'Chapter One 1. Background: Dual Axis Solar Tracker

[DUAL AXIS SOLAR TRACKER] June 2014
`Chapter One
1. Introduction
1.1. Background
One of the most important problems facing the world today is the energy problem. This problem is
resulted from the increase of demand for electrical energy and high cost of fuel. The solution was in
finding another renewable energy sources such as solar energy, wind energy, potential energy...etc.
Nowadays, solar energy has been widely used in our life, and it's expected to grow up in the next years.
Solar energy has many advantages:
1. Need no fuel
2. Has no moving parts to wear out
3. Non-polluting & quick responding
4. Adaptable for on-site installation
5. Easy maintenance
6. Can be integrated with other renewable energy sources
7. Simple & efficient
As the range of applications for solar energy increases, so does the need for improved materials and
methods used to harness this power source. There are several factors that affect the efficiency of the
collection process. Major influences on overall efficiency include solar cell efficiency, intensity of source
radiation and storage techniques. The materials used in solar cell manufacturing limit the efficiency of a
solar cell. This makes it particularly difficult to make considerable improvements in the performance of
the cell, and hence restricts the efficiency of the overall collection process. Therefore, the most
attainable method of improving the performance of solar power collection is to increase the mean
intensity of radiation received from the source. There are three major approaches for maximizing power
extraction in medium and large scale systems. They are sun tracking, maximum power point (MPP)
tracking or both.
1.2. Need For Solar Tracker
Each day, the sun rises in the east, moves across the sky, and sets in the west. Whenever the sun is
shining on us, it is sending energy in our direction. We can feel the heat from the sun, and we can see
objects that are illuminated by the light from the sun as it moves across the sky. However, if we could
get a solar cell to turn and look at the sun all day, then it would be receiving the maximum amount of
sunlight possible and converting it into the more useful energy form electricity. If we are located in the
tropics, we see that the sun appears to follow a path that is nearly directly overhead. However, for
locations north or south of the tropics (e.g., latitudes greater than 23.5 degrees), the sun never reaches
a position that is directly overhead. Instead, it follows path across the southern or the northern part of
the sky. If we could configure a solar cell so that it faces the sun continually as it moves across the sky
from east to west, we could get the most electrical energy possible.Tracking systems try to collect the
largest amount of solar radiation and convert it into usable form of electrical energy (DC voltage) .
WU ECeng 1
1.3. Problem Statement
Electrical energy from solar panels is derived by converting energy from the rays of the sun into
electrical current in the solar cells of the solar panels. The main challenge is to maximize the capture of
the rays of the sun upon the solar panels, which in turn maximizes the output of electricity. A practical
way of achieving this is by positioning the solar panels such that the rays of the sun fall perpendicularly
on the solar panels by tracking the movement of the sun. This can be achieved by means of using a solar
panel mount which tracks the movement of the sun throughout the day.
WU ECeng 2
1.4. Thesis Objective

1.4.1. General Objective
The objective of this thesis is to design and simulate an automated, dual-axis solar-tracking system with
an actual solar panel mount to be installed outdoors insideHawassa university compound (location:
7°03′36″N 38°28′12″E) in Hawassa, Ethiopia in order to optimize the efficiency of overall solar energy
output.
1.4.2. Specific Objective
Design an active azimuth-altitude dual-axis solar tracking that will have a minimum allowable
error of 4°.
Design a controller using embedded microcontroller.
To design DC motor control interfacing with driver circuit.
To design an electronic circuit to sense an electric intensity of light and to control DC motor
driver for the panel movement.
WU ECeng 3
1.5. Scope of The Thesis
The scope of this thesis:
Designing a mechanical structure for the mounting system.

Setting up of the tracking controller system using embedded micro-controller.
Designing of electrical power system for utilization to generate a 160 welectric power.
1.6. Relevance of Solar Trackers
For people living in remote communities, often in third world countries, access to grid-connected
electricity is not always possible. Often the nearest utility is a long distance from homes and the cost of
developing the infrastructure that would allow for access to the grid is prohibitive. Remote communities
in third world countries are of course not the only ones that suffer this dilemma. When we come to our
country Ethiopia is a large country with many farmers and communities that are remote from the local
grid and in these cases alternative sources of electrical power must be obtained.
1.7. Organization of The Thesis
This thesis consists of five chapters. In the first chapter the background of solar energy, the relevance
and need of solar tracker as well as the objective and scope of the project is stated. The second chapter
discusses theory regarding sun tracking, focuses on various methods and types of trackers. This
incorporates a review of relevant literature in the field of sun tracking. The third chapter deals with,
design of mounting structure, hardware and embedded software and a detail study of the PIC
microcontroller, microcode instructions used in this thesis, explains the control scheme used while
forming a solution to the problem and the design considerations undertaken in this process. Fourth
chapter gives an analysis of the design and total system simulation using matlab with discussions. Fifth
chapter concludes the report by discussing the effectiveness of the tracking system. It also suggests
some further research areas and future design proposals.
WU ECeng 4
Chapter Two
2. Literature Review
2.1. Solar Power Fundamentals
Solar panels are formed out of solar cells that are connected in parallel or series. When connected in
series, there is an increase in the overall voltage, connected in parallel increases the overall current.
Each individual solar cell is typically made out of crystalline silicon, although other types such as ribbon
and thin-film silicone are gaining popularity.
PV cells consist of layered silicon that is doped with different elements to form a p-n junction. The p-
type side will contain extra holes or positive charges. The n-type side will contain extra electrons or
negative charges. This difference of charge forms a region that is charge neutral and acts as a sort of
barrier. When the p-n junction is exposed to light, photons with the correct frequency will form an extra
electron/hole pair. However, since the p-n junction creates a potential difference, the electrons can’t
jump to the other side only the holes can. Thus, the electrons must exit through the metal connector
and flow through the load, to the connector on the other side of the junction [19][21].
Figure 2.1 (a) Equivalent Circuit of Solar Cell (b)The Schematic Symbol of Solar Cell
The equivalent circuit of a PV cell is shown in Fig. 1(a). It includes a current source, a diode, a series resistance and
a shunt resistance. In view of that, the current to the load can be given as:
In this equation,Iph is the photocurrent, Is is the reverse saturation current of the diode, q is the electron charge,
V is the voltage across the diode, K is the Boltzmann's constant, Tis the junction temperature, N is the ideality
factor of the diode, and Rs and Rshare the series and shunt resistors of the cell, respectively.
As a result, the complete physical behaviorof the PV cell is in relation with Iph, Is, Rsand Rsh from one hand and
with two environmental parameters as the temperature and the solar radiation from the other hand.
Because the PV cells generate a current, cells/panels can be modeled as DC current sources. The amount
of current a PV panel produces has a direct correlation with the intensity of light the panel is absorbing.
From simple drawing of the system below the normal to the cell is perpendicular to the cell’s exposed
face. The sunlight comes in and strikes the panel at an angle.
WU ECeng 5
The angle of the sunlight to the normal is the angle of incidence (θ).Assuming the sunlight is staying at a
constant intensity (λ) the available sunlight to the solar cell for power generation (W) can be calculated
as:
W = A λ cos(θ)
Figure 2.2: Angle of Incidence to Solar Cell [ Adrian 2010]
Here, A represents some limiting conversion factor in the design of the panel because they cannot
convert 100% of the sunlight absorbed into electrical energy. By this calculation, the maximum power
generated will be when the sunlight is hitting the PV cell along its normal and no power will be
generated when the sunlight is perpendicular to the normal. With a fixed solar panel, there is significant
power lost during the day because the panel is not kept perpendicular to the sun’s rays. A tracking
system can keep the angle of incidence within a certain margin and would be able to maximize the
power generated calculated the amount of power gained by tracking can come close to an ideal 57%
difference[20].
2.2. Solar Tracker
A solar tracker is a device that is used to align a single P.V module or an array of modules with the sun.
Although trackers are not a necessary part of a P.V system, theirimplementation can dramatically
improve a systems power output by keeping the sun infocus throughout the day. Efficiency is
particularly improved in the morning andafternoon hours where a fixed panel will be facing well away
from the suns rays. P.Vmodules are expensive and in most cases the cost of the modules themselves
willoutweigh the cost of the tracker system. Additionally a well designed system whichutilizes a tracker
will need fewer panels due to increased efficiency, resulting in areduction of initial implementation
costs.
2.3. Overview Of Existing Tracking Technology
As mentioned in the previous sub-section the absorption of light by a PV panel is dependent on its
angular position to the sun. A PV panel must be perpendicular to the sun for maximum solar absorption,
which is done by using a tracking system. Multiple tracking systems exist, which vary in reliability,
accuracy, cost, and other factors. A tracking system must be chosen wisely to ensure that the tracking
method increases the power gained instead of decreasing it.
WU ECeng 6
2.3.1. Tracking Configuration - Immobile Versus Mobile
Different power applications require different tracking systems. For certain applications a tracking
system is too costly and will decrease the max power that is gained from the solar panel. Due to the fact
that the earth rotates on its axis and orbits around the sun, if a PV cell/panel is immobile, the absorption
efficiency will be significantly less at certain times of the day and year. The use of a tracking system to
keep the PV cell/panel perpendicular to the sun can boost the collected energy by 10 -100% depending
on the circumstances [20].
If a tracking system is not used, the solar panel should still be oriented in the optimum position. The
panel needs to be placed where no shadow will fall on it at any time of the day. Additionally, the best tilt
angle should be determined based on the geographical location of the panel. As a general guideline for
the northern hemisphere, the PV panel should be placed at a tilt angle equal to the latitude of the site
and facing south. However, for a more accurate position and tilt angel a theoretical model of the suns
iridescence for the duration of a year is created and the angel and position is matched to the model.
Figure2. 3: Image showing a fixed-tilt module set-up. The angle theta is equal to the latitude of the location where
the module is to be stationed.
Using one axis of tracking can provide a significant power gain to the system. Practically one axis
trackers are placed into the following classifications: horizontal single axis tracker (HSAT), vertical single
axis tracker (VSAT), tilted single axis tracker (TSAT), and polar aligned single axis tracker (PASAT). A TSAT
at a tilt angle of 5° increases the annual collection radiation by 10% compared to a HSAT, a HSAT
increases the annual collection radiation by 15% to a VSAT, and finally a PASAT increases the annual
collection radiation by 10% over a HSAT. Thus for one axis a PASAT or TSAT configuration would collect
the most solar radiation[20].
For an additional power gain a dual-axis tracking system can be used. The percent gain from going from
a PSAT to a dual-axis system is small, but as long as the system doesn’t use more power than gained, it
still helps. There are two classifications for dual axis trackers: Tip-Tilt Dual Axis Tracker (TTDAT) and
Azimuth-Altitude Dual Axis Tracker (AADAT). The difference between the two types is the orientation of
the primary axis in relation to the ground. TTDAT’s have the primary axis horizontal to the ground and
AADAT’s have theirs vertical. The azimuth/altitude method seems to be largely used, based on its
reference in multiple research articles on tracking. In the article by Sefa et al. the following was stated;
“The results indicated that increases of electrical power gains up to 43.87% for the two axes, 37.53% for
the east–west, 34.43% for the vertical and 15.69% for the north–south tracking, as compared with the
WU ECeng 7
fixed surface inclined 32 to the south in Amman”[22]. A prototype AADAT was made in this project and a
picture of it can be seen in Figure 6.
Figure 2.4: (left) single axis (right) dual axis tracker.
2.3.2. Tracking Methods – Passive, Active and Chronological

