online_failure detection

Early detection of machinery failures using online
Condition Monitoring Systems (CMS)
PRUFTECHNIK Middle East FZE (Dubai) / Robert Schmaus / 28-04-2013

27/04/2013PRUFTECHNIK Middle East FZE (Dubai) – Robert Schmaus 2

Agenda
- Content
►Brief company profile
►Introduction to CM
►A typical plant
►Machine health monitoring
►CM technologies
►Level 1 & Level 2 strategy
►The right measurement tools
►Is online monitoring needed?
►Case studies
►Examples of online CMS installations
►Summary

COMPANY PROFILE
PRUFTECHNIK

PRÜFTECHNIK Group facts
► Location of headquarter: Ismaning (Munich), Germany
► Divisions:
● PRÜFTECHNIK Condition Monitoring GmbH
● PRÜFTECHNIK Alignment Systems GmbH
● PRÜFTECHNIK NDT GmbH (Nondestructive Testing)
► Founded in 1972 by Dieter Busch
► PRÜFTECHNIK Group CEO: Dr. Sebastian Busch
► Activities:
● Research & Development
● Production
● Sales and service of electronic and electrical instruments,
especially measurement and control systems.
► Patents & Trademarks: 300 patents & 150 trademarks worldwide
► Quality Management: ISO 9001 since 1994
► Staff: approximately 550 employees
► Sales turnover 2012: approx. 70 millions EURO
► Sales Channel Network in over 70 countries
PRÜFTECHNIK general HQs in Germany
Dieter Busch Dr. Sebastian Busch

Worldwide sales channel network
 PRÜFTECHNIK maintains an international sales network
covering more than 70 countries

PRÜFTECHNIK – a reliable long-term
partner
 Since the foundation of PRÜFTECHNIK
in 1972, the company has maintained a
reasonable growth of sales turnover,
company organization and employees.
 PRÜFTECHNIK invests over 15% of its
annual sales turnover in R&D and every
year in order to ensure product
enhancements and development of new
products.
 Today more than 80% of the turnover is
achieved with products which are no
more than 4 years old.
 A substantial role in this is played by
the development department which still
makes up 25% (a quarter!) of the
workforce.
The high export quota of approx. 80% is
a distinguishing feature of technically
qualified products

Main product launches for Condition
Monitoring
1980 VIBROSPECT®
1980 VIBROCORD® 1
1989 SYSTEM 2
1992 X-Tector
1993 VIBROTIP®
1995 VIBCODE®
1996 VIBRONET®
1997 VIBREX®
1998 VIBROCORD® 2
1999 VIBRONET® Signalmaster
2000 VIBSCANNER®
2003 smartSCANNER®
2004 VIBROWEB®
2004 VIBXPERT®
2005 VIBROWEB® XP
2005 VIBNODE®
2009 WearScanner®
2010 VIBXPERT® II
2011 VIBCONNECT®
2012 VIBGUARD®
VIBROSPECT® VIBROCORD® 1
SYSTEM 2®
VIBROCORD® 2 VIBSCANNER® smartSCANNER®
VIBXpert® IIVIBXpert® EXVIBXpert®

PRUFTECHNIK awards
Year Product Award
2010 VIBXPERT® II Product of the Year 2010 – Winner (Inżynieria i Utrzymanie Ruchu)
2010 VIBXPERT® II Product of the Year 2010 – Finalist (Plant Engineering Magazine)
2010 VIBXPERT® II TOP 2 in 2010 IMPOvation Awards
2010 VIBXPERT® II Honorable Mention in the 2011 / AHR Expo Innovation Awards Competition
2010 VIBXPERT® II Breakthrough Product of 2010 (Processing Magazine)
2010 VIBXPERT® II Product Innovation of the Year – 4th place (Pumps & Systems)
2010 PARALIGN® Industriepreis 2010 (Huber Verlag für Neue Medien)
2009 INCLINEO® Breakthrough Product of 2009 (Processing Magazine)
2009 INCLINEO® Product of the Year 2009 – Finalist (Plant Engineering Magazine)
2009 ROTALIGN® Ultra IMPOvation product technology award (IMPO Magazine)
2008 VIBXPERT® EX Product of the Year – Finalist (Plant Engineering Magazine)
2008 OPTALIGN® smart Designpreis 2008 Nominee (Designpreis der Bundesrepublik Deutschland)
2007 PARALIGN® Technology of the Year Award (AIMCAL)
2007 PARALIGN® Matteucci Award Winner (AIMCAL)
2007 OPTALIGN® smart Product of the Year – Gold Award (Plant Engineering Magazine)
2007 OPTALIGN® smart red dot design award (Design Zentrum Nordrhein Westfalen)
2007 OPTALIGN® smart Product of the Year – Finalist (Plant Engineering Magazine)
2006 ROTALIGN® Ultra Product of the Year – Silver Award (Plant Engineering Magazine)
2005 PARALIGN® Palme d’Or des trophées de l’innovation 2005 (ATIP)
2005 ROTALIGN® Ultra Best of NED' Problem-Solver (New Equipment Digest (NED))
2005 ROTALIGN® Ultra Breakthrough Product (Processing Magazine)
2003 smartSCANNER® Product of the Year – Finalist Plant Engineering Magazine)
2003 smartSCANNER® Breakthrough Product (Processing Magazine)
2002 VIBSCANNER® Product Of the Year Award (Fluid Handling Systems Magazine)
2001 VIBSCANNER® Product of the Year – Gold Award Plant Engineering Magazine)
2000 VIBSCANNER® Roter Punkt für Hohe Designqualität Design Zentrum Nordrhein Westfalen
1997 OPTALIGN® PLUS Product of the Year – Silver Award (Plant Engineering Magazine)
1996 OPTALIGN® PLUS IF Product Design Award (Industrie Forum Design Hannover)
1993 VIBROTIP® Roter Punkt für Hohe Designqualität (Design Zentrum Nordrhein Westfalen)
1992 TURBALIGN® Product of the Year – Bronze Award (Plant Engineering Magazine)

04/02/2013PRUFTECHNIK Middle East FZE (Dubai) - Robert Schmaus 11
Some of our target markets
Oil & Gas Off Shore Power
Wind
Marine
Chemical Paper / Films
Sugar
TextileCement
Mining
Steel
Water
ServicesAutomotive

