Professional Documents
Culture Documents
Portable Lorentz Force Eddy Current Testing System With Rotational Motion
Portable Lorentz Force Eddy Current Testing System With Rotational Motion
Motion
Reinhard Schmidt1 , Jan Marc Otterbach1 , Marek Ziolkowski1,2 , Hartmut Brauer1 and Hannes Toepfer1
1 Advanced Eletrocmagnetic Group, Technische Universität Ilmenau, Ilmenau, Germany
2 Applied Informatics Group, West Pomeranian University of Technology, Szczecin, Poland
In this study we investigate a new portable Lorentz force eddy current testing (PLET) system using a diametrically magnetized
cylindrical magnet rotating around its axis of mass. In order to compare the Lorentz force and the electromagnetic torque used for
defect detection, three different model approaches are considered - the weak reaction approach and the quasi-static approach as
approximating approaches and the transient simulation as a reference one. The results are validated by measurements.
Index Terms—Eddy currents, finite element analysis, nondestructive testing, rotation of permament magnets, torque measurement
h i 0.5
(p)
∇ · J = ∇ · σ(−∇φ − µ0 v × ∇(ψm ) = 0. (5)
Force Fx (mN)
0
After separating φ and assuming a homogeneous conductivity
σ in the specimen apart from the defect, (5) changes to −0.5
5 mm
h i −1 10 mm
σ∇2 φ = −µ0 σ ∇ψm (p)
(∇ × v) − v(∇ × (∇ψm (p)
)) . (6) 15 mm
20 mm
−1.5
0 15 30 45 60 75 90
For the rotating specimen ∇ × v = −2 ωmag ez , which leads to
Angle ϕ (◦)
the final governing WRA equation for the electrical potential 0.5
φ
∇2 φ = 2µ0 ωmag ∂ψm
(p)
/∂z. (7) 0
Force Fy (mN)
−0.5
In case of QSA, the problem is described by the electrical
potential φ and magnetic vector potential A, and the following −1
5 mm
equation 10 mm
−1.5
15 mm
1 20 mm
∇×( ∇ × A − M) = σ(−∇φ + v × (∇ × A)). (8) −2
µ0 0 15 30 45 60 75 90
Angle ϕ (◦)
Additionally, the continuity of the current density ∇·J = 0 has
2.06
to be taken into account. For both approaches, the boundary
condition
n·J=0 Torque Tz (mNm)
(9) 2.04
5 mm
10 mm
15 mm
has to be fulfilled on the surface of the specimen in order to 20 mm
2.02
prevent current flowing out the specimen.
The resulting Lorentz force F acting on the magnet is
calculated using Newton’s third law by integrating the force 2
0 15 30 45 60 75 90
density fspec = J × B in the conductive specimen Angle ϕ (◦)
Z
F = −Fspec = − J × B dV. (10) Fig. 2. Lorentz force components Fx and Fy and torque Tz acting on the
V permanent magnet for different defect radii rdef as a function of the angle ϕ.
Solid lines denote WRA results and circles the results of QSA. Used setup:
The electromagnetic torque T acting on the permanent magnet Dmag = 20 mm, Hmag = 40 mm, µ0 M = 1.43 T, ωmag = -5 s−1 ,
is determined as h = 1 mm, R = 60 mm, H = 10 mm, Ddef = 4 mm, σ = 21 MS/m.
Z
T = −Tspec = − r × (J × B) dV. (11)
V V. C ONCLUSION AND OUTLOOK
We presented a new portable Lorentz force eddy current
IV. R ESULTS testing system using a diametrically magnetized magnet. The
In the simulation, the defect is rotating. The position of the studied WRA and QSA methods show good agreement for the
defect can be described by time-varying angle ϕ(t) = −ωmag t. analyzed angular velocity. Furthermore, the resulting signals
As the magnet is fixed and its magnetization is assumed to depend strongly on the defect’s position rdef . It allows to find
be in x-direction, the angle ϕ describes the angular position an optimal PM distance to the defect to maximize the signal
of the defect with respect to the x-axis. The diametrical perturbation due to the defect. In the full paper, the results of
magnetization of the permanent magnet leads to a periodical the WRA and QSA are compared with the transient solution
signal with an angular period of 180◦ . As a consequence, the and the numerical results will be validated by measurements.
force components Fx and Fy and the torque Tz acting on the
R EFERENCES
magnet can be expressed as a function of the angle ϕ for
different positions rdef of the defect (Fig. 2). For angles ϕ [1] H. Brauer and M. Ziolkowski, ”Eddy Current Testing of Metallic Sheets
with Defects Using Force Measurements,” Serbian Journal of Electrical
up to 90◦ the solid lines represent the results calculated by Engineering, Vol. 5, No. 1, pp. 11-20, 2008.
the WRA and the circled points represent the QSA results in [2] K. Weise, ”Advanced Modeling in Lorentz Force Eddy Current Testing”,
Fig. 2. First, it can be noticed that differences between the doctoral thesis, Technische Universität Ilmenau, Ilmenau, Germany, 2016.
[3] Y. Tan, X. Wang, and R. Moreau, ”An innovative contactless method
results of both methods are small. Furthermore, the change for detecting defects in electrical conductors by measuring a change in
of Lorentz force components and the torque strongly depends electromagnetic torque,” Measurement Science and Technology, vol. 26,
on the position rdef of the bypassing defect, i.e. its relative no.3, pp.1-8, 2015.
[4] T. Shiozawa, ”Phenomenological and Electron-Theoretical Study of the
position to the magnet. The perturbations of the torque Tz are Electrodynamics of Rotating Systems,” Proc. IEEE, vol. 61, pp.1694-1702,
the largest for the defect located at rdef = 10 mm. 1973.