17-Jan-12

What is the Plan?

• 1.5 days of classroom training
- ELCID theory of operation
- Excitation calculations
- ELAN software
- Analysis and interpretation

• 0.5 day of practice

Why So Much Training for One Device?

1
 17-Jan-12

Questions?

EL CID
The modern way to test
stator core insulation

You will be asked to spot the

deliberate mistake in this picture later.

History

A method of detection of faults in core interlamination

insulation by electromagnetic means was developed by
the Central Electrical Research Laboratory of the UK
C.E.G.B. (now RWE npower plc).

The method uses only a small fraction of rated excitation

to generate fault currents within the core body which are
sensed by a pick-up coil. This avoids the testing problems
usually found with high excitation, yet still gives an
accurate indication of damaged areas along tooth tips and
walls, as well as possible sub-surface damage.

2
 17-Jan-12

Older El Cid
El Cid, Military Leader

Born: 1043 A.D.

Birthplace: Bivar, Castile, Spain
Died: 10 July 1099
Best known as: Spanish military
and folk hero
Name at birth: Rodrigo Diaz de
Bivar
The technique is called
Electro-magnetic Core Imperfection
Detection (EL CID).
The full rights to the technology were acquired by
ADWEL who have continued extensive development
of the product. EL CID equipment is now
manufactured and supplied by IRIS Power.

Do Not Go to the Wrong Place …

SI versus Imperial
• 1 metre = ?? cm = ?? mm
• 1 inch = ?? cm = ?? mm

3
 17-Jan-12

SI versus Imperial Conversions

• 1 metre = 100 cm = 1000 mm
• 1 inch = 2.54 cm = 25.4 mm
• ¼ inch = 6.35 mm
• 1 foot = 30.48 cm = 304.8 mm
• 1 metre = 39.37 inches

Stator Cores

Stator Cores
• Provide low reluctance paths for working magnetic fluxes
• Support stator winding
• Provide primary heat removal from indirect cooled stator
winding
• Contain air ducts for heat removal and air/gas flow
throughout stator
• Hysteresis losses contribute to temperature rises and
machine in-efficiency
• Cores must be capable of withstanding the operating
forces: mechanical and magnetic

4
 17-Jan-12

Construction of Cores
• Made from magnetic steel laminations
• Assembled on cage frame and axial clamping plates,
including teeth supports
• Inside space blocks used to separate core packets and
form air/gas ducts
• Assembled separately or in outer frames/casings

Stator Laminations
• Made from thin (14-19 thou or 0.355 to 0.483 mm) low
loss silicon steel lamination rings mounted on core bars
and clamped between endplates
• The laminations have slots punched in them to
accommodate the stator winding
• In large turbine generators, 10 to 24 laminations per
layer are used to form 360-degree core
Each layer is staggered to increase mechanical integrity
and reduce shaft voltages

Stator Laminations
• Each lamination is insulated on both sides
• Mill applied or OEM applied insulation
• Organic or inorganic coatings
• Enamels or film coatings
• Temperature rating 300-400°C
• Grain oriented with the rolling direction

5
 17-Jan-12

Core Assembly
• Laminations assembled on key bars in a fabricated
frame
• During stacking process pressed at intervals to assure
even pressure
• Pressure maintained by core clamping plates with
fingers, at each end
• Key bars may be spring mounted to isolate core
vibration from foundation

Photograph source: rcgroups.com website

Core Design

Core Design

6
 17-Jan-12

Why Test Stator Cores?

• Mechanical, electrical and environmental stresses can cause
a breakdown in the inter-laminar insulation on stator core
assemblies
• Insulation breakdown causes
fault currents to be set up as
illustrated. These fault
currents create hot spots
which can cause further
deterioration to the core. If
left unchecked, this can lead
to damage to the stator
winding and the machine as a
whole
• EL CID is a simple, cost effective, preventative maintenance
tool for checking core lamination integrity

Core Damage Example 1

Core Damage Example 2

7
 17-Jan-12

Core Damage Example 3

Core Meltdown

Common Causes of Damage

• Relaxation of core clamping
• Failure of inter-laminar insulation
• Thermal creeping
• Loose debris
• Rotor rub
• Loose coils and wedges

8
 17-Jan-12

Requirement for Core Testing

• Service/Repair
– Assess machine condition
– Determine fault location
– Assess severity of fault
– Monitor and verify repairs
• Manufacturing/Quality Control
– Quality assurance/control
– Baseline results f achine owner
– Acceptance testing

