8/3/2019 ELE366 LAB1
1/15
ELE366 EE08U155 ARSAL JAVID
1
ELE 366 Lab Experiment 1: Numerical Solution OfLaplaces Equation
Introduction
This experiment consists of lab simulation which provides a clear understanding of Laplaces
Equation. Geometry is usually used to describe the solutions to the Laplace equation,
however, real world problems are solved numerically. Solving problems numerically
provides a solution to a greater accuracy. Solving problems hypothetically may not be able
to provide an ideal solution. In order to deal with numerical solutions, computers can be
used as they are powerful enough to perform iterations.
The experiments consists of the derivation and application of Laplace equation. Thisequation is a partial differential equation named after Pierre-Simon Laplace and is used in
the field of electromagnetism and fluid mechanics. This is used due to involvement of
gravitational, electric and fluid potentials.
In order to gain a deeper understanding of Laplaces Equation, an experiment was
performed using Microsoft Excel. This software provided us with a numerical of the
geometric shapes that turn up in the course.
8/3/2019 ELE366 LAB1
2/15
ELE366 EE08U155 ARSAL JAVID
2
Experiment 1
Assuming that = :
22
+22
= 0
The first term represents the rate of change of Potential V with respect to x, whilst the
second term represents the rata of change of V with respect to y.
The rate of change of potential points 1 and 0 is given by:
10
1 0
The rate of change between points 0 and 2 is:
02
0 2
The second order differential equation can be found by similar reasoning to be given by:
2
2
10
02
8/3/2019 ELE366 LAB1
3/15
ELE366 EE08U155 ARSAL JAVID
3
The second order differential equation can be given with respect to y in a similar manner
using points 3 and 4:
10
02
+
30
04
This can be rearranged to give:
0 = 0.25(1 + 2 + 3 + 4)
8/3/2019 ELE366 LAB1
4/15
ELE366 EE08U155 ARSAL JAVID
4
Experiment 2
The diagram above was implemented using a predefined 29 by 29 grid based on the five point
average formula. The convergence of this formula was then examined by performing iterations of
this formula. It will be assumed that the potential is zero at every location except the electrodes.
8/3/2019 ELE366 LAB1
5/15
ELE366 EE08U155 ARSAL JAVID
5
2.2)
After 20 iterations:
Computation Region after 20 Iterations
8/3/2019 ELE366 LAB1
6/15
ELE366 EE08U155 ARSAL JAVID
6
Magnitude of Electric Field after 20 iterations
Laplace Representation after 20 iterations
8/3/2019 ELE366 LAB1
7/15
ELE366 EE08U155 ARSAL JAVID
7
After 100 iterations:
Computation Region after 100 Iterations
Magnitude of Electric Field after 100 iterations
8/3/2019 ELE366 LAB1
8/15
ELE366 EE08U155 ARSAL JAVID
8
Laplacian Representation after 100 iterations
The converged solution:
Computation Region after 101 iterations
8/3/2019 ELE366 LAB1
9/15
ELE366 EE08U155 ARSAL JAVID
9
Magnitude of Electric Field after 101 iterations
Laplacian Representation after 101 iterations
8/3/2019 ELE366 LAB1
10/15
ELE366 EE08U155 ARSAL JAVID
10
As convergence is performed, the voltage pattern around these boundaries (layout of electrodes)
stabilises. Hence, the pattern stabilises as the number of iterations increases. As the number of
iterations come closer to that of number of iterations required for convergence, the grid becomes
similar to the problems faced that require the use of the five point average equation. Therefore, it
can be seen that there is no charge stored inside the conductor as any existing charges are attached
at the electrodes that were used in the spreadsheet. It can be seen that the charge propagates
quickly in certain directions and slowly in opposite direction. Any cell not represented by an
electrode is a vacuum cell. Each of these store a five point average formula. If the solution to the
Laplacian is equal to the average of the four surrounding cells. Each Vaccum cell is recalculated
repeatedly as iterations.
Free charges exist outside the boundaries drawn in Figure 1 due to the fact that two positive and
negative charges are repelling each other and moving the sets towards the boundaries around the
edge of the spreadsheets represented by the purple area on the diagram showing the potential
distribution. At 20 iterations, it can be seen that the charge stored inside the electrodes merge with
the charge initially on the electrodes which is why it can be seen that the values represented in the
red and black change significantly between 20 and 100 iterations. This also applies to the charges
contained outside the boundaries, as the iterations take place, these charges combine with the
charge already existing on the boundary around the edge of the screen and increases the number of
positive charges. The change in charge at a point on a boundary is proportional to the charge density
in the area.
In this particular instance, 101 iterations were performed for convergence which is slightly more
than the previous number of 100; therefore this value could be used if required. Despite this, it is
important to note that the value required for convergence varies due to the sizes of the electrodes
in cells. Any zeros in this computation region can be be derived using Gauss s law. The boundary
conditions guarantee that any field line that leave the top of the universe re-enter at the bottom of
the universe. Consequently, there is not net flux flowing into the universe (this can be seen as the
edges are displayed with the value of zero). Therefore, the accuracy of Gauss s law depends wholly
on the structure of the operator (a + b + c + d -4w).
8/3/2019 ELE366 LAB1
11/15
ELE366 EE08U155 ARSAL JAVID
11
2.3) Using the converged solution, The equi-potential lines between 0 Volts and 100 Volts in
intervals of 10 Volts is given as:
The electric field lines for the given configuration can be given as:
By looking at both graphs, it can be seen that Electric Potential is known to be perpendicular to the
electric field which is poof of one of the basic electromagnet concepts in physics.
8/3/2019 ELE366 LAB1
12/15
ELE366 EE08U155 ARSAL JAVID
12
Experiment 3
3.1) The 5 point average formula will be implanted with the boundary conditions specified in part 2
using the same 29 by 29 grid. The diagram shows the initial implementation of the problem with the
potential of the electrodes set at 100 Volts.
Computation Region with Cross-shaped electrode defined
8/3/2019 ELE366 LAB1
13/15
ELE366 EE08U155 ARSAL JAVID
13
Computation Region with Cross-shaped electrode defined once convergence is performed
Equipotential Surface Line of cross shaped Electrode
8/3/2019 ELE366 LAB1
14/15
ELE366 EE08U155 ARSAL JAVID
14
Electric Field Lines of Cross Shaped Electrode
8/3/2019 ELE366 LAB1
15/15
ELE366 EE08U155 ARSAL JAVID
15
3.3) In order to calculate the capacitance per unit length, the Gaussian surface needs to be found.
This can be done by calculating Gaussian Surface area for one side of the surface:
= 3 + 5 + 8 + 10 + 13 + (2 16) + (2 19) + (2 21) +(2 23) + (2 24) + (5 25) + 14 + 11 + 9 + 7 + 4 + 2 = 456
456 10 3 = 0.456
= 456 10 3 8.85 10 12 = 4.0356 10 13 = 4.0356 10 12 4 = 1.614 10 11 C
We know that : = and that V = 100 Volts
= 1.614 1011
100 = 1.614 10 13 1