Numerical Solution of the Boundary Problem for 2D Laplace Equation for Electrostatic Potential in Given Geometry Jared Land,

Jeremiah Land, Sheharyar Khan

Department of Physics, McMurry University, Abilene, TX 79697

Introduction
The purpose of this project was to further our understanding of computational physics and the implementation of numerical methods in order to study 2D capacitors of various configurations and find space distributions of electric potential, electric field, and charge density. The use of the C programming language was initially chosen because of the group members familiarity with it. The experimental grid configuration was adopted from the experiment performed in the introductory physics lab. The grid was segmented and spaced into a thirty by thirty two cells; the extra two were the grounding plates for the configuration. The geometry of the grid consisted of two grounding plates separated by a distance with a rhombus, where the potential was applied, located in the center. The grid spacing was approximately one square centimeter. The computational portion of this project was separated into two sections the first was creation of the numerical code for the implementation of a finite difference method to solve 2D Laplace equation for a parallel plate capacitor with finite length pates placed inside of a grounded box. The second section was the actual numerical simulation for our experimental configuration.

Results
Potential (V)

Parallel plate Capacitor Potential Simulation A

100 80 60 40 20 0 -20 1 4 -40 -60 -80 -100 80-100 60-80 40-60 33 25 17 7 10 13 16 1 19 22 25 28 31 34 37 40 9 20-40 0-20 -20-0 -40--20 -60--40 -80--60 -100--80

The Simulation of Experimental Results: Electric Field

Experimental Electric Field

Electric Field (V*cm^-1)

Simulation A : parallel plate capacitor were the plates were made from a very thin conductive material with a potential of + V0 and V0
Electric Field of Parallel Plate Capacitor Simulation A
Electric Field (V*cm^-1 3500 3000 2500 2000 1500 1000 500 0 1 4 7 10 13 16 19 1 22 25 28 31 34 37 40 Position (cm) 31 21 11 3000-3500 2500-3000 2000-2500 1500-2000 1000-1500 500-1000 0-500

The graph generated from the experimental data was jagged especially near the center, more points would have smooth this some. The flat boundaries were attributed to the Neumann boundaries of the plate
Simulated Electric Field
70 Electric Field (V*cm^-1) 60 60-70 50 40 30 20 10 0 1 3 5 7 9 11 13 15 17 7 19 21 23 25 27 29 1 25 19 13 50-60 40-50 30-40 20-30 10-20 0-10

1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 1 4 7 10 13 16 1 19 22 25 28 31 Position (cm) 1.6-1.8 1.4-1.6 1.2-1.4 1-1.2 0.8-1 0.6-0.8 25 17 9 0.4-0.6 0.2-0.4 0-0.2

Position (cm)

The program generated expected results for the electric potentials and electric fields. It indicates semi uniform electric field between the plates and maxima at the thin plates with peaks at the corners.

The Electric field generated by our program produced a graph was a close fit to our experimental near the center rhombus, the jaggedness of the center was attributed to the derivatives of potential being a maximum.

Method
The Numerical Method Calculation of Electric Potential Poissons/Laplaces equation for 2D \ Components calculated individually

Simulation B: The implementation of charge density applied to the plates which were to be considered a thin dielectric material with uniform charge densities of and on the plates 100
Potential (V)

Parallel Plate Capacitor Simulation B

Position (cm)

Electric Field of Parallel Plate Capacitor Simulation B

Electric Field (V*cm^-1 3500 3000 2500 2000 1500 3000-3500 2500-3000 2000-2500 1500-2000 1000-1500 500-1000 0-500

Magnitude calculated

Approximated by Taylor expansion

Finite difference method approaches Jacobi Iterates through entire grid, simplest and slowest Gauss-Seidel Updates iterated potentials to improve convergence speed. Successive over-relaxation Similar to Gauss-Seidel, but includes a relaxation parameter to increase speed of convergence A value of 100 volts was used in the program to determine which method would provide the best results The Experimental Method Experimental plate was segmented into a 32x30 centimeter area, two sides being grounding strips. The center, were a potential of 10V was applied, of the plate was a rhombus, and was approximated using a step pattern. A digital multi meter was used to measure the points
Grid spacing for Experimental Plate Photo of experimental setup