Some worldwide references
FEWA - United Arab Emirates

Introduction

Predictive Maintenance helps to
► Increase machine reliability
► Ensure machine availability
► Secure productivity
► Decrease maintenance costs
► ... And finally helps to optimize profits
Increase profitability in your plant
Years
Productivity
Costs
PROFIT

Time
Why CM?
Starting
point of a
failure
Evolution
into
vibration
Wear
particles in
oil
Increased
noise
Heat
development
Functional
failure
 CM shows the optimal time for intervention on machines

Example of a successful CM pilot project
 CM Vibration Analysis concept was started in 1990. The concept was realized
in 1991 and covered approximately 250 pumps.
 Usually 100 maintenance stops were required due to failures on pumps
 Main problems were the bearings
 Anual repair costs: 160,000 USD (without production losses, etc)
 After implementation of a CM strategy the number of repairs per year
decreased
Numberofrepairs

A typical plant

So... how are your machines doing?
Conveyors
KilnCooler PrehomogenizationRaw material storageRoller MillPacking Cement silo
QuarryCrusherPreheaterClicker siloCement mills Raw meal homogenization
12
4
3
5
7
8
9
10
11
12
13
14
You know how your production is doing. Fine.
 But what about the health of each machine involved in production?
6
Simplified example of a cement plant

So... how are your machines doing?
Conveyors
KilnCooler PrehomogenizationRaw material storageRoller MillPacking Cement silo
QuarryCrusherPreheaterClicker siloCement mills Raw meal homogenization
12
4
3
5
7
8
9
10
11
12
13
14
Condition monitoring allows to detect problematic machines before a damage unexpectedly
interrupts the production process
Condition monitoring tools identify machines at an early deterioration stage and classifies them
into different categories reflecting severity of health
GREEN = long term operation is OK / YELLOW = restricted operation allowed / RED = machine did
run into alarm status; must be stopped immediately
6

Machine health
monitoring

Breathing frequency
Condition Monitoring = Health Monitoring
Breathing trend
Oxygen trend
Vein pressure trend
Arterial blood pressure trend
Electro cardiogram trend
Oxygen in blood
Values of
artificial
respiration
Central vein pressure
Blood pressure
Alarms Heart rate
- Applied to human being  health monitoring
► Measuring parameters such as blood pressure, temperature, etc
► Trending values when required
► Analyzing changes over time
Control panel
TRENDS
VALUES

Condition Monitoring?
- Monitoring machinery health
► DEFINITION: Condition monitoring is the process of monitoring parameter of
machinery health or machinery condition, such that a significant change is
indicative of a developing failure.
► Rotating machinery reveales its condition via parameters such as:
● Vibration
● Temperature
● Oil samples
● Sound
● Motor current
● Etc
► Such machine parameters are measured with the aid of diverse technologies

Condition Monitoring technologies
Infrared thermography
19%
Oil analysis
14%
Other
9%
Ultrasound
7%
Motor current
analysis
3%
Vibration
analysis
48%
Process parameters ≠ condition monitoring parameters

Level 1 and Level 2
strategy

Level 1 & Level 2
Level 1
 Where is the problem?
Detection of machines with problems
Level 2
 What is the problem?
Analysis: finding the root cause of
problems
Example:
Level 1 measurements detect changes in the machine
condition
Measurement value trends increased over a time period
Changes in machinery health occur: vibration levels are
higher, machine condition is deteriorated
Example:
Analysis of FFT spectra and envelope spectra
Find source of problems e.g. unbalance,
misalignment
Diagnose and correct problem, e.g. balance rotor,
align machine shaft, etc

Level 1 & Level 2
Level 1
Detection
Level 2
Analysis
Example:
Performing vibration measurements periodically allows
to detect a machine with poor condition
Vibration levels increased compared to previous
measurements
Machine condition is clearly deteriorated – but why?
 Level 2
Overall value measurement trend
Example:
Analysis of FFT spectra and envelope spectra
Find source of problems e.g. unbalance,
misalignment
Diagnose and correct problem, e.g. balance,
align machine shaft
Vibration
value
time
Prewarning level
Alarm level
Warning level

Typical Reliability based Maintenance work flow
Define and perform measurements
of Level 1 & 2, depending on the
machine and the typical failures
expected to be detected (fast failure
evolution / online measurements /
Offline measurements / etc.)
Detection: trending of collected
measurement data of overall values.
Intelligent alarming is required here.
(Level 1)
Analysis: when relevant changes
occur to the values in the trends,
analyze Level 2 measurements
Diagnostic: based on the
interpretation of Level 2
measurements, formulate a
diagnostic and action
Correction: select the appropriate
correction method to correct the
failure and eliminate the failure root
cause*
1) DEFINITION
2) DETECTION
3) ANALYSIS
4) DIAGNOSTIC
5) CORRECTION





Example gearbox: RPM, bearing
types, number of teeth, number of
stages, etc.
Example: elevated overall vibration
value on motor drive side
Analyze the increased vibration peaks
and the characteristic frequencies : 1x,
2x, gear mesh = RPM x number of
teeth, etc.
Diagnose and describe the failure as per
Level 2 measurements – formulate the
root cause of failures
Misalignment  align with laser alignment
Imbalance  perform dynamic filed balancing
*After the correction phase, it is necessary to recheck the machine condition
again, to ensure that there is a real vibrational enhancement of the machine

The right
measurement tools

Tools used for vibration analysis
- Condition monitoring / vibration analysis
► Vibration data collectors can be:
● Portable data collector and analyzer
■ Mobile instrument and sensors
■ Fixed installed sensors connected to junction boxes
● Online Condition Monitoring Systems (online CMS)
● Machine protection
► Powerful software platform:
● Storage of collected measurement data
● Analysis and diagnosis of problems
● Reporting capabilities
● Interface to e.g. CMMS (SAP, Maximo) etc

Offline measurements

Portable data collector (offline) 1
Manual data collection with
portable instrument on a
large number of machines
Machine (pump)
Measurement
sensors attached to
machines via
magnetic bases
Collected data is
stored on a
software platform
(database),
analyzed and
printed into reports
VIBXpert® II instrument

