ENGINEERING OF
METALLURGICAL
AND MATERIALS
Development of a Magnetic Sensor for
Detecting and Sizing Paraffin
Deposition Inside Pipelines
Professor Dr. João Rebello (UFRJ)
MSc. Claudio Camerini (Petrobrás)
Eng. Ivan Vicente Janvrot (Pipeway)
Dra. Maria Cristina López Areiza (LNDC/UFRJ)
Motivation
ENGINEERING OF
METALLURGICAL
AND MATERIALS
Pipe
Paraffin
The paraffin deposit reduces
the oil flow effiency in the
pipe, by reducing its inner
diameter.
VENKATESAN, R., NAGARAJAN, N.R., PASO, K., et
al, “The strength of paraffin gels formed under static
and flow conditions”, Chemical Engineering Science,
v.60, pp. 3587 – 3598, 2005.
Motivation
ENGINEERING OF
METALLURGICAL
AND MATERIALS
Many companies use PIGs
(Pipeline Internal Gate).
This equipment
is used in several operations:
defect detection, cleaning and
paraffin removal from the pipe
inner wall.
These operations must be
periodic to avoid flow
obstruction in the pipe.
Typical distribution of sensors: called crown
Cleaning PIG
Geometrical PIG
Magnetic PIG
Objectives
ENGINEERING OF
METALLURGICAL
AND MATERIALS
Development of a magnetic based
methodology for:
-detecting and sizing paraffin deposits
on the pipeline inner wall and;
-detecting and sizing corrosion attack
areas located underneath the paraffin
deposits.
Objectives
ENGINEERING OF
METALLURGICAL
AND MATERIALS
-to develop a specific sensor to be
adapted to the equipment called
Pipeline Internal Gate (Pig), for
detecting and sizing such paraffin
deposition areas.
References
ENGINEERING OF
METALLURGICAL
AND MATERIALS
Currrently used techniques for metal loss detection due to corrosion
attack based on magnetic principles:
The typical technique is Magnetic Flux Leakage (MFL)
Magnetic Flux lines during
a MFL inspection :
(a) Low magnetization
level, not suitable;
(b) Medium magnetization
level;
(c) High magnetization
level, suitable.
NESTLEROTH, B., “Technology Status Assessment for Circumferential MFL for SCC”, Final Report on RESEARCH
PROGRAM PR39420, 1999.
Source: www.netl.doe.gov/scng/
Reference
ENGINEERING OF
METALLURGICAL
AND MATERIALS
-Different carbon steels can exhibit rather different magnetic responses
NESTLEROTH, B., “Technology Status Assessment for Circumferential MFL for SCC”, Final Report on RESEARCH
PROGRAM PR39420, 1999.
Source: www.netl.doe.gov/scng/
Reference
- How to detect defects and corroded areas using MFL and PIG’s?
ENGINEERING OF
METALLURGICAL
AND MATERIALS
Defect detection
consists in finding
the areas in the pipe
where there is a
MFL due to metal
loss.
One can see the
drive section, the
brushes for
cleaning and
magentic contact,
the sensors and
device for data
adquisition.
NESTLEROTH, B., “Technology Status Assessment for Circumferential MFL for SCC”, Final Report on RESEARCH
PROGRAM PR39420, 1999.
Source: www.netl.doe.gov/scng/
Reference
ENGINEERING OF
METALLURGICAL
AND MATERIALS
Alternative ways of magnetic induction can be
considered.
Besides the longitudinal orientation of
the magnetic flux lines, it is also possible
to use a perpendicular magnetic field.
GLADCHTEIN, R. S., “Solução do Problema Inverso Magnetostático Utilizando Elementos Finitos e Técnicas de
Otimização”, PhD Theses, PUCRio, 2002.
Reference
ENGINEERING OF
METALLURGICAL
AND MATERIALS
Perpendicular magnetic induction was succesfully used for crack
detection and sizing
Based on Gladchtein´s work, Boechat et all (2009) developed a
new magnetic sensor with perpendicular magnetic flux, called by
the authors, Internal Corrosion Sensor (ICS). The focus was on
the detection and sizing of small internal defects in thick walls
pipelines.
BOECHAT et Al., “Development of a magnetic sensor for detection and sizing of internal
pipeline corrosion defects”, NDT&E International , Vol 42 , Pag, 669–677, 2009.
Reference
ENGINEERING OF
METALLURGICAL
AND MATERIALS
-It was confirmed the big potential of the magnetic vertical induction
as technique for detection of small internal defects
BOECHAT et Al., “Development of a magnetic sensor for detection and sizing of internal
pipeline corrosion defects”, NDT&E International , Vol 42 , Pag, 669–677, 2009.
Methodology
ENGINEERING OF
METALLURGICAL
AND MATERIALS
1.
2.
3.
4.
To develop computational models for the
magnetic flux density behavior perturbation in
the pipelines wall. These values were obtained
by simulations with FEMM and OPERA software.
The focus was to obtain the best signal contrast
for the sizing of paraffin deposits.
To design, to build and to test a prototype probe.
To search for the best parameters of the probe
sensor Hall aiming to attain high sensitivity.
To set up calibration curves for paraffin deposits
sizing.
OPERA
3D
Cobham
FEMM
Finite Element
Method Magnetics
Operational principle
ENGINEERING OF
METALLURGICAL
AND MATERIALS
Steel
Operational principles
ENGINEERING OF
METALLURGICAL
AND MATERIALS
Steel
The operational principle consists in creating a constant
magnetic field close to the internal wall of the pipe, using
a permanent magnet.
The magnetic density flow lines are measured by a Hall
sensor, placed between the magnet and the internal pipe
wall.
The presence of paraffin reduces the density of the
magnetic flow detected by the Hall sensor.
Computational Model
ENGINEERING OF
METALLURGICAL
AND MATERIALS
Sensor position
OPERA 3D - Cobham

