Lyft Training V1.2R1 Letter

Pulsed Eddy Current Training
Version 1.2
Pulsed Eddy Current Training
Version 1.2R1
3/20/2017
Training Scope
This course is intended to prepare prepares the participants to perform pulsed

eddy current (PEC) inspections with the Lyft® system.
Participants will learn to:

• Perform complete inspection projects.
• Carry out tasks necessary before and after performing inspections, such as:
• Creating and managing inspection projects, components, and scan zones.
• Selecting appropriate probes and instrument configurations to achieve optimum performance.
• Interpreting and evaluating results.
• Generating inspection reports.
• Understand the breadth, intricacies, and limitations of PEC.
20/03/2017 © Eddyfi 2009–2017 2
V 1.2.R1
Training Scope
Agenda Overview
Day 1
Session 1: Pulsed Eddy Current Theory
Session 2: Overview and Operation Workflow
Day 2
Session 3: Applications and Conditions Impacting PEC Signals
Session 4: Experiments with Special Conditions
Day 3
Session 5: Inspection Procedures
Session 6: Exams
20/03/2017 © Eddyfi 2009–2017 3
Application Scope
PEC measures wall thickness with liftoff
• Without liftoff, UT inspection is • 3 options to the problem:

straightforward. • Low-frequency eddy current array
But liftoff is often a constraint: (LFECA)
• Non-metallic pipe protection • Remote-field array (RFA)
• External corrosion product • Pulsed eddy current (PEC)
• Corrosion under insulation (CUI) • Eddyfi chose PEC: more versatile !
• Marine growth
Corrosion scab Coating and wrapping CUI
20/03/2017 © Eddyfi 2009–2017 4
V 1.2.R1
Application Scope
PEC measures wall thickness with liftoff
20/03/2017 © Eddyfi 2009–2017 5
PEC Technique Strengths and Weaknesses
PEC Strengths
• Direct contact with the surface or surface preparation are unnecessary.

• Measures the full wall thickness.
• Sensitive to near-side and far-side defects.
• Inspect through insulation, coatings, paints, concrete fireproofing, marine growth.
• Inspect through aluminum, stainless steel, and galvanized steel weather jackets.
• Inspect structures featuring chicken wire and rebars.
• Unaffected by water or ice.
• No costly or unnecessary insulation stripping.
• No safety issues (unlike X-rays) and performed on working components.
• Easy to use, fast learning curve.
20/03/2017 © Eddyfi 2009–2017 6
V 1.2.R1
PEC Technique Strengths and Weaknesses
PEC Weaknesses
• Affected by the edge effect (about the size of a probe’s footprint) near metallic
structures.
• Unable to discriminate between near-side and far-side defects.
• Impossible to detect small pitting.
• Under sizes flaws smaller than the probe’s averaging area (mitigated with the
compensated wall thickness tool—see later in the training).
• Difficult to use on elbows smaller than 200 mm (8 in) in diameter.
• Difficult to use on very fast materials (with characteristic decay rates less than 3 ms).
20/03/2017 © Eddyfi 2009–2017 7
Session 1
Pulsed Eddy Current Theory
20/03/2017 © Eddyfi 2009–2017 8
V 1.2.R1
Session 1 Agenda
In session 1, participants will learn:
• What is the PEC acquisition cycle.

• What is an A-scan and how the decay rate is derived from this curve.
• How C-scans summarize information from multiple A-scans.
• How the footprint and averaging area are defined and how they influence the resolution
and the detection limit.
• How to select the right probe according to the component.
20/03/2017 © Eddyfi 2009–2017 9
Session 1 Agenda
• What is an A-scan
• How C-scan
• Information
fields
20/03/2017 © Eddyfi 2009–2017 10
V 1.2.R1
PEC Acquisition Cycle
PEC analyzes the transient eddy current in a conductive component after a

sharp electromagnetic transition
The first phase is the emission (the pulse). The probe injects magnetic fields
that penetrate and stabilize in the component’s thickness.
Time
Excitation pulse
Time
Magnetic response
20/03/2017 © Eddyfi 2009–2017 11

The second phase is the cutoff. It induces strong eddy currents into the component
when the magnetic field emission is abruptly terminated.
Time
Excitation pulse
Time
Magnetic response
20/03/2017 © Eddyfi 2009–2017 12
V 1.2.R1

The third phase is the reception. The magnetic sensors measure the eddy
current decay rate as they diffuse in the material thickness.
3
Time
Excitation pulse
Time
Magnetic response
20/03/2017 © Eddyfi 2009–2017 13
The overall length of a cycle depends on the material’s time constant
Pulse width Relaxation time
Driver
Time
Acquisition cycle
Active measurement time
Receiver
Time
Magnetic field decay related to
material thickness and properties
A thicker wall, more permeable and more conductive materials leads to a longer time
for the magnetic field to stabilize and has a slower magnetic field decay.
A single cycle can take from 50 ms to 1 s.
20/03/2017 © Eddyfi 2009–2017 14
V 1.2.R1
This three-step process is repeated cyclically
1. Emission 2. Cutoff 3. Reception
Time Time Time

Excitation pulse Excitation pulse Excitation pulse
Time Time Time

Magnetic response Magnetic response Magnetic response
20/03/2017 © Eddyfi 2009–2017 15
A-Scan
An A-scan shows the amplitude of a probe’s response according to the elapsed

time after the pulse is cut off
The log-log scale is used by Lyft shows the log-lin scale by default
conventional PEC systems
Log-Log Scale Log-Lin Scale

Voltage —Log Scale (v)
Voltage—Log Scale (v)
1 10 100 0 50 100 150

Time—Log Scale (ms) Time—Lin Scale (ms)
20/03/2017 © Eddyfi 2009–2017 16
V 1.2.R1
A-Scan
Early in an A-scan, faster eddy currents from the material’s surface predominate
The decay rate follows a power law that produces a:

• Straight line in a log-log scale.
• Rapid drop in a log-lin scale.


∝ ∝
1 10 100 0 50 100 150
Time—Log scale (ms) Time—Lin Scale (ms)
20/03/2017 © Eddyfi 2009–2017 17
A-Scan
Later in an A-scan, eddy currents diffusing through the material’s volume

predominate
The decay rate follows an exponential relation that produces a:
• Rapid drop in a log-log scale. Material’s
• Straight line in a log-lin scale. time constant

∝ ∝
1 10 100 0 50 100 150

20/03/2017 © Eddyfi 2009–2017 18
V 1.2.R1
Decay Rate
The decay rate is inversely proportional to the material’s time constant (

) over
the exponential portion of the signal
Component
1
Thinner material = faster decay = smaller
2
3
Thicker material = slower decay = larger
4


∝
1 10 100 0 50 100 150

20/03/2017 © Eddyfi 2009–2017 19
Decay Rate
On a log-lin scale, a thinner wall leads to a steeper slope
• The curve used to fit is more complex Log-Lin Scale

than a simple line, but the basic principle
remains.
Thin Wall
• The analysis method is insensitive to Thick Wall
liftoff variations because it looks at the
signal’s shape.
• Liftoff variations change the amplitude
but not the shape of the curve.
• Up to a certain point: Great liftoff
0 50 100 150
variations can lead to saturation and low
Time—Lin Scale (ms)
signal, covered in session 3.
20/03/2017 © Eddyfi 2009–2017 20
V 1.2.R1
Decay Rate
The acquisition region is reduced on Lyft® to improve the acquisition rate and
minimize effects of weather jackets
Log-Lin Scale
10
Voltage - Log scale (v)

1 Lyft region of interest
The region is defined during
system autoset.
0.1
If component’s properties
change significantly, signal
can get outside the box and a 0.01
new calibration is needed.