Portable data collector (offline) 2
- Junction box (“SwitchBox”)
► Permanent sensors mounted on machines are connected to junction boxes
► Data is still manually collected but from a safe location
► Ideal for machines in hazardous or restricted areas
► Faster data collection than with mobile sensor and data collector

- Multiplexer support
(Automatic Switchbox)…
- just a step before
implementation of online CMS

► Concept of a standard (manual)
switchbox:
● safe, reproducible and reliable
measurement of
● remote or hard to reach
locations
► VIBXPERT® II multiplexer support:
● The above features PLUS
● Automatically switching through
the channels – up to 54 per line
● "One click measurements"
→ click once – have a coffee 
► A time saving feature for safe,
reliable and fast measurements
► First step towards implementation of
online systems (CMS)
…
Intelligent MUX measurements

► The feature can also be used in
intrinsically safe environment
► All that is needed is the ATEX
version of the multiplexers,
sensors, cables…
► … and a limiting device.
► The instrument connects then in
the non-EX area and enters the
ATEX area via the limiting device
VIB 3.550
Limiting
device
Multiplexor VIB 8.306
EX* with according
sensor cards
(VIB 8.314EX for ATEX
vibration sensors)
Vibration sensors VIB 6.122DEX
Intelligent MUX measurements

Online Condition
Monitoring systems
(CMS)

Online data collection
 Sensors are permanently mounted on critical machines
 Online CMS automatically collects data and handles alarms
 Monitoring ‘around the clock’
Simplified example of a cement plant
Database
Sensors
& Online
CMS
LAN /
Ethernet
Internet

RPM, temperature, process
parameter,etc
Sample main functions of an online CMS
VIBNODE®
online CMS
Switch off
Communication
with DCS / SCADA
Via Filedbus
Measurements, alarms, etc.
MACHINE PROTECTION PREDICTIVE MAINTENANCE
PLC
OMNITREND®
Software Platform
LAN / Ethernet
SENSORS
CABLES

CMS interfaces and functions (examples)
LAN/Ethernet
12 analog inputs
(e.g. accelerometers) 2 RPM
signals
2 digital inputs
(“trigger” for turn on/ turn off the
machine) TTL 5V…+30V
Control room
Visualization of the
measurement results
Example: Profibus, Modbus
Control panel
Visualisation of the
measurement results
e.g. stop of machine
Optional (VIB 7.112)
Two outputs 4-20mA
3 digital outputs
+30V / +0.1A
e.g. Acustic alarms
e.g. visual alarms
e.g. For stopping the machine
(*it is not a protection system)
RS232/1 serial communication
Wireless Transfer of
measurement data and
access toCMS
Notebook
RS232/2 serial communication
(System Fieldbus: Modbus, Profibus)
SPC/PLC

Sample online CMS architecture
Measurements
Processing
Storage
Communication
VIBNODE®
VIB 7.150-SET-SO
Analog inputs 1 to 12, configurable as standard per jumper to
also ICP (voltage) sensors, +/- 5V or +/- 20 mA.
Digital inputs:
Analog inputs:
Power supply.:
A_1-BA11101 analog input 1
A_6-Bn11106 analog input 6
A_12-Bn11112 analog input 12
N_1-BN11201 Puls/RPM 1
N_2-BN11202 Puls/RPM 2
24V230VAC / 24V DC 24 V-Switch24VSW+
-BD11205 System OK (FET)SYS-FET
-BD11206 Digital output 1OUT_1
Digital outputs:
Ethernet TCP/IP
RJ45, 10MBit/s
Ethernet
Communication:
Interface relay
Power supply
VIB 5.964-0,8
(optional)
Phone connection analog /
GPRS (optional)
RS 232/1
Fieldbus Profibus-DP
(optional) or other ‘protocol
converter’. Other
converters available upon
request.
RS 232/2
Fieldbus-Module
VIB 5.920-PDP
(optional)
Modem
VIB 5.93x-xxx
(optional)
Cabinet IP65 (Larger cabinet dimensions: approximately 300 x 200 x 130 mm)
IN_1-BD11203 Digital input 1
IN_2-BD11204 Digital input 2
Interface relay
Interface relay
Interface relay

OMNITREND® software platform
Representation of machines,
including measurement
points and measurement
tasks, with individual
settings for each point and
tasks
Visual detection of alarms
highlighted with colored flags :
red (alarm), yellow (warning), etc.
Problem analysis, diagnostic
and identification of the root
cause of problems
Setting of narrow-band
alarms (etc) for accurate
fault monitoring and
diagnostic trends allow
predictive view of health of
machines.
OMNITREND® is used also
to configure all tasks etc. in
the online CMS.

3-D Spectra, waterfall diagram

CMS in action in water plant in Germany
VIBNODE® CMS interior (example)
VIBNODE® Online CMS (large cabinet)
Vibration sensors*
*Location and number of sensor varies depending on machine characteristics

SENSOR INFORMATION
A SHORT OVERVIEW*
*) for additional information please refer to existing documentation such as but not limited to brochures, catalogs, manual, technical notes, etc.

TANDEM-Piezo®, the 2-in-1 sensor
Coverage of standard
vibration frequencies
(range e.g. 1Hz – 20kHz))
1) VIBRATION 2)
BEARING
CONDITION
PUMP
CAVITATION
Coverage of higher
frequencies
(sensor resonance @ 36kHz)
ATEX
option

Options for mounting sensors on
machines
Option 1:
Type VIB 6.1xx sensor with
90° base screwed directly
onto machine
Option 2:
90° base screwed into
adapter which is bonded
onto machine
Option 3:
flat base bonded directly
onto machine

Screw-in sensor
Bonded adapter
Bonded sensor Screw-in sensor

Examples of bonded adapters
ICP type
CLD type CLD type

Typical mounting of CLD vibration sensor
Clamp VIB 6.720
Clamp VIB 6.721
Protective cap VIB 6.700
Sensor VIB 6.122 type
Dust cap sleeve VIB 6.722
The VIB 6.122 accelerometer
sensors can be
1) screwed into drilled/tapped
holes – or – 2) screwed on
adapters as this one; adapter only
is bonded to the machine.
Or, 3) additionally special
“bonded sensors “(mounted with
special glue/Loctite) can be
mounted without drilling the
machine e.g. VIB 6.102R sensors.
Different cable options for EX
or non-EX areas

Accelerometer sensor VIB 6.122R / VIB 6.122 DEX* CLD type
Dimensions
Mounting instructions
*Instead of threaded sensors, bonded sensors available on request!

