the areas wh

the rubbing occurred in the hydrogen compressor. T

sectional drawing will allow us to visualize the c

cerned areas in a better way.

The following pictures provide evidence of the pr

lems analyzed through the vibration plots, where our t

ory of rubbing was proven. The drive end probes (Dwere the key for the investigation, since the problem w

reected best in that area. During disassembly much da

age was observed in stationary and rotary parts, howe

most of the severe problems were noticed near the s

tion (non-drive end), where rubbing at the inter-stage l

yrinth seals (rst stages) and fractures at the diaphra

were found. By analyzing the evidence, we suspect t

the shaft was internally misaligned & bowed during

eration causing a pivoting effect in the DE side where


Figure 8: Orbit plots before and after the rub.

Figure 7: Rotor with Rubbing problem (May 2015).


21/28

Great mindsdiscuss ideas

Average mindsdiscuss events

Small mindsdiscuss people

Quote: Eleanor Roosevelt, 1884-196

April 11-15, 2016

DiscoverNEW Ideas

Las VegasConference

The

reliabilityconference.com | 888.575.1245 | 239.333.2500 | [email protected]

+

Produced by names you trust

Save $100

FREE AppleTVUse code: VI Register by 3/31/16

South Point Hotel and Casino

Las Vegas, Nevada


22/28


case history

increase of vibration levels was observed.

Piece of suction strainer found attached at impeller

eye. Solid deposits found inside and outside the

cylindrical inner bundle (cartridge casing) where rotor

and stationary internals are assembled.

ROOT CAUSE DETERMINATION

So we knew that a significant rub had occurred but

why? What caused or initiated this problem? From

figure 10 below notice that the shaft centerline position

had moved over time. This was observed as evidence

of internal misalignment between the rotor and station-

ary diaphragms, which caused an upward trajectory

from July 2013 to February 2015.

Damage was observed in both stationary and rotating parts during

the site inspection. Damage was observed at the inter-stage laby-

rinth seals, diaphragms and balance drum; contact between these

components created a rub, resulting in high vibration levels. The

contact between these stationary and rotating parts also generated

heat from friction causing thermal growth changes. These chan

in thermal growth altered the internal alignment, reducing cleara

enough to generate the contact.

From our inspections of the dismantled machine, a possible r

cause of this problem could have been the continuous formation

deposits in the compressor seals or ingress of foreign debris l

pieces of the strainer found attached to the rst stage impeller. O

Piece of suction strainer found attached at impeller eye.

Severe rub at discharge of lab. seal at 1st stage impeller

Solid deposits found inside and outside the cylindrical inner bundle

(cartridge casing) where rotor and stationary internals are assembled.

Cracks were found on diaphragm (top and bottom half).

Figure 9: Compressor sectional drawing showing key areas of trouble.


23/28

case history

accumulation of these deposits exceeded the available

clearance, a friction rub could be developed at one point of

the shaft surface, causing an increase of the shaft surface

temperature and subsequent shaft bowing.

REFERENCES:

1. Bently Nevada Corporation “Machinery DiagnosticCourse”. Figure 3.

2. Arthit Phuttipongsit (Principal Engineer) “The Morton Effect and Light Rubs in Rotating

Machinery”. ORBIT Vol.27 No.1 2007.

BIBLIOGRAPHY:

Ronald L. Eshleman “Machinery Vibration Analysis” Vol-

ume II: Analysis and Correction. Vibration Institute 2002.


All inter-stage labyrinths have heavy contact mark

Rub observed at balance drum area

Solid deposits on labyrinths seals.

Piece of suction strainer lost in 2013 found

trapped between case and bundle.

Figure 10: Shaft centerline position from July 2013 to February 2015.


24/28


newly certied

CATEGORY I

Ian Kendrick; David R. Craft; Kevin R.