Accepted Test Methods

1) High Power Ring Flux Test - LOOP test
– Core is energised to 80-100% flux level
– Large (3 - 4 MVA) single-phase power supply required
– Safety concerns with high voltage and current
– Expensive thermal sensing equipment
2) ELectromagnetic Core Imperfection Detector – ELCID
– Core is energised to just 4% flux level
– Low power require ts (1 - 3 kVA)
– No safety concern due to high voltage or current

Accepted Test Method – LOOP Test

Power POWER
Cables SOURCE

9
 17-Jan-12

Accepted Test Method – EL CID

Now, which method do
you like?

POWER
SOURCE

Required Excitation Levels

Loop Test
80-100%
(of rated flux density)

EL CID

4%

Advantages of EL CID
• Low excitation power - 4%
• Fast, portable, therefore easy to set up
• Low manpower requirements
• Significant reduction in safety hazards
• Instant interpretation of test results
• No risk of further damage
• Ability to re-test during maintenance cycle

10
 17-Jan-12

Advantages of EL CID

Fault Current

Fault Current

11
 17-Jan-12

Excitation and Induced Voltages

Across Laminations

Induced voltage lags flux (and

excitation current) by 90 degrees

Flux produced by Excitation

Voltage induced across one pair of laminations
Voltage induced across damaged laminations
Voltage induced along complete core length

Measuring Fault Current

with a Chattock Potentiometer
The Chattock potentiometer measures the Magnetic
Potential Difference (m.p.d.) between its ends.

Positioning the Chattock Coil

12
 17-Jan-12

Understanding Fault Current Magnitude

4% 100mA
equates to 5-10°C on
LOOP/HFRT Test

Typical Excitation System

Typical Hydro Generator Excitation System

Typical Hydro-generator
EL CID Excitation

13
 17-Jan-12

Excitation Computation
• The machine operating voltage was induced by normal
(100%) flux, and EL CID test should be conducted at 4%
flux.
• The 4% flux is determined by measuring the voltage
induced on a Single Turn Trace winding around the core
= Vt
• However, the machine winding number of Turns per
phase in series (Tp), which generates the operating
voltage, must be known. It should be noted that the
induced winding voltage is caused by a total turn in the
machine winding, ie down one slot/bar and back up
another

Excitation Computation
Simplified formula for 2 bars in one slot:

slots x turns
Tp =
phases x parallels

Stator Winding Design 1

• 36 slots, two parallels per phase, 2 bars/slot, 3 phases
• Turns per Phase (Tp) = 36 x 1 / (3 x 2) = 6

14
 17-Jan-12

Stator Winding Design 2

• 36 slots,
single parallel
2 bars/slot,
3 phases

• Tp = 36 x 1 / (3 x 1)
= 12

Bars or Coils
• A winding may be comprised of series of single
conductor bars, connected at each end to form a
complete turn, or instead a complete coil with 2 or more
turns embedded inside the insulation.
• It is not always easy to determine the turns in a coil by
inspection; reference to drawings or spare bar is
needed.
2 Bars = 1 turn

2 Turn Coil

Multi-turn Coil

15
 17-Jan-12

Cross-section of Multi-turn Coil

Inside a Multi-turn Coil

• Copper strand and

insulation
• Strand bundle
• Turn insulation
• Coil bundle
• Groundwall insulation

One Reason Calculation Could Be Wrong Is …

Tp - the number of turns

in series per phase is
increased by 3 for 3
turns per coil

16
 17-Jan-12

Excitation Calculation

Voltage Rating (kV) Vp-p

No. of Slots S
No. of Windings (Bars or Coils) / Slot W (Normally = 2 but not always!)
No. of Parallel Circuits / Phase C
No. of Series Turns / Coil T (Always = 1 for a bar)
No. Turns / Phase Tp = S x W x T/(3 x C x 2)

Excitation Calculation
• For applying the correct level of excitation, it is necessary to know
the rated rms voltage along a single length of the core, for the
winding, Vr.
Vp-p
• This is given by: Vr =
2 x K x 1.73 x Tp
Vt = 0.04 Vr
where Vp-p is the rated phase-to-phase voltage in volts
Tp is the number of turns in series per phase (computed
earlier)
K represents the combined spread and short pitch factor. A
value of 0.92 is used for all generators.
The 2 in the denominator allows for the fact that the core voltage is
generated down each side or half of the turn.

