Electromagnetic Fields Lab 1 – Report

Harry Moss
10478272
Group F31

Objectives
The objectives of the experimental and simulation study were the
following:

1. To explore how electric fields can be mapped by measuring

various voltage signals.
2. Find the characteristics of electric fields in a curved resistor path in
addition to a parallel-plate setup.
3. To model the electric field in the real world and also in a simulation
software in order to compare the two.
4. Learn basic skills for clear, accurate data recording and
interpretation.

Supporting Theory
Kirchhoff’s Current Law states that in all steady-state conditions, the
total current passing through any closed surface is always = 0.

From circuit theory, we can conclude the following:

∑ it 0¿

m=1

(1)

Which states that the total currents flowing into a node = 0.

In a 1D case, such as a wire, the following relationship is valid:

I
J=
AS
(2)
With J being the current density (A/m²), I being the current (A), and A S
being the cross-sectional area (m²).

In order to get a 2D case, integrating equation (2) is required to get a

surface integral.
❑
I =∫ J ∙ d S
S

(3)

Where S is now the surface area (m²).

For a closed loop, this can be summarised with the equation:

∮ J ∙ d S=0
S

(4)

Where J and S are now both in vector form.

Equation (4) is an example of a closed-surface integral, and is the vector

form of Kirchhoff’s Current Law.

For E and V, the relationship is fairly similar. Initially, for a 1D case, we

have:

V =−E ∙l
(5)

Where V is voltage (V), E is the electric field (V/m) and l is the distance
(m) between 2 set points in the field. Notice the negative sign in this
equation at the field moves in the direction of the potential drop.

Again, from circuit theory, the total voltage in a closed loop can be
summarised with:
B
V AB=−∑ E ∆ l
A

(6)
(note this is an ideal case)
Which can of course be compiled into an integral upon taking smaller
and smaller steps of ∆l:
B
V AB=−∫ E∙ d l
A

(7)

This, just like equation (3), this is another line integral. Notice also E and
l are now in vector form.

If you swap the limits and integrate equation (7), you find that they
cancel out perfectly in any closed path.

This leads to the following closed-loop line integral:

∮ E ∙ d l=0
(8)

Similar to the findings for equation (4), equation (8) is an example of

Kirchhoff’s Voltage Law, given it’s summing all the individuals potential
differences in a closed loop.

This leads to the conclusion of the relationship between E and V, as

equation (7) and (8) can be adapted into:
B

∫ E ∙ d l=V A −V B =−V
A

(9)

Where V is the electric potential (V).

Questions

1. They are parallel at the surface as the electric field is uniform/has

a linear potential drop between the electrodes. The electrodes
themselves are also equipotentials, as they are fantastic
conductors ∴ effectively no potential drop across their surface.
2. Since we have a closed loop, we can use:

E ∙ d l=0
(10)
When we convert (10) from vector form to scalar form, we get:
 E ∙ dl∙ cos ( α )=0

(11)

∴ α must always = 90 degrees.

∴ equipotentials must always be normal to the electric field lines,
and therefore cross insulating boundaries at 90 degrees also, since
electric fields cross them at 0 degrees (i.e. they don’t) given they are
non-conducting surfaces.

3. The ‘height’ component (or Z-axis) of the electrodes being used in

this experiment are negligible in comparison to their respective
‘length’ and ‘width’ (X and Y axes) components. The electric field
in the Z direction is therefore small due to a very small surface
area S present in this plane.

Experimental and Computational Methods

Please refer to the lab script – no additional data is needed.

Results

The results for my experiment are shown below.

 Graph of potentials measured at various distances on the
parallel plate capacitor (measured)
Potential Difference from 0 V terminal (V) 1.2

0.8

0.6

0.4
Line 1 (centre) (V) Line 2 (edge) (V)

0.2

0
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100

Measurement from 0 V plate (mm)

(a)

Graph of potentials measured at various distances on the

parallel plate capacitor (simulated)
1.2
Potential Difference from 0 V terminal (V)

0.8

0.6

0.4
Line 1 (centre) (V)

0.2

0
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100

Measurement from 0 V plate (mm)

(b)

(c)

(d)
(e)

(f)
For our measured values, in graph (a), we measured the varying
potentials on 2 lines (all is explained fully in the lab script) which can be
seen in the positively-correlated trends of the 2 lines.