Tracking
In active tracking, the position of the sun in the sky during the day is continuously determined by
sensors. The sensors will trigger the motor or actuator to move the mounting system so that the solar
panels will always face the sun throughout the day. This method of sun-tracking is reasonably accurate
except on very cloudy days when it is hard for the sensor to determine the position of the sun in the sky
thus making it hard to reorient the structure.
Unlike active tracking which determines the position of the sun in the sky, a passive tracker moves in
response to an imbalance in pressure between two points at both ends of the tracker. The imbalance is
caused by solar heat creating gas pressure on a ‘low boiling point compressed gas fluid that is driven to
one side or the other which then moves the structure. However, this method of sun-tracking is not
accurate.
A chronological tracker is a timer-based tracking system whereby the structure is moved at a fixed rate
throughout the day. The theory behind this is that the sun moves across the sky at a fixed rate. Thus the
motor or actuator is programmed to continuously rotate at a slow average rate of one revolution per
day (15 degrees per hour). This method of sun-tracking is very accurate. However, the continuous
rotation of the motor or actuator means more power consumption and tracking the sun on a very
cloudy day is unnecessary[23].
2.4. Literature Survey
Daniel A. Pritchard had given the design, development, and evaluation of a microcomputer-based solar
tracking and control system (TACS) in 1983. It was capable of maintaining the peak power position of a
photovoltaic (PV) array by adjusting the load on the array for maximum efficiency and changed the
position of the array relative to the sun. At large PV array system installations, inverters were used to
convert the dc electrical output to ac for power grid compatibility. Adjustment of the inverter or load for
maximum array output was one function performed by the tracking and control system. Another
WU ECeng 8
important function of the system was the tracking of the sun, often a necessity for concentrating arrays.
The TACS also minimized several other problems associated with conventional shadow-band sun
trackers such as their susceptibility to dust and dirt that might cause drift in solar alignment. It also
minimized effects of structural war page or sag to which large arrays might be subjected during the day.
Array positioning was controlled by Q single-board computer used with a specially designed input
output board. An orderly method of stepped movements and the finding of new peak power points
wasimplemented. This maximum power positioning concept was tested using a small two-axis tracking
concentrator array. A real-time profile of the TACS activity was produced and the data analysis showed a
deviation in maximum power of less than 1% during the day after accounting for other variations [16].
A. Zeroual. had designed an automatic sun-tracker system for optimum solar energy collection in 1997.
They used electro-optical sensors for sun finding and a microprocessor controller unit for data
processing and for control of the mechanical drive system. This system allowed solar energy collectors
to follow the sun position for optimum efficiency. It had a modular structure which facilitates its
application to different systems without great modifications. The system had been applied to control a
water heating parabolic solar system for domestic uses. Many parameters had been controlled for
system security such as temperature, pressure and wind velocity. The system had been tested for a long
period in variable illumination. The result showed that it operated satisfactorily with high accuracy [18].
EftichiosKoutroulishad given the microcontroller based photovoltaic maximum power point tracking
control system in 2001. Maximum power point tracking (MPPT) was used in photovoltaic (PV) systems to
maximize the photovoltaic array output power, irrespective of the temperature and irradiation
conditions and of the load electrical characteristics. A new MPPT system had developed, consisting of a
Buck-type dc/dc converter, which was controlled by a microcontroller-based unit. The PV array output
power delivered to a load was maximized using MPPT controlsystems, which consisted of a power
conditioner to interface the PV output to the load, and a control unit, which drove the power
conditioner such that it extracted the maximum power from a PV array. It was used to directly control
the dc/dc converter, thus reducing the complexity of the system. The resulting system had high-
efficiency, lower-cost [17].
S. Armstrong had investigated the effectiveness of maximum power point tracking (MPPT) and
proposed a quantitative measure of MPPT efficiency in 2005.Used a vector methodology to track the
direction and path of the sun throughout the day, the optimal solar tracking angle and angle of
incidence of the sun’s rays were derived. The solar array’s output power was monitored, under sunny
sky conditions, with and without the use of maximum power point tracking in order to study
thedifference in efficiencies and to quantify the benefits of maximum power point tracking. He
presented results for the efficiency of MPPT under fixed horizontal solar panel conditions and optimal
solar tracking. It had been shown that solar panel east-west tracking combined with maximum power
point tracking provided optimum amount of available energy at any time [15].
CemilSungur had given the electromechanical control system of a photovoltaic (PV)panel tracking the
sun on the axis it moved along according to its azimuth angle in 2007. In this system, Programmable
Logic Controls (PLC) was used instead ofphotosensors which were widely used for tracking the sun. The
WU ECeng 9
azimuth angle of the sun from sunrise to sunset times was calculated for each day of the year at 37.6
degrees latitude in the Northern hemisphere, the location of the city where the experiment was
conducted. According to this azimuth angle, the required analog signal was taken from the PLC analog
module and sent to the actuator motor, which controlled the position of the panel to ensure that the
rays fell vertically on the panel. After the mechanical control system of the system was started, the
performance measurements of the solar panel were carried out. For this, the necessary measurements
were implemented when the solar panel was in a fixed position. Afterwards, the panel was moved on a
single axis according to the azimuth angle and the necessary measurements were performed. The values
obtained from the measurements were compared and the necessary evaluations were conducted [19].
2.5. Summary
From these literatures, it was shown that building a solar tracker was feasible. Further investigations
could lead to an upgrade to the tracking technology. But the dual axis solar tracker is still the best
approach over other tracking system, because it can simultaneously actuate the panel to move in two
directions to locate the position of the sun with little or no margin of error and provides the solar panels
with extensive freedom of movement. We have seen several advantage and disadvantages of tracking
system based on their design and application. In this project an automated azimuth-altitude dual-axis
solar tracker will be built for the selected particular location.
WU ECeng 10
Chapter Three
3. System Design
The purpose of a solar tracker is to accurately determine the position of the sun. This enables solar
panels interfaced to the tracker to obtain the maximum solar radiation. With this particular solar tracker
a closed-loop system was made consisting of an electrical system and a mechanical system.
Figure 3.1:General block diagram of the system
Major functional blocks of tracking system are shown in figure 3.1. For the purpose of clarity, the east-
west of the tracker will be called the “horizontal tracking” while the angular height tracker will be
referred to as “vertical tracking”. The tracker will be based on an embedded system circuit using a
microcontroller (MCU). Two light dependent resistors (LDR) are used for each degree of freedom. LDRs
are basically photocells that are sensitive to light. Software will be developed which would allow the
MCU to detect and obtain its data from the two LDRs and then compare their resistance. The two LDRs
will be positioned in such a way, so that if one of the two comes under a shadow, the MCU will detect
the difference in resistance and thus actuate the motor to move the solar panel at a position where the
light upon both LDRs is equal. Limit switches are added to the system. When the limit switch is triggered
at the end of the day, the tracker will move back to its original position. Two separate but identical
circuits will be utilized for both axes.The system is then connected to a battery bank via a charge
controller and DC voltage from the solar panels is converted to AC voltage through an inverter and
supplied to the necessary load.
WU ECeng 11
3.1. Mounting System
The mounting system refers to the structure which holds the solar panels. Since the purpose of the
mounting system is to point the solar panels towards the sun throughout the day, the structure consists
of movable and fixed parts.
To decide how the tracker would move and the mounting structure should be sited up in this project, it
is important to consider the movement of the sun in the sky throughout the year Hawassa. The sun path
diagram below shows the annual variation of the path of the sun in Hawassa
Figure 3.2: Sun path diagram for Hawassa, Ethiopia
From the sun path diagram, the movement of the sun in the sky throughout the year in Hawassa can be
divided into three different scenarios. As the sun rises from the East to sets to the West, the sun path
may move in the Southern or Northern region respectively, or it may move almost directly overhead.
If the path of the sun is in the Northern region, the structure must be able to track the sun from East to
West in anti-clockwise direction. If the path of the sun is in the Southern region, the structure must be
able to track the sun from East to West in clockwise direction. If the sun is moving overhead, only the
axis which tracks the angular height of the sun will move. In all three situations, there must be a way to
turn back the tracker to its original position after it has followed the movement of the sun from morning
to dusk. To achieve this, limit switches are added to the system. When the limit switch is triggered at the
end of the day, the tracker will move back to its original position. Taking the above three scenarios into
consideration the initial design of the structure are shown in the diagram below:
WU ECeng 12
Figure 3.3: (a) initial sketch (b) front and side View
3.1.1. Design Criteria
The design of the structure and the materials used to make the mounting system is based on a set of
criteria. As stated in the scope of the project, the project aims to generate 160w of power from the
installed panel. Therefore before designing the mounting structure the area of the solar panel used to
generate the desired power must be determined.
3.1.1.1. Major system components to design solar power
Solar PV system includes different components that should be selected according to the system type,
site location and applications. The major components for solar PV system are solar charge
controllerinverter, battery bank, auxiliary energy sources and loads (appliances).
WU ECeng 13
• PV module – converts sunlight into DC electricity.

• Solar charge controller – regulates the voltage and current coming from the PV panels going to
battery and prevents battery overcharging and prolongs the battery life.
• Inverter – converts DC output of PV panels or wind turbine into a clean AC current for AC
appliances or fed back into grid line.
• Battery– stores energy for supplying to electrical appliances when there is a demand.
• Load – is electrical appliances that connected to solar PV system such as lights, radio, TV, computer,
refrigerator, etc.
Solar PV system sizing
Step 1. Determine power consumption demands

The first step in designing a solar PV system is to find out the total power and energy consumption of all
loads that need to be supplied by the solar PV system as follows:
1.1 Calculate total Watt-hours per day for each appliance used.
Add the Watt-hours needed for all appliances together to get the total Watt-hours per day which
must be delivered to the appliances.
1.2 Calculate total Watt-hours per day needed from the PV modules.
Multiply the total appliances Watt-hours per day times 1.3 (the energy lost in the system) to get
the total Watt-hours per day which must be provided by the panels.
Step 2: Size the PV modules

Different size of PV modules will produce different amount of power. To find out the sizing of PV
module, the total peak watt produced needs. The peak watt (Wp) produced depends on size of the PV
module and climate of site location. We have to consider “panel generation factor” which is different in
each site location. For Hawassa, the panel generation factor is 3.43. To determine the sizing of PV
modules, calculate as follows:
2.1 Calculate the total Watt-peak rating needed for PV modules

Divide the total Watt-hours per day needed from the PV modules (from item 1.2) by 3.43 to get
the total Watt-peak rating needed for the PV panels needed to operate the appliances.
2.2 Calculate the number of PV panels for the system

Divide the answer obtained in item 2.1 by the rated output Watt-peak of the PV modules available
Increase any fractional part of result to the next highest full number and that will be the
number of PV modules required.
Result of the calculation is the minimum number of PV panels. If more PV modules are installed, the
system will perform better and battery life will be improved. If fewer PV modules are used, the system
may not work at all during cloudy periods and battery life will be shortened.
WU ECeng 14
Step 3: Inverter sizing

An inverter is used in the system where AC power output is needed. The input rating of the inverter
should never be lower than the total watt of appliances. The inverter must have the same nominal
voltage as the battery.
For stand-alone systems, the inverter must be large enough to handle the total amount of Watts you will
be using at one time. The inverter size should be 25-30% bigger than total Watts of appliances. In case of
appliance type is motor or compressor then inverter size should be minimum 3 times the capacity of
those appliances and must be added to the inverter capacity to handle surge current during starting.For
grid tie systems or grid connected systems, the input rating of the inverter should be same as PV array
rating to allow for safe and efficient operation.
Step 4: Battery sizing

The battery type recommended for using in solar PV system is deep cycle battery. Deep cycle battery is
specifically designed for to be discharged to low energy level and rapid recharged or cycle charged and
discharged day after day for years. The battery should be large enough to store sufficient energy to
operate the appliances at night and cloudy days. To find out the size of battery, calculate as follows:
4.1 Calculate total Watt-hours per day used by appliances.

