Beruflich Dokumente
Kultur Dokumente
a r t i c l e
i n f o
Article history:
Received 28 May 2014
Received in revised form 15 August 2014
Accepted 21 August 2014
Available online 1 September 2014
Keywords:
Electromagnetic forming
Analytical model
Velocity measurements
a b s t r a c t
In this paper, an analytical model is used to design a coil, called a Uniform Pressure Actuator (UPA), for use
during electromagnetic forming (EMF) and magnetic pulsed welding (MPW) by combining and extending
past efforts by other researchers. The UPA offers increased forming efciency and repeatability, as well
as a robust design. Magnetic pressure applied to the workpiece and workpiece velocity are predicted to
ensure impact conditions are sufcient for MPW. The UPA is constructed and tested experimentally to
validate the accuracy of the analytical model, as well as verify assumptions made during modeling. The
coupling coefcient introduced in the magnetic analysis is experimentally determined and compared to
previous researchers values. Workpiece velocities for various energy levels, workpiece thicknesses, and
materials with various conductivities and densities are compared to analytical predictions and show good
agreement for the initial acceleration in the process. Workpiece velocity measurements were obtained
using a Photon Doppler Velocimetry (PDV) system, which provides a robust method for measuring velocities with submicron displacement and temporal resolution in the nanosecond range. Uniformity of the
workpiece deformation is also examined, which is an advantage of the UPA.
2014 Elsevier B.V. All rights reserved.
1. Introduction
Lightweight sheet metal components and assemblies formed
and joined electromagnetically are of interest in various industries
such as automotive, aerospace, and electronics. In electromagnetic
forming (EMF) and magnetic pulsed welding (MPW), a capacitor
bank is discharged into a coil which creates a magnetic eld in
close proximity to the workpiece. Eddy currents are induced in the
workpiece, which create an opposing magnetic eld and a repulsive
Lorentz force. The workpiece to accelerate away from the coil and
plastically deforms at high velocity rates, typically on the order of
100300 m/s. Benets of EMF include higher strains prior to failure, more uniform strain distributions, reductions in wrinkling,
active control of springback, and the possibility of local coining and
embossing (Psyk et al., 2011).
If a second, stationary workpiece is impacted at a critical velocity and angle, large pressures are achieved, and a solid state weld
is produced at the interface (i.e., a MPW process). Joining can be
achieved even between base metals of vastly different material
properties, and bond strength can be stronger than the parent components (Zhang et al., 2011). This allows joined components in an
Corresponding author. Tel.: +1 603 862 1811; fax: +1 603 862 1865.
E-mail address: bkinsey@unh.edu (B.L. Kinsey).
http://dx.doi.org/10.1016/j.jmatprotec.2014.08.019
0924-0136/ 2014 Elsevier B.V. All rights reserved.
assembly to be tailored to a specic function according to material properties, and assembly weight is reduced from removing
fasteners.
Typical control parameters of the EMF and MPW processes
include the charge energy of the capacitor bank, the coil geometry, the standoff distance between the workpiece, coil, and die,
etc. Typical measurements of the processes include primary current in the coil and induced current, velocity and displacement of
the workpiece.
Fig. 1 shows a schematic of the MPW process (Blakely, 2008).
Past applications and modeling of EMF and MPW have typically focused on crimping and expansion of tubular workpieces.
However, for lightweight automotive applications, forming and
welding of at sheets are of interest. Development of work coils
and process analysis for sheet forming has been limited due to
its complexity (Psyk et al., 2011). According to Daehn, this is
caused by a gap between experts in sheet metal forming and
pulsed power applications (Daehn, 2006). While some nite element analysis (FEA) packages exist that are capable of modeling
these processes, there is a lack of simplied analytical modeling
efforts, especially for sheet metal workpieces. Analytical modeling
is attractive for its simplicity and cost in effectively determining
e.g., an optimal coil design and workpiece velocities. Such modeling
acts to bridge this gap and promote electromagnetic applications in
manufacturing.
