PRELIMINARY STUDY OF MAGNETIC FLUX LEAKAGE ON TUBE INSPECTION

Noorhazleena Azaman, Ilham Mukriz Zainal Abidin and Nurul A'in Ahmad Latif Leading Edge Non Destructive Testing Group (LENDT), Industrial Technology Division (BTI),

Malaysian Nuclear Agency (Nuclear Malaysia), Bangi, 43000 Kajang, MALAYSIA hazleena@nm.gov.my

Abstract

Magnetic Flux Leakage (MFL) is an advanced NDT technique which has the inspection capability in wall loss detection and measurement of sharp defects such as pitting, grooving and circumferential cracks in ferromagnetic samples. The working principle of MFL involves the induction of magnetic field in the part to be tested and the response or signal produced is analysed to determine the presence and characteristics of defects. In this paper, simulation and experimental work on wall loss detection in steel tube using MFL were carried out. The simulation was performed using Comsol software and followed by experimental work using MFL system for validation. The results from the simulation and experiment indicates that variation of the groove defect affect the magnetisation and the output of the MFL signal is related to change of flux caused by the detection of wall loss.

Abstrak

Magnetic Flux Leakage(MFL) merupakan satu teknik NDT termaju yang mempunyai keupayaan pemeriksaan bagi pengesanan penipisan dinding dan pengukuran kecacatan tajam seperti pitting, grooving dan retak lilitan dalam sampel feromagnet. Prinsip kerja MFL melibatkan aruhan medan magnet di bahagian yang akan diuji dan tindak balas atau isyarat yang dihasilkan dianalisa untuk menentukan kewujudan dan ciri-ciri kecacatan. Dalam kertas kerja ini, simulasi dan ujikaji pengesanan penipisan dinding dalam saluran tiub keluli menggunakan MFL telah dijalankan. Simulasi ini dilakukan dengan menggunakan perisian Comsol dan diikuti dengan eksperimen menggunakan sistem MFL bagi tujuan pengesahan keputusan. Hasil keputusan daripada simulasi dan eksperimen menunjukkan bahawa perubahan kecacatan alur menjejaskan pemagnetan dan output isyarat MFL mempunyai kaitan dengan perubahan fluks yang disebabkan oleh pengesanan kehilangan dinding. Keywords /Kata kunci: Magnetic flux leakage, nondestructive testing, ferromagnetic, magnetic field /Magnetic flux leakage, ujian tanpa musnah, feromagnet, medan magnet 1.0 INTRODUCTION

Magnetic flux leakage (MFL) technique is widely used for non-destructive testing for detection and evaluation of surface a sub-surface defects in ferromagnetic objects such as oil and gas pipelines, storage tank floors and wire ropes [1]. This technique detect gaps, cracks and flaws by the help of the magnetic field supplied by a current source. The aim of non-destructive testing methods is to obtain some of the information about the specimen or object under test without any physical impact to the material. In this technique, only ferromagnetic materials have been used in this measurement and it is also well known that the ferromagnetic materials drive the magnetic flux. The principle behind this technique is a ferrite core is used to magnetise the ferromagnetic material components such as steel and used amplitude of detected MFL signals to determine empirically the flaw volume. In MFL, permanent magnets or DC electromagnets are applied in generation of magnetic field in order to magnetise the ferromagnetic specimen under inspection to saturation [2]. MFL is used to estimate the amount of material loss of corroded steel or at area where there is missing metal and the magnetic field will leaks from the steel. In an MFL tool, a magnetic detector or magnetic sensor such as using coil or a hall sensor is placed between the poles of the ferrite core to detect the leakage field. One of the advantage of MFL technique is its ability to simulate the leakage field from defect using COMSOL Multiphysics software. The simulation enables the study of field or defect interactions that will helps better

understanding and effective utilisation of the MFL technique. length, width, and shape of the defecthandled using this technique.

Below (equation 1) is the mathematical eq(ɸ is magnetic flux, B is the magnetic fieldbetween magnetic field and vector perpendicular to the area.field passes (flows) through an area and the unit for the magnetic flux is below shows the effects of magnetic flux when the presence of flaw in tube.