Grider; Tariq Salim Al-Sultan; Jeremy Boyer;

Mubarak Mohammed Al-Khaldi; Mubarak

Abdullah Aldossari; Joe Overbaugh; Shawn

Paul Morgan; Donald Robb; Ali Abdullah

Al Ousif; Mubarak Aali Alghamdi; Lloyd

Adam Moritz; Vishnu Itwaru; Kaitlin Saranda

Spak; Selwyn Fridey; Nathaniel Parr; Alberto

Izquierdo Leal; Alberto Conde Romero; Felipe

Martin Arroyo; Jesus Gomez Calderon; Luis

Amores Garcia; Angel Luis Garcia Gragera;

David Carrell Marcos; Jesus De La Fuente

Miguel; Enrique Velasco Montoro; MatthewSean Adamchick; John Wayne Borelly;

Edgar Godinez; Manuel Ramirez Lucero;

Troy E. Scott; Luvern Curt Andrist; Russell

Alan Davis; Taylor North Fry; Justin Miller;

Chase Lee Kiser; Shane Lockwood; Stephen

Wayne McGinn; James Wesley Palmer; Trent

Wesley Pol; Jason Christopher Stapley; Jason

R. Ungar; Dongyang Hu; Sivasubramanaian

Malli Sivaramaiyer Mothilal; Guo Long

Oh; Parthiban Palaniappan; Diego Gerardo

Barquero Espinoza; Kendal Varela Carvajal;

Fabian De Jesus Loria; Donald Nelson Moreno;

Juan Diego Ugalde; Jeffrey Luna Galera;Olger Robles Solano; Khisbullah Hudha;

Muhammad Anas Bin Abu Bakar; Mat Rashid

Rasnijeery; CW Ahmad Saiful Bin CW Zakaria;

Leo Ventura Punongbayan; Randy Dinso

Nakila; Vicente Jr. Pabilan Trinidad; Elmer

Leonardo Arias Cantreras; Marco Gomez

Carias; Elmer Ulises Gonzalez Hernandez;

Mauricio Ernesto Quionenz Monroy; Oscar

Huber Ramirez Gomez; Imad ur Rahman

Khan; Stephen Owen; Rogelio Diaz Le

Fabricio Doldan Novo; Jose Alberto Ga

Garrido; Roberto Ibanez Rico; Jose AntoMarquez Secilla; Jose Maria Melion Ala

Docner Andrey Schorbooth Contreras; Lor

Pastor Sopelana; Jose Octavio Gomez Pon

Pedro Reyes Lopez; Luciano Ivan Aco

Ramirez; Rafael Dekar Chavez Molano; Ma

Antonio Izquierdo Licona; Alvaro Ang

Lara; Victor Hugo Boyzo Torres; Rey Da

Coronado Castillo; Pedro Alberto Coron

Marquez; Juan Carlos Duran Gastelum; Ju

Cesar Lopez Miranda; Cesar Enrique Agu

Diaz; Uriel Rodriguez Chairez; Carlos Mig

Tobon Plascencia; Teboho Timmy Kutoa

Sandile Ralph Mkhize; Alejandro Pa

Bellmunt; Sergi Desunvila Benaiges; J

Pablo Dodero Bendana; Ivan Aliaga Due

Jose Manuel Iglesias Garcia; Jose M

Martinez Trinidad; Jesus Miguel Portab

Cerezal; Angel Ramirez Martinez; Josep S

Calanda; Oriol Save Sanchez; Wilson Alb

Martinez Vargas; Ashleigh Brianne Gutterid

Michael Bruno Mitrovic; Peter John Richa

Stuart William Hodge; Russell St

Lemmens; Michael Adam Klomp; Christop

James Woodward; Juan Jesus Gomez Tex

Anthony Wayne Bradford; Shawn Tye WaKris R. Witteveen;

CATEGORY II

William R. Ballenger; Bruce W. Blankensh

Jerry L. Martin; Luis F. Ovalle; Alfred Will

DeVaux; Deon Foonk; Johannes Will

Ferreira; Ahmad Adnan Al-Ammar; Has

WELCOME

TO OUR NEW

COLLEAGUES

CERTIFIED INDIVIDUALS

November 26, 2015 -

March 7, 2016

Please note: Every attempt is

made to alphabetize the names

of individuals by surname. We

apologize if any names are out of

order. If you notice a misspelling,

please contact the Institute

headquarters at information@

vi-institute.org.

