Eddy Currents in Copper Disk Simulation
Self-guided learning
This example of eddy currents in a copper disk contains a short description of the theory, detailed information on how to construct the model, and the fully constructed model ready for you to download.
When working through the example, you may notice some small differences in your model compared to ours – this is usually simply due to the fact that you are using a different software version.
where E denotes the electric and B the magnetic field. Gauss's law
links the electric field electric charge p via the permitivity of free space ε0.
In the absence of electric charge, a change in the magnetic field is the only source for the electric field. The governing equations are given by:
Mathematically this is identical to magnetostatics,
where μ0 denotes the permeability of free space.
As such, Faraday-induced electric fields - that is fields generated solely by a change in the magnetic field - are determined by [1] in the same way that magnetostatic fields are determined by [2]. The Maxwell–Faraday law [3] in integral form is therefore closely related to Ampère’s law [4]:
[1]
[2]
[3]
(Maxwell–Faraday law in integral form)
[4]
(Ampère’s law from magnetostatics)
Furthermore, if an electric field is induced in a conductor this generates a current, as shown by Ohm’s law,
where J denotes the current field and σ the conductivity, it follows that, if a conductor is placed within a region of space where we have a magnetic field that changes over time, a current will be induced as well.
For a time harmonic magnetic field
where ω=2πν denotes the angular frequency, the time derivative is given by
For a circular, conducting surface orthogonal to e_{z} we can exploit the symmetry to solve for the electric field using the Maxwell–Faraday law (the same basic calculation can be found in [5]):
[5]
[6]
Fig. 2 shows a schematic representation of the symmetry involved.
By using Ohm’s law one immediately arrives at the current distribution,
or
if we neglect the magnetic susceptibility of the material. In the case of copper, with μr≈1 this is a justified approach.
Question
What direction does the current have when the magnetic field decreases in z-direction / increases in z-direction?
The current will in turn induce a magnetic field. What is the qualitative shape of this magnetic field? As the current distribution changes so will the field induced by the current, which in turn influences the current. Why can we neglect this in our consideration?
The Model
A circular copper disk of radius 2 mm and depth 0.5 mm is defined. We also define a time harmonic homogeneous magnetic source field which is oriented orthogonally to the circular surface of the disc. In accordance with the source field, the boundary conditions in the z-direction are set to ‘magnetic’. With the cylindrical disc it is possible to activate symmetry conditions which help to reduce mesh size and therefore calculation time. The basic setup is shown in Fig. 3.
Parameter | Value | Description |
Freq | 50 Hz | Evaluation frequency |
Hz | 1e03 | Amplitude factor |
By setting up a curve along the radius of the disc we can directly evaluate the induced current distribution in the disc and compare the results with the analytical expression. For this purpose, the corresponding post-processing templates are used.
The “Template Based Postprocessing” window offers a variety of features. To evaluate the field on a curve choose “2D and 3D Field Results” in the first drop down menu and “Evaluate Field on Curve” in the second. Browse for results, select the Conduction Current Density, set “Component” to “Abs” and “Complex” to “Mag” to get the magnitude of the conduction current (the main steps are shown in Fig. 4). The default name for the result is “curve1_Cond. Current Dens. (Freq)”, indicating that the conduction current density is evaluated along curve1 at frequency “Freq”. However, barring a few special characters, you can enter an arbitrary result name. When comparing simulation results with analytical results it might e.g. make sense to use “Simulation” instead of “curve1_Cond. Current Dens. (Freq)”. To evaluate the template click on the “Evaluate” button in the “Template Based Postprocessing” window.
As we can see from eq. (4) we need the conductivity of the disc in order to evaluate the analytical expression for the conduction current density. To find the respective value open the material folder in the Navigation Tree (look to the left hand side) and double-click on Copper (annealed) (cf. fig. 5 a)) to open the Material Parameters dialog. You will find the conductivity value in the “Conductivity” tab (cf. fig. 5 b)). Next, set up a new parameter named “conductivity” in the Parameter List and enter the value in the “Expression” column. To set up the parameter simply double-click on “<new parameter>” in the Parameter List on the bottom of the CST Studio Suite window (cf. Fig. 5 c)).
There are several ways to plot the analytical expression. In this example we will use the option “0D or 1D Result from 1D Result (Rescale, Derivation, etc.)”. To do so, select “General 1D” in the first and “0D or 1D Result from 1D Result (Rescale, Derivation, etc.)” in the second drop down menu. Fig. 6 shows the resulting dialog and required input. First, mark the “1D(C)” (upper left corner of the “Specify Action”-frame), indicating that you are interested in generating a continuous result. “Axes: Scale with Function f(x,y)” should already be preselected. In the “1D Result”-frame select the previous template. In figure 6 the name is given as “Simulation”, the default would be “curve1_Cond. Current Dens. (Freq)”. Back in the “Specify Action”-frame choose “y-axis” in the “Apply to” field. This indicates that we will be using the x-axis from the selected result and scaling the y-axis of the plot. In the “f(x,y)”-field enter the expression we have derived above in eq. (2). You can simply copy and paste the formula from the caption of fig. 6, but if you used different parameters, you will have to modify the formula accordingly. The constants “pi” and “mue0” do not need to be defined separately, as they are part of a set of predefined values, meant to ease calculations.
Analytical expression: conductivity * π * Freq * x * 1.0-3 * µ0 * Hz
Keep in mind that we are using mm as length unit. Therefore a factor of 10−3 has been added to the expression.
So far the templates plot the field results vs. the curve coordinates, i.e. the length of the curve. This can be bothersome if the curve was set up not from the center outward, but from the rim to the center. In this case you can either set up a new curve, or e.g. replot with the axis coordinates. To do the latter a second post processing task has to be performed. Select “General 1D” in the first drop down menu and “0D or 1D Result from 1D Result (Rescale, Derivation, etc)” in the second. In the “0D or 1D Result from 1D Result” dialog check “1D(C)” and select “Parametric X-Y Plot”. Then select the x-coordinates of the curve as x-values (Field Along Curves\curve1\Coordinates\X) and the results for the y-axis as shown in Fig. 7.
Click on “Evaluate All” in the “Template Based Postprocessing” window to evaluate all post processing steps. The corresponding results can be found in the navigation tree under “Tables\1D Results\”. Select the whole folder in order to show all contents or make a subselection in order to show only specific results.
You can change axis labels and plot title, but in order to so, first you must establish a new subfolder, e.g. named “Compare Results”, in “1D Results” in the navigation tree and copy the results there. Fig. 8 shows the comparison between the simulation results and the analytical expression for the conduction current. Axis labels and title can be changed in the “Home” tab using the “Properties” button in the “Edit” section.
As we can see the analytical and numerical results show a very good agreement.
Questions and Further Tasks
What direction does the current have when the magnetic field decreases in the z-direction / increases in the z-direction?
The current will in turn induce a magnetic field. What is the qualitative shape of this magnetic field? As the current distribution changes so will the field induced by the current, which in turn influences the current. Why can we neglect this in our consideration?
References
[1] = David J. Griffiths , Introduction to Electrodynamics, fourth edition, Example 7.7, p. 317