0.001
1 21 41 61 81 101 121 141
Time - Linear scale (ms)
20/03/2017 © Eddyfi 2009–2017 21
Decay Rate
not only depends on wall thickness, but also on material permeability () and
conductivity (σ)
Material’s
time constant

∝ , , σ
Log-Lin Scale

Lyft does not automatically discriminate ∝

between the effects of , σ, and the wall
thickness, but: Decay Rate
• Within a component, σ and should be fairly
constant.
• You can see odd signals with fast decay rates
but high amplitudes.
0 50 100 150
• When in In doubt: double check with an Time—Lin Scale (ms)
inspection port.
20/03/2017 © Eddyfi 2009–2017 22
V 1.2.R1
Decay Rate
Furthermore,
depends on the wall thickness within the averaging area
General wall loss

Average wall thickness
Minimum wall thickness
Averaging Area
Irregular wall loss (such as corrosion under insulation)

Average wall thickness

Averaging Area
Very localized corrosion (such as pitting) Average wall thickness

Averaging Area
20/03/2017 © Eddyfi 2009–2017 23
Decay Rate
Furthermore,
depends on the wall thickness within the averaging area
Averaging Area
The averaging area ≈ area covered

with 90% of the emitted magnetic 90% of the
flux. Total magnetic flux
It is commonly referred to as
the surface seen by the probe.
20/03/2017 © Eddyfi 2009–2017 24
V 1.2.R1
Footprint and Averaging Area
The footprint is the main variable used in describing the signal distribution in
space
Averaging Area (AvgA)

Footprint (FP)
The footprint is defined as the area
on the component’s surface 90% of the
containing 50% of the magnetic flux. Total magnetic flux
50% of the
total magnetic
flux
.
20/03/2017 © Eddyfi 2009–2017 25
The footprint and the averaging area are closely related
Averaging Area Averaging Area (AvgA)

Footprint Footprint (FP)
100
50% of the 90% of the
total Total magnetic flux
magnetic flux
50
.
20/03/2017 © Eddyfi 2009–2017 26
V 1.2.R1
The footprint is defined along the axis of the PEC probe
The signal amplitude

• Reaches maximum at the center of the probe’s magnetic element.
• Decreases axially and transversely following a bell-shaped curve.
–6 dB
(50%)
Footprint
20/03/2017 © Eddyfi 2009–2017 27
The footprint and averaging area increase with:
Probe Size
The larger the probe, the larger the footprint at the probe’s interface.
Coating/Insulation Thickness
The signal diverges as it travels toward the
component under test.
Weather Jacket Material

For example, galvanized steel.
20/03/2017 © Eddyfi 2009–2017 28
V 1.2.R1
The footprint can be estimated with the probe size and the liftoff
. !
LO is the liftoff: insulation, weather jacket, coating thickness, etc.
FP0 is the footprint at liftoff = 0
35 mm 62 mm 100 mm
1.38 in 2.44 in 3.94 in
FP0
~18° FP
20/03/2017 © Eddyfi 2009–2017 29
The footprint is defined to ensure there is at least 50% signal overlap between
each grid point
It is equal to the full width at

half maximum (FWHM).
It also corresponds to the

minimum grid resolution to
obtain 100% coverage. –6 dB
(50%)
FP
Grid Spacing Grid Spacing
20/03/2017 © Eddyfi 2009–2017 30
V 1.2.R1
C-Scan
The footprint is defined to ensure at least 50% signal overlap between each grid
point
Physical Grid Resulting C-Scan
FP
90 81 65 77
FP
95 75 58 68
100 85 62 73
20/03/2017 © Eddyfi 2009–2017 31
C-Scan
The footprint is defined to ensure at least 50% signal overlap between each grid
point
Physical Grid Resulting C-Scan
FP
90 81 65 77
FP
95 75 58 68
100 85 62 73
20/03/2017 © Eddyfi 2009–2017 32
V 1.2.R1
C-Scan
A C-scan is a simplified representation of all the analyzed A-scans meant to ease

data visualization
The remaining wall thickness is 58%

from the nominal value.
Log-Lin Scale

∝ 90 81 65 77
95 75 58 68
100 85 62 73
0 50 100 150
Time—Lin Scale (ms)
20/03/2017 © Eddyfi 2009–2017 33
C-Scan
The Lyft system recommends three standard acquisition resolutions that are
defined according to the footprint
Grid-mapping mode Dynamic mode Dynamic mode

For fast screening For fast screening For high-resolution scans
FP/5
FP
FP/2
FP
FP
FP/2
20/03/2017 © Eddyfi 2009–2017 34
V 1.2.R1
Resolution and Detection Limit
The system’s resolving power relates to the footprint, not the sampling
Inspection
Reality 98 96 95 96 98
Results
96 85 75 85 96
Smaller 95 75 55 75 95
100 100 100 100 100 96 93 91 93 96
Footprint 96 85 75 85 96
100 100 100 100 100 93 87 83 87 93
98 96 95 96 98
100 100 25 100 100 91 83 75 83 91
100 100 100 100 100 93 87 83 87 93
100 100 100 100 100 96 93 91 93 96

96 95 94 94 93 92 94 95 95 97
94 93 90 89 88 87 90 93 93 96
93 91 87 83 81 82 85 88 92 96
Grid
Higher 94 90 84 80 77 75 81 85 91 94
Component Sampling 95
96
91
92
85
87
82
84
75
79
76
81
80
84
85
88
90
92
95
95
96 94 88 86 84 84 85 89 94 96
Field intensity 98 97 95 94 93 95 96 97 98 98
20/03/2017 © Eddyfi 2009–2017 35
The smallest detectable defect corresponds to a volume approximately 15% of

the footprint volume
Deep Defect, Small Diameter Shallow Defect, Large Diameter
FP Volume
Volume of Smallest
Detectable Defect
This rule of thumbs is valid down to defects making "⁄# of the FP’s surface.
It was determined without a weather jacket under typical conditions.
20/03/2017 © Eddyfi 2009–2017 36
V 1.2.R1
The smallest detectable defect corresponds to a volume approximately 15% of

the footprint volume
To find the smallest diameter: To find the shallowest depth:

• Minimum volume ratio: VolRatio(%) = 15% • Minimum volume ratio: VolRatio(%) = 15%
• Specific defect depth: DefDp(%) • Specific defect diameter: DefDiam
,-.(/% *+
2
%&'() *+ %&'1% ,-.(/%
%&'1% %&'()
20/03/2017 © Eddyfi 2009–2017 37
PEC measurements are disturbed by nearby geometry variations such as

nozzles, flanges, or the end of a structure
Edge Effect Mass Effect
Measurements begin to vary from a distance of approximately