4.2 Divide the total Watt-hours per day used by 0.85 for battery loss.
4.3 Divide the answer obtained in item 4.2 by 0.6 for depth of discharge.
4.4 Divide the answer obtained in item 4.3 by the nominal battery voltage.
4.5 Multiply the answer obtained in item 4.4 with days of autonomy (the number of days that you
need the system to operate when there is no power produced by PV panels) to get the required
Ampere-hour capacity of deep-cycle battery.
Battery Capacity (Ah) = Total Watt-hours per day used by appliances x Days of autonomy
(0.85 x 0.6 x nominal battery voltage)
Step 5: Solar charge controller sizing

The solar charge controller is typically rated against Amperage and Voltage capacities. Select the solar
charge controller to match the voltage of PV array and batteries and then identify which type of solar
charge controller is right for your application. Make sure that solar charge controller has enough
capacity to handle the current from PV array.
For the series charge controller type, the sizing of controller depends on the total PV input current
which is delivered to the controller and also depends on PV panel configuration (series or parallel
configuration).
According to standard practice, the sizing of solar charge controller is to take the short circuit current
(Isc) of the PV array, and multiply it by 1.3
Solar charge controller rating = Total short circuit current of PV array x 1.3
WU ECeng 15
The desired produced power from the PV in this particular project isused for the following electrical
appliance utilization found in St Gabriel mini supermarket located in student complex building behind
the auditorium building:
• One 18 Watt fluorescent lamp with electronic ballast used 4 hours per day.
• One 60 Watt fan used for 2 hours per day.
• One 75 Watt refrigerator that runs 24 hours per day with compressor run 10 hours and off 14
hours.
1. Determine power consumption demands
Total appliance use = (18 W x 4 hours) + (60 W x 2 hours) + (75 W x 24 x 0.42 hours)
= 948Wh/day
Total PV panels energy needed = 948 x 1.3
= 1,232.4Wh/day.
2. Size the PV panel
2.1 Total Wp of PV panel capacity = 1,232.4 / 3.43

needed
= 159.3Wp
2.2 Number of PV panels needed = 159.3 / 180
= modules
Actual requirement = 2 modules
So this system should be powered by at least modules of 180 Wp PV module.
Chosen PV module is SUNTECH STP-18OS, its maximum power current (mppI) =5A
Therefore, the PV modules are connected in series to have a system voltage of 24v and will be placed on
the mount as shown in the figure below:
3. Inverter sizing
Total Watt of all appliances = 18 + 60 + 75 = 153 W
For safety, the inverter should be considered 25-30% bigger size.
The inverter size should be about 190 W or greater.
∴Thus, a total of 190/200=0.95=1inverter is needed
The inverter chosen in this design is APOLLO S-211Cwhich has a power of 200W.
4. Battery sizing
Total appliances use = (18 W x 4 hours) + (60 W x 2 hours) + (75 W x 10 hours)
Nominal battery voltage = 12 V
Days of autonomy = 3 days
WU ECeng 16
Battery capacity = [(18 W x 4 hours) + (60 W x 2 hours) + (75 W x 10 hours)] x 3

(0.85 x 0.6 x 12)
Total Ampere-hours required 461.77 Ah/day

Maximum number of continuous cloudy days expected 1.5 days
461.77Ah / day ×1.5days = 692.65Ah
Divide the total ampere-hour 0.8 to maintain a 20% reserve after deep dischargeperiod.
692.65 Ah/0.8=865.81Ah
VISION IEL12150
Therefore the battery chosenisVISION
The total number of batteries in parallel required.
∴865.81Ah/150Ah =5.018≈ 6 batteries
Number of batteries required in series= 48/12 = 4
Finally the total number of the chosen battery needed forthe system.
∴6*4= 24 batteries
5. Solar charge controller sizing

PV module specification
Pm = 180Wp
Vm = 36Vdc
Im = 5 A
Voc = 44.8 A
Isc = 5.29 A
Solar charge controller rating = (2 strings x 5 A) x 1.3 = 13 A
So the solar charge controller should be rated 13 A at 12 V or greater.
Therefore the chosen charge controller isLEONICS SOLARCON SPT 2420
Secondly, the structure must be able to support the weight of the solar panels which are mounted on it.
In this project, 2solar panels are used.
Weight of 1 solar panel = 15.5kg

Weight of 4 solar panels = 2*15.5=31Kg
Besides that, the column and the base of the structure should also be able to support the weight of the
frame, which is estimated to be about 70kgs. That gives a total weight of slightly more than 100kgs.
Thirdly, since the structure will be erected outdoors, the structure must be able to withstand the
elements of nature, most notably the effects from the sun (heat), rain (water) and the wind (air). Of
utmost concern will be effect of wind load on the structure when wind load is acting upon the solar
panels. Based on the recorded maximum wind speed data of Hawassa which is 6mph. and assuming the
wind flow is acting perpendicularly upon the maximum area of the solar panels, the wind load is
calculated using the generic formula:
WU ECeng 17
Maximum wind speed in Hawassa = up to 6mph

For design purpose, consider wind speed up to 8mph
Wind load: Force, F = A x P x Cd
A = the projected area of the item
= 2.553m2
P, Wind pressure (Psf) = 0.00256 x V2(V= wind speed in Mph) = 0.00256 x 82
= 0.16384 = 0.8 Kg/m2
Cd , Drag coefficient, = 2.0 for flat plates.
∴Wind load force = A x P x Cd = 2.553m2x 0.8 Kg/m2x 2.0
= 4.0848Kg ≈40.848 N
The calculation shows that the material chosen to make the structure would need to be able to
withstand a perpendicular acting force of 40.848 N.
Finally, the movable parts of the structure must be able to rotate to follow the movement of the sun
throughout the day. A double-axis solar tracker means the tracker must be able to rotate along the
vertical axis to follow the movement of the sun from East to West, and also rotate along the horizontal
axis to follow the position of the sun’s angular height in the sky. In that manner, the structure would be
able to point the solar panels towards the position of the sun in the sky.
3.1.2. Material And Design
Having taken into account the criteria for the mounting system, ordinary structural steel was chosen to
be the material for the structure. The ultimate tensile strength is about 50000 psi while the yield
strength is about 30000 psi. Based on the strength of steel, the structure would be able to withstand the
effect of wind load as well as the other weight loads placed on it.
After discussion with the 4th year mechanical Engineering students it was decided that the design was to
be optimized and the design shown in figure 8 was proposed and agreed upon:
From the diagram, it can be seen that at each end of the frame, a pillow block bearing is used to enable
the frame to rotate along the horizontal axis. A metal shaft which goes through the center of the frame
connects the two bearings. Also, the DC motor controlling the vertical movement is placed on one side
of the frame. The upper part of the column is designed to hold the other DC motor which controls the
horizontal movement. The lower part of the frame was then attached to the upper part of the column
using two flange bearings which are connected by a shaft. The bearings are used to enable the structure
to turn about the vertical axis.
WU ECeng 18
Figure 3.4: Finalized design of the structure
3.1.3. Motor
There are three possible alternatives to actuate the solar tracker: a stepper motor, a further geared
motor, or hydraulics. Both the stepper motor and further geared motor provide a significant decrease in
the overall cost. Hydraulics might be better in power efficiency when moving large panel arrays, but it is
uncertain if it would decrease cost.
The motor chosen for the proposed system is 12V DC geared motor with 60 rpm, coupled with a worm
gear. The gear ratio of the worm gear is 1:25. The main reason for the selection of the geared motor is
that it will consume very less power for a very small span of time. Also the torque required to rotate
heavy PV arrays is sufficient enough, which has a torque of about 12.5Nm to 20Nm.To justify the use of
this motor, the following calculation was done to ensure that it has enough power to rotate the
structure:
Assuming that the mass of the structure, m to be moved is 100kg, gravitational acceleration, g = 10ms-2,
the radius of the worm wheel, r = 10cm = 0.1m, and taking values of coefficient of friction of the bearing
= 0.001 and the coefficient of friction between the worm and the worm-wheel (steel and steel) =0.42,
the torque needed is calculated:
WU ECeng 19
Torque = F × r, where F = mgcf