When do I need online
CMS?

Offline measurements cannot detect
intermittent problems
But I already measured
this machine and did not
find any irregularities...!!!*
*Hmm... How to find problems that do no show up while performing offline measurements?

100% of all machines in
the plant
A
Why online? “ABC Analysis”
Do I need to monitor all machines with online CMS?
 Less relevant machines, low or none impact on production  C
 Critical machines (also auxiliary critical machines!)  B
 Very critical machines for production  A
B C

Online
protection
systems
Routine inspections
Why online? The right CM toolCosts–lossofproduction,etc
(*time till failure / time between measurements)
$$$
$$
$
Deterioration Sudden failures
Measurements
with portable data
collectors
Online Condition
Monitoring
months weeks days hours minutes secondsyears t*

time
Deterioration
Machine failure types
standard
Fast deterioration
Very fast deterioration
Warning limit
Alarm limit
Interval time between measurements
Online / “real time” (protection)
Online / multiplexed CM
Offline / CM intermittent
portable data collector
Failures with very fast deterioration: internal bearing race and rolling elements in bearings, broken shafts, broken fan blades, etc
Failures with fast deterioration: external bearing race, wear in gear tooth, etc
Failures with standard deterioration: general wear in machines, unbalance and misalignment, etc.
Machine failure
Why online? Evolution of failures…

Justification for online CMS
- Some relevant topics are
► Increasing trend towards online CMS
► Machine is critical to the production process
► Detection of rapid evolution of failure
► Operating with a known problem
► Operating beyond original design
► Savings increase profit of plant
► Quality Control
► Reducing overtime costs
► Reducing equipment downtime
► Any many more...

 Online CMS are used in every industry having
rotating machinery working in its process. Especially
the so-called critical machinery that has a strong
impact on productivity is monitored with online CMS.
 Online condition monitoring is performed in various
industries, such as but not limited to: cement plants,
mining plants, power generation companies, paper
mills, marine industry, wind turbine farms, oil & gas
sector, etc.
Market trends clearly show an increase of machinery health monitoring using
online CMS.
Worldwide more than 1.5 billion USD are spent for Condition Monitoring
solutions.
Approximately 50% of investments – roughly 750 millions USD – represent
acquisition of vibration monitoring equipment
Around 75% of that portion being online CMS, the remaining 25% represent
portable instruments such as vibration data collectors.
Worldwide online CM vs. portable CM
-Increasing trend towards online CMS

- Machine is critical to the production process:
► Comparing the potential cost of a damaged machine in terms of lost downtime is an
excellent way to justify investments and compute the time frame for ROI.
► For example, a process pump is not expensive; however, its impact on production time
may be around $140,000 per hour for a plant.
► Therefore, the value of only one unplanned failure could justify a $60,000 online system.
► ROI would be achieved by early detection of problems and avoiding only one unplanned
failure.
 Pumps in an underground mine in Brazil:
A series of pumps with SIEMENS motors are
monitored with the VIBRONET® Signalmaster
online CMS.
Pumping water out of the mine protects all
machines, ensures an uninterrupted production
and is a must for personnel safety.
Even more critical are those pumps directly
involved in the mining process – such as pumps
working with the SAG mills.

Multiplexor VIB 8.306
cable VIB 90005
VIB 6.122R sensors
Original photo of pump in Brazilian mine
Subsurface mining – pump monitoring

time
Deterioration
Vibration: Warning limit
Vibration: Alarm limit
Fast failure evolution
= Measurements with route-based
portable data collector
Slow failure evolution
No detection! Detection possible.
- Detection of rapid evolution of failure:
► Route-based data collection is sufficient for most machinery, with the average collection
interval of 4 weeks. But, there are some machines where failure evolution can happen
quickly, even within hours. A faulty inner race for example propagates very quickly. Early
detection is critical and 4-week collection intervals are simply not enough to capture the
fault as seen on the graphic.

- Operating beyond original design:
► most processes are running near capacity
► The trend in industry is to increase output capacity by 10-15%.
► This often puts machines just beyond design specifications
► Sometimes, this means running into a resonance speed
► Condition monitoring identifies how close the unit is to ultimate performance

- Operating with a known problem:
► After detection of a problem or fault, the next question is, "How much time do we have
before we must shut down" or "Can we make it to the next scheduled intervention?"
► Trending the fault condition reveals information that is important to estimate the time
before failure.
► Vibration value trends clearly detect variations in the health of the machines as shown in
the graphic.
► Additional analysis reveals the root cause of the problem that can be eradicated.
time
Deterioration
Vibration: Warning limit
Vibration: Alarm limit
Machine can easily run
till next scheduled stop
Machine has an increased
failure probability – stop
immediately
MACHINE 2
MACHINE 1
Scheduled maintenance stop

- Quality control:
► Another feature of online monitoring is fast feedback on product quality.
► For example, on a roll process, chatter between rolls and nips can cause
variations in the thickness of the rolled product, producing a "bar" pattern on
the sheet. This fault is commonly known as barring.
A propagating failure equals to decreased
quality of final product, not only in the paper
industry.

- External specialists can help
► In case required external specialist can access the online CMS
► Integrated email server send out notifications in case of alarms
► External specialist can be a “second opinion”
► Plant can outsource the entire monitoring of critical machines
► External specialist can help if plant just lost its specialist until a new one is recruited
VRM KILN MILLS
ONLINE CMS
MACHINES
ANALYSIS
MAINTENANCE OPERATORS
ETHERNET
EXTERNAL
SPECIALIST
SCADA HMI

Do plants go for CM?