one footprint from the center of the probe.
20/03/2017 © Eddyfi 2009–2017 38
V 1.2.R1
Selecting the Right Probe
A smaller probe yields a better resolution (smaller footprint), but tolerates less
liftoff
20/03/2017 © Eddyfi 2009–2017 39
Probe Selection Table — First-Generation Probes
20/03/2017 © Eddyfi 2009–2017 40
V 1.2.R1
Probe Selection Table—Second-Generation Probes
20/03/2017 © Eddyfi 2009–2017 41
Probe Selection Table—Offshore Probes
20/03/2017 © Eddyfi 2009–2017 42
V 1.2.R1
Footprint Evaluation Table—First-Generation Probes
20/03/2017 © Eddyfi 2009–2017 43
Footprint Evaluation Table—Second-Generation Probes
20/03/2017 © Eddyfi 2009–2017 44
V 1.2.R1
Minimum Detectable Defect Diameter as a function of the Defect Depth and the
Probe’s Footprint
Minimum Defect Depth
10% 20% 30% 40% 50% 60%
mm In mm in mm in mm in mm in mm in mm in
40 1.6 49 1.9 35 1.4 28 1.1 24 1.0 22 0.9 20 0.8
50 2.0 61 2.4 43 1.7 35 1.4 31 1.2 27 1.1 25 1.0
60 2.4 73 2.9 52 2.0 42 1.7 37 1.4 33 1.3 30 1.2
70 2.8 86 3.4 61 2.4 49 1.9 43 1.7 38 1.5 35 1.4
80 3.1 98 3.9 69 2.7 57 2.2 49 1.9 44 1.7 40 1.6
F
90 3.5 110 4.3 78 3.1 64 2.5 55 2.2 49 1.9 45 1.8
O
O 100 3.9 122 4.8 87 3.4 71 2.8 61 2.4 55 2.2 50 2.0
T 110 4.3 135 5.3 95 3.8 78 3.1 67 2.7 60 2.4 55 2.2
P 120 4.7 147 5.8 104 4.1 85 3.3 73 2.9 66 2.6 60 2.4
R 130 5.1 159 6.3 113 4.4 92 3.6 80 3.1 71 2.8 65 2.6
I
140 5.5 171 6.8 121 4.8 99 3.9 86 3.4 77 3.0 70 2.8
N
T 150 5.9 184 7.2 130 5.1 106 4.2 92 3.6 82 3.2 75 3.0
160 6.3 196 7.7 139 5.5 113 4.5 98 3.9 88 3.5 80 3.1
170 6.7 208 8.2 147 5.8 120 4.7 104 4.1 93 3.7 85 3.3
180 7.1 220 8.7 156 6.1 127 5.0 110 4.3 99 3.9 90 3.5
190 7.5 233 9.2 165 6.5 134 5.3 116 4.6 104 4.1 95 3.7
200 7.9 245 9.6 173 6.8 141 5.6 122 4.8 110 4.3 100 3.9
Defects smaller than the probe averaging area are undersized but detected if their diameter is
larger than the tabulated Minimum Defect Diameter.
20/03/2017 © Eddyfi 2009–2017 45
Session 2
System Overview and Operation

Workflow – Hands-On
20/03/2017 © Eddyfi 2009–2017 46
V 1.2.R1
Session 2 Agenda
In session 2, participants will become familiar with Lyft. Specifically, they will
learn about:
• The Lyft system:
• Instrument, probes and accessories
• The inspection workflow:
• Instrument Configuration
• Component preparation/marking
• Calibration, SmartPULSETM and Survey Mode
• Data acquisition / scan
• Data analysis
• Reporting
• The compensated wall thickness algorithm.
20/03/2017 © Eddyfi 2009–2017 47
Inspection Workflow
The goal of the first hands-on session is to go through the inspection workflow
with Lyft and its probes
• Instrument configuration
• Project creation
• Component creation
• Scan zone creation
• Calibration
• SmartPULSE
• Survey mode
• Data analysis
• Reporting
20/03/2017 © Eddyfi 2009–2017 48
V 1.2.R1
Inspection Workflow
• Project creation Project
• Scan zone creation Component
• Calibration Scan Zone
• SmartPULSE and Setup
• Survey mode
Data
• Data analysis
• Reporting
20/03/2017 © Eddyfi 2009–2017 49
Inspection Workflow
• Calibration
• SmartPULSE
• Survey mode
• Data analysis
• Reporting
20/03/2017 © Eddyfi 2009–2017 50
V 1.2.R1
Inspection Workflow
• Calibration
• SmartPULSE
• Survey mode SmartPULSE Survey mode

• Data analysis
• Reporting
20/03/2017 © Eddyfi 2009–2017 51
Inspection Workflow
FP
• Component preparation/marking FP
• Calibration
• SmartPULSE
• Survey mode
• Data analysis
• Reporting
20/03/2017 © Eddyfi 2009–2017 52
V 1.2.R1
Inspection Workflow
Calibration Color Compensated
point palette Wall Thickness
• Calibration
• SmartPULSE
• Survey mode
• Data analysis
Report Entry Indication
• Reporting
20/03/2017 © Eddyfi 2009–2017 53
Inspection Workflow
• Calibration
• SmartPULSE
• Survey mode
• Data analysis
• Reporting
20/03/2017 © Eddyfi 2009–2017 54
V 1.2.R1
Calibration, SmartPULSE, and Survey Mode
Lyft calibration is a 3-step process combined in the SmartPULSE algorithm
= + +
WT Repeatability
SmartPULSE PEC Autoset Calibration Optimization
20/03/2017 © Eddyfi 2009–2017 55
PEC Autoset adjusts the pulser settings to avoid saturation and improve the
signal-to-noise ratio
• Instrument internal gain
• Line filter frequency
• Pulse start time and length
20/03/2017 © Eddyfi 2009–2017 56
V 1.2.R1
WT Calibration defines the reference wall thickness
The reference WT is typically 100%

of the nominal, but it can be a
different percentage.
Wall Thickness Calibration cannot

be used alone to avoid saturation
(PEC Autoset required).
20/03/2017 © Eddyfi 2009–2017 57
The Repeatability Optimization step evaluates the system noise
• The averaging is suggested by

default to the least yielding
1% repeatability
• The user can choose:
• A higher averaging to
increase data quality.
• A lower averaging to
increase the acquisition
speed.
20/03/2017 © Eddyfi 2009–2017 58
V 1.2.R1
Survey Mode is used to quickly validate the best area to calibrate on
It simply deactivates the c-scan

and data is not saved.
It enables sampling without

recording nor changing the scan
zone settings.
20/03/2017 © Eddyfi 2009–2017 59
Here are some tips to find the best point to calibrate on:
1. Always try to use the thickest area Calibrate here

with the lowest insulation/coating
thickness.
2. Avoid defects.
3. Stay away from other metallic
Calibrate here
structures / edges
(at least 1 footprint).
4. Avoid jacket overlaps and straps.
5. Stay away from sources of vibration.
6. Use Survey Mode.
20/03/2017 © Eddyfi 2009–2017 60
V 1.2.R1
Compensated Wall Thickness
The actual signal contains information from the defect and the nominal wall
thickness
Actual signal
Signal - Log Scale (V)
2 12 22 32 42 52
Time - Log Scale (ms)
20/03/2017 © Eddyfi 2009–2017 61
thickness
Actual signal = Average contribution of nominal

2 12 22 32 42 52
20/03/2017 © Eddyfi 2009–2017 62
V 1.2.R1
thickness
Actual signal = Average contribution of nominal + defect
2 12 22 32 42 52
20/03/2017 © Eddyfi 2009–2017 63
The Compensated Wall Thickness isolates the defect contribution from the
signal
It uses the known nominal and actual signal

curves to find the defect curve.
2 12 22 32 42 52
It analyzes a defective region rather than a
single data point to increase the signal-to-
noise ratio.
• Needs a C-scan
• Finds one minimum WT for the whole
region
20/03/2017 © Eddyfi 2009–2017 64
V 1.2.R1
The compensated wall thickness improves results on lab samples
Sample Flat bottom holes

A B
Plate WT 0.5” (12.7mm)
Insulation 2” (50.8mm)
Defect Diameter Real WT C
A 3’’ 66%
B 6’’ 33%
C 3’’ 33%
20/03/2017 © Eddyfi 2009–2017 65
A B
Average C
Defect Diameter Real WT
WT
A 3’’ 66% 88%
B 6’’ 33% 62%
C 3’’ 33% 80%
20/03/2017 © Eddyfi 2009–2017 66
V 1.2.R1
A B
Average Comp. C
Defect Diameter Real WT
WT WT
A 3’’ 66% 88% 57%
B 6’’ 33% 62% 35%
C 3’’ 33% 80% 36%
20/03/2017 © Eddyfi 2009–2017 67
The compensated wall thickness improves results on real corrosion samples
Sample Corrosion Scab UT scan from inside

Pipe WT 0.25” (6.35mm)
Insulation None 29%
32%
PEC scan from outside—Average WT C-scan
52%
56%
20/03/2017 © Eddyfi 2009–2017 68
V 1.2.R1