= (mgcf) × r = (100kg × 10ms-2× 0.001 × 0.42) × 0.1m
= 0.042Nm (which is within the torque range of the motor)
3.1.4. Coupling system
Coupling system is a method of connecting the mounting structure to the shaft of the motor to rotate
the structure accordingly to the desired position. There are two options to couple the structure with the
motor:
I, Rubber Belt system
II, Worm Gear
If rubber belt is employed for the coupling system a sprocket is affixed unto the shaft of the motor, unto
one end of the shaft, besides the pillow block bearing and at the end of the shaft on top the flange
bearing. A chain is then put in place so that when the motors are powered, the structure can rotate.
However, there is one big problem with the system: the rubber belts will encounter the mechanical
damping that is not great enough to counteract the inertia of the platter when the power to the motor
terminated. This is not ideal for a solar tracking system as the tracker needed to move both slowly and
with minimal influence from inertia. This led to the next setup using worm gears.
Worm gears are capable of high speed-reduction ratios as well as ensuring that there is no inertial
backlash to the driving source.
Figure 3.5: Worm gear system locking.
A worm gear is used when a large speed reduction ratio is required between crossed axis shafts which
do not intersect. A basic helical gear can be used but the power which can be transmitted is low. A
worm drive consists of a large diameter worm wheel with a worm screw meshing with teeth on the
periphery of the worm wheel. The worm is similar to a screw and the worm wheel is similar to a section
WU ECeng 20
of a nut. As the worm is rotated the worm wheel is caused to rotate due to the screw like action of the
worm.
Worm gears provide a normal single reduction range of 5:1 to 75-1. The pitch line velocity is ideally up
to 30 m/s. The efficiency of a worm gear ranges from 98% for the lowest ratios to 20% for the highest
ratios. As the frictional heat generation is generally high the worm box is designed disperse heat to the
surroundings and lubrication is and essential requirement. Worm gears are quiet in operation. Worm
gears at the higher ratios are inherently self-locking - the worm can drive the gear but the gear cannot
drive the worm. A worm gear can provide a 50:1 speed reduction but not a 1:50 speed increase.
A worm is affixed unto the shaft of the motors and a worm wheel unto one end of the shaft, besides the
pillow block bearing and at the end of the shaft on top the flange bearing, the structure can rotate. The
gear ratio used is 1:25, 1 on the motor and 25 on the tracker.
3.1.5. Motor braking
After the power given to the DC motor, controlled by the microcontroller, is removed from the motor
the tracker is not required to move because the elapsed time power given to the microcontroller is
exactly the time required moving the tracker. But due to the inertial of the motor after the power is cut
the tracker may move therefore, an appropriate braking system must be employed to the system.
To perform this action an electro-mechanical clutch are incorporated in the control system. Therefore
the chosen clutch is WARNER electric ERD series.
ERD Series brakes are designed to safely keep the load in position.By applying voltage to the ERD,an
electromagnetic field is created those the springs force the armature to compress the friction carrier
against the mounting flange, thus stopping and holding the load.When power is removed, this causes
the armature plate to pull-in against helical compression springs, thus releasing the brake. Fully dynamic
friction material on the carrier allows for repeated braking cycles from full motor speed with no torque
fade.Technical specification is attached in the appendix.
3.1.6. Installation Site
For a solar power system to choose installation site, first, it is important to note that there are no
shadows cast upon the solar panel. Any shadows which are cast upon the solar panel can greatly affect
the output of electric current. It is therefore important to ensure that during day-time at the site for the
installation of the system, no shadows from trees, buildings or other tall objects such as poles are cast
onto the solar panels. Secondly,check the solar radiation of the installation site. This is done by taking
the longitude and latitude from Google earth and checking the insolation level from NASA surface
metrology and thus, the site selection for this project took into consideration those issues. The selected
site is at the open space behind Auditorium building.
WU ECeng 21
3.2. Electrical Control System
To produce a useful solar tracker the electrical system needs to give accurate control signals to the
mechanical system, be reliable, and have low power consumption and using digital microcontroller for
improved efficiency seemed ideal for providing smooth and accurate control of the mechanical system.
Thus, the electrical system consisted of power supply to the system, solar sensors, a digital
microcontroller and an H-bridge.
3.2.1. Motor Control
As described previously, the motor chosen was a DC motor. This motor can rotate in both directions by
reversing the direction of the current supplied. Because the power supply on the tracking system has
only one pole, a circuit was needed to switch the direction of the current to the motor. The most power
efficient way to accomplish this is with a circuit known as an H-bridge which a simplified schematic is
provided in Figure below.
Q1 Q2 Q3 Q4 Current path
Low High Low High
High Low High Low
Figure 3.6: Basic H-bridge Circuit
Here, two pairs of MOSFETs work in conjunction with each other to provide a current through the motor
in a certain direction. The upper MOSFETs are P-channel devices and the lower devices are N-channel.
Using P-channel devices negates the need to use a voltage doubler to get the gate voltage high enough
to turn the device on. When one P-channel MOSFET allows current to pass through and the N-channel
MOSFET diagonally across from it also does, the motor spins in one direction. If these are both turned
“off” and the other two MOSFETs turned “on”, the motor spins in the opposite direction. The table
WU ECeng 22
inFigure 12 above shows what current paths coordinate with which voltage values at the gates of the
MOSFETs.
For the H-bridge, the transistors chosen had to be both power efficient fit with the mantra of the project
as well as inexpensive and widely available for production reasons. MOSFETs were chosen for their high
power efficiency compared to other transistor technologies like the Bipolar Junction Transistor (BJT)
which actively draws current in addition to the current flowing through the device. Within the range of
MOSFETs there are also power MOSFETs with a different internal construction that give them high
current and/or high voltage capabilities. These power MOSFETs generally have a much lower on-
resistance than standard MOSFETs and therefor have a lower power dissipation and higher efficiency. To
further improve power efficiency, a complementary pair was chosen. Complementary transistor pairs
are two devices, in this case, one N-channel and one P-channel MOSFET, that are made in a way so that
they operate as near identical opposites of each other. For this project where the MOSFETs are acting as
switches, the pair switches at the same speed. Based on previous experience, this quality is especially
important to prevent the shorting of the H-bridge. If the P and N-channel MOSFETs on one side of bridge
switch at different speeds, one MOSFET can stay on for too long of a time and when the other MOSFET
turns on, the H-bridge can short and burn out one of the transistors. A complementary pair negates this
by ensuring that there is always some amount of resistance between the two power rails during
switching.
Another major consideration in choosing the MOSFETs was the power consideration. The internal on-
resistance of the MOSFET determines how much power is dissipated as heat when current is flowing
through the device. For maximum power efficiency, the amount of power dissipated has to be minimal
and therefore the resistance has to be minimal. Additionally, this dissipated power can damage the
MOSFET if the device is not properly cooled. For cost purposes, both heat sinks and active cooling
systems were avoided and MOSFETs were chosen that had a low enough power dissipation such that
they did not require additional cooling measures. For this the thermal resistance of the devices has to be
taken into account.
The thermal resistance is a measure of how many degrees in temperature the device will rise in relation
to the ambient temperature when dissipating a certain amount of power, measured in Watts. Because
there will be no heat sinks on the H-bridge, the total thermal resistance will be from the device junction
to the air. This measure is often included in device datasheets as RθJA. The lower this measurement, the
more power can be dissipated without additional cooling. To calculate the maximum allowable thermal
resistance, the equation is:
ΔT / P = Rθmax
Where P is the power dissipated by the device and ΔT is the difference between the ambient
temperature and the maximum operating temperature of the device. Knowing that the tracker will be
operating near the equator where record high temperatures have been recorded; the ambient
temperature in the worst case scenario can take the value of 50°C. Also assuming the MOSFET has an
WU ECeng 23
atypically low maximum operating temperature of 125°C, and then ΔT under the worst-case conditions
is 75°C. The power dissipated by the device can be calculated as the product of the internal resistance
and the current squared:
P = I2R
Knowing that the maximum current drawn by the motor during testing was ≈400mA, the equation to
calculate the maximum thermal resistance is now:
25°C / (0.160 * RON) = Rθmax
Power MOSFETs typically come in a non-insulated TO-220 package which has a typical thermal
resistance junction-to-air of ≈65°C/W. ThereforeRθmax> 65°C/W. plugging this inequality into the above
equation and solving for RON, the minimum on-resistance can be calculated as:
RON< 2.404 Ω
This calculation is a maximum rating that the power MOSFETs used in the H-bridge can have assuming
their maximum operating temperature is 125°C and the thermal resistance junction-to-air is 65°Taking
these specifications into consideration and keeping in mind that any components chosen have to be
available across the world to reduce repair costs, the power MOSFETs chosen was the complementary
pair of IRF530 N-channel MOSFETs and IRF9530 P-channel MOSFETs. This pair exceeds the required
specifications by a wide margin with a maximum current rating of 12A, maximum operating
temperature of 175°C and junction-to-air thermal resistance of 62.5°C/W. Most importantly, this is a
widely used pair of power MOSFETs ensuring that they are available to most parts of the world and have
a very low cost. Additionally, the specifications of the MOSFETs allowlarger motors to be used for
scaled-up tracking systems without significant electrical changes to the electrical system. Motors
drawing ≈2A of power would still not necessitate additional cooling of the H-bridge even at 50°C
ambient temperatures.
For many of the control systems, and in particular, the microcontroller, they could not send a signal all
the way to the power rail.Sending an input at the ground voltage to the P-channel MOSFETs to reach the
lowest on-resistance which gives the best power efficiency is the problem that will occur. For this
reason, an NPN BJT on the gate of the P-channel MOSFETs was used to allow the signal to reach ground.
In the same sense, an N-channel MOSFET needs to reach the power supply voltage to have the lowest
on-resistance. For this reason, an NPN BJT was also added to the gates of the N-channel MOSFETs. This
also meant that the inputs to the H-bridge were inverted; a high signal turned on the P-channel
MOSFETs and a low signal turned the N-channel MOSFETs on. In keeping with the idea of using widely
available parts, the common 2N3904 NPN BJT can be used in the H-bridge.
Below in Figure 3.7 is the schematic for the final H-bridge circuit block. The diodes connected across the
drain and source terminals of the MOSFETs are for the inductive kickback from the motor. This kickback
WU ECeng 24
is a high voltage spike which can easily damage the MOSFETs. The diodes are so that if the voltage gets
above a certain value, they effectively short the motor to the rails and the spike is eliminated. The
1N4004 rectifier diode is ideal for this application because of its high surge current capability and wide
availability.
Because the motor will be moving for a very small percentage of the day, the power used when the H-
bridge is “off” or not supplying any power to the motor needs to be minimal. The resistors in the circuit
were chosen so that when the H-bridge is off, the current draw is <5mA for efficiency. Testing the circuit
with the motor parameters in Multisim shows the total current draw when the H-bridge is “off” is 4.65
mA which is 55.8 mW. Higher value of resistors would lower the current draw but would also increase
the switching times of all the MOSFETs which would increase the transient current draw.
Figure 3.7: Revised H-bridge Schematic
When the motor is switching on and off as well as reacting to changes in its mechanical load, the current
drawn can change quickly and will drop the voltage the power supply can deliver. The change in the
voltage is called the ripple amount and can be alleviated with a capacitor to supply the surge currents.
To calculate the ideal value for the capacitor, the equation for the ripple in the voltage supply is:
VPP = IL / (f * C)
Where VPP is the peak-to-peak ripple voltage, IL is the peak current draw, f is the frequency of the
oscillations, in this case the PWM frequency and C is the capacitor’s value. The frequency of the PWM
used in the final circuit, supplied by the microcontroller, is 500Hz according to the datasheet. The
maximum load current is 400mA from the maximum measured current going to the motor during
testing. For a desired amount of ripple to be 0.5 volts, less than 5% deviation from the 12V power
supply, the optimal capacitor value can be calculated:
WU ECeng 25
0.5V = 0.4A / (500Hz * C)

C = 1600µF
This capacitor was added between the 12V power rail and ground for both H-bridges used, as seen in
Figure 3.8
Figure 3.8: Final H-bridge Schematic
3.2.2. Solar Sensor Array
After choosing and building the optimal design alternative, the control and feedback systems had to be
put into place to make the design function as a solar tracker. Although the actual task of tracking the sun
is not extraordinarily complicated, this particular project had a lot of things to consider since the tracker
needed to be able to track autonomously for an extended period of time. At the end of each day the
tracker needed to be able to reposition itself so it could find the sun in the morning and start tracking
again. These and other issues had to be accounted for when programming the tracker’s motion, timing,
and response to environmental stimuli.
In design of this project the sensor system consists of two sensors: one to determine the position of the
sun in the sky and another to determine the position of the sun’s movement from east to west. Each
sensor consists of two light sensors.
Figure 3.9: sensor placement
WU ECeng 26
The sensors were placed as shown in the diagram above. As illustrated in the next diagram, a shadow
will fall on one of the sensors when the sensor is not pointing directly toward the sun resulting in
difference of the level of output between the two sensors. This difference will be detected by the
microcontroller in the control system and will move the tracker accordingly so that both sensors are
pointing towards the sun.
Figure 3.10: The sensor responding once a shadow is cast on one LDR
To provide an accurate and reliable tracking of the sun, a solar sensor array was designed.The solar
sensors were primarily chosen based on linearity of the output response due to the change in the angle
of incidence. The length of barrier between the two sensors had to be chosen that obtained the highest
height response.
Based on the above criteria and availability, photo-resistors and photovoltaic cells were considered as
possible sensors. And for this project Cadmium Sulphate (CdS) light dependent resistors (LDRs) is
chosen.
3.2.2.1. Light dependent resistor
A Light Dependent Resistor (LDR) or a photo resistor is a device whose resistivity is a function of the
incident electromagnetic radiation. Hence, they are light sensitive devices. They are also called as photo
conductors, photo conductive cells or simply photocells. They are made up of semiconductor materials
having high resistance.
Working Principle of LDR
A light dependent resistor works on the principle of photo conductivity. Photo conductivity is an optical
phenomenon in which the materials conductivity (Hence resistivity) reduces when light is absorbed by
the material.When light falls i.e. when the photons fall on the device, the electrons in the valence band
of the semiconductor material are excited to the conduction band. These photons in the incident light
should have energy greater than the band gap of the semiconductor material to make the electrons
jump from the valence band to the conduction band. Hence when light having enough energy is incident
on the device more & more electrons are excited to the conduction band which results in large number
of charge carriers. The result of this process is more and more current starts flowing and hence it is said
that the resistance of the device has decreased. This is the most common working principle of LDR.
WU ECeng 27
Figure 3.11: (a) Characteristics of LDR (b)diagram of the light detection system used in this project
The relationship between the voltages at Voutis given by:
From this, it was clear when there was light on the photo-resistor, Vout would be larger.
3.2.2.2. Sensor Holder Design
The workings of the sensor have been described in above section of this thesis.The length of barrier
placed in between the two sensors had to be chosen that obtained the highest height response.
For the initial design, two photo-resistors can be used, each with a measured dark resistance of certain
resistance value.In order to determine a barrier length that provided the best accuracy and the best
response to a change in the angle of incidence, a testing setup with a 60 watt light bulb can be built. The
adjustable mount allowed for different barrier length. The light bulb can be positioned approximately a
foot from the adjustable sensor holder mount. And allow the light bulb to be rotated around the
mount, while maintaining the foot long distance. This setup is seen in Figure 3.12. The figure represents
the angle of incidence as θ and the barrier length
-5 +5
-θ+θ
Figure 3.12: Setup to Test for Ideal barrier length
WU ECeng 28
For testing the photo-resistors were each setup in a voltage divider with a 1 V source and a 5kΩ resistor,
Figure 3.13: Photo-resistor Voltage Divider Circuit
The angle of incidence from the bulb will be varied by 5° from -90° to 90° and the voltage difference
between the two photo-resistor voltage dividers can be measured and recorded for different barrier
length.
3.2.3. Limit switch
Practically, there must be a way to turn back the tracker to its original position after it has followed the
movement of the sun from morning to dusk. To achieve this, limit switches are added to the system.
When the limit switch is triggered at the end of the day, the tracker will move back to its original
position.
Three waterproof magnetic contact switches are used as limit switches for the system. Two are used for
the horizontal movement, while the other one is used for the vertical movement. The switches are
configured at the ‘normally open’ position. When the switches come in contact with a magnet, the
switch is turned on.
The switches and magnets are placed in such a way that they will come in contact when the frame is at
the position of facing the path of the setting sun. The switch is placed at a permanent position, while the
magnet is attached unto the frame, so thatas the frame rotates, the magnet will rotate together until it
comes in contact with the limit switch.
3.2.4. Digital System
Because of the drawbacks present in the analog control system, a microcontroller-based, digital control
system is employed in the project. The microcontroller allowed a minimal control circuit complexity,
reduced power consumption and allowed for additional features to be introduced to the tracker.
WU ECeng 29
The microcontroller selected had to have at least four analog-to-digital converter (ADC) inputs to take
the four signals from the four sensors. It also had to have a minimum of four digital outputs, two for
both of the motor H-bridges. Since the horizontal and vertical axis are independently controlled by one
microcontroller each, have the number listed above are used to select the microcontroller. Finally, the
microcontroller used had to have a very low power consumption when active to keep the efficiency as
high as possible.
A thorough research was conducted on the various types of PIC MCUs available in the market.
Eventually, the criteria for selecting a particular PIC for this project boiled down to two basic elements:
cost and options. Although some PICs were half the cost of PIC16F84A, they were either an OTP (one-
time programmable) MCUs or didn’t offer the required functions. OTP PICs would be unfeasible as
constant changes would be required in developing and prototyping the code for the tracker. Other High-
end PICs were a bit expensive but offered wide variety of special features. Once again, such PICs were
beyond the scope and requirement of this project. PIC16F84A proved to be the best option in terms of
cost and the required features. PIC16F84A is an 18-pin enhanced FLASH/EEPROM, 8-bit microcontroller.
It is one of the most popular PIC microcontrollers in recent times.Since PIC16F84Ais an 8-bit
microcontroller, it offers 1024 by 14 flash program memory for storing a written code. It includes 36
bytes of data RAM, 64 bytes of non-volatile EEPROM data memory, 13 I/O pins, a timer, a watchdog, and
internal and external interrupt sources. The timer is 8-bits wide but can be programmed to generate
internal interrupts for timing purposes. PIC16F84A can be operated from a crystal or a resonator for
accurate timing. A resistor-capacitor can also be used as a timing device for applications where accurate
timing is not required. In order to run, PIC16F84A requires a regulated voltage between 2.0V to 5.5V.
3.2.4.1. PIC16F84A Architecture Detail
There are three types of PIC16F84A packaging designs available in the market: PDIP (18-Lead Plastic
Dual In-line), SOIP (18-Lead Plastic Small Outline), and SSOP (20-Lead Plastic Shrink Small Outline). PDIP
type packaging will be used for the solar tracker embedded design. Figure20illustrates the PIC16F84A
PDIP design and table 4 shows the name and pin positions.
Figure 3.14:PIC16F84A Pin Positions
WU ECeng 30
Pin No Name Function