Case studies

Garadagh Cement (Holcim) - Azerbaijan
- Accurate online monitoring*
► The online CMS at Garadagh Cement in
Azerbaijan performs measurements and
saves them on the OMNITREND® software
platform that displays graphics in form of
measurement value trends, spectra, time
wave form plots, etc.
► Trends of measured data from monitored
compressors and proper alarm settings
enable very quick detection of problematic
machines.
► On the right top figure the trend of values
recorded with a vibration sensor (picture
bottom right) clearly shows an increase of the
vibration on an air compressor.
► Once an anomaly is detected on a machine,
additional measurement data such as spectra
is analyzed to find the root cause of the
problem.
► After analysis, a diagnosis is done to define
the root cause problem that is then
eradicated.
Online CMS data visualization (trend values)
Vibration sensor mounted on air compressor
*Source: ZKG Magazine, Holcim, PRÜFTECHNIK

Garadagh Cement (Holcim) - Azerbaijan
► Garadagh Cement decided to use the
VIBRONET® Signalmaster online CMS from
PRÜFTECHNIK as a modern tool to securely
increase machine reliability to help secure
production availability, but also to reduce the
maintenance costs of monitored machines.
► In this case, the maintenance costs for 9
compressors and for a period of only five
months totaled 51,803 EUR.
► The objective was to reduce these costs to 14
500 € – a reduction of a factor 3.6! But the real
sum required for maintenance costs then only
reached 2,134 EUR – an overall reduction of
maintenance costs by a factor of 24.3!
► In other words: maintenance costs have been
reduced to only 4 %, a great result.
Product uptime ensured with online CMS*
Maintenance cost savings
Reduced costs for unplanned stops
*Source: ZKG Magazine, Holcim, PRÜFTECHNIK

Oil & Gas applications

Monitored machinery
- Drilling Machine
(Top drive / DDM (Derrick Drilling Machine))
► Variable RPM
► Low speeds
► Variable loads
► Many transient events can
occur
Without Drilling Machine
 NO OIL !

Monitored machinery
► Measurement locations

Case Study Drilling Machine
Case study I
► Instrument used: PRUFTECHNIK – Germany
► Software: OMNITREND®
► Machine measured: DDM 650 AC (Derrick Drilling
Machine)
► Measured on site, East Azeri, Azerbaijan  BP.
► Operating conditions:
● Zero load / Main shaft speed = 269 /
Intermediate shaft speed = 1093
/ Motor speed = 2300RPM.

- Case study I
► Vibration analysis
● BPFI visible in acceleration time -waveform.
380 400 420 440 460 480 500 520 540 560 580 600 620 640 660 680 700 720 740 760 780 800 820 840 860 880 900 920 940
t [ms]
-1,0
-0,8
-0,6
-0,4
-0,2
0,0
0,2
0,4
0,6
0,8
a [g] SWIVEL 9 SWIVEL RADIAL 1012 VXP Gear time > 600 - 1s 04.11.2006 06:50:38
M
(436,25)BPFI
BPFIHarmonic#2
0,014
a op [g] SWIVEL 9 SWIVEL RADIAL 1011 env_400hz-4000L 04.11.2006 06:50:38
FI
#2
#3
#4
#5
#6

380 400 420 440 460 480 500 520 540 560 580 600 620 640 660 680 700 720 740 760 780 800 820 840 860 880 900 920 940
t [ms]
-1,0
-0,8
-0,6
-0,4
-0,2
0,0
0,2
0,4
0,6
0,8
M
(436,25
BPFIHarmo
0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400
f [Hz]
0,000
0,001
0,002
0,003
0,004
0,005
0,006
0,007
0,008
0,009
0,010
0,011
0,012
0,013
0,014
a op [g] SWIVEL 9 SWIVEL RADIAL 1011 env_400hz-4000L 04.11.2006 06:50:38
(64,11)BPFI
BPFIHarmonic#2
BPFIHarmonic#3
BPFIHarmonic#4
BPFIHarmonic#5
BPFIHarmonic#6
BPFIHarmonic#7
BPFIHarmonic#8
- Case study I
► Vibration analysis
● BPFI in envelope FFT, with family of sidebands.
► This was reported as a level 3 defect. That means you have some time left before you
need to replace the bearing, but needs follow up measurements.
► Decision was made to replace the bearing before they started on next drilling job.

- Case study I
- The Bearing:
► Findings:
● Heavy / sharp marks on the inner
race. Contamination and water.
► Cause:
● Most likely vibration damages
during heavy “yaring”.
● Problems with seals
(water in the oil)

- Vibration Analysis of Drilling Machine
2nd Example:
► Instrument used: PRUFTECHNIK – Germany
► Machine measured: DDM 750 AC 1M (Derrick Drilling Machine)
► Measured in Workshop – Factory Acceptance Test (MAERSK JACKUP 3)
● Zero load / Main shaft speed = 216 / Motor speed = 1762RPM.

- DDM 750 AC 1M
► Needed:
■ High resolution time signals
■ High resolution spectrum
measurements
□ More than 100,000 lines
of resolution possible
also with VIBXPERT®

- Vibration Analysis:
0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 950 1000
t [ms]
-2,5
-2,0
-1,5
-1,0
-0,5
0,0
0,5
1,0
1,5
2,0
2,5
a [g] DDM 750 AC 1M, (FAT) 3 RIGHT PINION UPPER 1012 VXP Gear time > 600 - 1s 20.12.2007 13:13:18
M
(203,48)FTF
FTFHarmonic#2
FTFHarmonic#3
FTFHarmonic#4
FTFHarmonic#5
FTFHarmonic#6
FTFHarmonic#7
FTFHarmonic#8
FTFHarmonic#9
0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 950 1000 1050 1100 1150 1200 1250 1300 1350 1400
f [Hz]
0,000
0,005
0,010
0,015
0,020
0,025
0,030
0,035
0,040
0,045
0,050
0,055
0,060
0,065
0,070
0,075
0,080
0,085
a op [g] DDM 750 AC 1M, (FAT) 3 RIGHT PINION UPPER 1008 env_3200hz-6400L 20.12.2007 13:13:25
(13,45)FTF
FTFHarmonic#2
FTFHarmonic#3
FTFHarmonic#4
FTFHarmonic#5
► Vibration analysis of
the machine during
FAT, detected
impacts from the
cage (train) in the
pinion upper
bearing.
► Visible both in
acceleration time
waveform and
envelope FFT.
► Bearing was
replaced and new
measurements
performed.
► See next slide.