Newly Certied Individuals

2016


25/28


newly certied

Yahya Alamri; Khaled Saad Al-Ghamdi; Mark Anthony Silvallana

Banal; Willman R. Pinto Flores; Francisco Gerardo Sosa De Los Santos;Francisco Javier Sanchez Olivares; Jesus Tellez Reyes; Raul Leana

Ortiz; Angel Maria Fernandez Rojas; Wilson Garcia Beltran; Donald

H. Matchett; Ben James Drew; Brett Anthony Esler; Ali Mohammad

Al-Ja’afar; Julio Cesar Roque Cormilluni; Leonid Ramirez Sierra; Gino

Blommaert; Sindre Karlsson; Abdullah Al-Abood; Ammar Abdullah

Abussonoon; Mohammed Essa Al Ahmed; Mohammed Sanad Al-Awwad;

Ibrahim Saad BuBshait; Prabu Velayutham; Gareth Ewan Coffee; Robert

George Maybin; Juan Carlos Sotelo; Christian Alexander Smith; Vladimir

Derkach; Nicolas Martinez; Dustin K. Dixon; Ademola Olarinde;

Mohammed Ali Al-Qahtani; Mohammed Abdulraouf Ahmed; Ahmad

Mohammad Alamri; Bernard Chan Poh Khin; Vipin Chandrasekhar;

John Paul Anderson; Edwin Evaristo Tapia Chuquimamani; Yahia A

Abdelhamid Ismail; Hongxing Wang; Gian Carlos Reyes; Elia Doretto;Eduardo Aguilar Armendariz; David Martinez Cantera; Jorge Angel

Saenz Serdio; Juan Carlos Serrano Orozco; Carlos Galan; Cesar Orlando

Perdomo Venegas; Ed Freddy Arzapalo Barrera; Juan Manuel Rodriguez

Biminchumo; Marco Antonio Luna Duenas; Jaime Bwrayan Mori Torrejon;

Carlos Omar Peralta Lamaure; Mario Alejandro Sime Odar; Victor Jesus

Ticse Carhuallanqui; Diego Alonso Valenzuela Medina; Omar Tawfiq

Albassam; Chellaiah Ganesh; Philippe Athanasiadis; Matias Canedo;

Luc Y. L. Fromaigeat; Michael Albert Hafner; Victor Lara; Yves Kwame

Mayor; Frederic Micco; Hans Claudius Reiss; Stefano Trono; Kyle

Wayne Davis; Jayde Kalei Jarnagin; Dean Michael Kelley; Alexander

Michael LeBlanc; Daniel Lucas Stein; Kyle Patrick Thayer; Britta

Haalboom; Mohd Farhan Abd Jabar; Azlee Bin Anis; Ruslan Bin Miah;

Ahmad Faris Bin Mohd Shukri; Sivananthan Thanabalan; Ove Sagen

Adsen; Petter Hausler; Jaroslaw Nowak; Agnieszka Tkaczyk; Zhenyou

Zhang; Sung-Won Han; Jung-Bin Im; Jeong-Yoon Kim; KyuEun Kim;

Min Seo; Sung-Jin Song; Anandvinoth Swamykannu; Piyanath Jaisuk;

Pongkul Akaravoramas; Poramit Chantasa; Pinyo Sinpoonkird; Somjin

Thanananthasiri; Krisada Vimonmangkang; Siwat Yuthong; Ian Struan

Peter Andrews; Nakikus Goliath; Daniel Humberto Bohorquez Velandia;

Charles Eduardo Correa Rodriguez; Alvaro Hernan Rojas Carvajal;