1 RA2 – PORTA bit 2 Second pin on port A. Has no additional function.
2 RA3 – PORTA bit 3 Third pin on port A. Has no additional function
3 RA4/T0CK1 Fourth pin on port A. TOCK1 which functions as a
timer is also found on this pin
4 MCLR – Master clear Reset input and Vss programming voltage of PIC
5 Vss – Gnd Ground of power supply
6 RB0/INT – PORTB bit 0 Zero pin on port B. Interrupt input is an extra function
7 RB1 – PORTB bit 1 First pin on port B. No additional function.
8 RB2 – PORTB bit 2 Second pin on port B. No additional function.
9 RB3 – PORTB bit 3 Third pin on port B. No additional function.
10 RB4 – PORTB bit 4 Fourth pin on port B. No additional function
11 RB5 – PORTB bit 5 Fifth pin on port B. No additional function.
12 RB6 – PORTB bit 6 Sixth pin on port B. 'Clock' line in program mode.
13 RB7 – PORTB bit 7 Seventh pin on port B. 'Data' line in program mode.
14 Vdd + V supply Positive power supply of +2.0V to +5.5V
15 OSC2 Pin assigned for connecting with an oscillator
16 OSC1 Pin assigned for connecting with an oscillator.
17 RA0 – PORTA bit 0 Second pin on port A. No additional function.
18 RA1 – PORTA bit 1 First pin on port A. No additional function.
Table 3.1:PIC16F84A Pin description
Microcontroller Features:
RISC CPU Features:
• 10,000 erase/write cycles EnhancedFLASH
Program memory typical • Only 35 single word instructions to learn
• 10,000,000 typical erase/write cycles EEPROM • All instructions single-cycle except for
Data memory typical program branches which are two-cycle
• EEPROM Data Retention > 40 years • Operating speed: DC - 20 MHz clock input
• In-Circuit Serial Programming™ (ICSP™) - via DC - 200 ns instruction cycle
two pins • 1024 words of program memory
• Power-on Reset (POR), Power-up Timer • 68 bytes of Data RAM
Oscillator Start-up Timer (OST) • 64 bytes of Data EEPROM
• Watchdog Timer (WDT) with its own On-Chip • 14-bit wide instruction words
RC Oscillator for reliable operation • 8-bit wide data bytes
• Code protection • 15 Special Function Hardware registers
• Power saving SLEEP mode • Eight-level deep hardware stack
• Direct, indirect and relative addressing modes
Peripheral Features: • Four interrupt sources:
- External RB0/INT pin
• 13 I/O pins with individual direction control - TMR0 timer overflow
• High current sink/source for direct LED drive - PORTB<7:4> interrupt-on-change
- 25 mA sink max. per pin - Data EEPROM write complete
- 25 mA source max. per pin
• TMR0: 8-bit timer/counter with 8-bit
Programmable prescaler
WU ECeng 31
3.2.4.2. PIC Software Development Tools

This section will focus on the development tools required to write, debug, compile and then program
the code onto the PIC16F84A PICBasic language will be used for the software development of the solar
tracker. Following is the list of parts required to develop, program, and run the PIC:
PICBasic IDE
PICBasic compiler
Programmer Software
JDM Programmer
4 MHz Crystal Oscillator
The process required for successfully programming PIC16F84A is included in the following diagram 3.1.
PIC Basic Pro language IDE for code A compiler to

to write the code developing & debug compile &convert
the code to HEX file
Upload binary code to Programmer software that

the microcontroller converts HEX file into Binary
Diagram 2: PIC Programming Procedure
The advantage associated with PICBasic when compared with Assembly or C is that it is more user-
friendly, easier to master, and its compilation speed is 20 to 100 times faster than the other two.
PICBasic Pro also allows the programmer to use simple and shorts commands for certain function,
whereas that same function requires pages of code when assembly or C language is used.
PICBasic PRO Program structure
As like any other well written program, PICBasic program should be structured in a manner so that it is
comprehensible and maintainable by any other programmer. This approach is essential when working at
firm, where the written may need to be reviewed, maintained or upgraded by different programmers.
Following steps are required in order to obtain a well-structured PICBasic Pro code:
Include a header at the beginning of the code briefly explaining the function of the program. The
header should consist of information such as, the authors name, date, program file name, etc.
WU ECeng 32
Frequent use of comments is an important aspect of good coding style as it makes the code
comprehensible for others. Comments should be used to describe the commands used in the
code.
Use logical symbols in order to refer topins or variables, instead of just naming them Pin2 or B2.
PICBasic PRO IDE and compiler
Although PICBasic code can be successfully written in a windows notepad, this may not be the preferred
way to write the code. In order to develop any code for any application, it is important to use An
Integrated Development environment (IDE) developed for the particular language. The IDE offers
features such as: full syntax highlighting of the code with context sensitive keyword help and syntax
hints. The IDE code explorer should allow the programmer to automatically jump to include files,
defines, constants, variables, aliases and modifiers, symbols and labels, which are contained within the
source code.
The compiler on the other hand compiles the written code and informs of any errors in case the
complication process is not successful. Compiler is essential for error correction and debugging. Most
IDEs available in the Market have a built in compiler capability. Apart from compiling the code, the
compiler also converts the PICBasic Pro code into a HEX format file which can later be programmed onto
the PIC MCU using a programmer.
Microcode Studio
Microcode Studio was used for the software development in this project. Unlike many PIC programming
and compiling software available in the market. Microcode Studio offers both a comprehensive
integrated development environment along with a PICBasic Pro compiler.
Development Suite can be done in following 3 simple steps:
1. Write the code in PICBasic using the Proton IDE

2. Use the built in Proton plus Compiler to compile and obtain the HEX format of the written code.
3. Test the code using the Virtual Simulator, or upload the code directly to the PIC using Proton
IDE's programmer integration feature.
3.2.4.3. Implementation of PIC in the Solar Tracker
The microcontroller’s function is to compare the voltage levels of the two sensors for one axis of
rotation and send the appropriate signal to the H-bridge to move the motor in a certain direction. To
reduce the amount of jitter the system has, the microcontroller takes the two voltage values of the
sensors and finds the difference between the two. The difference is compared to some reference value.
If it is greater than the positive reference value, the microcontroller sends a signal to the H-bridge to
move the tracker in one direction. If the difference is less than the negative reference, the
microcontroller sends a signal to move the tracker in the opposite direction.The ADC of the
microcontroller is 10 bits and effective over a 5V range which has a resolution of VRes ≈ 4.88mV. This
WU ECeng 33
means that the sensors can have a maximum of half this value, or 2.44mV, between them before the
microcontroller can detect the change.
Knowing that the sensor used have a maximum output voltage of 1V and assuming a linear relationship
between angle of incident and voltage produced, the resolution in angle the microcontroller can
produce is:
θRes = cos-1(1 - ½ VRes / 1) = 4o
Given the maximum error of 180°, this gives the system an error of 2.22%
Setting the reference to zero will reduce the error to the minimum but it will also introduce significant
jitter when a shadow passes over the sensors. Setting the reference to a value greater than zero allows
some hysteresis into the tracker but also increases the error. Setting the reference, for example, to 5
means there is a maximum allowable difference between the two sensors of 5 x 4.88mV = 24.4mV.
θRes = cos-1(1 - ½ VRes / 1) = 8o
gives the system an error of 4.44%.
3.2.4.4. PIC16F84A Software Design
The software for the solar tracker will be developed using the concepts mentioned above in section
3.2.2 solar sensor array. The basic concept is that two CDS cells will be connected to port RB2 and RB3 of
the PIC. The two photocells will be positioned on a small straight piece of wood or plastic. Another piece
will be mounted perpendicular to the straight piece, thereby dividing both the sensors. The concept is
that if both the photocells are equally illuminated by the sun, their resistance level will be the same. As
long as the resistance is the same with an error margin of ±5 points, the PIC will analyze this data and
thus will not generate any signal to actuate the motor. Whereas, if one of the sensors comes under a
shadow, then the PIC will detect this change and thus it will actuate the motor to move the sensor
module to a position where equal light is being illuminated on both of them
The flowchart used for programming the microcontroller is shown below. The program code along with
description and circuit diagram for each tracking axis is attached in the Appendix A.
WU ECeng 34
Figure 3.15:Program flowchart for the horizontal tracking controller
The time fixed (0.6 sec) in the program to actuate the motor when the microcontroller detected a
change is calculated as follows.
The speed of the motor =60rpm = 360oper second
The gear ratio of the worm gear = 1:25
So, when the motor actuated for 1 second (360o) the tracker will be rotated for about 14.4o(360/25).
Therefore in order to rotate the tracker 8o the motor must be actuated for about 0.6 seconds which is
approximately 8.6o .
WU ECeng 35
Figure 3.16:Program flowchart for the vertical tracking controller.
3.2.5. Solar Tracker Control Circuit