► New measurements after bearing was changed:
0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 950 1000
t [ms]
-2,5
-2,0
-1,5
-1,0
-0,5
0,0
0,5
1,0
1,5
2,0
2,5
a [g] DDM 750 AC 1M, (FAT) 3 RIGHT PINION UPPER 1012 VXP Gear time > 600 - 1s 10.01.2008 12:57:25
M
(203,48)FTF
FTFHarmonic#2
FTFHarmonic#3
FTFHarmonic#4
FTFHarmonic#5
FTFHarmonic#6
FTFHarmonic#7
FTFHarmonic#8
FTFHarmonic#9
0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 950 1000 1050 1100 1150 1200 1250 1300 1350 1400
f [Hz]
0,000
0,005
0,010
0,015
0,020
0,025
0,030
0,035
0,040
0,045
0,050
0,055
0,060
0,065
0,070
0,075
0,080
0,085
a op [g] DDM 750 AC 1M, (FAT) 3 RIGHT PINION UPPER 1008 env_3200hz-6400L 10.01.2008 12:57:25
M
20.12.2007
10.01.2008
Blue color is the last
measurement.
The impacts from the
bearing cage
disappeared!

- The bearing:
► Cage: Slight wear on the cage and pockets,
corrosion and paint.
► Inner race: grayish wear pattern and
indentations
► Conclusion:
■ This is not a bearing you would want in
a new machine.

Sheaves (fast line / Line sheaves / Chrown block)
Monitored machinery

 Fast line sheave lifts up the drilling
machine
Fast line sheave
Drilling machine
Vibration analysis of fast line sheave

- Instrument used:
► Online Condition Monitoring System VIBRONET® Signalmaster
► PRUFTECHNIK – Germany
- Software:
► OMNITREND®, PRUFTECHNIK
- Machine measured:
► Fast line Sheave
► Measured on site, Eirik Raude, OceanRig.
- Operating conditions:
► The wire that goes thru this sheave is lifting and lowering the topdrive.
► Speed is never constant, and load depends on what's hooked up in the topdrive (can be
difficult for vibration analysis)

20.11.2008 21.11.2008 22.11.2008 23.11.2008 24.11.2008 25.11.2008 26.11.2008 27.11.2008 28.11.2008 29.11.2008 30.11.2008 01.12.2008
date
0,05
0,10
0,15
0,20
0,25
0,30
0,35
0,40
0,45
0,50
0,55
0,60
0,65
0,70
0,75
0,80
0,85
0,90
0,95
1,00
a [g] Fast line Inner 107 Ov erall ac c el. >120
0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800
f [Hz]
0,0000
0,0002
0,0004
0,0006
0,0008
0,0010
0,0012
0,0014
0,0016
0,0018
0,0020
0,0022
a op [g] Fast line Inner 1005 Bearing Env elope 22.11.2008 05:52:38
2
3
4
5
6
7
8
9
10
11
12
D
M
(49,72)BPFO
► Bearing trend/noise is slightly developing
► BPFO Pattern in envelope, but we can not be sure,
because exact speed is not known.

► Decision was made to replace the bearing during a rig move.
A similar case in the past caused 7 days downtime.
The bearing:
► Water / contamination.
► Indentations, probably
from mounting.
• Conclusion:
► Water in the lubrication caused
the raised noise. The
indentations on the outer race,
caused the BPFO in envelope.

► This is what happened to a
sheave bearing which did not
have vibration monitoring
equipment installed.
-
- Unnecessary
costs $$$
Another damaged sheave bearing:

Mud Pump
► Vibration Analysis of Mud Pump Motor:
► Main bearing is monitored online
■ Varial conditions (load, speed)

OMNITREND®
Software Platform
LAN/Ethernet
VIBRONET®
Signalmaster
VIB 3.550 Limiting devices
3 ‘string lines‘
with triaxial cable
VIB 90080
General ATEX architecture
Coaxial cable
VIB 90006 (blue)
Mud Pumps – monitored online
MUX 3
machine
7+8+9
MUX 2
machine
4+5+6
MUX 1
machine
1+2+3
Multiplexor VIB 8.306 EX
MUX 6
machine
16+17+18

Vibration Analysis of Mud Pump Motor:
► Instrument used: online CMS VIBRONET®
Signalmaster / PRUFTECHNIK – Germany
► Machine measured: Mud Pump and Motors
► Measured on site => Eirik Raude, Ocean Rig.
● Motor speed = 619 RPM / 89 Strokes
(approx. 50% load)
The drilling rig "Eirik Raude" (Ocean Rig) seen from
the top of the drilling tower.
Mud Pump
The drilling rig "Eirik Raude" (from Ocean Rig)

27.11.2007 25.12.2007 22.01.2008 19.02.2008 18.03.2008 15.04.2008 13.05.2008 10.06.2008 08.07.2008 05.08.2008 02.09.2008 30.09.2008 28.10.2008 25.11.2008
date
0,0
0,5
1,0
1,5
2,0
2,5
3,0
3,5
4,0
4,5
5,0
5,5
a [g] Motor A NDE 107 Ov erall ac c el. >120
0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800
f [Hz]
0,00
0,05
0,10
0,15
0,20
0,25
0,30
0,35
0,40
0,45
0,50
0,55
0,60
0,65
0,70
0,75
0,80
0,85
0,90
a op [g] Motor A NDE 1005 Bearing Env elope 13.12.2007 00:00:40
2
3
4 5
6
7
8
9
D
M
(37,02)BPFO
Mud Pump
► High bearing noise detected from day one,
after system was installed.
Vibration analysis of a mud pump motor:
► noise came down to normal level after
bearing was replaced.
► Detected BPFO in envelope spectra.

Mud Pump
► Current is passing through the
bearing during operation.
The Bearing
► The system has detected the same
problem in 5 mud pump motors out
of six (!).
► Same problem also detected on draw
work motors.
► Next time we detect bearing defects
in these motors they will install
bearings with ceramic balls, to solve
the problem.