252
E. Thibaudeau, B.L. Kinsey / Journal of Materials Processing Technology 215 (2015) 251263
that the thickness to skin depth ratio is sufcient. This ratio was
found to be lower than specied in past research efforts to achieve
effective magnetic coupling between the coil and workpiece. Also,
the magnetic coupling coefcient was shown to be dependent on
the geometry of the coil and workpiece and independent of the
workpiece thickness and material properties.
2. Uniform pressure actuator
Prof. Glenn Daehns group from The Ohio State University developed multiple generations of UPA as shown in Fig. 3. Initially, their
construction consisted of soft copper windings around an insulating mandrel. Further generations moved to a stronger design, with
a thicker coil cross-section, machined out of a solid block of high
strength copper alloy. The coils were then potted in urethane for
electrical insulation.
Kamal (2005) presented an EMF coil design methodology comprised of an analytical model and FE magnetic modeling in Maxwell
2D. The deformation of the workpiece was solved separately in
LS-DYNA. The analytical model predicted rigid body workpiece
acceleration and is attractive for its simplicity and cost in effectively
determining an optimal coil design.
A large coil turn count results in a stronger magnetic eld, and
thus higher magnetic pressure. However, a small turn count allows
for a shorter rise time to peak current (Kamal, 2005). Therefore, it is
important to specify whether the coil design is to create maximum
pressure or maximum sheet velocity. Forming shallow features or
embossing requires larger magnetic pressures, whereas deep features with large deformations require higher velocities to create
large inertial effects and high impact pressures (Kamal and Daehn,
2007).
For the model presented here, the research by Kamal (2005) and
Al-Hassani (1975) is combined and extended. The goal is to determine, for a given sheet geometry, a coil geometry that maximizes
pressure or sheet velocity, while maintaining structural integrity
and uniform pressure on the workpiece.
Compared to Kamal and Daehn (2007), an alternative design
process is conducted that consists of an analytical model and a
FEA structural analysis. Analytical electromagnetic solutions from
Al-Hassani (1975) are implemented to determine the pressure distribution on the workpiece. The analytical model is then used
to evaluate the electromagnetic coupling coefcient for various
materials and thicknesses. Lastly, the modeling effort is veried
experimentally by measuring workpiece velocity for various energy
levels, workpiece thickness, and materials.
The analytical model calculates the initial workpiece acceleration of the EMF/MPW process. The magnetic pressure distribution
on the workpiece and rigid body workpiece acceleration are solved
for a given UPA geometry, workpiece length, and coil turn count.
A FE mechanical analysis determines if the resultant geometry is
strong enough to withstand the predicted magnetic pressure. If the
mechanical simulation shows material failure, the analytical model
is used to modify the geometry. For a given workpiece length, a
larger conductor cross-section, and thus lower turn count, produces
a more robust coil.
3. Analytical model
Fig. 2. Cross-section of the uniform pressure actuator (Kamal and Daehn, 2007).
E. Thibaudeau, B.L. Kinsey / Journal of Materials Processing Technology 215 (2015) 251263
253
Fig. 3. The Ohio State University UPA coils, (a) generation 2 and (b) generation 3 (Kamal, 2005).
M=k
Lc Lw
(1)
where Lc and Lw are the inductance of the coil and the workpiece,
respectively, and k is the geometric dependent coupling coefcient.
The coupling coefcient represents the leakage of magnetic ux
from the alignment of the two inductors, with a value between
0 k 1. Jablonski (1976) reduced the mutually coupled model by
assuming that the workpiece inductance was negligible (i.e., Lw 0
thus removing this term from Equation (1) not causing M = 0). This
simplication is valid when the inductance of the workpiece is
small, which is the case for typical tubular and sheet workpieces.