Pipe WT 0.25” (6.35mm)
Insulation None 29%
32%
PEC scan from outside—Compensated WT C-scan
30%
24%
20/03/2017 © Eddyfi 2009–2017 69

Pipe WT 0.25” (6.35mm)
Insulation 2.5” (63mm) 29%
32%
PEC scan from outside—Average WT C-scan
75%
74%
20/03/2017 © Eddyfi 2009–2017 70
V 1.2.R1

Pipe WT 0.25” (6.35mm)
Insulation 2.5” (63mm) 29%
32%
PEC scan from outside—Compensated WT C-scan
40%
42%
20/03/2017 © Eddyfi 2009–2017 71
CWT Rules
The Compensated Wall Thickness algorithm works best on isolated defects

surrounded by near-nominal walls
Optimal Conditions Sub-optimal Conditions
Flange
Defect embeded in
large corrosion
20/03/2017 © Eddyfi 2009–2017 72
V 1.2.R1
CWT Rules
The CWT algorithm will be inactive or will not converge if:
• The selection box does not contain at least 3 rows, 3 columns and 9 valid data points.
Only 8 points
3x1 grid
Valid Invalid
20/03/2017 © Eddyfi 2009–2017 73
CWT Rules
• The deepest point of the selection is on an edge (including the impact of the edge effect).
Valid Invalid
20/03/2017 © Eddyfi 2009–2017 74
V 1.2.R1
CWT Rules
• Defects are barely detectable or the signal is too noisy.
Valid Invalid
20/03/2017 © Eddyfi 2009–2017 75
CWT Rules
• Defects are barely detectable or the signal is too noisy.
• The scan sampling is lower than 1FP along the index axis and ½FP along the scan axis.
⅓ FP 1 FP
Valid Invalid
20/03/2017 © Eddyfi 2009–2017 76
V 1.2.R1
CWT Rules
The CWT tool can work with sub-components
• Sub-components must be calibrated

• The box cannot include sub-component boundaries
Sub-component
calibration point
Valid Invalid
20/03/2017 © Eddyfi 2009–2017 77
Session 3
Applications and Conditions

Impacting PEC Signals
20/03/2017 © Eddyfi 2009–2017 78
V 1.2.R1
Session 3 Agenda
In session 3, participants will learn:
• What are edge and mass effects, and specific considerations regarding elbows and
small pipes.
• The impact of aluminum and galvanized steel weather jackets.
• How to identify and mitigate vibration sources.
• What is metallic component interference and how it impacts PEC signals.
• The effects of liftoff variations within a scan.
• The system’s speed limitations.
20/03/2017 © Eddyfi 2009–2017 79
Edge and Mass Effects
PEC measurements are disturbed by nearby geometry variations such as

nozzles, flanges, or the end of a structure
Edge Effect Mass Effect
Measurements begin to vary from a distance of one footprint from the center of a probe’s coils.
20/03/2017 © Eddyfi 2009–2017 80
V 1.2.R1
Edge effect: When the probe approaches an edge, an increasing portion of the
averaging area is outside the component
Edge Effect
Indication
Plate
edges
This results in a decreasing measured wall thickness.
20/03/2017 © Eddyfi 2009–2017 81
Mass effect: When the probe approaches a flange, the strong and slow flange
signal covers a portion of the pipe
1. Pipe wall: Far + Thin

→ Weak, fast-decaying signals
2. Flange: Close + Thick Insulation
→ Strong, slow-decaying signals
3. Within one FP distance, flange signal 2
hides pipe signal
Flange
1
Pipe wall
This results in an increasing measured
wall thickness.
3
20/03/2017 © Eddyfi 2009–2017 82
V 1.2.R1
Mass effect: Example of corrosion close to the flange
Nominal signal
Flange signal
2
1
20/03/2017 © Eddyfi 2009–2017 83
Mass effect also applies to:
• Pipe saddles • T-pipes

• Supports • Any other area interfering
• Welded I-beams with the averaging area
• Nozzles
20/03/2017 © Eddyfi 2009–2017 84
V 1.2.R1
The CWT can take mass effect into account
The CWT algorithm will assume nominal

on a perpendicular line from the
bounding box and compensate the
measurement.
For this reason, the box should be larger

than the defect !
20/03/2017 © Eddyfi 2009–2017 85
Other components only interfere if they cross the averaging area
This pipe does not interfere with the probe.
PEC area
20/03/2017 © Eddyfi 2009–2017 86
V 1.2.R1
Other components only interfere if they cross the averaging area
The averaging area is slightly wider on the sides of a probe—take advantage of this to avoid
interfering components.
Averaging area
Interfering component,
edge or mass
20/03/2017 © Eddyfi 2009–2017 87
Elbows
Elbows are complex components that require special attention
Pipe wall and elbow wall

properties may be different
Tight intrados radius

Smooth extrados radius
Welds
20/03/2017 © Eddyfi 2009–2017 88
V 1.2.R1
Elbows
The best place to calibrate is usually on the side of an elbow
In case of saturation or weak signals around elbow circumference,

separate elbow inspection in two scan zones
20/03/2017 © Eddyfi 2009–2017 89
Elbows
Lyft integrates an elbow component to handle the shape variation between the
intrados and the extrados
3453 6 78
20/03/2017 © Eddyfi 2009–2017 90
V 1.2.R1
Elbows
Lyft integrates an elbow component to handle the shape variation between the
intrados and the extrados
3453 6 78
20/03/2017 © Eddyfi 2009–2017 91
Elbows
Inspecting with the probe perpendicular to the pipe axis will yield better results
The probe is less affected by liftoff variations due to:
The probe casing and cable The component’s radius
For same thickness, expect different

measurements along elbow radius
20/03/2017 © Eddyfi 2009–2017 92
V 1.2.R1
Small pipes
Pipes with an outer diameter as small as 25.4 mm ( 1 in) can be inspected with
Lyft
• The footprint and minimum volume concepts are still valid.
• As the pipe diameter decreases, the footprint wraps around the circumference and
stops to expand in the transverse axis.
NPS 2 NPS 4 NPS 6
20/03/2017 © Eddyfi 2009–2017 93
Small pipes
When the liftoff is high enough for the footprint to wrap around the pipe, far-
side defects can be detected. This is the small pipe regime.
• Defect A is detected on the NPS 2 pipe with increasing liftoff
opposite side with enough liftoff. 100
Small Pipe Regime
• After 3 inches of LO, the signal 95
Measured Remaining WT (%)
decreases due to the SNR ratio.

90
85
Front side
0°
80
Defect
75
Front Side
70
Opposite Side
65
Opposite side 0 1 2 3 4 5
180°
Lift off (inches)
20/03/2017 © Eddyfi 2009–2017 94
V 1.2.R1
Small pipes
Small pipes—Example on a pipe with a 50.8 mm ( 2 in) OD
90° defects 360° defects
WL 60% 60% 20% 20% 60% 60% 20%
LO: 1“
Nominal WT: 0.154" (~5/32")
20/03/2017 © Eddyfi 2009–2017 95
Small pipes
Lyft takes the pipe curvature into account - it defines a circumferential footprint
Simple footprint translation

leads to oversampling.
The circumferential footprint

projects the angle
on the outer surface.
20/03/2017 © Eddyfi 2009–2017 96
V 1.2.R1
Aluminum Weather Jackets
Aluminum weather jackets create an additional fast decaying signal on the A-

scan
Lyft adds a delay to the reception phase of the PEC cycle to avoid the fast decaying
aluminum.
Aluminum jackets may prevent accurate measurements on thin walls (WT ≤3mm; ≤1/8in)
Early saturation when adding 0.02”

(0.5mm) aluminum on component
20/03/2017 © Eddyfi 2009–2017 97
Galvanized Steel Weather Jackets
Galvanized steel weather jackets have a strong negative impact on PEC signals
The ferromagnetic nature of galvanized steel (GS):