Draw works, Winches

Draw works, Winches
► Online Monitoring
■ Variable conditions (load, speed)

Floating Production Units
 A floating production unit does not have anchors.
 To stay in position it uses thrusters.
 For this type of drilling rigs the thrusters has to
be count as “critical machinery“
Thrusters are
keeping the rig in
the right position
on the water
surface
Thruster

Thruster Monitoring
Thrusters are keeping the
floating rig in the right position
on the water surface and count to
the critical machinery
Thrusters can be
monitored
continuously with
online CMS

Thrusters – examples of failures

Oil Rig Jack Bates –Top Drive
monitoring since 2007 (Australia)
VIBRONET® Signalmaster online CMS

Oil Rig Jack Bates –Top Drive
monitoring since 2007 (Australia)
SENSOR 1:
Motor NDE Horizontal
Bearing# 6326 MC4
Variable Speed: 200-1200rpm
SENSOR 2:
Motor DE Horizontal
Bearing# 5588 30C
SENSOR 3:
Gearbox Input Axial (High Speed)
Pinion 24 teeth (mounted directly to Motor Shaft)
Low Speed Spur Gear 40 teeth/23 teeth
High Speed Spur Gear 34 teeth/32 teeth
Spur Gear Lower Bearing = McGill BR-1457 Cylindrical Roller
Bull Gear 114 Teeth
MULTIPLEXER JUNCTION
BOX INSTALLATION
VIBRONET® Signalmaster online CMS

Reference letter: Ocean Rig

References

Maersk (Denmark)
- Online CMS on FPSO
► Online condition monitoring
in EX environment on a FPSO
vessel using VIBRONET®
Signalmaster online CMS
► Additional sensors will follow
in the near future
► Easily expandable by just
adding:
1. Additional multiplexers
2. Cables
3. Sensors
● …. without the need of
buying a new “online
system”

Oil & Gas: tank farms
 In this example
three (3) VIBRONET®
Signalmaster online
CMS units cover a
high number of
decentralized
machines.
 The intelligent
string line concept
together with the
field multiplexer
make it an ideal
system for this kind
of ATEX application.
 Several plants can
be monitored and
information can be
stored on a central
database for further
analysis and
diagnosis.




PLC
Example: power signal, RPM, temperature,
etc
Example: alarm signals, CMS status, etc.
Oil & Gas: refinery pumps
VIB 8.306 EX
field
multiplexors
Triaxial cable VIB 90080
for string lines
LAN/Ethernet
Distances between last sensor on a string line
and the CMS can be 800m. If the online CMS is
located in the middle of a plant it could cover
a total distance of1600m
string line 3
VIBRONET®
Signalmaster
(one per string line)
string line 1 string line 2

Oil & Gas: fuel pumping station
LAN / Ethernet
Internet access
OMNITREND®
software
VIB 6.122DEX vibration sensors
VIB 6.610EX temperature sensors
Triaxial cable
VIB 90080
Online CMS:
VIBRONET® Signalmaster
VIB 8.306 EX field
multiplexors
(one per string line)
string line 1

MINING applications

Triaxial cable
VIB 90080
Mining: underground mine pumps
VIB 8.306 field
multiplexer
Original photo of pump in Brazilian mine
Control Room
External expert
Internet
PLC
Signals: Motor
current, RPM, etc.
Signals: alarm status,
CMS status, etc.
VIBRONET®
Signalmaster string 3
string 2
string 1

GUI – BWE drive 1
YE 1202
M
YE 1102
YE 1101
GE 1207
YE 1203
YE 1204
YE 1106
YE 1201
SE 1007
YE 1105
EE 1114
140° 0°
85°
SE 1113
180° 180°
280°
45°
135°
145°
M1
left
right
Conical gearbox
side
Planetary
gearbox side
Vibration
Gear mesh
Bearings
Machine data
Parameters
operation
Online trends
Alarm System status Trend Diagnostic Back Start
VIBRONET - Online Condition Monitoring
Gearbox – drive 1
Connection status
YE - Acceleration
GE - Displacement
SE - RPM
EE - Load

GUI machine level – BWE 546 – SRs 2000
-
LubCos_Ölzustandsüberwachung
A
P
-BA_Beschleunigungssensor
-BT_Temperatursensor
-BN_Drehzahlsensor
-
BM_DMS_Momentenmessung
-BP_Drucksensor
M
Momentanwerte
Maschinendaten
BA1101
BA1102
BA1103
BA1104
BT1201
BT1202
BT1203
BT1204
BM1105
Schwingung Kegelritzelwelle oben radial
Schwingung Kegelritzelwelle oben axial
Schwingung Kegelritzelwelle unten radial
Schwingung Kegelritzelwelle unten axial
Temperatur Kegelritzelwelle oben Lager 1
Temperatur Kegelritzelwelle unten Lager 2
Temperatur Kegelritzelwelle unten Lager 1
BT1205 Außentemperatur
Abtriebsmoment
BP1106
BP1107
BT1206
Öldruck vor Filter
Öldruck nach Filter
Ölsumpftemperatur
BN1301 Eingangsdrehzahl
LubCos Ölzustandsüberwachung
Temperatur Kegelritzelwelle oben Lager 2
Warnung Signalanalyse Zurück StartseiteAlarm System
Normal
Warning
Alarm
Open Circuit
Disabled
BE1303 Kupplungsschlupf (Temperaturüberschreit.)
VIBROWEB XP - Online Condition Monitoring
Schaufelradbagger 546 – SRs 2000 TAKRAF GmbH
links
rechts
Kegelradseite
Planetenseite
Antrieb 1
(oben)
M
Belchatow / Polen
Antrieb 2
(unten)APP
Ölfilter

Belt conveyor system – typical defect
types
Unbalance Misalignment Wear
Wear Gear mesh
problem
Broken Teeth
Unbalance
Cracks Deformation
WearCracks
Bearing Defect Wear
Electrical
Unbalance Electrical
Motor
Gear Box
Drive Belt
Pulley

PRUFTECHNIK Middle East FZE (Dubai) – Robert Schmaus 111
References: Brazil - Jaguar Mining Inc
* With variable conditions: load and RPM
BALL MILLS:
 24 hours a day
 7 sensors per mill
 2,5MW Siemens motor
Vibration sensor
PRUFTECHNIK VIBROWEB XP
27/04/2013