Excitation Calculation - Turbine Generator

Consider a 23kV, three phase, two bars/slot, two parallels, 660MW
generator with 42 slots.

42 x 2 x 1
Tp = = 7 series turns per phase
3x2x2

Since Vp-p = 23 kV,

23000
Vr = = 1032 V
2 x 0.92 x 1.73 x 7

Then the standard Vt for xcitation is 1032 V x 0.04 = 41.2 V

Simplified formula: Vt (V) = 12.56 U (kV) / Tp

17
 17-Jan-12

Calculation Example
• What is the Tp value for a 36-slot, two parallel paths, two
bars per slot, single turn (bar) machine?

• Assuming 2 bars per slot,

36 x 1
Tp = = 6 series turns per phase
3x2

Calculation Example

What is required Vt for a 13.8 kV machine with single parallel, 36 slots,

2 bars per slot (to the nearest volt)?
36 x 2 x 1
Tp = = 12 series turns per phase
3x2x1
13800
Since Vp-p = 13.8kV, Vr = = 361 V
2 x 0.92 x 1.73 x 12

Then the single-turn voltage for 4% excitation,

Vt = 361 V x 0.04 = 14.45, or
Vt = 12.56 x 13.8/12 = 14.44 V using the simplified formula

Calculation Example - Hydro Generator

Consider a 13.8kV, three phase, four parallels 100MW hydro generator
with 360 slots and 2 turns/coil, and 2 coils per slot.
360 x 2 x 2
Tp = = 60 series turns per phase
3x4x2
Since Vp-p = 13.8 kV,
13800
Vr = = 72 V
2 x 0.92 x 1.73 x 60

Then the single-turn voltage for 4% excitation,

Vt = 72 V x 0.04 = 2.89 V

18
 17-Jan-12

Calculation Based on Core Size

• And other machines where the winding is not known.
This assumes that a mean flux of ~0.9 T rms is used.
• f is the frequency
l is the core length in metres
d is the depth of the core distance in metres from slot
base
• Vt ≈ 0.226 f l d (metric)

• If l and d are measured in inches:

• Vt ≈ 0.000146 f l d

Calculation Based on Core Size

• Calculate the single-turn voltage for the following motor:
– f = 60Hz
– l = 40 in (1.016 m)
– d = 4 in (0.1016 m)
• Vt ≈ 0.226 x 60 x 1 x 0.1 V = 1.399 V (metric)

• If l and d are measured in inches:

Vt ≈ 0.000146 x 60 x 40 x 4 V
= 1.4 V

Sanity Check for Voltage

• When installing the excitation system, if the excitation cable is taken
around the core more than once (for example motors and hydros),
then remember to divide the measured voltage by the number of
cable turns around the core.
• After computing the single turn voltage VT, divide it by the core
length in metres and express the value as volts per metre.
• For medium to large 2-pole turbine generators, it should be 4-6 V/m
• For typical hydro generators, it should be around 1-3 V/m
• For small to medium motors, it should be around 0.7-2 V/m
• These values are NOT absolute, but if the answer lies outside these
ranges, double-check the computation before accepting the result.

19
 17-Jan-12

Reality Check for Voltage

• Two methods were used to calculate VT on a 13.8 kV
winding, with 396 slots, 2 m high core (effective 1.65 m)

• Winding based calculation VT = 3.83 V

(Tp = 44)
• Flux based calculation VT = 3.84 V
(Design flux known, effective core height used)
• Flux based calculation VT = 4.34 V
(2 m core height, 0.9 T)
• Flux based calculation VT = 3.54 V
(Effective core height, 0.9 T)

Switched Excitation System

• Designed for rapid installation of excitation windings
• Switched number of 2/3/4/5/6 turns, key locked, (can
obviate need for variac/Powerstat)
• Rated up to 20 A (4 to 6 turns) or 32 A (2 to 3 turns)
• Integrated trace loop winding and current test loop
• Two excitation leads in kit (10 m and 20 m) - can be
extended
• Hydro excitation system provides windings in series
around core
• Rotor-in-place flat cable option
• Plastic connectors prevent core shorting accidents