1. Attenuates PEC signals
2. Adds a mechanical vibration synced with the pulses
Without galvanized steel With galvanized steel
20/03/2017 © Eddyfi 2009–2017 98
V 1.2.R1
1. Signal attenuation forces the unit to use a higher gain and averaging
• Assume an equivalent increase of 40 mm (1.5 in) of liftoff per 0.5 mm (0.02 in) of galvanized
steel.
• Lyft 1.2 suggests another probe when the gain is too high.
Insulation / Coating thickness
mm 0.00 6.35 12.70 19.05 25.40 38.10 50.80 63.50 76.20 88.90 101.60 127.00 152.40 177.80 203.20 254.00 304.80
in 0.000 0.250 0.500 0.750 1.000 1.500 2.000 2.500 3.000 3.500 4.000 5.000 6.000 7.000 8.000 10.000 12.000
3.18 0.125
6.35 0.250
9.53 0.375 PEC-025-G2

12.70 0.500
Wall Thickness
15.88 0.625
19.05 0.750
PEC-089-G2
25.40 1.000
31.75 1.250
PEC-152-G2
38.10 1.500
50.80 2.000
63.50 2.500
76.20 3.000
101.60 4.000
20/03/2017 © Eddyfi 2009–2017 99
2. GS vibrates under the probe, causing noise on the A-scan
• PEC pulses induce magnetic forces and electrostriction in GS.

• This noise cannot be averaged out because it is synced with PEC pulses.
• The dynamic mode is not recommended on galvanized steel jackets.
A-scan from vibrating galvanized steel
A-scan from static

galvanized steel
20/03/2017 © Eddyfi 2009–2017 100
V 1.2.R1
A probe shoe and pressure on the probe increases the mechanical coupling
between the probe and galvanized steel, damping vibrations
The shoe is secured to the probe with 4 screws.
20/03/2017 © Eddyfi 2009–2017 101
Vibration Sources
PEC measurements are influenced by mechanical vibrations of inspected

components
Potential vibration locations include:
• Vibrations caused by nearby pumps and rotating
machines.
• Spectral content of vibrations typically concentrated
around a few discrete frequencies.
• Pipes affected by flow-generated vibrations and

cavitation near elbows, tees, and other flow-disrupting
structures.
• Flow disruptions may cause turbulence generating low-
frequency, wide-bandwidth vibrations.
20/03/2017 © Eddyfi 2009–2017 102
V 1.2.R1
Vibration Sources
High-frequency vibrations do not affect measurements, despite being visible on

A-scan
Vibration period
A-scan length
High frequency = Period less than A-scan length/20

20/03/2017 © Eddyfi 2009–2017 103
Vibration Sources
Low frequency vibrations cause periodic modulation of A-scan that are typically
not in sync with measurement repetition rate
Linear amplitude scale
Nominal signal
Vibration-affected signal
The resulting C-scan is noisy
0.000 4.500 9.000 13.500
14.745 91.9 74.1 117.7 84.8
Logarithmic amplitude scale 9.830 18.9 36.1 73.4 89.1
4.915 83.6 75.0 45.4 64.8
0.000 40.6 112.9 115.4 39.6
20/03/2017 © Eddyfi 2009–2017 104
V 1.2.R1
Vibration Sources
Low frequency vibrations may also cause occasional / intermittent saturation at

start of A-scan
Saturation
Vibration-affected signal
Nominal signal
20/03/2017 © Eddyfi 2009–2017 105
Vibration Sources
Increasing averaging filters out signal modulated by external (not synced)

vibrations
0.000 4.500 9.000 13.500
14.745 91.9 74.1 117.7 84.8
9.830 18.9 36.1 73.4 89.1
4.915 83.6 75.0 45.4 64.8

More averaging
0.000 40.6 112.9 115.4 39.6
0.000 4.500 9.000 13.500
14.745 93.8 95.0 95.2 94.7
9.830 98.3 97.7 98.2 97.6
4.915 96.9 97.0 97.7 98.8
0.000 100.5 100.1 101.2 100.5
20/03/2017 © Eddyfi 2009–2017 106
V 1.2.R1
Metallic Component interference
Other metallic components can interfere with PEC signals
Chicken wire
• Thin cross-section, ~ 9.1 mm (¾ in) wire mesh wrapping insulation.
• Generally no problem. If loose, it can vibrate and induce noise.
Applying pressure and using the shoe
might help for these vibrations too!
Welded wire mesh (concrete reinforcement )

• ~3.2 mm (1/8 in) cross-section wire and ~ 38.1 mm (1.5 in) mesh opening.
• Generally no problem when far from probe. Problematic if near probe.
20/03/2017 © Eddyfi 2009–2017 107
Other metallic components can interfere with PEC signals
Rebars (concrete reinforcement )

• ~ 12.7 mm (½ in) cross-section bars and ~ 152.4 mm (6 in) mesh opening
• Generally no problem when far from probe. Problematic if in contact.
Adding artificial liftoff might help!
Heat tracers
• Steam tubes and electrical heating copper wires.
• Generally non-ferromagnetic alloys in contact with components, so no
interference.
• Electrical tracers may interfere if the current is high and the frequency is
different from the line frequency.
20/03/2017 © Eddyfi 2009–2017 108
V 1.2.R1
Rebar example: Stronger signal (more volume or closer to probe) creates more
interference
9:;&;:< ∝ 1/; 2 9:;&;:< ∝ Vi
FP volume (Vp)
Interfering volume (Vi)
20/03/2017 © Eddyfi 2009–2017 109
Rebar example: Stronger signal (more volume or closer to probe) creates more
interference
Rebar far from the probe Rebar close to the probe

No impact on the C-scan Appreciable impact on the C-scan
20/03/2017 © Eddyfi 2009–2017 110
V 1.2.R1
Signal with decay time close to region of interest creates more interference
Fast decay Slow decay
Thin weather jacket
3.2 mm (1/8 in) welded wire mesh

Fine chicken wire
12.7 mm (½ in) OD rebars
Pipe flanges
Inspected plate
Repads
Thick SS tracer
Corroded plate
Faster decay rates have less impact than similar or slower decay rates
20/03/2017 © Eddyfi 2009–2017 111
Liftoff Variation Effects
The nominal liftoff can vary significantly due to many factors
(3)
(1) (2)
Sagging insulation Damaged weather jackets Near elbows and tees Uncontrolled cover,
e.g., marine growth
(1) www.cimesupply.com/products/thermo12.php
(2) microscopicminute.wordpress.com/tag/asbestos/
(3) www.foundocean.com/en/what-we-do/products/marine-growth-prevention/
20/03/2017 © Eddyfi 2009–2017 112
V 1.2.R1
PEC Autoset adjusts the amplification level to 70% of the full-scale for optimal
tolerance to liftoff variations
• Significant liftoff reduction according to calibration may lead to A-scan saturation
• High liftoff increase may reduce signal amplitude below noise level
Saturation
Nominal
Low signal
20/03/2017 © Eddyfi 2009–2017 113
Actions must be taken when dealing with saturation or low signals
Liftoff reduction Liftoff increase
Saturation
Low signal
• Create a new scan zone. Calibrate on a low liftoff • Create a new scan zone. Calibrate on a high liftoff
section (thin insulation). section (thick insulation).
• Before inspecting, create a “survey” scan zone. • Increase averaging at expense of longer
Search for a region with the lowest liftoff and measurement times.
maximum wall thickness on the sample under test.
20/03/2017 © Eddyfi 2009–2017 114
V 1.2.R1
Speed Limitations
The maximum scan speed is reached when the instrument can no longer supply
data samples quickly enough
The maximum speed is influenced by:
• Pulse repetition frequency (PRF) Pulse duration
Pulses per seconds. Active measurement time