(a) Without flaw

Figure 1: Schematic diagram of general MFL set up and effects of magnetic 2.0 MFL SYSTEM AND EXPERIMENTAL SET UP MFL system was developed in this work and used hall sensor to detect leakage fluxes from defects in tubetechnique is capable to detecting the defects up to MFL technique and the results will be published elsewhere. The this system is shown in figure 2. The sensor probe consists of a core elctromagnetFigure 3 shows MFL sensor probe where it consists of electromagnet to apply magnetising field and hall sensor to detect the vertical component of leakage field. This measurement system is based on a National Instruments Data Acquisthe computer through an USB interface) and National Instruments LabVIEW software package. The leakage of magnetic flux is measured by a detecting coil or a sensor and its output is connected to the NIthe measured signal can be post-processed by LabVIEW procedures and function. In this work, simulation and the MFL system were developed and the test were carried out on a steel tube grooves with different depths as shown on figure According to this technique, MFL technique produced the flux leakage variations at the defect regions and the flux leakage can be read using Hall sensor. A number of simulations were carried out in order to evaluate the performance of the tube inspection uMATLAB. The tube containing different geometry of defect with various depth were used as a test specimen. In order to measure the magnetic flux density distribution, z component of magnetic fluxdefect can be characterised by utilising the raw data which is obtain from COMSOL Multiphysics software. The various simulation studies and analyses made in this work is for comparing the resultant signal of magnetic flux leakage for defective and non-defective part. In this research, MFL system was developed and the MFL sensor probe was fabricated. While the probe scanned along the tube using the MFL system, the data of flux density were being stored and displayed in the form of images using LabVIEW and MATLAB.

Ferrite core

N

tion of the MFL technique. The detected leakage field depends on the depth, defect. Tubing up to 406 mm in diameter and 19 mm

is the mathematical equation that relates to the magnetic field, area and angle of the surfaceis the magnetic field, A is the area that magnetic field pass through and

between magnetic field and vector perpendicular to the area. Magnetic flux is a measure how much the magnetic field passes (flows) through an area and the unit for the magnetic flux is Tesla, Tm2 or Wb (Weber).below shows the effects of magnetic flux when the presence of flaw in tube.

ɸ = BA cos θ ………………………………….

(a) Without flaw (b) With flaw

Schematic diagram of general MFL set up and effects of magnetic flux on flaw tube.

MFL SYSTEM AND EXPERIMENTAL SET UP

MFL system was developed in this work and used hall sensor to detect leakage fluxes from defects in tubecapable to detecting the defects up to 19mm. The test were carried out on a steel tube using this

technique and the results will be published elsewhere. The MFL system and the experimental setup for the . The sensor probe consists of a core elctromagnet, coppe

shows MFL sensor probe where it consists of electromagnet to apply magnetising field and hall sensor to detect the vertical component of leakage field.

This measurement system is based on a National Instruments Data Acquisation card (NI-DAQ card connected to the computer through an USB interface) and National Instruments LabVIEW software package. The leakage of magnetic flux is measured by a detecting coil or a sensor and its output is connected to the NI

processed by LabVIEW procedures and function.

In this work, simulation and the MFL system were developed and the test were carried out on a steel tube as shown on figure 4 and table 1 below.

According to this technique, MFL technique produced the flux leakage variations at the defect regions and the flux leakage can be read using Hall sensor. A number of simulations were carried out in order to evaluate the performance of the tube inspection using COMSOL Multiphysics software and the signal displayed in MATLAB. The tube containing different geometry of defect with various depth were used as a test specimen. In order to measure the magnetic flux density distribution, z component of magnetic flux defect can be characterised by utilising the raw data which is obtain from COMSOL Multiphysics software. The various simulation studies and analyses made in this work is for comparing the resultant signal of magnetic flux

defective part.

In this research, MFL system was developed and the MFL sensor probe was fabricated. While the probe scanned along the tube using the MFL system, the data of flux density were being stored and displayed in the form of

ages using LabVIEW and MATLAB.