Switched Excitation Parts

Excitation Cable Switched Control Box Rotor-in-place

20
 17-Jan-12

Switched Excitation System for Turbo

Mains Supply
Trace Voltage
Excitation Cables Measurement

Current Measurement
and Reference
Transducer
Turns Selection
Switch

Switched Excitation System for Turbo

Switched Excitation System for Hydro

Excitation
Cables Excitation Control
Boxes

To other Excitation
windings Supply input

Extension Cables Series Connection Boxes

Boxes

21
 17-Jan-12

Switched Excitation System for Hydro

Reality…

The number
of groups will
depend on
the number of
core joints,
normally one
per sector

Excitation Power Calculation

For a winding with NW turns, the excitation current in an
individual turn IWt is given by:

IWt = A-turns / NW

and the total excitation winding voltage, Vw by:

VW  Vt x NW

The approximate total apparent excitation power

requirement is simply:

VA  VW * Iwt  Vt x A-turns

Excitation Current Calculation

• To achieve 4% of rated voltage requires the order of 2 to 15 A-turns
per metre circumference of core
• The 660 MW generator has an outer core diameter of 3 m and a diameter at the base
of the slots of 2.2 m, giving a mean diameter of 2.6 m. The mean circumference is
8.16 m and the A-turn requirement will thus probably be in the range of 25 - 80 A-turns.

• Od = 3 m, Id = 2.2 m
• 3 + 2.2 = 5.2/2 = 2.6 m
• Md = 2.6 m 2.6m
• Mc = 2.6x3.14 = 8.16 m
3m
• 8.16 x 2 = 16.32 A-turns
• 8.16 x 15 = 122 A-turns
2.2m
• Why is this useful?

22
 17-Jan-12

Current Calculation Example

• For the 660 MW generator being considered,
Vt = 41.2 V and IW < 80 A-turns
• The total power requirement will be < 3.3 kVA
• In this example, requiring VT = 41.2 V. A six-turn winding
powered from a 240 V supply would give almost the
recommended 4% value.
• The current would be < 80 A-turns/6 turns i.e. < 13.3 A
• A 120 V mains supply would require a three-turn winding
but the current would be increased to a maximum of 27 A

To Conclude First Part …

1. Calculate Tp
2. Calculate Vt
3. Calculate expected current
4. Determine number of turns in excitation winding
5. Determine the length of wire required (for hydro)
6. Read ELAN software manual version 5.5;
section 8 STEP-BY-STEP ELCID TEST GUIDE

23
 21/03/2012

Original Analogue EL CID

First Generation

Plotter Distance Encoder and SPU Sensors etc

Digital EL CID – Model 601

Second Generation

• Insulated case allows the unit to be carried into the

energised core with no risk of shorting out laminations.
• PC allows direct view of scans and storage of multiple
scans/slot

1
 21/03/2012

Digital EL CID – Model 601

Second Generation

EL CID Evolution

It is time for the question, what is

wrong here?

Everything Is in the Boxes

2
 21/03/2012

Unmistaken Connections

Signal Processing Unit Front Panel

Off/Measure/Setup
Signal
switch
inputs/outputs
OK button

Rotary control
Mains input
Softkey buttons

Boot Up Screen

Screen after
turn-on

Wait 30
seconds

3
 21/03/2012

Measure Screen

• Basic measure
screen
• Single or dual
signal input
• Change signal
range with
Range softkey
• Frequency
shows
Reference
frequency
• X-distance will
follow PC, but
can be zeroed
for manual use

Setup Screen

• Basic Setup
screen
• Select function
to be set up with
soft keys
• Start selected
function with
right-hand
softkey

X Calibration
• Set Trolley or RIV
to give correct
distance over
actual core
surface
• Pulses from
encoders are
calibrated in
“pulses per metre”
• Follow on-screen
instructions, press
OK button to
complete
• Return to Setup
screen with
SETUP softkey

4
 21/03/2012

Signal Calibration

Signal Calibration

Signal Calibration

• Set Chattock
calibration %
and Phase
calibration to
give +1.0 A
PHASE and 0
mA QUAD
• Follow
instructions on
screen, press
OK button to
complete
• Return to Setup
screen with
SETUP softkey

5
 21/03/2012

Phase Calibration

• Done once per

test, before
first slot
• In the case of
core splits on
hydros, repeat
for every split
• In case of
unavoidable
circulating
currents, do
not do it!