Appendix Bshows the schematic for the solar tracker control circuit. A voltage supply of 12V is applied to
the circuit which is then passed through a 5V voltage regulator. The regulated voltage is then supplied to
the PIC. Notice that the LDRs are connected to port RB2 and RB3 of the PIC. The output which drives the
motor is obtained from port RB0 and RB1. Pins RA0 and RA1 are connected to 5V power source. MCLR
reset input port is also connected to 5V supply via 10k resistor. In order to supply power to the PIC, PIN
5 or Vss is set to ground whereas PIN 14 or Vdd is connected to positive 5V.The RA0, RB6 and RB7 are the
ports which drive the LEDs to show the microcontroller is on, the motor is actuating clockwise and
motor is actuating anticlockwise respectively. The limit switches are connected to RB4 and RB5 for the
horizontal tracking and to RB4 for the vertical tracking. The output for RA3 drive the relay coil which
energize and de energize the clutch to mechanically break the motor.
WU ECeng 36
However, if the limit switch is triggered while the tracker is moving there must be a system in addition
to the functionality of the microcontroller to halt the movement of the tracker because any movement
of the tracker after the activation of the limit switch will cause the tracker to pass beyond the east or the
west limit for the horizontal tracker and the vertical limit for the vertical tracker. Therefore in order to
overcome this problem an off-delay timer is incorporated to the circuit as shown in the figure below.
Figure 3.17:Off delay timer with the limit switches for the horizontal tracking
Whenever, either of the limit switch is pressed the data will be fed to the microcontroller as well as it
also energizes timer coil in order to cut the power fed to H-bridge drive circuit to stop the motor. Here
coil of the relay is connected to the normally close contact of the timer. For the vertical tracking the
system is similar to that of the horizontal tracking except only one limit switch is used.
4 MHZ Crystal Oscillator
The external clock determines MCU fundamental operating characteristics; therefore it needs to be
selected wisely. It was realized that faster is not always better in terms of MCUs operating speed and
programming execution. The drawbacks associated with high frequency clock are: their requirement for
large power consumption and also possibly the electromagnetic interference. All the instructions
embedded within the MCU which require proper timing depend on the external clock. Therefore, if the
clock speed is stable and accurate the operation of instructions will stable and accurate as well.
Considering all the factors mentioned above, a 4MHZ oscillator clock was selected. 4MHZ oscillator is
considered an optimum choice when working with applications such as Solar Tracking.
Figure 3.18: XT configuration
WU ECeng 37
Chapter Four
4. Final System
An overall system simulation for the tracker were created on proteus after deciding what and how to
implement each main component of the tracker. The mechanical components consisted of the DC
motors, the worm gears and the solar panel sensor array each described above. Electrically, the tracker
used the PIC16F84A microcontroller, the H-bridge circuit and rechargeable batteries to supply the
power.
4.1. Total System Simulation
To understand how the tracking system would operate before it was constructed; an ideal simulation
was created for one axis of rotation. Only one axis was simulated because under ideal conditions, both
axes would operate identically. MATLAB was chosen as the medium for the simulation because it could
more accurately simulate the processes of the microcontroller.
The first part of the simulation was to collect the specifications of the system. The major portion of the
simulation would rely on an accurate representation of how the sensors act during the day. The final
LDR sensor used had an approximate maximum output voltage of 1V as verified through calculation. The
angle between the two sensors is 180° which means that the angle between the normal to the faces of
the LDRs, denoted as gamma (γ), is also 180°. If the angle of the sun to some reference point denoted as
alpha (α), the voltage across one LDR can be ideally represented in an equation:
V1 = 0.5 cos(α + γ/2) + 0.5V
V1 = 0.5 cos(α + 90o) + 0.5V
Inversely, the voltage across the opposing panel can be ideally represented as:
V1 = 0.5 cos(α -90o) + 0.5V
Because cosine is an even function, the voltage difference between the two panels will be zero at α = 0°.
To incorporate the angular position of the tracker (p) according to the same reference point as the sun,
the equations were modified to be:
V1 = 0.5 cos((α-P) + 90o) + 0.5V
V1 = 0.5 cos((α-P) - 90o) + 0.5V
Now given an angle of the sun and the tracker, the voltage across the panels can be calculated.
Second, the speed of the motor had to be calculated. The calculated speed at 100% duty cycle was 60
rpm or 6.28 rad/s. Given the duty cycle and the gear reduction ratio of 25:1, the speed of the tracker is
approximately 0.84 rad/s.
WU ECeng 38
For the simulation, five vectors where created. The first vector is the time vector to show the transient
nature of the system. The second is the vector that denotes the angular position of the sun over the
duration of the time vector. The rate at which the sun changes its position can be roughly calculated as
360° in a day or 7.272 x 10-8rad/ms.The position vector is initially set at the reference point and is
modified as the simulation progresses. The last two vectors are for the voltages of the panels over time.
Each one is calculated from the present value of the sun and the position of the tracker.
The simulation runs the changing sun position through the solar panel equations to get the voltages on
the panels at the current position. Simulating the microcontroller, the difference between the two
voltages is compared to some reference value that can be changed. If the difference is greater than the
reference or less than the negative reference, the simulation increments the position vector at the rate
of the tracker speed, 0.84 rad/s in the direction correspond to which sensor is getting more radiation. If
the difference is in between the positive and negative values of the reference, the position is
maintained.
After then simulation has run for the specified amount of time, one last vector is created as an error
vector. The error is the difference between the sun and the tracker. This vector was converted into
degrees for readability.
Using the reference of 5 * 4.88mV = 24.4mV, gamma in radians which is 1.57 rad and the speed of 0.84
rad/s, the simulation plots the voltages on the panels, the position of the sun and the tracker next to
each other as well as the error of the tracker all over 500 seconds:
WU ECeng 39
Figure 4.1: Simulation Results over 500 seconds
The simulation shows that according to an ideal situation, one axis of the tracker works as expected. It
follows the sun and maintains a position never more than approximately 1° from the sun. However, the
movement of the tracker is stepped in small increments. Zooming in to an interval of 15 seconds, the
simulation shows this:
WU ECeng 40
Figure 4.2: Simulation Results over 15 Seconds
These zoomed in graphs show the tracker jumps in position approximately every 7 seconds. This is a
result of the tracker forming a “safe zone” that is 0.9945° off of the sun’s position. When the tracker is
within this safe zone, the system is at equilibrium and the tracker does not move. When the sun moves
outside of the safe zone, the tracker sees the error and responds by turning the motor on in that
direction. The motor spins the tracker back into the safe zone and is moving faster than the tracker can
respond so it moves it to approximately 0.965° from the sun, well within the safe zone. The tracker then
is at equilibrium until the sun moves too far again and the cycle repeats.
To examine the inner workings of the tracker, it is best to also look at how the panel voltages are
reacting over time. Modifying the simulation code to output the solar panel voltages as well as the
difference between the two over an interval of 15 seconds, the result is this:
WU ECeng 41
Figure 4.3: Simulation of the sensors Voltages over 15 seconds
LDR 1 is the sensor that is closest to the sun during this simulation. If the direction of the sun’s
movement is reversed, Panel 2 would be the closer. Therefore the simulation is correct in showing that
as the sun moves, the voltage on LDR 1 is increasing and the voltage on LDR 2 is decreasing. The
difference between these two slowly increases until it reaches 24.4mV at which point the tracker begins
to move. The stepped results are still an effect of the motor speed described above.
The simulation shows excellent results that clearly correlate almost exactly to the theoretical
calculations done before construction. The MATLAB code for this simulation is included in Appendix D.
After verifying the simulated performance, the next step was to construct a working prototype to
measure the actual results and compare them with the theoretical.
4.3. Electrical Power System
4.3.1. System Connection
The overall electrical connection for the system is given by the diagram below.As seen, the solar panels
are connected to the charge controller, which is then connected to the battery bank. The battery bank is
then connected to the inverter and also another wire is connected from the battery to power the
controller circuit and the motors (represented by the orange coloured wire in the Figure 4.4). A detailed
explanation of the connection is given in section 4.4 of this thesis.
WU ECeng 42
Figure 4.4: Electrical connection for the system
4.4. Parts
4.4.1. Solar Panels
Two solar panels are connected in series to give it a total output voltage of 72V. The reason they are
connected in series is so that the output voltage from the solar panels should be higher than the voltage
of the battery bank, which is 48V to enable the charging of the battery bank.
Figure 4.5: The solar panels
WU ECeng 43
4.4.2. Charge Controller
The wire from the solar panel is then connected to the charge controller. The charge controller is used
to maintain proper charging voltage on the batteries. It prevents overcharging of the batteries should
the input voltage from the solar panels rises.
Figure 4.6: The charge controller
4.4.3. Battery Bank
Four 12V batteries are connected in series to create a total output voltage of 48V. The wire from the
charge controller is then connected to the battery bank which enables the batteries to be charged. From
the battery bank, a wire is connected to the inverter. Then another wire is place on oneof the 12V
batteries to power the 12V DC loads such as the motors as well as the controller circuit.
Figure 4.7: The battery bank
4.4.4. Inverter
The inverter converts the DC voltage from the battery to AC voltage to power AC loads.
Figure 4.8: The inverter
WU ECeng 44
4.4.5. Isolator
The isolator is put in place to make maintenance work on the system easier.
4.4.6. Cut-off timer
The cut-off timer is a 24-hour based timer which allows electric current to pass through at pre-set times.
The timer is powered using AC voltage from the inverter. The purpose of the timer is Upon inspection, it
was discoveredthat the sensor may responded to the street lightings along roadside of the main gate
and those behind the auditorium building which were lit after sunset. To overcome the problem, a cut-
off timer was placed to cut-off the electricity supply to the controller circuit between just before sunset
(around 6pm) until 8am the next morning. This timer is adjustable.
Figure 28: 4.9:-hour based timer switch
4.5. Cost Analysis
Although a tracker might be highly efficient and provide a good average energy gain compared to an
immobile system or a single axis system, this is not enough yet to claim the system is beneficial over its
alternatives. The added tracking component to the system must provide enough power gain that there
is either an immediate payback or the payback period is less than the lifetime of the system. Thus the
initial cost of the system, the lifetime, and the payback period need to be found to determine the cost
effectiveness of the tracker system.
The total cost of the project was $833.7 which did not include shipping. The prices values were taken
from several component distributers. The list of parts used and the respective prices can be seen in
Table 8. The overall price for a single tracker, cost $833.7.
WU ECeng 45
Part Description Quantity Manufacturer Unit Price total

