Ebook: Electromagnetic Nondestructive Evaluation (XVI)

This volume contains selected papers from the 17th International Workshop on Elecromagnetic Nondestructive Evaluation (ENDE), which was held in Rio de Janeiro, Brazil from July 29 to August 1, 2012.

ENDE workshop is an important event for all scientists with interests in nondestructive testing. The first workshop took place in 1995 in London UK, and has been followed by workshops held in various parts of the world, but this is the first time this workshop series has come to Brazil and to a Latin American country.

The aim of the workshop is to bring together scientists from universities and research institutions conducting in-depth research into the basics of electromagnetic nondestructive evaluation on the one hand, and engineers presenting practical problems and industrial applications on the other hand. ENDE workshops also aim to provide a forum for discussing recent developments in the growing field of electromagnetic nondestructive evaluation methods. The workshop emphasizes both the basic science and early engineering developments in the field, so experts from theoreticians to research engineers in industry are encouraged to participate.

The workshop was jointly organized by Federal University of Rio de Janeiro, in cooperation with the Brazilian Society for Nondestructive Testing and Inspection (ABENDI), which is a NGO supported by Brazilian Companies, and whose focus is on the certification and training of people in NDT.

Eighty seven participants from Argentina, Brazil, Colombia, France, Germany, India, Italy, Japan, Korea, Poland, Slovak Republic, Spain, United Kingdom, United States, South Africa and Switzerland, were officially registered. There were eighty seven oral and poster presentations in all, one Keynote Conference and five Invited Lectures. From the papers submitted for publication, thirty six peer reviewed articles have been accepted and appear in these proceedings.

Keynote Conference

C.S. Camerini. Exploration of oil and gas on the Brazilian coast and the application of ENDs for integrity inspection.

Invited Lectures

1. F. Kojima, Reliability assessment of EMAT-NDE system for pipe wall thinning management.

2. N. Bonadeo, Electromagnetic NDT studies at Tenaris R&D: Mathematical modeling and experiments.

3. A. Tamburrino, Noniterative methods for real time imaging of conducting materials.

4. R. Sikora, Artificial intelligence systems in NDT.

5. T. Jayakumar, Advances in electromagnetic NDE techniques for materials evaluation.

I would like to express my sincere gratitude to all attendees, for their presentations and manuscripts, and the reviewers, for their time and energy which was necessary to produce a high quality volume of work, and the session chairs that help the ENDE 2012 organizers. Thanks are due to the members of the standing committee, and especially to the co-editors of this book, Professors Fumio Kojima and Tomasz Chady.

I am indebted to the students and researchers at the LNDC, Non Destructive Testing, Corrosion and Welding Laboratory – UFRJ, for their fantastic help and organizing efforts. Lastly, Dr Maria Cristina Lopez Areiza deserves my special recognition and acknowledgement for the hard work and invaluable assistance put forth during the preparation of the ENDE 2012 workshop.

Financial support coming from the Brazilian Agencies, CNPq (National Council for Scientific and Technological Development) and CAPES (Coordination for Education of Graduated Personnel) is acknowledged.

Joao M.A. Rebello

ENDE 2012 Chairman

A reliability assessment method for pipe wall thinning management using electromagnetic acoustic transducer (EMAT) is discussed. In the series of laboratory experiments, we collect the measurement data for carbon steel test pipes fabricated to simulate flow-accelerated corrosion (FAC). Those data are used for the reliability assessment of EMAT based NDE system. The reliability of the system can be derived from the so-called probability of detection (POD). The POD evaluation is carried out and evaluated using a signal response model. The final part of this paper is devoted to the applicability and the validity of the POD based assessment method considered here.

The purpose of this paper is to compare performances of various sensing elements in eddy current non-destructive inspection. A new eddy current testing probe is designed to compare detection and resolution capabilities of different sensors. Three magnetic sensors, specifically GMR, AMR and Fluxgate, and a standard induction coil are used as a sensing element. The excitation part of the probe is realized by two circular coils connected in series, but magnetically opposite to drive uniform eddy currents. The comparison of these sensors is realized on non-magnetic conductive plate where EDM notches with different depths are introduced. The numerical simulations and experimental measurements are performed under the same conditions for this purpose. The results are presented and discussed in the paper.

Eddy current techniques are often used to detect and, in some advanced applications, to characterize fatigue surface-breaking cracks appearing in mechanical parts. One of the major difficulties in this approach is to reliably separate the crack contribution to the signal from the others (liftoff, edge effects, material property variations). This is particularly prominent in weld inspection. We present here a novel sensor consisting in an array of 200 micro-Hall sensors on a CMOS chip aligned over a length of 2 mm with a 10 μm pitch. This sensor, which we call a “high-resolution magnetic camera”, gives a direct measurement of the perturbation of the local magnetic field caused by the modification of the Eddy currents flow due to the surface-breaking crack. We show that using this technique, a crack has a very distinctive signature which can be easily distinguished from all other contributions. Finite element simulations are used to understand the signal response of different cracks. In addition, this system can perform “micro-magnetoscopy” on magnetic materials by giving a fine picture of the local variations of the magnetic field caused by the presence of a defect.