Two Inputs
• Normally use INPUT 1 input only
• Use both for RIV double-ended to allow the entire core
length to be scanned, with one Chattock at front, another
at rear
• Calibrate each Chattock separately
• Selection of Signal +1/+2 or +1/-2 allows you to reverse
INPUT 2 so that the polarity of the two Chattocks is the
same, which is vital for easy interpretation

Input Selection

• Normally use
Input 1, use both
for RIV double-
ended
• Select the Signal
Selection with
sofkeys
• softkey wi ll
allow selection of
either:
+1, +1/+2, +1/-2
• Press the OK
button to confirm
the selection

6
 21/03/2012

Other Controls

• Select Trolley or
RIV for X-axis
(distance) input
• Set bar graph
levels
• Select
Frequency
(Auto/50/60Hz),
use Auto unless
supply frequency
is very unstable

Other Controls

• Manual
adjustment of all
calibration
parameters
possible with the
rotary control
• Must confirm all
changes with OK
or values revert!
• Battery bar
shows
approximate
charge, red
colour if
discharging

System Features
• Chattock Calibration Unit: Allows each test to be carried
out at the same sensitivity
• Distance Transducer: X-axis measured with universal
calibrated encoders mounted on the manual Trolley or
RIV
• External Reference Coil: Clamped on the excitation
cable to free up valuable core space. Phase
adjustments made with the touch of a single button
• Dual Signal Capture Simultaneous capture of PHASE
and QUAD provides valuable information
• RF Noise Immunity: Digital processing allows superior
immunity to noisy environments

7
 21/03/2012

System Features
• Large LCD display with on-screen guidance
• Two Chattock inputs for RIV double-ended testing
(Second Chattock polarity +/-)
• Improved accuracy/resolution from 12 bit to 16 bit
• 200% overload margin
• USB interface for simple connection to PC
• Easy user setup

More System Features

• Faster operation up to full nominal speed

(100 mm/s @ 50 Hz or 120 mm/s @ 60 Hz)
• How many millimetres per minute?
• Common trolley and RIV input - separate calibrations
• On-screen colour bar graphs for QUAD current
• Internal battery providing 4 hours of operation time
• Robust locking connectors
• Rugged insulated case

Test Procedures
• SAFETY FIRST
• Check machine windings (ideally all open, else one end
earthed/grounded, the other open)
• Position excitation winding and set voltage level
• Calibrate chattock and SPU
• Calibrate X-axis
• Set up software settings in ELAN
• If no circulating currents, complete a Phase Calibration
on a defect-free section of core
• Short practice before main tests
• Back-up test data after the test

8
 21/03/2012

Safety in Use
• All instructions in the User Manual should be followed
• Follow the local risk assessment procedures
• Ensure the machine is safely isolated from all sources of
hazard
• Always use at least 2 staff –
1 skilled operator and 1 safety assistant, who may be
from the station

Steps to a Safe Test

1. Ensure that a properly authorised Permit to Work/Test
is in place
2. Ensure that an agreed, written and understood Test
Plan is in place
3. Ensure all Personal Protective Equipment is functional
and used (such as hard hat, safety boots, safety
glasses, ear plugs, knee pads, gloves)
4. Ensure that the machine is fully isolated from all
sources of hazard

Steps to a Safe Test

5. Ensure that work area is safe, including access to and
from the bore, platform barriers, any cleaning
fumes are dissipated
6. Check the isolation/earthing of any machine windings
7. Check that all the test equipment is present and
functional
8. Set up the EL CID in the operator area, with proper
support for equipment (tables etc) and cables routed
to prevent trip hazards
9. Check the excitation winding and trace loop is safely
installed and insulated, and temporary power sources
are insulated and operating within their ratings

9
 21/03/2012

More Steps to a Safe Test

10. Add slot number indicators with white tape if not
already marked.
11. Ensure that no metal objects have been left inside the
bore since these could cause arcing damage to the
core
12. After positive confirmation feedback that the test area
is clear and safe, and all other agreed risks are
protected, commence the test. Take regular breaks or
rotate staff to prevent tiredness causing accid nts.
13. Review the actual execution of the test in a post-test
meeting and learn from any incidents.