Magnet coercivity= 35mgoe

Isotropic and linear material settings

Steel 1020 permeability=> μ = 760

Steel conductivity= 5.8 MS/m

Steel size= 100x100x5 mm

Nodes = 71435

Mesh elements= 216096
FEMM – 2D
Sensor position
Study of Magnetic Sensitivity
Sensitivity was evaluated as function of the probe
sample distance
Magnet
B(T)
Magnet
B(T)
Field aquisicion
point (Hall Sensor)
STEEL
STEEL
0.090
LB (H=0)
0.085
0.080
H: Distance from the pipe
internal wall and the Hall sensor
LB: Value of B(T) measured by
the Hall sensor when in contact
with the steel plate .
B (T)
ENGINEERING OF
METALLURGICAL
AND MATERIALS
0.075
0.070
0.065
0.060
Magnetic Flux
Density at the Hall
sensor position
0
2
4
6
H (mm)
8
10
Study of Magnetic Sensitivity
Sensitivity was evaluated as function of the probe
sample distance
Magnet
B(T)
Magnet
B(T)
Field aquisicion
point (Hall Sensor)
STEEL
STEEL
Even for big distance (8 to 10
mm) the magnetic flux
density is high enough. This
means that the parameters
selected are appropriate. The
figure shows that the Hall
sensor is not in a magnetic
saturated condition.
0.090
LB (H=0)
0.085
0.080
B (T)
ENGINEERING OF
METALLURGICAL
AND MATERIALS
0.075
0.070
0.065
0.060
Magnetic Flux
Density at the Hall
sensor position
0
2
4
6
H (mm)
8
10
Level of magnetism
ENGINEERING OF
METALLURGICAL
AND MATERIALS
Magnetic flux was measured B(Gauss) experimentally and
compared with OPERA 3D simulation results.
Gaussmeter
H
Spacer´s Height
Point of the magnetic flux
density measurement,
B(Gauss).
Simulation Validation
ENGINEERING OF
METALLURGICAL
AND MATERIALS
Comparison of B (Gauss) for different
spacer´s values (H).
∆
H
Spacer´s Height
(mm)
B (G)
(Experimental)
B (G)
(OPERA3D)
to 2mm
(Error)
8
185
198
6,5%
7
220
230
4,3%
6
246
270
8,8%
5
295
319
7,5%
There is a good correspondance between the experimental and
simulated results for the magnet initially chosen (35 MGOe).
The maximum error in the table is lower than 10 %.
Interference Among Probes
The interference between the sensors signals was evaluated in the
X and Y directions in the crowns.
B
LB
X direction
Moving sensor
Fixed sensor
1.00
0.98
B/LB
ENGINEERING OF
METALLURGICAL
AND MATERIALS
0.96
0.94
Y(mm)
0.92
0.90
X(mm)
0
20
40
60
80
100
120
140
160
X (mm)
The crowns
X(mm)
The influence can be
neglected
for
X
greater than 70mm
Interference Among Probes
•Vertical interference among sensors.
ENGINEERING OF
METALLURGICAL
AND MATERIALS
Z (mm)
1
2
3
Z (mm)
1
2
3
Y (mm)
1.08
A
Probe 2
Probe 3
1.06
B
Probe 1
Probe 3
1.06
1.04
B/LB
B/LB
1.08
1.02
1.04
1.02
1.00
1.00
0
2
4
6
Z (mm)
8
10
0
2
4
6
Z (mm)
8
10
Interference Among Probes
•Vertical interference among sensors. From figure A the moving sensor
is placed at the extremity of the
Z (mm)
crown. There is a clear
interference on sensors 1 and 2,
1
2
3
but not for the sensor 3. When
the moving sensor is in a middle
position in the crown, figure B,