High PRF = Fast inspection. Pulse width Relaxation time
PRF decreases with wall thickness, liftoff
and metallic interfering components
• Averaging (Avg)
More noise = More averaging = Slower
inspection
FP
• Grid resolution (GR)
Higher spatial resolution = Slower FP
inspection speed
20/03/2017 © Eddyfi 2009–2017 115
Speed Limitations
The maximum scan speed is simply the pulse rate times the grid resolution,
divided by the averaging
F GF PRF: Pulse repetition frequency

?@A BCDDE Avg: Averaging
GR: Grid resolution
For instance, knowing that:

• After performing PEC Autoset, the pulse rate is 10 Hz
• The scan axis resolution is set to 12.7 mm/sample (½ in/sample)
• Repeatability Optimization returned an averaging of 4 samples
We can calculate the maximum scan speed:
F GF H. I
?@A BCDDE K. I LL/M . HNO/M
J
20/03/2017 © Eddyfi 2009–2017 116
V 1.2.R1
Speed Limitations
Exceeding the maximum speed in dynamic scan mode leads to lower averaging
and missing data points
• Consider 5 consecutive acquisition points with an averaging of 2.
• Blue boxes = Pulse cycles. The faster the probe goes, the longer the box is.
1 2 3 4 5
Slow scan 1 1 2 2 3 3 4 4 5 5
Faster scans
Near optimal speed 1 1 2 2 3 3 4 4 5 5
Too fast: Less averaging 1 1 2 2 3 4 4 5
Too fast: Missing data 1 2 4
20/03/2017 © Eddyfi 2009–2017 117
Speed Limitations
In addition to a warning on the instrument and the red LED on the probe,
overspeed events can be detected by analyzing scans
Unstable/Unusual ending curves Missing data points

on the A-scan on the C-scan
Signal overspeed System overspeed

The probe left before capturing the entire The instrument did not have time to start at
signal decay. least one pulse during the time interval the
probe was on the region of interest.
20/03/2017 © Eddyfi 2009–2017 118
V 1.2.R1
FAQ
Frequently Asked Questions
• Wet insulation, ice, water, marine growth effects: Negligible

• Scale effects: Scale and corrosion materials result in later signals and they do not interfere
with wall thickness measurements.
• Weld effects: Blind to most welds, but not when welds are very thick or have large HAZ.
• Typically, carbon steel permeability and resistivity increase with temperature. If the
temperature is stable, the effect is negligible on defect detection and sizing. However, the
minimum measurable thickness increases with temperature.
• Superimposed plates: Two, 12.7 mm (½ in) thick plates do not measure like one, 25.4 mm
(1 in) thick plate
• Beware of magnetized carbon steel: PEC results are affected by magnetization, so properly
demagnetizing components is recommended if such a risk is present.
20/03/2017 © Eddyfi 2009–2017 119
Session 4
Conditions Influencing PEC

Signals – Hands-On
20/03/2017 © Eddyfi NDT Inc. 2009–2016 120
V 1.2.R1
Session 5
Inspection Procedures
20/03/2017 © Eddyfi 2009–2017 121
Inspection Workflow
This inspection workflow is meant to help you develop a procedure

Know all component characteristics (geometry, nominal wall thickness, insulation, weather jacket, temperature). Initial
Obtain work permit/access to the site.
Validate inspection scope, select probes, and calculate footprint/averaging area and smallest detectable defects. Preparation
Perform systems check. On-site

Mark the structure. Preparation
Create reference scan zone, find best calibration point with survey mode and verify signal quality.
Perform final calibration. Calibration
Record a few points at calibration position in reference scan zone, verify signal quality and repeatability.
Create inspection scan zones by duplicating reference scan zone (keeps calibration).
Adjust scan zone parameters and scan definition. Inspection
Perform scan, verify data quality, and adjust scan parameters as necessary.
Recalibrate wall thickness, adjust color palette to optimize contrast, and discard invalid ponts. Analysis &
Identify and mark defects, use CWT tool if available.
Update scan zone and generate report. Report
20/03/2017 © Eddyfi 2009–2017 122
V 1.2.R1
Inspection Workflow
Initial Preparation – Make sure you know all component characteristics
• Basic component characteristics Initial

Preparation
• Wall thickness (WT)
• Insulation thickness (CT)
On-site
• Weather jacket material and thickness (JT)
Preparation
• Other structures in the insulation
• Rebars, chicken wire, etc.
Calibration
• See session 3 for details about effects of other structures on signals
• Different geometries in component
• Elbows, Ts, straight sections of a pipeline are all different geometries and Inspection
should therefore be different components
Analysis &
Report
20/03/2017 © Eddyfi 2009–2017 123
Inspection Workflow
Initial Preparation – Make sure the scope of work is clear
• Know all components that need to be inspected Initial

Preparation
• Must inspect entire component or only specific areas?
• Determine origin (datum) of components and know how you will mark the On-site
structure Preparation
• Determine scan zones you must create
• Calculate the offset from origin along X and Y axes
Calibration
• Evaluate the size along X and Y axes
• Determine the scan direction during data acquisition
• Grid mode: any direction is suitable. Inspection
• Dynamic mode: make sure the selected direction allows good encoder contact
with the surface. For example, avoid scanning along circumference axis of small
pipes. Analysis &
Report
20/03/2017 © Eddyfi 2009–2017 124
V 1.2.R1
Inspection Workflow
Initial Preparation – Example of components creation
Initial
Preparation
On-site
Preparation
Calibration
O
Inspection
Analysis &
Report
20/03/2017 © Eddyfi 2009–2017 125
Inspection Workflow
Initial Preparation – Example of determining the scan zones for the bottom of a
pipe
Initial
Preparation
On-site
Corresponding scan area
Preparation
0°
Circumference along Y axis
Calibration
Scan Zone 1 Scan Zone 2 Scan Zone 3
Offset X = 0 Offset X = 1 × length Offset X = 2 × length
180° Offset Y = 90° Offset Y = 90° Offset Y = 90°
Size X = Length
Size Y = 180°
Size X = Length
Size Y = 180°
Size X = Length
Size Y = 180°
Inspection
Analysis &
360° Report
Length along X axis
20/03/2017 © Eddyfi 2009–2017 126
V 1.2.R1
Inspection Workflow
Initial Preparation – Example of determining the scan zones for the lower
portion of a pressure vessel
Initial
Preparation
Corresponding scan area

On-site
Preparation
Height along Y axis

Calibration
Scan Zone 1 Scan Zone 2
Offset X = 0 Offset X = Half circ.
Offset Y = 0 Offset Y = 0
Size X = Half circ. Size X = Half circ.
Size Y = Half height Size Y = Half height Inspection
O O
Circumference along X axis
Analysis &
Report
20/03/2017 © Eddyfi 2009–2017 127
Inspection Workflow
Initial Preparation – Get the permits and check the working temperature
• Work permits Initial

Preparation
• Lyft is not intrinsically safe—make sure you have the proper permits if you
must use the system in a hazardous environment
On-site
• Temperature Preparation
• Surfaces in contact with PEC probes should remain under 70 °C (158 °F )
• If the surface temperature is higher, use the optional shoe for use up to 120 °C
(248 °F ) Calibration
Inspection
Analysis &
Report
20/03/2017 © Eddyfi 2009–2017 128
V 1.2.R1
Inspection Workflow
Initial Preparation – Choose the right probe
• Follow the table for your probes Initial

Preparation
• First-generation probes (025/089/152)
• Second-generation probes (025-G2/089-G2/152-G2)
On-site
• Tank floor probe (TF)
Preparation
• Splash zone and underwater probes (SZ-089/SZ-089-G2)
• Assume an equivalent increase of 40 mm (1.5 in) of liftoff per 0.5 mm
(0.02in) of galvanized steel. Calibration
• Lyft 1.2 suggests another probe when the gain is too high after Autoset.
Inspection
Analysis &
Report
20/03/2017 © Eddyfi 2009–2017 129
Inspection Workflow
Initial Preparation – Calculate essential variables
Footprint Initial
Preparation
• To determine the scan’s resolution and the edge/mass effects limit
On-site
Averaging Area Preparation
• To evaluate when defects are underestimated
Calibration
Smallest Detectable Defect
• To find the shallowest or smallest detectable defect
Inspection
Use the formulas or the tables available in session 1. Analysis &