Excitation Cable Installation

• Always try to run the excitation cable down the core
middle

Magnetic Trolley

Chattock Adjus tments for width (4) Adjus tments for bore radius (2)

Clamp s crews for Chattock (2) Clamp nuts for s ide guidance rails (4)
Screws for cable clamp plate (2)

10
 21/03/2012

Trolley Remote Control

• A simple interface allows the following functions from the
Trolley Operator:
• Start slot recording
• Stop slot recording
• Stop and restart current slot recording at beginning
• A dedicated distant viewable PC screen (“Afar”) allows
progress and result appraisal by operator
• Speed limit: 0.1 m/s or 0.36 kph or 0.22 mph! at 50 Hz

Speed Limit

≈ 4 in/s

Methods of Scanning
Alternate Single Scan Single Scan Reverse
Slot Direction Slot Direction

1 Forward → 1 Backward ←
2 Forward → 2 Backward ←
3 Backward ← 3 Backward ←

Single Scan Double Scan

1 Forward → 1 Forward →
Backward ←
Forward →
2 Forward → 2 ←
Backward
Forward →
3 Forward → 3 ←
Backward

11
 21/03/2012

Alternate Single Scan – Slot Change

End of Slot n Start of Slot n+1

Method of Scanning

Method of Scanning

12
 21/03/2012

Method of Scanning – Hydro Generator

Method of Scanning Step Iron

• Set timebase mode to

a comfortable speed
(10-25 mm/s ≈ 0.5-1”/s)
• Hand scan obstructed
areas
• Record results in
separate file or use
over-record traces
function

Scanning Step Iron

13
 21/03/2012

Importance of Step Iron Scans

EL CID Testing with RIV

Robotic Inspection Vehicle • Multi-purpose
Available initially to enhance transportation platform for
ELCID testing, now used for EL CID test sensors in
wider applications relevant to particular, and also other
other developments payloads associated with
stator core
inspection/testing
complementary to EL CID
• Primary function to
provide faster EL CID
testing at consistent
scanning speeds
• Less operator fatigue

Main RIV Features

RIV attaches to stator surface
by permanent magnets
• Propelled by self-steering
tracks
• Magnetically self aligning by
sensing stator teeth edges
• Optionally deployed
mechanical guides provide
means on maintaining
horizontal alignment
• Complementary standard
payloads include:
– Wedge Tightness Test
Module (WTD)
– Video Camera

14
 21/03/2012

Generator Cross-
Cross-section

Entrance Limitations

Windows …

15
 21/03/2012

Windows: Folders and Files

• All data is held in files, which are organised into folders

Windows: Software

• ELAN program is
started by single-
click on Start
menu
• Files are opened
by double click on
file

Windows: Data and Backup

• All data is saved in (or other user folder set in Options)
C:\Documents and Settings\username\My Documents\Elan\
• ALWAYS back up your data OFF the PC after test and during
an extended test. If test data/software crashes, Recovery.dec
file is in \Elan\Tempdata\

16
 21/03/2012

ELAN 5 Features
• Integrated test and analysis functions
• Test information saved for both machine and test parameters
• Can null QUAD offset and mask zero/missed data-points
• Single slot Display Mode for detailed inspection with zoom
and cursor, plus simultaneous QUAD and PHASE display
• Integrated test results stored in open XML format, with import
conversion for old formats
• Display and recording of QUAD or PHASE values, with
optional dc (separate for QUAD and PHASE) and zero
removal
• Addition of notes to slot data allows annotation of analysis

ELAN 5 Features
• Remote operation allowing one-man operation
• Normalisation to 4% or any other excitation level
• Ability to select or de-select displayed slots from Exciter
or Turbine, plus choice of best alternative slot results (a,
b or c etc)
• User selection of display and printout trace colours
• Machine and test parameters can be reviewed at any
time
• Ability to copy a previous test set-up for a new test, and
to continue any previous test left unfinished

ELAN 5 Features
• Support for USB and RS232
• Chattocks automatically switched for RIV (Forward or
Backward)
• Extensive excitation calculation assistance
• Ability to merge data from different tests
• Can over-test a trace and add step-iron areas
• User offset of Backward traces with respect to Forward
• Copy Slot List, Peaks and Notes to Clipboard for reports
• False-colour core visualisation

17
 21/03/2012

ELAN Opening Screen Options

Opening Screen – Start Test

• Setup a New unit, Copy a previous test, or Edit current
one

Machine Parameters

18
 21/03/2012

Excitation Calculation
The software supports three computation methods for
Single Turn Voltage:
1. If you have already computed Tp, then simple calculation
of Single Turn Voltage from machine voltage, or …
2. If you do not know Tp, but do know winding details,
turns/coil parallels etc, or …
3. If you do not know winding details, but do know core
geometry and assumed operating flux levels …

Excitation Calculation
1. Calculate
Single Turn
Voltage Vt
from turns
per phase Tp

Excitation Calculation
2. Calculate Vt
from winding
details

19
 21/03/2012

Excitation Calculation
3. Calculate Vt
from core
geometry

Test Parameters

Testing
The test must first be commanded to Start
Nothing is being recorded yet, but the EL CID output Hot key s
is visible on PC, Trolley button is active etc.