Z (mm)
the moving sensor does
interfere on the signal sensors 1
1
2
3
and 3 but in the same way. This is
the real case in the PIG crown.
ENGINEERING OF
METALLURGICAL
AND MATERIALS
1.08
A
Probe 2
Probe 3
1.06
B
Probe 1
Probe 3
1.06
1.04
B/LB
B/LB
1.08
1.02
1.04
1.02
1.00
1.00
0
2
4
6
Z (mm)
8
10
0
2
4
6
Z (mm)
8
10
Material Loss
Is it possible to detect defects beneath the paraffin layer?
From the simulated results below it is possible to define 4 mm as the
maximum paraffin thickness enabling crack detection.
1.05
1.00
H = 0mm
H = 2mm
H = 4mm
H = 6mm
0.95
Field
aquisicion
point (Hall
Sensor)
B/LB
ENGINEERING OF
METALLURGICAL
AND MATERIALS
0.90
0.85
0.80
Wax
H
0.75
-30
B : Magnetic flux density
Defect depth = 3 mm
-20
-10
0
10
20
30
Distance to the center of the defect (mm)
B and LB Magnetic flux density of the
sensor WITH and WITHOUT paraffin layer
Excessive Material
A weld bead could represent what is called “excessive material”
or a bump on the pipe inner surface.
FEMM – 2D
L
L
What is the effect of the
size of the probe tip on the
sensor response when it
moves over a weld bead?
The smaller the probe tip
the higher the possibility
of surface scractching.
L= 4mm is the ideal probe
tip dimension.
1,05
1,00
B/LB
ENGINEERING OF
METALLURGICAL
AND MATERIALS
0,95
0,90
L = 16mm
L = 8mm
L = 4mm
L = 2mm
0,85
0,80
-30
-20
-10
0
X (mm)
10
20
30
Study of excessive material
The excessive material could be interpreted as a weld.
FEMM – 2D
L
L
10 mm
1,05
It was studied the effect
of the size of the base of
the sensor in the
response when passing
through an excessive
material metal area.
1,00
B/LB
ENGINEERING OF
METALLURGICAL
AND MATERIALS
0,95
0,90
L = 16mm
L = 8mm
L = 4mm
L = 2mm
0,85
0,80
-30
-20
-10
0
X (mm)
10
20
30
Prototype Probe Construction
ENGINEERING OF
METALLURGICAL
AND MATERIALS
B
MLX90242: Linear Hall Effect Sensor
NdFeB magnets:grade N35
Material of the Structure: Stainless Steel (non magnetic)
Prototype Probe Construction
ENGINEERING OF
METALLURGICAL
AND MATERIALS
Prototype Testing
ENGINEERING OF
METALLURGICAL
AND MATERIALS
A carbon steel plate with known defects was used.
The focus was not to reconstruct the shape of the defect but to
evaluate the ability of the sensor to detect the different defect depths.
Depth of the circular defects = 3,6,9,12,15 mm
Displacement direction
Dielectric material used as Wax
H=1mm of wax´s deposition
H=0, without Wax
deposition
H=4mm of wax´s deposition
Prototype Probe Test
ENGINEERING OF
METALLURGICAL
AND MATERIALS
A carbon steel plate with defects was used. The focus was to assess
the ability of the sensor to detect defect at different depths.
Circular defect depth = 3, 6, 9, 12, 15 mm
Scanning direction
Dielectric material used to simulate paraffin
Thickness H
H=1mm
H=0, without paraffin
layer
H=4mm
3,0
3,0
3,0
Hall Sensor Sensibility
2,9
2,8
22,5mV/mT
2,8
2,7
2,6