Report
20/03/2017 © Eddyfi 2009–2017 130
V 1.2.R1
Inspection Workflow
On-site Preparation – Perform a systems check
• The systems check is performed to: Initial

Preparation
• Validate that the probe and equipment are functioning properly
• Comply with standards (ISO standard under development)
On-site
• The systems check should be performed: Preparation
• With the probe used for inspection
• On a known sample
Calibration
• Perfect if same thickness and insulation (similar conditions)
• For practicality, can be performed on different plate thicknesses for which
probe behavior is known (can be small to be easier to carry)
• Often Inspection
• At the beginning and end of each inspection/work shift
• At least every four hours Analysis &
• If probe is changed Report
20/03/2017 © Eddyfi 2009–2017 131
Inspection Workflow
On-site Preparation – Perform a systems check
• To perform a systems check : Initial

Preparation
1. Create a systems check scan zone with a significant name.
2. Create a grid mode setup for the systems check scan zone.
On-site
3. Calibrate the system with SmartPULSE.
Preparation
• If the reference WT is significantly different from component WT:
• Perform PEC Autoset, WT Calibration, and Repeatability Optimization separately.
• Manually type Mag. Time Constant in Autoset for faster calibration. Calibration
4. Perform initial reading—3 to 5 measurements on the same row.
• Change row for following systems check.
Inspection
If the reference plate wall thickness is different from the component WT, the displayed
WT is incorrect. However, WT(%) is accurate. To see the correct WT of the acquired data,
create a subcomponent covering the scan zone with a different nominal WT. Analysis &
Report
20/03/2017 © Eddyfi 2009–2017 132
V 1.2.R1
Inspection Workflow
On-site Preparation – Mark the component
• Markings are meant to help inspectors scan straight and according to Initial
the setup resolution. Markings can be: Preparation
• Eddyfi’s grid mat
• Masking tape On-site
Preparation
• Permanent marker pen
• Etc.
Calibration
Inspection
Analysis &
Report
20/03/2017 © Eddyfi 2009–2017 133
Inspection Workflow
Calibration – Create a reference scan zone
• Create a reference scan zone to: Initial

Preparation
• Perform a survey scan and find the
best location to perform calibration.
• Evaluate inspection conditions/ On-site
validate if a dynamic scan is possible. Preparation
• Maintain a record of the calibration
and propagate it to all scan zones. Calibration
• We suggest:
• Use the full component size.
• Use a standard name such as RefZone. Inspection
• Use the grid scan mode.
• Use the screening grid size (FP × FP). Analysis &
Report
20/03/2017 © Eddyfi 2009–2017 134
V 1.2.R1
Inspection Workflow
Calibration – Perform initial calibration
1. Always try to use the thickest area Calibrate here Initial

with the lowest insulation/coating Preparation
thickness.
On-site
2. Avoid defects. Preparation
3. Stay away from other metallic
Calibrate here
structures / edges
(at least 1 footprint). Calibration
4. Avoid jacket overlaps and straps.

5. Stay away from sources of vibration . Inspection
6. Perform SmartPULSE.
Analysis &
Report
20/03/2017 © Eddyfi 2009–2017 135
Inspection Workflow
Calibration – Use the Survey Mode to find the optimal reference point
• The C-scan is disabled in Survey Mode. All the information is shown in the Initial
information fields . Preparation
• Take readings at different points to locate the area with the strongest
On-site
signal. Look at the shape of A-scans (saturation, noise level) Preparation
Calibration
Inspection
Analysis &
Report
20/03/2017 © Eddyfi 2009–2017 136
V 1.2.R1
Inspection Workflow
Calibration – Use the Survey Mode to find the optimal reference point
Watch out for signal saturation. In this case, perform SmartPulse again in Initial
this area and continue looking for optimal reference point Preparation
On-site
Preparation
Calibration
Saturation creates a flat signal at Inspection

the beginning of an A-scan
Analysis &
Report
20/03/2017 © Eddyfi 2009–2017 137
Inspection Workflow
Calibration – Perform final calibration on the optimal reference point
Initial
Preparation
• Stop Survey Mode and perform
the final calibration. On-site
Preparation
• Start acquiring data.
• Capture a few (5–10) data points
Calibration
at the reference point.
• Check the signal quality and
repeatability.
Inspection
• Having data is necessary to save
the scan zone.
Analysis &
Report
20/03/2017 © Eddyfi 2009–2017 138
V 1.2.R1
Inspection Workflow
Inspection – Create inspection scan zones
Create a new scan zone by Initial

duplicating the reference point Preparation
scan zone. Check ‘’Keep
calibration’’. On-site
Preparation
Calibration
Inspection
Analysis &
Report
20/03/2017 © Eddyfi 2009–2017 139
Inspection Workflow
Inspection – Review scan resolution
• If Scan Mode or Scan Size were modified, verify the grid resolution. Initial
Preparation
• Tap Scan on the Setup ribbon.
• Adjust Resolution according to the new scan zone (this does not delete the
calibration). On-site
Preparation
Calibration
Inspection
Analysis &
Report
20/03/2017 © Eddyfi 2009–2017 140
V 1.2.R1
Inspection Workflow
Inspection – During acquisition, continuously validate data quality
Good data should have Initial

Preparation
• Fairly linear decay rate in A-scan (in Log-Lin scale)
• Initial amplitude between 1V & 10V
On-site
Preparation
Calibration
Inspection
Analysis &
Report
20/03/2017 © Eddyfi 2009–2017 141
Inspection Workflow
Good data may contain low-amplitude and high-frequency noise that does Initial
not affect decay rate evaluation (signal slope) such as: Preparation
On-site
Preparation
Oscillations Spikes Calibration
Inspection
Increasing averaging is always a good way to reduce such noise.
Analysis &
Report
20/03/2017 © Eddyfi 2009–2017 142
V 1.2.R1
Inspection Workflow
Bad data points may show: Initial

Preparation
Saturation
On-site
Preparation
Calibration
Vibration noise
Inspection
Overspeed effects
Analysis &
Report
20/03/2017 © Eddyfi 2009–2017 143
Inspection Workflow
If data quality is bad during acquisition, you can: Initial

Preparation
• Measure at these coordinates again in grid mode
• Move the probe back to overwrite bad data in dynamic mode On-site
Preparation
• Increase averaging
• In dynamic scan, increasing averaging may force you to switch to Calibration
grid scan mode
• Perform SmartPULSE again
Inspection
• The line filter frequency is re-evaluated. It may improve the
signal quality.
Analysis &
These changes erase previous data—perform as early as possible. Report
20/03/2017 © Eddyfi 2009–2017 144
V 1.2.R1
Inspection Workflow
Analysis and Reporting – Recalibrate WT on existing data
Once data acquisition is complete, you may perform WT calibration again if: Initial
Preparation
The thickness at a
A thicker area was found specific point is known
On-site
Preparation
Calibration
Inspection
Analysis &
Report
20/03/2017 © Eddyfi 2009–2017 145
Inspection Workflow
Analysis and Reporting – Recalibrate and WT using a subcomponent
If nominal WT is different from originally defined for the component: Initial

Preparation
Create a Sub- On-site

component Preparation
covering the
entire scan
zone Calibration
Define new
Inspection
nominal WT
Analysis &
Report
20/03/2017 © Eddyfi 2009–2017 146
V 1.2.R1
Inspection Workflow
Analysis and Reporting – Discard bad data that could not be re-acquired
Initial
Preparation
• Move cursor on bad data On-site