Begin slot recording Hot key b

End slot scan recording Hot key e

You can Goto any slot between recording

Hot key g
Previous slot data is over-recorded

You can Re-record any slot (a-z)

Hot key r
Previous slot data is saved

You can Delete any slot you make a mistake Hot key d

At end of test, Finish Hot key f

20
 21/03/2012

Test Screen
• Screen shows current slot ready to or being recorded in
white, with previous 2 slot traces and next one to come

Test Commands

G oTo Slot
ReDo Slot
Delete Slot Tes t
End Slot Recording
Begin Slot Recording
View Tes t Progres s
End Tes t
Begin Tes t

Afar Test Screen

21
 21/03/2012

Test Progress

End of Test
• Back up test file to off-PC memory, such as USB stick
• Review test results to ensure all slots tested, and entire
core covered
• Check if the core step-iron area (if any) has been
adequately covered, maybe carry out a local area test
• From review check if any obvious defects found and
investigate further

22
 16/02/2012

Analysis – File Open

Review Header Data

Is this the right test?

1
 16/02/2012

Slot Test Results

Trace Selection for Analysis/Display

• Select first/last slots
• Test data source
• Choice of units
• Length of core to trace
• Correct any Backward
offsets
• Excitation
normalisation
to 4% or other voltage
• DC removal
• Trace scale/Autoscale
• Trace colours

Single Trace – No DC Removal

2
 16/02/2012

Single Trace – DC Removed

Dual Slot Comparison

Peaks Analysis

3
 16/02/2012

Notes

Data Display – Normal Traces

Core Visualisation

4
 16/02/2012

Printing
• Previous slot
selections plus:
– Select first/last
slots and number
of slots on each
page
– Optional printout of
header page
– Printout colours
selectable separate
to display

Printing

Data Export to Excel

• Right-click
on trace

• Exports all
slots shown
on screen,
up to 20

5
 16/02/2012

Data Inspection in XML File

Trace Insert
• Allows insertion of step-iron traces or merge part tests.

EL CID for Rotor Bar Testing

6
 16/02/2012

EL CID for Rotor Bar Testing

Fault Magnitude

4% 100mA
equates to 5 - 10 C

Interpretation of Data

QUAD signal from

fault within Chattock
span is always
opposite polarity to
PHASE signal

7
 16/02/2012

Typical Fault Indication

Substantial faults always show a PHASE signal increase

EL CID Signal and Thermal Response to Faults

Results from trials at Woodleaf, 1997

Test Result Correlation

15°C

Woodleaf-1
Sept 1997
Slot 51 = 319mA

8
 16/02/2012

How does EL CID Compare

with Ring Flux for Buried Faults?
• EL CID is known to detect buried faults quite well.
However there is concern that buried faults are not so
well detected by the Ring Flux Test, no matter which
name is used to describe it:
– Stator Core High Energy Test
– Rated Flux Test
– High-power Lamination Insulation Test
– High-flux Test
– High Flux Ring Test
– Loop Test
– Ring Flux Test
– Flux Test
• How do they compare in practice?

Applied Resistive Faults Results

• 50mm resistance
wire pressed on
lamination edges
on slot base
• In EL CID test, slot
base shows only
16% attenuation

QUAD Signal
Tooth tip -325 mA
Slot base -270 mA
Difference -16%

Ring Flux Test Comparison

• 30 mm resistance wire fault was pressed against the slot
base surface = 520 mA of EL CID indication at 4% flux
• Thermocouples placed alongside fault, on tooth tip and
core body
• Core excited to 0.8 T rms, typical of Ring Flux
• Temperatures allowed to stabilise for 2 hours
• Fault current on EL CID ranged from 12.9 A to 10 A at
0.8 T

9
 16/02/2012

Ring Flux Results

• Slot base temp
rise +37°C
• Tooth-tip temp 80.0
70.0
rise only +6°C
60.0
Temperature °C