For the simulated values, we unfortunately couldn’t obtain values for the
‘Line 2 (edge)’ during the experiment, due to the simulation software
crashing which even the demonstrators couldn’t fix. Luckily there is still
Line 1, which offers a good comparison to our measured values for
these different potentials between the parallel plates.

(c) and (d) are simulations to show the varying potentials across the two
field setups. It is interesting to compare (d) specifically to (f) and how a
simulation can differ from results in the real world.

(e) has also been added to aid in visualising the current flow in the
‘resistor’ circuit.

Discussion

Do the measurements and simulations agree?

In favour:
The results in (c) and (d) clearly show a positive, upward trend with
a very close average upon picking several points:
 For Line 1 (centre), at 0mm, measured = 0.002 V, simulated
= 0.0019 V
 At 50 mm, measured = 0.43 V, simulated = 0.46 V
 At 100 mm, measured = 0.96 V, simulated = 0.97 V
These results show a very close average when looking at the
graph broadly as above, however it is clear to see visually there
are few distortions between these 3 points.

In addition, although I cannot sadly comment too much on the Line

2 (edge) due to lack of simulated results, it is fairly clear to see that
this line hugs the Line 1 very closely, which suggests the results
are valid for the measured results.

In terms of (d) and (f) – up to 0.7 V in (f) – do agree for the most
part, with clear, parallel equipotentials being present for both.

These points suggest they do agree to a reasonable extent.

Not in favour:
The results in (c) for Line 1 do have a single sharp spike – from
90-95 mm distance from the 0 V plate. This is contrary to the
simulated results, which show a far smoother relationship at this
point.

Furthermore, in (f), there is a clear difference at 0.8 V onwards that

the equipotentials differ quite significantly from the simulated graph
in (d), of which has far more parallel equipotential lines than what
is seen in (f), of which begins to both the closer one gets to the
positive terminal.
The simulation further backs up its own findings with the graph of
(e), which suggest a very smooth 2D current flow that matches the
equipotentials in (f) perfectly.

These points suggest they do not agree.

Do the measurements and simulations agree with basic theory?

In favour:
The fact that both graphs for the centre lines on (a) and (b) display
a strong, reasonably linear upward trend suggests that they all
agree on uniform field theory between a parallel-plate capacitor.

In addition, for the measured value of Line 2 in (a), there is a clear

incline on the start and end of the measurements. This makes
sense, as this is where the reading that is being taking is very
close to the electrode plates, meaning there is almost certainly
going to be some edge effect occuring, and therefore the readings
should be sightly distorted anyway in coming into close proximity
with the fringing field (which it does). This agrees that we are
dealing with a radial field at the edges. This can be further backed
up by looking at the shape of the equipotentials in graph (c).

As explained earlier, up to 0.7 V, graphs (d), (e) and (f) agree with
one another. This supports basic theory with the earlier scalar-form
equation (11): E ∙ dl∙ cos ( α )=0

As we can tell that due to α (= 90 degrees), the equipotentials

should always be normal to the insulating surfaces, which they are
for nearly the whole path.

Not in favour:
 That being said, one interesting point to note is the behaviour
beyond 0.7 V in the measured graph (f). This doesn’t appear to
support basic theory, as there shouldn’t be a reason why there
would be such a curved equipotential this close to the centre of the
electrode (as discussed previously with equation (11), the
equipotentials should always be perpendicular to insulating
surfaces, and be parallel to very close electrodes).

Another point to note is the strange spike from Line 1 (centre) in

graph (a) – according to basic theory, this uniform field should be
perfectly linear (i.e. a perfectly straight line), due to crossing the
parallel equipotential lines exactly collinear at regular intervals.
This isn’t the case right at the end of this line, where it jumps up
suddenly.

Errors in the measurement and models

There were a few clear errors in my results that are covered below:
1. Lack of evidence for a simulated Line 2 (edge)
2.

Conclusion

To summarise the key points,