• Tap Discard/Keep Preparation
• Selected data is grayed
and no longer considered
Calibration
• A-scan remains visible
• Tap Discard/Keep again
on same coordinates to Inspection
reactivate data
Analysis &
Report
20/03/2017 © Eddyfi 2009–2017 147
Inspection Workflow
Analysis and Reporting – Optimize color palettes
Modify color palette settings to improve contrast of small WT variations: Initial

Preparation
Tap Palette on
On-site
the Home
Preparation
ribbon
Calibration
Select C-scan Edit the color
to modify range and
pattern Inspection
Analysis &
Report
20/03/2017 © Eddyfi 2009–2017 148
V 1.2.R1
Inspection Workflow
Analysis and Reporting – Update scan zones
Data is always saved, but not modifications made to the setup such as: Initial
Preparation
• WT calibrations on existing points
• Subcomponents
On-site
• Color palette modifications
Preparation
• Layout modifications
To save these, tap Update Scan Zone on the Home ribbon.
Calibration
Inspection
Analysis &
Report
20/03/2017 © Eddyfi 2009–2017 149
Inspection Workflow
Analysis and Reporting – Report defects
Initial
Preparation
Compute CWT
if possible On-site
Preparation
Short tap: Add

Calibration
indication
Fully contain
Inspection
the indication
with cursors
Analysis &
Long tap: Select
Report
indication type
20/03/2017 © Eddyfi 2009–2017 150
V 1.2.R1
Inspection Workflow
Analysis and Reporting – Report defects
Initial
Preparation
Activate the
report tab on
On-site
the Layout
Preparation
ribbon
Calibration
Inspection
Review list of
indications
Tap (and pin) Analysis &
the report tab Report
20/03/2017 © Eddyfi 2009–2017 151
Inspection Workflow
Analysis & Report – Generate Report
Create a spreadsheet report for the current component: Initial

Preparation
Fill in the
On-site
report Preparation
summary
Calibration
Inspection
Tap Finish
Tap Generate
Report Analysis &
Report
20/03/2017 © Eddyfi 2009–2017 152
V 1.2.R1
Inspection Workflow
Analysis and Reporting – Generate a report
The report contains: Initial

Preparation
• Inspection summary
• Report summary
• Component details On-site
• Scan zone list Preparation
• Reported indications list
• C-scan (color-coded cell format) Calibration
• Screen captures of reported indications
Inspection
Analysis &
Report
20/03/2017 © Eddyfi 2009–2017 153
Inspection Workflow
Analysis and Reporting – Export data and reports
Projects and components can be Initial

transferred to computers and other Lyfts: Preparation
On-site
Preparation
Via USB Via Wi-Fi
Calibration
1. Insert USB key in any instrument 1. Enable Wi-Fi on instrument.
port. 2. Connect instrument and
2. Export to USB. computer on the same network. Inspection
3. Import from USB on other 3. In Lyft Pro software:
instrument or on computer via • Import from instrument to Lyft Pro. Analysis &
Lyft Pro software. • Export from Lyft Pro to instrument. Report
20/03/2017 © Eddyfi 2009–2017 154
V 1.2.R1
Inspection Workflow
Analysis and Reporting – Take advantage of Lyft Pro capabilities
Lyft Pro allows you to analyze data on a computer, with additionnal features: Initial
Preparation
Calibration propagation between
scan zones in the same component On-site
Preparation
Right-click options Calibration

in C-scan
Inspection
Compensated wall
thickness C-scan Analysis &
Report
20/03/2017 © Eddyfi 2009–2017 155
Advanced System Features
Buttons help you perform faster inspections

Calibration
SmartPULSE Actions Button (combinations) Calibration Actions Button (combinations)
Open SmartPULSE + Open PEC Autoset +
dialog box dialog box
Open WT Calibration +
Start SmartPULSE
dialog box
Select averaging value or Open Repeat. Optimiz. +
dialog box
Apply selected
averaging Start / Execute
Quit without applying Apply and close PEC
+
averaging + Autoset /WT calibration
Optimization: Select or
PEC Autoset: Get A-scan

Optimization: Apply and close
20/03/2017 © Eddyfi 2009–2017 156
V 1.2.R1
Advanced System Features
Buttons help you perform faster inspections

Acquisition
Scan Control Probe Button (combinations)
Start acquisition
Acquire measurement All shortcuts are listed

point (grid mode) directly on the instrument!
Change row
Stop acquisition +
20/03/2017 © Eddyfi 2009–2017 157
V 1.2.R1
The information in this document is accurate as of its publication. Actual products may differ from
those presented herein.
© 2017 Eddyfi. Eddyfi, Lyft, SmartPULSE, and their associated logos are registered trademarks of
Printed on 100 % Eddyfi in the United States and/or other countries. Eddyfi reserves the right to change product
www.eddyfi.com [email protected] post-consumer fibers offerings and specifications without notice.

Lyft Training V1.2R1 Letter

Uploaded by

Copyright:

Available Formats

Lyft Training V1.2R1 Letter

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Lyft Training V1.2R1 Letter

Uploaded by

Copyright:

Available Formats

What is the purpose of the training?

What is the purpose of the training?

What are the main topics covered in the training?

What are the main topics covered in the training?

Pulsed Eddy Current Training

This course is intended to prepare prepares the participants to perform pulsed

Participants will learn to:

20/03/2017 © Eddyfi 2009–2017 2

20/03/2017 © Eddyfi 2009–2017 3

PEC measures wall thickness with liftoff

• Without liftoff, UT inspection is • 3 options to the problem:

Corrosion scab Coating and wrapping CUI

20/03/2017 © Eddyfi 2009–2017 4

PEC measures wall thickness with liftoff

20/03/2017 © Eddyfi 2009–2017 5

PEC Technique Strengths and Weaknesses

• Direct contact with the surface or surface preparation are unnecessary.

20/03/2017 © Eddyfi 2009–2017 6

20/03/2017 © Eddyfi 2009–2017 7

Pulsed Eddy Current Theory

20/03/2017 © Eddyfi 2009–2017 8

In session 1, participants will learn:

• What is the PEC acquisition cycle.

20/03/2017 © Eddyfi 2009–2017 9

20/03/2017 © Eddyfi 2009–2017 10

PEC analyzes the transient eddy current in a conductive component after a

20/03/2017 © Eddyfi 2009–2017 11

PEC Acquisition Cycle

PEC analyzes the transient eddy current in a conductive component after a

20/03/2017 © Eddyfi 2009–2017 12

PEC analyzes the transient eddy current in a conductive component after a

20/03/2017 © Eddyfi 2009–2017 13

PEC Acquisition Cycle

The overall length of a cycle depends on the material’s time constant

Pulse width Relaxation time

Active measurement time

This three-step process is repeated cyclically

1. Emission 2. Cutoff 3. Reception

Time Time Time

Time Time Time

An A-scan shows the amplitude of a probe’s response according to the elapsed

Log-Log Scale Log-Lin Scale

Voltage—Log Scale (v)

1 10 100 0 50 100 150

20/03/2017 © Eddyfi 2009–2017 16

The decay rate follows a power law that produces a:

Log-Log Scale Log-Lin Scale

Voltage—Log Scale (v)

20/03/2017 © Eddyfi 2009–2017 17

Later in an A-scan, eddy currents diffusing through the material’s volume

Log-Log Scale Log-Lin Scale

Voltage—Log Scale (v)

1 10 100 0 50 100 150

20/03/2017 © Eddyfi 2009–2017 18

The decay rate is inversely proportional to the material’s time constant (

Log-Log Scale Log-Lin Scale

Voltage—Log Scale (v)

1 10 100 0 50 100 150

20/03/2017 © Eddyfi 2009–2017 19

Lyft does not automatically discriminate ∝