The EMF machine is largely represented by a capacitance, but
internal resistance and inductance are also included. Conversely,
the coil is primarily an inductance element, but an internal resistance is also included. Resistance and inductance values of the
EMF machine, (Rm and Lm ) can be determined experimentally by
recording the electrical response to a shorted load. Resistance and
inductance values for the coil (Rc and Lc ) are calculated from the
geometry for the purposes of UPA design, but they can also be determined in the same manner as the EMF machine once the coil exists
(i.e., Rc = R Rm ), since the resistive elements are in series.
The inductance of the coil can be calculated as that of a rectangular current sheet, with corrections for non-conducting space
and nite conductor cross-section (Grover, 1946). The inductance
is determined from that of the coil geometry only; therefore, effects
of the return path and workpiece are neglected (Jablonski, 1976).
Since the current is oscillating, the skin effect produces an exponentially decaying current density distribution, J, in the conductor
cross-section from the surface current density, Js . The distribution
is governed by:
J = Js ed/
(2)
254
E. Thibaudeau, B.L. Kinsey / Journal of Materials Processing Technology 215 (2015) 251263
where d is the depth into the conductor, and the skin depth, , is
dened as:
2
0 r
(3)
L
Aeff
(4)
ip dt + ip R + L
dip
=0
dt
(5)
ip (0) = 0,
(6)
Vco
1 2
C/Len t sin n
1 2 t
(7)
(8)
C
L
F = J B
(9)
(10)
It is this Lorentz force acting as a body force that creates workpiece acceleration and deformation. Since we are concerned with
the force in the thickness direction, y, the current density, J, is
through a partial derivative in that
related to the magnetic eld, H,
direction (Psyk et al., 2011):
J = H
y
(11)
B = H
(12)
H = 1 (H )
F = H
2
y
y
(13)
y1
Pm =
y2
1
2
2
= (Hgap
Hpen
)
Fdy
2
(14)
1
2
Hgap
2
(15)
1
2
kHgap
2
(16)
ii
nip
(17)
where ii is the induced current in the workpiece, and n is the number of turns in the coil.
E. Thibaudeau, B.L. Kinsey / Journal of Materials Processing Technology 215 (2015) 251263
255
Fig. 6. (a) 2D geometry imposed on actual geometry, and (b) 2D magnetic eld.
i
2
gy
(g y)2
g+y
(g + y)2 + x2
(18)
i
2
g+y
(g + y)2 + x2
(19)
ip dxdy
a2
(20)
depends on the gap distance between the coil and the workpiece.
In this way, the electromagnetic-mechanical problem is solved in
a loosely coupled manner, as was represented in Fig. 4.
The pressure distribution across the width of the workpiece is
averaged and applied to the surface. Applying Newtons second law
determines the workpiece acceleration, A(t). Velocity and position
are then determined from the integration of A(t). The incremental
change in position of the workpiece from the initial gap distance is
then used to calculate a new magnetic pressure to apply.
While the analytical model described above can develop the
theoretically optimal coil geometry, a robust design is limited to
mechanical and electrical requirements. For example, high voltages
in the EMF process require space for insulation, and the mechanical
loads on the coil from accelerating the workpiece require additional
support.
Voltage potentials are generated between each turn of the coil
from the transient current, as well as between the coil and surrounding return path. This requires adequate spacing for insulation,
which was determined from the thickness and dielectric strength of
the insulator. Assuming a linear relationship, the required spacing,
bs , is found from:
bs =
e
d
(21)
where d is the dielectric strength of the insulator, and e is the voltage potential that is being insulated.
The coil must withstand the forces used to accelerate the sheet,
as well as forces generated by the coil on itself. A static mechanical
FE simulation was performed in SolidWorks to estimate the stresses
and deections of the coil geometry loaded by a simplied case of
the magnetic pressure predicted in the analytical model.
The loading for the FE analysis was the magnetic pressure
generated from the workpiece interaction. Other forces are generated by the coil on itself, but were not included for simplicity.
An attractive force exists between each of the parallel sections
of the coil, where current is owing in the same direction.