Tactile sense is an essential sense in our daily lives, and we usually operate or sense objects using our tactile sense. Recently, many touch devices are being produced. They are so intuitive and easy to use that we use them on smartphones or tablet PCs. This is associated with an expectation that combinations of tactile sensor and display can contribute to various applications. Therefore, we are studying a magnetic type tactile sensor which has a simple and tough structure. The feature of the sensor is a softness of touch area. The sensor measures three-axis displacement generated on the surface. In this paper, we describe the details of the structure and show the prototype sensor. The hysteresis characteristic and the repeat accuracy are shown through the laboratory experiment. We verify that the sensor has low-error in measuring displacement.

In this study, the feasibility of using magnetostrictive sensors for measuring magnetic fields was explored. The sensors were fabricated using optical fibers containing Bragg gratings attached to a magnetostrictive material in bulk or in the form of particulate composites. The responses under effects of uniform and gradient magnetic field were analyzed. They have shown a strong anisotropy regarding the shape of the magnetostrictive material when it is in bulk form in comparison to when it was made of the particulate composites.

In this study, defect shape evaluation based on the magnetic flux leakage (MFL) technique for inclined defects was conducted. A magnetic dipole method was used to represent the defect surface, and the Tikhonov regularization technique was used to reconstruct the magnitude of magnetic dipole moments. The result of the inverse analysis using magnetic flux obtained from numerical calculation showed good agreement with shape of the original defect. In case of inverse analysis using magnetic flux obtained from experimental measurement, the L-curve did not indicate clear kink point. However, applying the optimal regularization parameter defect shapes could be reconstructed with acceptable errors.

From the electrical point of view, CFRP composites are characterized by an anisotropic uniaxial conductivity posing difficult problems for their electromagnetic description and numerical modeling. In this paper, the previous computational model, based on a magneto-quasi-stationary integral numerical formulation, is suitably extended to take into account the electro-quasi-static contribution as well as the magneto-quasi-static part, mainly dueto the presence of low values of the conductivity perpendicular to the fibres. The formulation is well suited for a broad spectrum of frequencies being possible to take into account low frequency applications as well as cases in which wave propagation is no more negligible. A simple example is analyzed showing the high potential of the proposed approach.

The Magnetic flux leakage (MFL) method is commonly employed for the non-destructive evaluation (NDE) of steel plates used in the construction of oil storage tanks and pipelines, where large areas have to be covered within a short time. Due to fundamental characteristics of the MFL method, very narrow and deep pipe-type defects can produce very similar signals to wide and shallow lake-type defects. This inherent uncertainty of MFL can cause dangerous misinterpretations of the measurement data as crucially deep defects could remain unnoticed.

This paper proposes a numerical method for the computation of the worst-case (WC) solution candidate to the inverse MFL problem in terms of defect depth, which produces MFL signals differing from the original data only within the limits of the measurement uncertainty. A fully nonlinear magnetostatic FEM model is used to create Taylor series expansions of the model error w.r.t. the multi-parametric surface profile in 2D, which is then iteratively changed into the WC estimate. The algorithm is explained and its effectiveness is illustrated for a particular simulation example, showing the resulting WC depths under different conditions.

This paper reports the most recent investigations on the detection of deep lying defects in nonmagnetic materials using Lorentz force eddy current testing (LET). Whereas classical eddy current testing techniques are limited due to the frequency dependent skin depth LET is expected to overcome this particular disadvantage by exploiting the induction of eddy currents by moving an electrical conductor in the vicinity of a permanent magnet. The detection of deep lying defects is examined at relatively high testing velocities and with high spatial resolution. Without target-oriented improvements defects in a depth of up to 6 mm under the surface can be detected and localized.

Real-time imaging is a either a mandatory or an attractive feature in practical applications of Nondestructive Evaluation of structures and materials. To date, real-time imaging in commercial systems is achieved only by means of calibration charts or pre-computed databases containing the “signatures” corresponding to a finite number of defect configurations and inspection parameters. Situations beyond these sets of signatures cannot be treated and may cause misinterpretations of the data. This paper is a review on the electromagnetic imaging of defects in conducting materials by Noniterative Imaging methods. The class of Noniterative Imaging methods has attracted considerable interest because is suitable for real-time imaging. The paper discuss in detail the Monotonicity Imaging Method (MIMe), that is a specific Noiterative Imaging method successfully applied in Electrical Resistance Tomography and Eddy Current Tomography.

In this work, we perform three-dimensional finite element simulations of a solid electrically conducting bar with pre-defined material defects moving across static magnetic field lines of a permanent magnet. This study is motivated by the novel non-destructive testing technique called Lorentz force eddy current testing (LET), which is based on the measurement of the Lorentz forces acting on the magnet. In particular, we perform parametric studies to quantify the effect of various parameters on perturbations of the Lorentz force, such as: defect depth, testing speed, electrical conductivity, and magnet position. The analyses provide reference results to understand the feasibility and testing capabilities of LET. The experimental validation of the numerical results is presented as well.