Hall Sensor Sensibility
Hall Sensor Sensibility
2,8
22,5mV/mT
22,5mV/mT
2,6
2,6
2,4
2,4
B
None (0mm)
None (0mm)
2,3
2,2
B
Paraffin wax deposit height
2,4
2,2
Paraffin wax deposit height
1mm
1mm
2,1
2,0
B
2,5
Paraffin wax deposit height
2,2
2,0
2,0
1,8
1,8
4mm
4mm
1,9
1,8
1,7
500
600
700
Tempo
3,0
2,9
2,8
2,7
2,6
2,5
2,4
2,3
2,2
2,1
2,0
1,9
1,8
1,7
500
550
4000
Tempo
600
2000
Tempo
Sensitivity Analysis
SENSOR
SENSOR
WAX
ENGINEERING OF
METALLURGICAL
AND MATERIALS
STEEL
STEEL
3,0
3.0
3.0
Hall Sensor Sensibility
2.9
2.8
22,5mV/mT
2.8
2,8
22,5mV/mT
22,5mV/mT
2.7
2.6
Hall Sensor Sensibility
Hall Sensor Sensibility
2.6
2,6
2.4
2,4
B
None (0mm)
None (0mm)
2.3
2.2
B
Paraffin deposit height
2.4
Paraffin deposit height
2.2
Paraffin wax deposit height
2,2
1mm
1mm
2.1
2.0
B
2.5
2.0
2,0
1.8
1,8
4mm
4mm
1.9
1.8
1.7
600
700
Hall Sensor Sensibility
3,0
Hall Sensor Sensibility
55mV/mT
2,9
55mV/mT
2,9
2,8
2,8
Paraffin wax deposit height
2,7
2,7
direto
direto
direto
2,6
2,5
1mm
1mm
1mm
2,5
1
1
Hall Sensor Sensibility
2,3
2,3
2,2
2,2
2,1
2,1
2,0
2,0
1,9
1,9
1,8
1,8
1,7
1,7
550
2,6
Paraffin wax deposit height
2,4
2,0
1,8
600
600
700
800
800
3,0
72,5mV/mT
2,9
2,8
2,7
2,6
2,5
Hall Sensor Sensibility
72,5mV/mT
3,0
2,8
2,9
Hall Sensor Sensibility
2,7
2,8
72,5mV/mT
2,6
2,7
2,5
2,6
2,5
2,4
1
2,4
Paraffin wax deposit height
2,2
2,1
direto
direto
direto
2,0
1,9
2,4
2,3
Paraffin wax deposit height
2,2
2,1
1mm
1mm
1mm
2,0
1,9
1,8
1,8
350
400
Time
450
2,3
2,2
Paraffin wax deposit height
2,1
2,0
4mm
4mm
4mm
1,9
1,8
1,7
1,7
900
Tempo
Tempo
Hall Sensor Sensibility
2,3
4mm
4mm
4mm
2,2
Tempo
2,9
55mV/mT
2,8
2,4
500
1
3,0
Paraffin wax deposit height
2,6
2,4
3,0
Tempo
Tempo
1
3,0
2000
4000
Tempo
B
500
400
450
500
Tempo
550
1,7
200
300
Tempo
400
First tests with array system
ENGINEERING OF
METALLURGICAL
AND MATERIALS
Setup ready at Pipeway Company.
Sensors Array
Nylon blocks used
to simulate
paraffin
Results
ENGINEERING OF
METALLURGICAL
AND MATERIALS
1.00
Experimental data
Simulated [FEMM] data
Simulated [OPERA] data
B/LB
0.95
0.90
0.85
0.80
H: Different heights of nylon blocks (to simulate paraffin)
0.75
0.0
0.5
1.0
1.5
2.0
h [mm]
2.5
3.0
3.5
4.0
4.5
Conclusions
ENGINEERING OF
METALLURGICAL
AND MATERIALS
A prototype probe for detection and sizing
paraffin height at the inner surface of steel
pipelines was designed.
It also showed to be able to detect defects
on the pipeline located beneath the paraffin
layer.
Future works
ENGINEERING OF
METALLURGICAL
AND MATERIALS

To evaluate the answer of an array of
sensors in a precision setup.
ENGINEERING OF
METALLURGICAL
AND MATERIALS
Questions?
Professor João M. A. Rebello
[email protected]
LNDC - Laboratory of Nondestructive Testing, Corrosion
and Welding
Department of Metallurgical and Materials Engineering,
Federal University of Rio of Janeiro, Rua Pedro Calmon
SN, Cidade Universitária. CEP 21941-596, Rio de Janeiro,
Brazil.