Since this would create a compressive force on the insulation, it
would not likely lead to failure, but could contribute to failure
stresses. Additionally, a repulsive force exists in each turn itself
across the coil where current is owing in the opposite direction. Due to the greater distance, this is a weaker force than
that of the workpiece magnetic pressure, and acts in the opposite
direction.
4. Model results for coil design
The nal coil design was driven by the analytical model initially,
where an efcient design for the coil geometry to achieve maximum
forming pressure or maximum workpiece velocity was predicted.
However, coil designs were also limited by electrical and mechanical requirements. Results from the analytical model are presented
256
E. Thibaudeau, B.L. Kinsey / Journal of Materials Processing Technology 215 (2015) 251263
here, which include electrical, magnetic, and mechanical predictions. These results are those of the nal coil design, before any
modications are made based on experimental results (e.g., tuning
coil resistance and inductance parameters from an experimentally
measured primary current). The nal design consists of a six turn
coil and a spacing between each turn of 3 mm which provides a
safety factor of 2.0 for the dielectric strength of the selected potting material. Additionally, a 1 mm thick, Al 6061-T6 workpiece is
assumed in the analysis results.
Electrical analysis determines the electrical parameters of the
coil, and the electrical response of the EMF machine and coil circuit.
Machine capacitance was 360 F, inductance was 0.1 H, and resistance was 4.4 m, as determined from a shorted load experiment.
From the nal geometry of the coil, the inductance was 0.9 H
and resistance was 1.2 m. The predicted circuit response at 100%
energy for the 12 kJ Magneform 7000 JA is shown in Fig. 7. Key electrical response parameters were a peak current of 146.8 kA, and a
rise time to peak current of 28 s. For the Al 6061-T6 workpiece,
the skin depth is 1.1 mm for this experimental setup.
The magnetic and mechanical analysis predicts the magnetic
eld between the coil and workpiece, the pressure exerted on the
workpiece, and the workpiece velocity. The mean magnetic pressure across the workpiece and the workpiece velocity are shown in
Fig. 8. The peak value of magnetic pressure developed was 23.1 MPa,
which produced a peak workpiece velocity of 432 m/s.
The experimentally determined coupling coefcient, k, acts on
the magnetic pressure prediction to correct for the superconducting workpiece assumption (i.e., 0). Effects of k, as well as effects
from the upper and lower bound magnetic analyses are shown on
the magnetic pressure in Fig. 9.
To optimize the coil for maximum sheet velocity, and thus MPW,
the effect of turn count was investigated for a given coil design and
workpiece length. To allow for various turn counts, the conductor width was varied such that the total length of the coil equals
the workpiece length. Fig. 10 shows both predicted peak velocity
and pressure during the process against the coil turn count. The
pressure shown is the temporal maximum of the average pressure
across the workpiece width, which occurs at a different turn count
Fig. 9. Effect of k value (Kamal and Daehn, 2007) on the predicted magnetic pressure
compared to the upper and lower bound solutions.
Fig. 10. Peak pressure and velocity generated for a given turn count.
E. Thibaudeau, B.L. Kinsey / Journal of Materials Processing Technology 215 (2015) 251263
257
Fig. 12. FE analysis mesh and symmetry of workpiece, with 10 mm die case shown.
Fig. 11. Predicted pressure distribution across workpiece width varying with time.
(For interpretation of the references to color in gure legend, the reader is referred
to the web version of the article.)
Fig. 13. Dynamic FEA comparison for center workpiece velocity with two die radii
values.
affected by the clamped edge conditions. For both die radii in the
FE simulations, the velocity matches almost exactly with the data
from the analytical model for the initial acceleration as shown in
Fig. 13. Therefore, the FE simulations validate the rigid body motion
assumption at the center of the sheet in the analytical model.
To determine if the rigid body assumption holds over the length
of the workpiece, a deformed plot with resultant velocity is shown
in Fig. 14. Since the mean pressure was used as the load case, no
effect of the widthwise (turn to turn) distribution is observed. This
could be included with additional model complexity, but would not
affect the rigid body assumption validation. The workpiece velocity
is uniform except near the die region.