This paper presents 3D finite element simulations of the In-Service Inspection of ferromagnetic steam generator tubes of Fast Breeder Reactors. 3D FEM models, resulting in large scale electromagnetic eddy current computations with up to 5,100,000 2nd order tetrahedrons are analyzed using parallel simulations with 1024 CPUs on Fujitsu BX900 and using an in-house developed code 3D-RFECT. Validations of FEM simulations with experimental measurements for helical tubes are conducted for a large model of SG tubes support plate and three SG tubes attached to it at various frequencies. ISI of SG tubes using ECT and detection of defects for large SP models are validated trough 3D FEM simulations.

A fast simulation tool of Eddy Current Testing (ECT) inspections, based on a Boundary Element Method (BEM), is proposed in this paper. Simulated signals consist of impedance variation of a coil due to multiple narrow cracks located in conductive planar multilayered structures (MLS). The cracks can have various shapes or orientations and can lie in any layer of the structure inspected. After a presentation of the method, validations results, consisting of comparison of simulated signals with experimental data and results obtained with finite element calculations, are presented. In all the problems tested, a very good accuracy and a very low computation time have been observed. Further perspective and conclusions of this work are detailed at the end of the paper.

A modal decomposition of 3D current sources is introduced in the finite integration technique (FIT) discretisation scheme for the solution of eddy-current testing problems in rotationally symmetric work-pieces. The original 3D problem is reformulated in this way in calculating the vector magnetic potential profile on the ρ – z plane for a number of angular modes. The interest of such decomposition relies on the simpler and conformal discretisation of the work-piece, the reduction of the numerical noise, as well as the smaller size of the linear system, which allows the simultaneous treatment of different probe positions.

We present an innovative method for case depth estimates, using Barkhausen noise, for AISI 1020 steel into which carbon had been diffused for various distances. Quenching after a short heat treatment resulted in a martensitic surface layer on top of a ferrite/pearlite nucleus. Case depths were measured by traditional destructive techniques. Barkhausen noise measurements were made and both the RMS Barkhausen pulse envelopes and the fast Fourier transforms (FFT) were obtained by numerical calculation. The areas of the Barkhausen pulses lend themselves to the construction of a calibration curve for determining the case depths. The FFT amplitudes were obtained as a function of frequency, and were associated with distance from the sample surface via the classical skin depth equation. Case depths can be estimated from a normalized power index, a quantity consistent with details of the sample microstructure. A knowledge of the material properties is especially important for determining the usefulness of this technique.

The paper presents a newly developed magnetic flux leakage transducer designated for a hollow axle testing. The transducer allows measuring three components of the magnetic residual field. Results of measurements obtained for planar and cylindrical samples with artificial defects are presented.

The paper carried a brief overview of artificial intelligence algorithms applicable to nondestructive testing. It focuses on three methods: artificial neural networks, fuzzy logic and rough sets. Selected examples of applications of these methods in digital radiography are given.

This paper is concerned with aging degradation of cable insulation used in airplane and power plants. Material degradation is associated with dielectric property. To prove availability of evaluating dielectric properties, first, a relationship between Q-factor and dielectric permittivity is identified. Secondly, to develop a method of identifying dielectric permittivity, mathematical model of the nondestructive testing for system is simply described by Maxwell's equations in two spatial dimensions. Thirdly, numerical scheme is developed using the finite-difference time-domain method (FDTD). Finally, parameters of dielectric property are estimated using information through numerical experiments by computational algorithm based on Markov Chain Monte Carlo (MCMC) method.

A three-dimensional topographic reconstruction system of the forge pattern surfaces through laser triangulation has been purposely designed and evaluated. With this work, the results of the topographic study of the forge pattern subjected to ion implantation and/or coated with Ti/TiN layer are being presented. Further, the system permits not only to monitor carefully the wear's performance of the pattern but also find the localization of the points at its highest efforts arising from the forge process as well as any defects caused by the variation of the treatment the piece have been exposed to.

This work presents practical results from a high lift-off stress measurement application on SAE 1060 steel using the Electromagnetic Acoustic Resonance (EMAR) technique. Resonance peak attenuation values were obtained for lift-offs up to 9.45 mm. A curve of the system's performance regarding lift-off variation is presented. Other sources of attenuation are discussed as well. Mechanical sources of attenuation, e.g., due to bonded materials, may affect the performance in a way that is indistinguishable from the magnetic coupling losses, thus reducing the maximum attainable lift-off.

Crack depth measurements using alternating current potential drop (ACPD) technique on shallow and finite length cracks are prone to errors. The influence of length of notch on depth measurements in stainless steel (SS) was addressed by the authors using an ‘Equivalent Resistive circuit model’ based on the parallel path length available for the current to flow around the defect, in addition to the path available below the defect. In addition to the path length, due to changes in electrical conductivity of the material, the variation in area of cross section needs to be considered. Working in this direction, a modified model is proposed which takes into account the length and area of cross section through which the current flows in the additional path. The proposed model is able to account for the observed variations in the notch depth with length in SS, brass and aluminum having conductivities of 1.41, 18.73 and 31.51 MS/m respectively.