PIC 16F84A 2 Intel corp $4.17 $8.34
DC geared motor 2 Surplus Trader $59.00 $115
3.76" Worm Gear, 180 Teeth, Bronze 2 Stock Drive $48.5 $97
0.75" Worm, Steel 2 Stock Drive $10.71 $11.42
1" Stainless Steel Thermoplastic Pillow 2 Stock Drive $12.7 $25.4
Block Bearing SSUCP205-16-TP
1-1/4" Four Bolt Flange Bearing 1 Stock Drive $13.28 $13.28
180W Cheap PV Solar Module 2 SUNTECH $126 $252
150ah lead acid battery 24 TAICO $23 $46
48v dc to 230v ac inverter 1 LEONICS $160 $160
MPPT solar charge controller 20A 1 Leonics $75 $75
Parts for PCBs mouser $30.26 $30.26
Total $833.7
Table 4.1: cost of AADAT
The lifetime of the system is determined by the element that has the highest likelihood to fail, which is
the motor. An average DC motor used for light loads, below 30Amps/in2, has a lifetime from 720-2,000
hours [11]. To use a specific number, the midpoint of 1375 hours is used. It was assumed that daylight
exists for 12 hours of a day and the tracker moves every 7 seconds for a half second, based on
calculations in Section 3.3.1. This means that in a day the tracker moves for about 2880 seconds. In a
year it will move for about 292 hours. Thus the motor has a lifetime of about 4 years.
The Ethiopia energy Information Administration calculated that about 1,080 kWh of electric energy is
consumed on average by an Ethiopia household within a year. An assumed energy goal from solar
energy of 25% of the total means a required 270 kWh produced in a year. One calculator (alstore.com)
predicts that 12 panels at 200 Watts would be required, a total of 2,400 Watts. Assuming an ideally
mounted immobile system, about 70% of the radiation would be absorbed. Based on the background
the tracker system would allow for 95% of radiation. If these percentages are used as a multiplier, then
the resulting power would be roughly 1,680 Watts and 2,280 Watts. This means that an extra six panels
are required for the immobile system, while only one extra panel is required for the tracker system.
Immobile: (2,400-1,680)/(200x0.7) = 5.14 panels
Tracker: (2,400-2,280)/(200x0.95) = 0.63 panels
This results in a 50% increase in price on solar panels for the immobile system, while only an 8% increase
in price for the tracker system.
WU ECeng 46
Chapter Five
5. Conclusion And Future Considerations

The completion of this project has led to several conclusions to be made about this solar tracking system
as well as solar tracking systems in general. Several recommendations were also outlined for future
consideration in the continuing development of solar tracking systems
5.1. Conclusion
Concepts related to the solar energy have constantly been under heavy research and development. The
basic objective is to optimize the energy produced from photovoltaic cells, by making the overall
systems more efficient and cost effective. Solar tracking plays a major role in this file of solar energy
optimization
It has been proven through research that solar tracking system with single-axis freedom can increase
energy output by approximately 20%, whereas the tracking system with double-axis freedom can
increase the output by more than 40%. Therefore the objective of this project was to develop and
simulate a solar tracking system with both degree of freedom and which detected the sunlight using
sensors. The control circuit for the solar tracker was based on a PIC16F84A microcontroller. This PIC was
the brain of the entire tracking system, and it was programmed to detect the sunlight through the
sensors and then actuate the motor to position where maximum sunlight could be illuminated onto the
surface of the solar panel.
This project proved to be very beneficial in terms providing valuable knowledge and hands-on
experience related to the development of 2-degree solar tracking system from scratch. The task of
building the tracker required constant debugging and troubleshooting, in both the hardware and
software aspect of the project.
Perhaps the most important conclusion to be made from this project is the total cost for this tracking
system is low, less than $833 in parts for each tracker but not in mass quantities. This means that the
system can be built for a very low cost and most importantly; this system would be within the financial
reach of many developing country communities. Based on the simulations, test results and cost analysis
this project has met its original goals. To improve the efficiency of this tracking system, however, this
project has several future recommendations for future study in solar trackers.
5.2. Future Considerations
Extensive research and development projects are being carried out today in order to optimize solar
tracking systems. There is no doubt that such tracking systems provide an efficient way to increase the
energy output and thus reduce overall cost. Following are some modifications which can be applied to
the tracker in order to make it more efficient:
WU ECeng 47
• There are a variety of solar sensors that were never tested or implemented. This primarily
includes photodiodes and phototransistors, but also includes different photo-resistors and
photovoltaics. Further investigation in the dimensions and the spacing between the sensors
should be conducted. Further investigation should also be done with the divider mount and the
mount with a collimator for the sensors.
• The mechanical system in the future should be designed and analyzed properly in simulation
software such as CAD. This should be done by an experienced mechanical engineer who would
be able to quickly determine errors and poor mechanical design. The design of the solar tracker
prototype could be modified so that it is more portable and efficient.
• Clouds and obstructions can result in misalignment. One way to improve this error would be to
calculate the approximate position of the sun using a location and time-based algorithm and
then using the sensor array to reduce the error to a minimum. A second way to solve this error
is to only check the angle of the sun and adjust the tracker once every few minutes or so.
WU ECeng 48
6. Reference
1. Andreas, K. 2002 Wind load. http://k7nv.com/notebook/topics/windload.html (accessed April 2,

2014)
2. Appleyard, D., 2009. Solar Trackers: Facing the Sun. Renewable Energy World
3. http://www.renewableenergyworld.com/rea/news/article/2009/06/solar-trackers-facing-the-
sun.html (accessed May 12, 2014)
4. Avallone, E. A. and T Baumeister. Eds. 1996. Standard Handbook for Mechanical Engineers, 10th
ed. New York: McGraw-Hill. Quoted in Messenger and Ventre, 2004, 167
5. Beardmore, R. 2010a. Rolling Bearing Friction.
6. http://www.roymech.co.uk/Useful_Tables/Tribology/Bearing%20Friction.html (accessed May
25, 2014) ----- 2010b. Coefficient of Friction.
7. http://www.roymech.co.uk/Useful_Tables/Tribology/co_of_frict.htm#coefficients (accessed
May 25, 2010)
8. Sun path diagram. Image. 2014. http://www.gaisma.com/en/location/hawasaa.html
9. Yearly trends: weather averages and extremes. 2010.
10. http://www.myweather2.com/City-Town/Ethiopia/hawassa/climate-profile.aspx?month=4
(accessed May 24, 2014)
11. Goetzberger, A., Hebling, C., Schock, H.. “Photovoltaic materials, history, status and outlook”.
Materials Science and Engineering: R: Reports. 2002.
12. Mousazadeh, H., Keyhani, A., Javadi, A., Mobli, H., Abrinia, K., Sharifi, A.. A review of principle
and sun-tracking methods for maximizing solar systems output”. Renewable and Sustainable
Energy Reviews. January 2009. Pages 1800, 1800, 1804, 1806, & 1812.
13. Krauter, Stefan. “Solar Electrical Power Generation: Photovoltaic Energy Systems”. Springer.
2006. Pages 21-22.
14. Qazi, M. A. 2009, Software development of solar tracking microcontroller, Undergraduate
Thesis, Curtin University of Technology Sarawak Campus.
15. S. Armstrong and W.G Hurley “Investigating the Effectiveness of Maximum Power Point
Tracking for a Solar System”, IEEE Conference on Power Electronics, 2005 Page(s):204 – 209.
16. Daniel A. Pritchard, “Sun Tracking by Peak Power Positioning for Photovoltaic Concentrator
Arrays” IEEE Transactions on Control System, Volume3, Issue3, Aug1983,Page(s):2-8.
17. Omar Aliman, Ismail Daut, Muzamir Isa and Mohd Rafi Adzman, “Simplification of Sun Tracking
Mode to Gain High Concentration SolarEnergy” American Journal of Applied Sciences, 2007,
Page(s):171-175.
18. A. Zeroual, M. Raoufi , M. Ankrim and A.J. Wilkinson,“Design and construction of a closed loop
Sun Tracker with MicroprocessorManagement” , International Journal on Solar Energy, Vol. 19,
1998, Page(s): 263-274.
19. CemilSungur “Sun –Tracking System with PLC Control for Photo-Voltaic Panels” International
Journal of Green Energy, Vol.4, 2007, Page(s):635–643.
20. Goetzberger, A., Hebling, C., Schock, H.. “Photovoltaic materials, history, status and outlook”.
Materials Science and Engineering: R: Reports. 2002. 4
WU ECeng 49
21. Mousazadeh, H., Keyhani, A., Javadi, A., Mobli, H., Abrinia, K., Sharifi, A..“A review of principle
and sun-tracking methods for maximizing solar systems output”. Renewable and Sustainable
Energy Reviews. January 2009. Pages 1800, 1800, 1804, 1806, & 1812. 5
22. Krauter, Stefan. “Solar Electrical Power Generation: Photovoltaic Energy Systems”. Springer.
2006. Pages 21-22. 6
23. Mehleri, E., Zervas, P., Sarimveis, H., Palyvos, J., Markatos, N.. “Determination of the optimal tilt
angle and orientation for solar photovoltaic arrays”. Renewable Energy. April 2010. Page 2469.
WU ECeng 50
Appendix A: Microcontroller C Code
The following C code was used in the PIC microcontroller for the horizontal tracking.
'****************************************************************
'* Name : Horizontal Tracking.BAS *
'* Author : Selamawit, Adane& Adisu *
'* Notice : Copyright (c) 2014 *
'* : All Rights Reserved *
'* Date : 12-Jun-14 *
'* Version : 1.0 *
'****************************************************************
Include "bs1defs.bas"
Start:
Symbol CDS1 = B0
Symbol CDS2 = B1
Symbol Diff = B6
TRISB.0 = 1 'set pin RB0 as input (sensor1)
TRISB.2 = 0 'set pin RB2 as output
TRISA.0 = 0 'set pin RA0 as output
HIGH PORTA.0 ' set pin RA0 to high to show uC is on
Low PORTB.2 'pin RB2 set to low to h-bridge
Low PORTB.3 'pin RB3 set to low to h-bridge
Low PORTB.6 'pin RB2 set to low CW led indicator
Low PORTB.7 'pin RB3 set to low ACW led indicator
low PORTA.3 ' pin RB3 set to low DENERGIZE CLUTCH
Low PORTB.4 'pin RB4 set to low (limit-switch1)
Low PORTB.5 'pin RB5 set to low (limit-switch2)
Pot PORTB.0,255,CDS1'read sensor1
If PORTB.4 = 1 Then Back1 'if limit-switch1 is pressed, go to 'Back1' subroutine
If PORTB.5 = 1 Then Back2 'if limit-switch2 is pressed, go to 'Back2' subroutine
If CDS1 = CDS2 Then Start 'if equal go back to 'Start'
If CDS1 > CDS2 Then Check 'if greater then move to 'Check' subroutine
If CDS1 < CDS2 Then Check 'if lesser then move to 'Check' subroutine
Back1: 'Back1 subroutine
Sleep 12600 'let the motor rest for about 1 hours before turning back
High PORTB.3 'move the motor clockwise
high PORTB.6 ' LED show the motor is moving CW
pause 5000 'allow the motor to turn clockwise for 5 seconds
LOW PORTB.3
high PORTA.3 'energize Clutch
PAUSE 2000
GoTo start
Back2: 'Back2 subroutine
WU ECeng 51
High PORTB.2 'move the motor anti-clockwise