In the FEA module of SolidWorks, a static pressure was applied
to the coil equal to that of the peak magnetic pressure of 29.4 MPa
as determined in the analytical model for a coil with 6 turns. To
restrain the coil, a xed boundary condition was applied to the
opposite side of the coil. The coil was meshed with quadrilateral tetrahedral elements. The coil material is UNS C18000, with
a modulus of 130 GPa and yield strength of 607 MPa (Alloy Physical
Properties, 2012). Element size was 1.13.2 mm with a maximum
aspect ratio of 4.0, and the mesh contained 52,666 elements. Since
all that was of interest in this model was yielding, SolidWorks was
an acceptable FEA package.
Table 1
Johnson-Cook parameters for Al 6061-T6 (Lesuer et al., 2001).
A (MP)
B (MP)
n
C
324
114
.42
0.002
m
0
Tmelt (K)
Ttransition (K)
1.34
1
925.37
294.26
Fig. 14. Deformed workpiece velocity from dynamic FEA, 10 mm die case, t = 64 s.
258
E. Thibaudeau, B.L. Kinsey / Journal of Materials Processing Technology 215 (2015) 251263
Results show that in this simplied case, the coil can withstand
the forming pressure, with a safety factor of 1.32 based on the yield
stress of the material. Since the repulsive force between each side
of the coil is not included, this creates a conservative estimate, even
with a small safety factor. See Fig. 15 for the von Mises stress in the
coil, and the locations of the boundary condition (shown in green)
and applied load (shown in orange).
6. Actuator construction
Based on the design analysis, but also considering other electrical and mechanical considerations, a UPA was constructed, which
consists of a six turn coil and a spacing between each turn of
3 mm. Wire electric discharge machining (EDM) was used to cut
the coil from a solid billet of UNS C18000 (2Ni1Si1Cr96Cu), as
shown in Fig. 16a. This high performance copper alloy was selected
for its high yield strength (607 MPa) and low electrical resistivity
(3.592 108 m).
A deformable workpiece area of 48.1 mm 127 mm was
selected for the UPA geometry, but the actual workpiece width is
larger to contact the return path. The cross-section of the conductor is 12.3 mm 12.3 mm, which was determined in the static FE
mechanical simulation to withstand the forming pressures.
Two Cu 110 leads extend from the coil for attachment to the
capacitor bank, with a 25.4 mm 4.8 mm cross-section. Brazing
and heat treating of the coil and leads assembly was performed
to increase the strength and conductivity of the C18000 alloy.
The return path consists of an Al-6061 T6 channel, with a 3 mm
gap between itself and the coil to allow for electrical insulation.
The geometry of the return path allows eddy currents generated
in the sheet to ow in a circuit, and its close proximity to the
coil allows for further eddy current generation. This increases the
Fig. 16. Pictures of (a) coil, and (b) the complete UPA assembly with PDV probe.
E. Thibaudeau, B.L. Kinsey / Journal of Materials Processing Technology 215 (2015) 251263
259
Fig. 18. Experimental setup showing (a) the exterior and (b) the interior (i.e., in the forming box) details.
ii
nip
(22)
where ii is the induced current in the workpiece, and n is the number of turns in the coil.
A slightly modied experimental setup was necessary for measurement of the induced current in the workpiece. A 1 mm 6061-T6
workpiece was manually deformed to create a small channel for
the Rogowski coil to surround the workpiece. Photographs of this
experimental setup are shown in Fig. 20. Measurements of the primary and induced currents were recorded with this setup and a
3.6 kJ discharge. Higher energy tests were not performed due to
concerns of deforming the workpiece and damaging the Rogowski
coil. The average of three experiments is shown in Fig. 21, with bars
representing the range of values from the three experiments.
Fig. 19. (a) Experimentally determined coil parameters, (b) modied prediction from experimentally determined parameters.