Magnetic

flux

Tube

Magnetic sensor

Ferrite core Ferrite core

S S N

The detected leakage field depends on the depth, wall thickness can be

ic field, area and angle of the surface. is the area that magnetic field pass through and θ is the angle

measure how much the magnetic or Wb (Weber). Figure 1

…………………………………. (equation 1)

flux on flaw tube.

MFL system was developed in this work and used hall sensor to detect leakage fluxes from defects in tube. MFL mm. The test were carried out on a steel tube using this

experimental setup for the , copper coil and hall sensor.

shows MFL sensor probe where it consists of electromagnet to apply magnetising field and hall sensor

DAQ card connected to the computer through an USB interface) and National Instruments LabVIEW software package. The leakage of magnetic flux is measured by a detecting coil or a sensor and its output is connected to the NI-DAQ card, and

In this work, simulation and the MFL system were developed and the test were carried out on a steel tube having

According to this technique, MFL technique produced the flux leakage variations at the defect regions and the flux leakage can be read using Hall sensor. A number of simulations were carried out in order to evaluate the

sing COMSOL Multiphysics software and the signal displayed in MATLAB. The tube containing different geometry of defect with various depth were used as a test specimen. In

density was used. The defect can be characterised by utilising the raw data which is obtain from COMSOL Multiphysics software. The various simulation studies and analyses made in this work is for comparing the resultant signal of magnetic flux

In this research, MFL system was developed and the MFL sensor probe was fabricated. While the probe scanned along the tube using the MFL system, the data of flux density were being stored and displayed in the form of

Figure

Figure

Figure 4: Steel tube with three grooves

Power Amplifier

Multifunction Generator

L = 30mm

OD1=26.1mm

Groove 1 (d1)

Figure 2: Photo of the Magnetic Flux Leakage system

Figure 3: Fabricated MFL probe for tube inspection

: Steel tube with three grooves and different depth

Steel tube with different depth of groove

Computer

Differential Amplifier

DC remover

OD3=24.1mmOD2=25.1mm

L = 30mm

Groove 2 (d2) Groove 3

Steel tube with different

=24.1mm

Groove 3 (d3)

Table 1: Dimension of steel tube

Dimension of steel tube: Tube thickness : 2.1mm Outer diameter (OD) : 27.1mm Inner diameter (ID) : 22.9mm Length of groove (L) : 30mm

Defect Identification

Outer diameter (OD), mm

Thickness (mm)

Percent of groove (%)

Groove 1 (d1) 26.1 1.6 28

Groove 2 (d2) 25.1 1.0 52

Groove 3 (d3) 24.1 0.6 76

Three grooves with an outer diameter and inner diameter as table 1 above and different grooves of depth were used in this study that correponds to 28%, 52% and 276% material loss respectively. In this work, simulation and experimental were carried on. 3.0 RESULT AND DISCUSSION 3.1 Simulation results

COMSOL Multiphysics software has been used for 2D modeling. Figure 5 shows the geometry consists of ferrite core, coil and steel tube. The ferrite core is used for magnetic induction to magnetise the tube. The COMSOL Multiphysics combined with the MATLAB used efficiently in the research work, since the geometry, the physics, finite element mesh and the post processing can be realised by the functions of COMSOL.

a)

b)

c)

d)

Figure 5: (a) Signal for free defect, b) Signal for defect d1, c) Signal for defect d2, d) Signal for defect d3

The data displayed in COMSOL software were analyses together with MATLAB and the result are as shown from graph 1, 2 and 3. Graph 1 shows the signal for raw data from four simulation that has been done, while graph 2 are the differential signal for the three defects on tube subtract the data from free defect and lastly graph 3 shows the signal for the sensitivity of the sensor output at different sensor positions. From graph 2, the numbers of magnetic flux density at the line point of the sensor for defect groove 1 is higher compared to the defects 2 and 3. Its means the more number of magnetic flux passes through the area along the line at defect 1.