Table 2 Summary of Age, Gender, Family Size, and Seriousness of Current Relationship
1000 500 0 1 4 7 10 13 16 19 22 25 1 28 31 34 37 40 Posittion (cm) 31 21 11 Position (cm)

80 60 40 33 20 25 17 0 9 -20 1 4 7 10 13 16 1 19 22 25 28 31 34 37 40 -40 -60 -80 -100

80-100 60-80 40-60 20-40 0-20 -20-0 -40--20 -60--40

Discussion
The program generated expected results for the electric potentials. The electric fields generated were off in unit values, but generated expected shapes. The variation between the simulated and the experimental, was due primarily to difference in boundaries. The simulated results were within a grounded box and the experimental plate consisted of only two grounding sides.

Similar to simulation A , in both the electric potential -80--60 -100--80 and electric field

Parallel Plate Capacitor Simulation C

Illustration of Averages of neighboring potentials

Adding equations for x and y components respectively cancels odd terms and a central difference approximation is obtained

Simulation C: Plates considered to be made from a conductive material, as in the first simulation, but with a finite thickness of 2
Potential (V)

Electric Field of Parallel Plate Capacitor Simulation C

Electric Field (V*cm^-1 3500 3000 2500 2000 1500 1000 500 0 1 4 7 10 13 16 19 22 25 28 31 34 37 40 1 3000-3500 2500-3000 2000-2500 1500-2000 31 21 11 1000-1500 500-1000 0-500

100 80 60 40 20 0 -20 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 -40 35 37 39 -60 -80 -100

80-100 60-80 40-60 33 25 17 9 1 20-40 0-20 -20-0 -40--20 -60--40 -80--60 -100--80

Position (cm)

Substituting these into Poissons/Laplaces yields a finite difference form

Position (cm)

The program generated expected results for the electric potentials with constant values inside of the plates. The electric field showed a marked increase in intensity near the plates and a slight curve abound the edges of the plates.
Experimental Electric Potential
12 10 Potential (V) 8 6 4 2 25 17 1 4 7 10 13 16 1 19 22 25 28 31 Position (cm) 9 10-12 8-10 6-8 4-6 2-4 0-2

The code was most successful for the Gauss-Seidel method of finite differences. It provided reasonably fast conversion speeds and fair results of the data. The Jacobi method was implemented successfully in to the first section for parts a, b, and c, but was slower and less efficient than the Gauss-Seidel method. The Jacobi method provide very similar results to the Gauss-Seidel, there for the Gauss-Seidel method was chosen to implement into section two of the project, the simulation of the experimental plate. The Successive over-relaxation method was experimented with, but it proved to be problematic in its implementation. This was largely a troubleshooting problem, with more time we feel that it would have been implemented effectively and would have provided even faster results than the GaussSeidel and the Jacobi methods.

If grid spacing is equal Solving for electric potential U Iteration of the grid until the potentials converge and a numerical solution is obtained Calculation of Electric Field
Illustration of Electric Field Calculation from change in potential were potentials are diamonds and Electric Field points are circles

The Simulation of Experimental Results: Electric Potential The program generated a fairly good result of electric potential compared to our experimental data
Simulated Electric Potential
10 Potential (V) 8 6 4 2 0 1 4 7 10 13 16 1 19 22 25 28 31 Position (cm) 9 25 17 8-10 6-8 4-6 2-4 0-2

References
Landau Rubin H., Paez Manuel J., Bordeianu Cristian C., Computational Physics. Problem Solving with Computers

The boundaries on the experimental are bowed because of the two non grounded sides on the plate, these are the Neumann boundaries of the experimental plate