HIGH PORTB.7 ' LED show the motor is moving ACW
LOW PORTB.2
PAUSE 2000
GoTo start
Check: 'Check subroutine
Low PORTA.4
PAUSE 5000 'check if the difference is permanent or temporary
If CDS1 = CDS2 Then Start 'if equal go back to start of the program
If CDS1 > CDS2 Then Higher 'if greater, then move to Higher subroutine
If CDS1 < CDS2 Then Lower 'if lesser, then move to Lower subroutine
Higher: 'greater routine
Diff = CDS1 - CDS2 'subtract CDS1 with CDS2, then find the difference
If Diff > 5 ThenantiCW'is it within 5 points? if not, then move to AntiCW subroutine
GoTo start 'if within range go back to start of program
Lower:'lesser routine
If Diff > 5 Then ClockW'is it within 5 points?if not, then move to ClockW subroutine
ClockW: 'actuate the motor to move clock wise
High PORTB.3 'supply voltage to pin to move motor clockwise
HIGH PORTB.6 'LED show motor is moving CW
pause 5000 'let it turn for some time
LOW PORTB.3
high PORTA.3 ' energize clutch
pause 2000
GoTo start 'go back to start of the program to check for any changes
AntiCW: 'actuate the motor to move counter clock wise
High PORTB.2 'supply voltage to pin to move motor anti-clockwise
high PORTB.7 'LED show motor moving ACW
low PORTB.2
pause 2000
End
WU ECeng 52
The following C code was used in the PIC microcontroller for the vertical tracking.
'****************************************************************
'* Name : Vertical Tracking.BAS *
'* Author : Selamawit, Adane& Adisu *
'* Notice : Copyright (c) 2014 *
'* : All Rights Reserved *
'* Date : 12-Jun-16 *
'* Version : 1.0 *
'****************************************************************
Include "bs1defs.bas"
Start:
Symbol CDS1 = B0
Symbol CDS2 = B1
Symbol Diff = B6
HIGH PORTA.0 ' set pin RA0 high to show uC is on
Low PORTB.2 'pin RB2 set low to h-bridge
Low PORTB.3 'pin RB3 set low to h-bridge
Low PORTB.6 'pin RB6 set low CW led indicator
Low PORTB.7 'pin RB7 set low ACW led indicator
low PORTA.3 ' pin RB3 set low de energize the clutch
Low PORTB.4 'pin RB4 set low (limit-switch1)
If PORTB.4 = 1 Then Back 'if limit-switch1 is pressed, go to 'Back1' subroutine
If CDS1 = CDS2 Then Start 'if equal go back to 'Start'
If CDS1 > CDS2 Then Check 'if greater then move to 'Check' subroutine
If CDS1 < CDS2 Then Check 'if lesser then move to 'Check' subroutine
Back: 'Back subroutine
High PORTB.3 'move the motor clockwise
high PORTB.6 ' LED show the motor is moving CW
LOW PORTB.3
PAUSE 2000
GoTo start
Check: 'Check subroutine
Low PORTA.4
PAUSE 5000 'check if the difference is permanent or temporary
If CDS1 = CDS2 Then Start 'if equal go back to start of the program
WU ECeng 53
If CDS1 > CDS2 Then Higher 'if greater, then move to Higher subroutine
If CDS1 < CDS2 Then Lower 'if lesser, then move to Lower subroutine
Higher: 'greater routine
If Diff > 5 ThenantiCW'is it within points? if not, then move to AntiCW subroutine
Lower: 'lesser routine
If Diff > 5 ThenClockW'is it within 5 points?if not, then move to ClockW subroutine
ClockW: 'actuate the motor to move clock wise
High PORTB.3 'supply voltage to pin to move motor clockwise
HIGH PORTB.6 'LED show motor is moving CW
LOW PORTB.3
pause 2000
AntiCW: 'actuate the motor to move counter clock wise
High PORTB.2 'supply voltage to pin to move motor anti-clockwise
high PORTB.7 'LED show motor moving ACW
low PORTB.2
pause 2000
End
WU ECeng 54
Appendix B Circuit Diagram
1. Circuit diagram for the horizontal tracking
U2 LDR1
7805
LDR2
1 3 CDS1
VI VO
GND
B3 C2 CDS1
12V C3 0.1uF CDS2
2
0.1uF
C6
2.2uF 2 1 C7 CDS2
2 1
2.2uF
TORCH_LDR
TORCH_LDR
C4
22pF
X1 U1 D5
C5 CRYSTAL
16 17
R12
15
OSC1/CLKIN RA0
18
R11
OSC2/CLKOUT RA1
1
R9 1k
(1)
22pF
R10 4
RA2
2
10k LED-BLUE
MCLR RA3 RC 10k
LS1 10k 3
RL2(NC) RA4/T0CKI
6
RB0/INT CDS1
RL2(C1) 7
RB1 CDS2
LS2 8
RB2 A
RL2 RB3
9
B D6
OMIH-SH-124D 10 R13
RB4 LS1
11
RB5 LS2
12
RB6
13
1k D7
RB7 R14 LED-RED
Q8
2N3904 PIC16F84A 1k
LED-RED
R1
1k C1 Q2 R2
16uF 1k
Q1
IRF9640
RL1(NO)
D1 D2
1N4007 1N4007
RL1(C1)
IRF9640
RL1
RC OMIH-SH-124D
R5
1k
D9
Q7 LED-BLUE
2N3904 Q4
Q3
IRF630 D3 D4 Q6
1N4007 1N4007
R4
B R3 Q5 A
2N3904 IRF630 10k
10k
2N3904
WU ECeng 55
2. Circuit diagram for the vertical tracking
U2 LDR1
7805
LDR2
1 3 CDS1
VI VO
GND
B3 C2 CDS1
12V C3 0.1uF CDS2
2
0.1uF
C6
2 1 C7
2.2uF CDS2
2 1
2.2uF
TORCH_LDR
TORCH_LDR
C4
22pF
X1 U1 D5
C5 CRYSTAL
16 17
R12
15
OSC1/CLKIN RA0
18
R11
OSC2/CLKOUT RA1
1
R9 1k
(1) R10 4
RA2
2
10k LED-BLUE
22pF 10k
MCLR RA3 RC
LS1 3
RL2(NC) 10k RA4/T0CKI
6
RB0/INT CDS1
RL2(C1) 7
RB1 CDS2
8
RB2 A
9
RB3
10
B D6
RL2 RB4
11
LS1 R13
OMIH-SH-105L RB5
12
RB6
13
1k D7
RB7 R14 LED-RED
Q8
2N3904 PIC16F84A 1k
LED-RED
R1
1k C1 Q2 R2
16uF 1k
Q1
IRF9640
RL1(NO)
D1 D2
1N4007 1N4007
RL1(C1)
IRF9640
RC RL1
OMIH-SH-105L
R5
1k
D9
Q7 LED-BLUE
2N3904 Q4
Q3
IRF630 D3 D4 Q6
1N4007 1N4007
R4
B R3 Q5 A
2N3904 IRF630 10k
10k
2N3904
WU ECeng 56
Appendix C: MATLAB System Simulation Code
The following MATLAB code was used to simulate the entire system
Functionmqp (gamma,speed,ref) %take in values for angle between sensors,
%motor speed and comparator reference
%gamma is in radians, speed is in rad/s and
%ref is in volts
t = 0:0.001:500; %create a time vector, 500 seconds (500000 ms) long
sun = t*(7.272*10^-5);
gamma= 1.571;
speed=0.84;
ref=5;%create a sun position vector that changes over time
v1 = zeros(1,500001); %setup a vector for one sensor
v2 = zeros(1,500001); %setup a vector for the other sensor
pos = zeros(1,500001); %setup a vector for tracker position
n = 2; %initialize a indicator
while n <= 500001
v1(n) = 0.5*cos((gamma/2)-(sun(n)-pos(n-1)))+0.5;
%current value of v1 is based on angle between sensor and sun which
%is based on gamma and tracker position
v2(n) = 0.5*cos((gamma/2)+(sun(n)-pos(n-1)))+0.5;
%current value of v2 is based on angle between sensor and sun which
%is based on gamma and tracker position
if ((v1(n)-v2(n)) > ref) %if v1 is over the reference and v2 is not
pos(n) = pos(n-1) + speed/1000; %rotate towards v1
n = n + 1;%increment indicator
elseif ((v1(n)-v2(n)) < -ref) %if v2 is over the reference and v1 is not
pos(n) = pos(n-1) - speed/1000; %rotate towards v2
n = n + 1; %increment indicator
else%if the difference is between ref and -ref
pos(n) = pos(n-1); %do not move
n = n+1; %increment indictor
end%exit if
WU ECeng 57
end%exit while
error = (sun-pos)*(180/(pi)); %calculate the error vector between
%the sun and tracker
%Plot the sun and tracker movement over 15 seconds:
subplot(2,1,1),plot(t(300001:315000),sun(300001:315000),'black'), hold on,
plot(t(300001:315000),pos(300001:315000),'r'),title('Tracker Movement'),
xlabel('Time (s)'),ylabel('Angular Position (rad)'), hold off;
legend('Sun', 'Tracker');
%Plot the error vector over the same timeframe:
subplot(2,1,2),plot(t(300001:315000),error(300001:315000)),
title('Tracker Error'),xlabel('Time (s)'),ylabel('Error (degrees)');
%Additional Code used to plot voltages on the panels:
%To use, comment out the above plots and uncomment the lines below
%subplot(3,1,1),plot(t(300001:315000), v1(300001:315000)),
%title('Panel 1 Voltages'),xlabel('Time (s)'),ylabel('Voltage (V)');
%subplot(3,1,2),plot(t(300001:315000), v2(300001:315000)),
%title('Panel 2 Voltages'),xlabel('Time (s)'),ylabel('Voltage (V)');
%subplot(3,1,3),plot(t(300001:315000),v1(300001:315000)-v2(300001:315000)),
%title('Difference in Panel Voltages'),xlabel('Time (s)'),
%ylabel('Voltage (V)');
WU ECeng 58
Appendix D:
Physical Specifications of the PV module
WU ECeng 59

'Chapter One 1. Background: Dual Axis Solar Tracker

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

'Chapter One 1. Background: Dual Axis Solar Tracker

Uploaded by

Copyright:

Available Formats

[DUAL AXIS SOLAR TRACKER] June 2014

Solar energy has many advantages:

1.2. Need For Solar Tracker

1.3. Problem Statement

1.4. Thesis Objective

1.4.2. Specific Objective

1.5. Scope of The Thesis

The scope of this thesis:

Designing a mechanical structure for the mounting system.

1.6. Relevance of Solar Trackers

1.7. Organization of The Thesis

Figure 2.2: Angle of Incidence to Solar Cell [ Adrian 2010]

2.2. Solar Tracker

2.3. Overview Of Existing Tracking Technology

2.3.1. Tracking Configuration - Immobile Versus Mobile

Figure 2.4: (left) single axis (right) dual axis tracker.

2.3.2. Tracking Methods – Passive, Active and Chronological

2.4. Literature Survey

Figure 3.1:General block diagram of the system

3.1. Mounting System

Figure 3.2: Sun path diagram for Hawassa, Ethiopia

3.1.1. Design Criteria

3.1.1.1. Major system components to design solar power

• PV module – converts sunlight into DC electricity.

Solar PV system sizing

Step 1. Determine power consumption demands

Step 2: Size the PV modules

2.1 Calculate the total Watt-peak rating needed for PV modules

2.2 Calculate the number of PV panels for the system

Step 3: Inverter sizing

Step 4: Battery sizing

4.1 Calculate total Watt-hours per day used by appliances.

Step 5: Solar charge controller sizing

1. Determine power consumption demands

2. Size the PV panel

2.1 Total Wp of PV panel capacity = 1,232.4 / 3.43

Battery capacity = [(18 W x 4 hours) + (60 W x 2 hours) + (75 W x 10 hours)] x 3

Total Ampere-hours required 461.77 Ah/day

5. Solar charge controller sizing

Weight of 1 solar panel = 15.5kg

Maximum wind speed in Hawassa = up to 6mph

3.1.2. Material And Design

Figure 3.4: Finalized design of the structure

Torque = F × r, where F = mgcf

Figure 3.5: Worm gear system locking.

3.1.5. Motor braking

3.1.6. Installation Site

3.2. Electrical Control System

3.2.1. Motor Control

Figure 3.6: Basic H-bridge Circuit

25°C / (0.160 * RON) = Rθmax

Figure 3.7: Revised H-bridge Schematic

0.5V = 0.4A / (500Hz * C)

Figure 3.8: Final H-bridge Schematic

3.2.2. Solar Sensor Array

Figure 3.9: sensor placement

3.2.2.1. Light dependent resistor

Working Principle of LDR

The relationship between the voltages at Voutis given by:

3.2.2.2. Sensor Holder Design

Figure 3.12: Setup to Test for Ideal barrier length

Figure 3.13: Photo-resistor Voltage Divider Circuit

3.2.3. Limit switch