Graph 1: Graph from raw data

Graph 2: Graph from differential signal Graph 3: Graph from sensitivity of sensor output

i) Experimental results

Graph

The result from the graph 4 shows the higher signalmuch higher than others defects. measure smooth grooves. High amplitude signal gives the answer to this work where the leakage of magnetic flux is concentrated at the groove that have the high depth. determined the flaw volume.

4.0 CONCLUSION

The MFL is useful tool that allows the rapid monitoring of large surface areas and corrosion pits. The signals of the MFL density were analysed. The magnetic flux density were calculated using COMSOL software for the tube with different grooves and diameter. The results showa method to identify the flaws in ferromagnetic tubes of different dimensions. will increase with the rise in defect depth.and limitations of the MFL tool.

5.0 ACKNOWLEDGEMENTS

The authors wish to express their sincere thanks to for the knowledge and idea that has been given to work. 6.0 REFERENCES

[1] W. Sharatchandra Singh, S. Thirunavukkarasu, S. Mahadevan, BPCRCKMkhdhdTJk B.P.C. Rao, C. K. Mukhopadhyay and T. JayakumarModeling of Magnetic Flux Leakage TechniqeforDetectionofDefectsin Technique for Detection of Defects in Carbon Steel Plates' Excerpt from the Proceedings of the COMSOL Con

Graph 4: Signal of defect groove from MFL system

shows the higher signal amplitude is at defect groove 3 where the depth of groove is The hall sensors measure the absolute axial magnetic field are bet

High amplitude signal gives the answer to this work where the leakage of magnetic flux is concentrated at the groove that have the high depth. The higher amplitude of detected MFL signals

The MFL is useful tool that allows the rapid monitoring of large surface areas and corrosion pits. The signals of the MFL density were analysed. The magnetic flux density were calculated using COMSOL

erent grooves and diameter. The results shows that the 2D analysis of geometry is a method to identify the flaws in ferromagnetic tubes of different dimensions. The intensity of the leakage field will increase with the rise in defect depth. Finally the success of analysing MFL data depends on the capabilities

ACKNOWLEDGEMENTS

The authors wish to express their sincere thanks to Professor Dr Yong Li from University Jiafor the knowledge and idea that has been given to us and also everyone who has given extensive support for this

W. Sharatchandra Singh, S. Thirunavukkarasu, S. Mahadevan, BPCRCKMkhdhdTJk B.P.C. Rao, C. K. ay and T. Jayakumar, 'ThreeDimensionalFiniteElement Three-Dimensional Finite Element

Modeling of Magnetic Flux Leakage TechniqeforDetectionofDefectsin Technique for Detection of Defects ' Excerpt from the Proceedings of the COMSOL Conference 2010 India.

d3 d2 d1

3 where the depth of groove is ll sensors measure the absolute axial magnetic field are better to

High amplitude signal gives the answer to this work where the leakage of magnetic amplitude of detected MFL signals

The MFL is useful tool that allows the rapid monitoring of large surface areas and corrosion pits. The signals of the MFL density were analysed. The magnetic flux density were calculated using COMSOL multiphysics

that the 2D analysis of geometry is he intensity of the leakage field

he success of analysing MFL data depends on the capabilities

Yong Li from University Jiaotong, Xi'an China us and also everyone who has given extensive support for this

W. Sharatchandra Singh, S. Thirunavukkarasu, S. Mahadevan, BPCRCKMkhdhdTJk B.P.C. Rao, C. K. Dimensional Finite Element

Modeling of Magnetic Flux Leakage TechniqeforDetectionofDefectsin Technique for Detection of Defects ference 2010 India.

[2] Y Li, GY Tian, S Ward, (2006), Numerical Simulation On Magnetic Flux Leakage Evaluation At High Speed School of Computing & engineering, University of Huddersfield Queensgate, Huddersfield, UK, NDT & E International 39 (5), 367-373.

[3] Luming LI, Songling Huang, Peng Zheng and Karen Shi (2002), Evaluation of Surface Crakcks using

Magnetic Flux Leakage Testing, NDT Centre, Mechanical Engineering Depatment Tsinghua University, Beijing, China, 319-321.