260
E. Thibaudeau, B.L. Kinsey / Journal of Materials Processing Technology 215 (2015) 251263
Fig. 20. Modied experimental setup for measuring induced current in workpiece.
in Fig. 23. The bars represent the range of values from the three
experiments.
The analytical model is shown to accurately predict workpiece
velocity during the initial acceleration, up to approximately the rst
plateau in velocity. However, the model is limited in the inability
to predict velocity later in the forming process, as the rigid body
motion assumption is invalided and bending forces overcome the
inertial forces in the workpiece.
Photographs of the deformed workpieces are shown in Fig. 24 for
the three energy levels. The side view in Fig. 24b shows the relative
uniformity of the deformation across the six turns of the coil, which
is where the UPA gains its name. Some increased deformation is
9. Experimental validation
To verify the analytical model for workpiece velocity prediction,
three energy levels of 3.6 kJ, 6 kJ, and 8.4 kJ were used to accelerate and deform 1 mm thick 6061-T6 workpieces (see Fig. 23). The
analytical model predictions shown are those with the modied
electrical parameters (see Fig. 19a) and k = .45. Experiments were
repeated three times at each energy level, with the average shown
Fig. 21. Primary and induced currents in the coil and workpiece.
Fig. 22. Predicted velocities for varying k, with a 6 kJ discharge and 1 mm 6061-T6
workpiece.
Fig. 23. Experimental velocity results from a 3.6 kJ, 6 kJ, and 8.4 kJ discharge and
1 mm Al 6061-T6 sheets, with the analytical model predictions shown for comparison.
E. Thibaudeau, B.L. Kinsey / Journal of Materials Processing Technology 215 (2015) 251263
261
Table 3
Material properties for workpiece conductivity experiments.
Fig. 24. Deformed 1 mm 6061-T6 workpieces from a 3.6 kJ, 6 kJ, and 8.4 kJ discharge
energies, showing (a) front view (i.e., view normal to coil axis) and (b) side view (i.e.,
coil turns across photograph).
Fig. 25. Workpiece velocity with varying workpiece thicknesses and analytical
model prediction. Discharge energy was 6 kJ.
0.5
1
2
0.57
1.1
2.3
Material
Resistivity (m)
Conductivity
(% IACS)
Density
(kg/m3 )
Cu 110
Al 6061-T6
Cu 230
Cu 510
SS 321
1.72 108
3.99 108
6.15 108
11.9 108
72 108
100
43
27
14.5
2.5
8890
2660
8530
8860
8000
1.7
1.1
0.91
0.66
0.27
shown in Fig. 25. The bars represent the range of values from the
three experiments. The corresponding analytical predictions with
the modied electrical parameters and k = .45 are also shown in
Fig. 25 for comparison.
The 1 mm and 2 mm thicknesses show no change from the analytical model during the initial acceleration. Therefore, even a
ratio as low as = 1.1 is acceptable for determining the magnetic
pressure with the aforementioned assumption. This is lower than
the ratio of 1.52.0 recommended in past research, which is useful for predicting thinner workpiece velocities. This also shows that
the magnetic coupling (and therefore, k) remains constant above
some critical value of . The = 0.57 experiment produce a slower
velocity from the analytical prediction, which demonstrates that
smaller ratios do not fully contain the magnetic eld, and thus
the forming pressure are velocity are reduced.
The coupling coefcient, k, was measured experimentally with
an Aluminum workpiece, however, the magnetic coupling may
change with workpiece conductivity, since the skin depth depends
on conductivity. For example, previous research with a UPA (Kamal
and Daehn, 2007) showed a coupling coefcient of k = 0.70 for
Aluminum, and k = 0.55 for stainless steel (having resistivitys of
5.9 cm and 69.5 cm respectively). However, = .57 for the
Aluminum workpiece, and = .15 for the stainless steel workpiece
in their experimental setup. Therefore, the difference in k can be
attributed to low ratios.