Sulaiman Abdulaziz Al-Turaifi; Mohammed Abdulmoghni Al-Sulami;

Ahmed Abdullah Al-Ammar; Fahad Mohammed Aldawsari; Abdullah

Nuri Alkhudhayr; Mohammed Salem Badughaish; Wesley Carter; Brian

Joseph Quirk; Brian J. Tafolla; Jason W. Jones; Mark Mosher; Huey

Pierce Cheek; Thomas M. Werking; Jesus Alejandro Cante Garcia;

Joshua David Stewart; David John Starcevich; Dustin Matthew Rothe;

John Robert Myers; Douglas Jillson Simmons; Gene Carl Standley; Ilya

Igorevich Burdukov; Brian Patrick Lawburgh; Mahmoud Nasr Salem;

Nathan Charles Decker; Starsun Levi Hendrix; Jonathan Patrick

CATEGORY III

Juan Carlos Albornoz; Gary F. Carroll; Robert L. Martin, Jr.; James H.

Payne; Rick Croix; Matthew Tyler Gilbert; Charles Stevens; Louie

Jefferson Langford; Mike L. Fitch; Ramiro Flores; Ronald Fennell;Sean D. DeVinney; Steve Duncan; Terry Scroggins; Keith Conley;

Michael A. Szurkowski; Kiran K. Toram; Guruprasad Pandurangan; Ali

M. Al-Shurafa; Majed Hassan Saimaldahr; Faisal Kalim Qazi; Sankar R

Ganesh; Tareq Essam Ibrahim Saqr; Jiaqian Wang; Carl Robert Binns;

Joe Paul Dinaro; Zoran Racic; Shaomin Wang; Weizhong Jin; Ning

Yue; Jace Landthrip; Niranjan Vallabhram Panchal; Michael Placid

Joseph; Zhen Zhang; Han-Yuan Chu; Ke Tan; Jian Gao; Chengzhong

Gu; Jiang Wang; Thomas Jerimy McDowell; Shivakumar Swaminathan;

Michael Supplee; Michael Merten; Sudhar Rajagopalan; Ahmed Rifaat

Abdelhamid; Faisal Ahmed Youssef;

CATEGORY IV

Thomas J. Walter; Nicholas A. Jagan;


26/28

VI headquarter update


A fond farewell

Two and a half years ago I sat down to an interview with members of the

Vibration Institute’s Executive Committee to discuss my qualifications

as a Director and the challenges that lie ahead as the leader of an organi-

zation that serves the vibration community worldwide. In the time since, I’ve

been honored to work closely with the Board of Directors to make some good

and lasting changes that will continue to benefit our members and certifiedanalysts who call the Vibration Institute their professional association.

It is with mixed emotions that I’ve submitted my resignation from the Vibra-

tion Institute. Although just over two short years, I have thoroughly enjoyed

the time I have spent at the Vibration Institute. I am appreciative of the trust

the Board has given me, and have had a wonderful experience in my time with

the Institute. My only regret is that I will not have the opportunity to be part

of the continued progress being enjoyed by this time-honored organization.

I have been tapped to lead a school and nature center in my home town of

Eagle River, Wisconsin – an organization that is near and dear to my heart and

played a signicant part in forging my love of the outdoors in my youth. My

last ofcial day with the Vibration Institute was Friday, February 19, 2016.

In the interim, until a replacement Director can be found, I trust in the skills

and commitment of the Institute’s headquarters staff to be able to assist you in

anything you may need. I have enjoyed my time with the Institute, and getting

to know so many of you. I hope to see you on LinkedIn, and wish you all the

best in your careers.

Kindest regards,

Robin Ginner


27/28

AREYOUREALLY

COMPARING

THESAME

THING?

IMI Sensors stands behind every spec we publish.

If your

measurementsmatter, then

SPECMANSHIP

MATTERS.

IMI -SENSORS.com

I N D U S T R I A L M O N I T O R I N G

800.959 .446


28/28

vi1603 spring r2 final

Documents