• Thus a buried slot 50.0

base fault only 40.0
shows 16% of 30.0
Slot Base
temperature rise 20.0
Tooth Tip
10.0
at core surface. Bore Surface
0.0
0 20 40 60 80 100 120
Time (mins)

EL CID Signal and Thermal Response to Faults

Correlation of EL CID & HFRT results

700
650
600
EL CID
550 Signal Correlation boundary lines
500 (mA)
450
400
350
300
250
200
150
100
50
0
0 5 10 15 20 25 30 35 40 45 50 55 60
HFRT values (Deg C)

From CIGŔE Questionnaire 2003

Major Damage Example

10
 16/02/2012

Damage Example

Damage Example

Damage Example

11
 16/02/2012

Slot 174

Slot 174 After Repair

Slot 190

12
 16/02/2012

Slot 190 After Repair

Deep Seated Damage

Deep Seated Damage

13
 16/02/2012

Step Iron Scans

• Do not use DC removal

Factors That Modify EL CID Results

1. Applied excitation
2. Erection state of the machine under test
3. Constructional features of the stator core
4. Test environmental magnetic conditions
5. Circulating currents in stator winding
6. Inappropriate measurement range

Modified Signal Values

• Due to the level of magnetic induction (Excitation) being
incorrect
– Calculation mistakes
– Lack of appropriate source
• 2 - 10% of excitation level acceptable
• Re-scale to 4% available in software

14
 16/02/2012

Modified Signal Values

Sensor Too Close to Excitation Winding

Keeping Winding Central from Core

15
 16/02/2012

Keeping Winding Central from Core

Keeping Winding Away from Core

Keeping Winding Away from Core

16
 16/02/2012

Skewing of Trace
• Sloping trace axis
• Usually caused by
excitation sloping off
axis

Skewing of Trace

High Core End Values

• Usually caused by excitation winding too close to end of
core on its return

17
 16/02/2012

Curvature of Trace Axis

• Usually caused by proximity of rotor pole

Off-centre Excitation
Causing Circulating Currents

Environmental Magnetic Conditions

• NOT inter-lamination damage

18
 16/02/2012

Circulating Currents

Circulating Currents in Hydro Generator

PHASE current

300

WINDING
Slot Average with Winding
Splits in Core

INSTALLED
200
Slot Average without Winding

WINDING
Average Quad mpd across slots (mA)

100

REMOVED

QUAD current
0
0 35 70 105 140 175 210 245

-100

No Stator Winding
Stator Winding Installed
-200

-300
Slot Number

Solutions for Circulating Currents

• ONLY relevant if there are windings in the core
• In turbine generator cores – ensure that the excitation is
down the centre
• In hydro generator cores – if smaller, also try to obtain
central excitation. If too large, still push the windings a
little away from the core surface
• If possible, have the windings totally disconnected both
ends, but will normally have to accept connected at one
end (star point) and open other end

19
 16/02/2012

Solutions for Circulating Currents

• Any remaining circulating currents will appear as a
standing signal on top of the QUAD (and PHASE) signal
• In this case, use of DC removal will reset the trace to
approximately 0 mA and allow any local defect to be
identified and assessed
• If you suspect when first doing a quick check in the core
that there are circulating currents (substantially different
QUAD and PHASE readings on different slots), then do
not perform a Phase Calibration in the core, just rely on
the calibration completed in the Calibration Unit

Inappropriate Range

4A range 40 A range

Inappropriate Range

4A range 40 A range

20
 16/02/2012

Report Generation
Data analysis reports should include the following:
1. Record of customer, operator, test equipment and asset
2. Record of test parameters
3. Report of test operation/method and any area tested
specially, e.g. step iron tested with Timebase mode
4. Summary report with Peaks
5. Full plots of all QUAD traces
6. Observations and analysis of any defects determined
7. Recommendations for future and any corrective actions
• Several features have copy-to-clipboard function (Peaks, Notes, Select Slots)
• Report can be completed in a word processor by copying relevant ELAN windows,
scan traces, values, etc into text file (using Print Screen)

CIGŔE Report 257, 2004

“There seems to be general consensus that if an EL CID
test is performed and no damage is found, then the core is
defect free. EL CID has gained good credibility in its ability
to determine and locate the presence of faults and to verify
repairs when faults are found.”

EL CID test is a time, labour and power

saving method of detecting imperfections in
core laminations

21