To show that magnetic coupling is not material dependent for
greater than some critical value, two other materials were formed
with the UPA, and their velocities were compared to the analytical
prediction with k = .45 to observe any signicant difference. Materials with conductivities both higher and lower than Aluminum
6061-T6 were chosen. The alloys and their material properties are
listed in Table 3. All workpieces were 1 mm in thickness. Each material was formed three times, with a 6 kJ discharge, and the average
velocity values measured with PDV are shown in Fig. 26 along with
the analytical model prediction with k = .45. The bars represent the
range of values from the three experiments.
Results show accurate velocity prediction during the initial
acceleration with > 0.66, which shows that the magnetic coupling
is constant when varying the workpiece material above a critical
ratio. However, the measured velocity is lower than predicted for
the stainless steel experiment, where = 0.27. Since the materials
also have large differences in densities, this series of experiments
also shows that rigid body motion captures the initial acceleration
well, and that the acceleration process is highly dependent on inertial forces. For example, with the lighter Aluminum, the velocity
decreases rapidly in the later part of the forming process due to
decreased inertial forces.
The physical interactions of the acceleration process were simplied to solve for workpiece velocity analytically, while still
maintaining sufcient accuracy. The predictive power of the model
has possibilities for improvement, e.g., determining a coupling
coefcient without physical experimentation. There are also other
opportunities for renement of the model to improve accuracy if
necessary. For example, inclusion of the skin effect and proximity effect in the magnetic analysis would alter the current density
distribution in the coil and workpiece, which would produce a
262
E. Thibaudeau, B.L. Kinsey / Journal of Materials Processing Technology 215 (2015) 251263
Fig. 26. Workpiece velocity with varying workpiece conductivity and analytical model prediction. Discharge energy was 6 kJ.
E. Thibaudeau, B.L. Kinsey / Journal of Materials Processing Technology 215 (2015) 251263
Knight, W., 2012. Practical Continuous Functions for the Internal Impedance of Solid
Cylindrical Conductors. http://www.g3ynh.info/zdocs/comps/Zint.pdf
Lesuer, D.R., Kay, G.J., LeBlanc, M.M., 2001. Modeling large strain, high rate deformation in metals. In: Modeling the Performance of Engineering Structural Materials
II. Proceedings of a Symposium, pp. 7586.
Nopper, R., Niekrawietz, R., Reindl, L., 2010. Wireless readout of passive LC sensors.
Instrum. Meas. IEEE Trans. 59 (9), 24502457.
Ogata, K., 2004. System Dynamics, 4th ed. Pearson Prentice Hall, New Jersey.
Psyk, V., Risch, D., Kinsey, B.L., Tekkaya, A.E., Kleiner, M., 2011. Electromagnetic
forming a review. J. Mater. Process. 211 (5), 787829.
Song, F.M., Zhang, X., Wang, Z.R., Yu, L.Z., 2004. A study of tube electromagnetic
forming. J. Mater. Process. Technol. 151, 372375.
263
VanBenthysen, R., Thibaudeau, E., Kinsey, B.L., 2013. Effect of specimen planar area
on electromagnetic anging. J. Manuf. Process. 15 (2), 194200.
Watanabe, M., Kumai, S., 2009. High-speed deformation and collision behavior
of pure aluminum plates in magnetic pulse welding. Mater. Trans. 50 (8),
20352042.
White, J.F., 2004. High Frequency Techniques: An Introduction to RF and Microwave
Engineering. IEEE Press, Piscataway, NJ, pp. 221 (Chapter 7).
Xu, D., Liu, X., Fang, K., Fang, H., 2010. Calculation of electromagnetic force in electromagnetic forming process of metal sheet. J. Appl. Phys. 107 (12), 124907.
Zhang, Y., Babu, S.S., Prothe, C., Blakely, M., Kwasegroch, J., LaHa, M., Daehn, G.S.,
2011. Application of high velocity impact welding at varied different length
scales. J. Mater. Process. Technol. 211 (5), 944952.