References: Brazil - Jaguar Mining Inc
Problem:
 The shaft was breaking often, causing low machine availability and high maintenance costs
 Maintenance was performed in monthly basis (preventive), but the faults didn’t follow a pattern
 No information regarding the vibration of the machine
After the implementation of an Online Monitoring System:
 Maintenance labours are performed based on the machines vibrations
 The shaft hasn’t broken since, hence raising machine availability
 Positive feedback from the customer
Monitoring with
Online VIEW
27/04/2013

Some mining references
Mine Device product
Mafube VIBXPERT® II Coal
Phola VIBXPERT® II Coal
Kriel Vibnode Coal
Greenside VIBXPERT® II,
VIBXPERT® EX
Coal
Sishen VIBSCANNER® Iron
AngloAmerican Coal Mines
South Africa
 Drummond Company, Inc.
Colombia / VIBROTIP® + VIBCODE®
 Evolution Mining
Australia
Mine Device Application
Cracow 1 VIBROWEB®, 2
WEARSCANNER®
Ball mills
Pajingo 1 VIBROWEB®, 2
WEARSCANNER®
Ball mills
27/04/2013

Some mining references
 Xstrata
Australia
Mine Device Application
Abbot Point 2 VIBNODE® Stack reclaimer
Mount Isa 1 VIBROWEB®,
1 VIBXPERT® II
Underground jaw crusher
George Fisher 1 VIBROWEB®,
1 VIBXPERT® II
Vertical cone crushers
George Fisher 1 VIBROWEB®,
1 VIBXPERT® II
Vent fans
Zinc VIBNODE® Mobile crane analysis on
crab wheels
Copper 2 VIBROWEB®XP, 1 VIBNODE(R),
1 VIBXPERT®
Anode crane, fans
Oaky No. 01 12 VIBROWEB®, 2 VIBXPERT®II Conveyor belts
Okay North 5 VIBGUARD®, 2 VIBROWEB®XP Longwall monitoring
 BHP
Australia / 3 VIBROWEB®XP + 1 VIBXPERT® / Fans
27/04/2013

Mining trucks under control
PRUFTECHNIK Middle East FZE (Dubai) - Robert Schmaus 115
Stand Test with VIBXPERT®II Multiplexer support
 Measure during routine visits to the shop
 Motor is set to run always at the same constant
speed
 MUX and sensors can be installed permanently
on truck
13/03/2013

PRUFTECHNIK Middle East FZE (Dubai) - Robert Schmaus 116
SAG Mill Feed Conveyor: electric motor + FALK gearbox
Example of Conveyor Gearbox monitoring
LAN / Ethernet
Motor:
Power: 300kW / 402 HP
Speed: 1195 RPM
Gearbox:
Output shaft: 66 RPM
Online CMS
13/03/2013

References: special mining equipment*
 VATTENFALL Europe Mining – Welzow Süd Mine
Germany / online monitoring of Bucket Wheel
Excavators SRs 6300 & conveyor systems, since
1997
 Maritza Mine
Bulgaria / online monitoring of Bucket Wheel
Excavator SRs 4000+VR & conveyor systems
 The fourth largest bucket wheel excavator in the
world!
 Elektrownia Bełchatów Power Plant
Poland / online monitoring of Bucket Wheel
Excavators SRs 2000 & conveyor systems
 QNI – Quensland Nickel
Australia / port crane system online monitoring at
Yabulu refibery
* With variable conditions: load and RPM27/04/2013

RAW MILL 1 [8]
KILN
FAN
[6] KILN DRIVE
[8]
COOLING FAN
BOOSTER [5] KILN FAN [5]
COAL MILL [9]
CEMENT MILL 2 [7]
CEMENT MILL 3 [7]
DRY FAN [6]
SEPARATOR [7]
RAW MILL SEPARATOR 1 [8]
RAW MILL SEPARATOR 2 [8]
SEPARATOR [7]
CMS
CLINKER CRUSHER
POLYTRACK [5]
Cement: plant monitoring
RAW MILL 2 [8]
SEPARATOR [7]

1 2 3
4 5
6
7
8 9
10 11
12
Vertical Roller Mills (VRM) –
measurement locations & parameters (examples)
Typical defects
 Drive motor: imbalance, misalignment,
wear
electrical problems, bearing condition
 Gear box: gear mesh of pinion and
planetary stage; bearing condition at
drive shaft; oil supply
MILLINGSEPARATION
Measurement location examples
1-3 = Motor, bearing condition + vibration RMS
4-11 = Gearbox, input, bevel gear, planetary
stage
12 = Torque measurement
 Additionally: motor current
Parameters for gearbox:
- Spectra, envelope spectra and vibration
acceleration
- Perform band analysis
Grinding table: Temperature of friction bearing
Optional: mill feed, power, loads, etc
Oil supply:
- Temperature
- Pressure
FLSATOXVerticalRollerMill
Accelerometer sensors
Proximity sensors
Torque measurements

Gearbox input shaft
Example of vibration sensor mounted
on gearbox input shaft

Online visualization with OMNITREND® Online
View

OPTION: Torque measurement

Installation of Torque measurement point

Reference Letters

Reference letter: Ocean Rig

Oil & Gas: sample reference letter of BP

Summary
- Condition Monitoring is very important for a plant
► Increases plant’s reliability
► Detects machines with problems and helps finding source of failures
► Reduces the number of unnecessary machine stops
► Knowing the failure, repairs can be planned at the most effective time, parts
can be ordered and labor can be organized well in advance
► Ensures machine uptime, decreases loss of production
► Decreases costs for maintenance, costs for stock, reduces energy
consumption of machines... and much more...
► Prevents from safety and environmental issues
► Helps to increase PROFIT of plants
► And...

27/04/2013 128
PRUFTECHNIK Middle East FZE
- www.pruftechnik-me.com
- www.pruftechnik.ae
PRUFTECHNIK Middle East FZE (Dubai) – Robert Schmaus
We are close to you. Because we care.
PRUFTECHNIK Middle East / Dubai – UAE / Phone: +971 4 214 6386

online_failure detection

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Viewers also liked

Viewers also liked (18)

Similar to online_failure detection

Similar to online_failure detection (20)

Recently uploaded

Recently uploaded (20)

online_failure detection