Electrical Engineering and Computers Science Faculty [309350]

– [anonimizat]: [anonimizat]: Sef lucr. dr. ing. MACHEDON-PISU Mihai

2018

TRANSILVANIA UNIVERSITY OF BRAȘOV

Electrical Engineering and Computers Science Faculty

Electrical Engineering and Applied Physics Department

Bachelor degree: [anonimizat] –

Graduate: [anonimizat]: Sef lucr. dr. ing. MACHEDON-PISU Mihai

2018

Abstract – The presented paper is a [anonimizat]-axis method for high conversion efficiency of direct solar radiation.

As a [anonimizat], [anonimizat]. [anonimizat] a fast migration to renewable energy sources is required.

[anonimizat]’s major renewable energy resource. [anonimizat], [anonimizat]. As a fuel-[anonimizat] a major contribution to national energy security and carbon dioxide abatement. [1]

Today solar energy is seen as the most reliable renewable energy source. According to calculations the sun deposits 120,000 TW of radiation on the surface of Earth. The sun covers about 0.16% of the land on Earth. With 10% efficient solar conversion systems we can generate almost 20 [anonimizat]’s consumption. [anonimizat] [2].

The energy extracted from photovoltaic panels (PV) depends on the solar radiation. [anonimizat]. During the night and in the case of earth’s [anonimizat]. The sun tracker moves the solar collector (PV) to follow the sun trajectories and keep the PV panel oriented at the optimal tilt angle. Energy efficiency of the solar photovoltaic panel can be substantially increased by using solar tracking systems.

[anonimizat] (LDR) sensors and DC micro servomotors with gear arrangements. For the tracking (azimuth angle and altitude angle), microcontroller based control logic has been used.

INTRODUCTION

Solar radiation

Global distribution

Earth’s [anonimizat] a result the solar radiation that reaches earth varies over a year. Thus, [anonimizat], depending on this factor. It varies between a minimum of approximately 1471 x 10⁸ km at the perihelion (the point we are closest to the Sun), which is from January 2 to 5, and a maximum of 1521 x 10⁸ km at the aphelion (point where we are the furthest from the sun), from July 3 to 5.

Figure 1.1. The maximum and the minimum point distance from the Sun.

[anonimizat] 1325 W/m² and 1412 W/m². The global mean value for radiation density is :

E0 = 1367 W/m²

Once it enters the atmospheric layers, the direct radiation decreases more in intensity, as a result of phenomena such as reflection, dispersion and absorption. At the ground level, in the afternoon, on a clear sky day, the intensity of direct radiation can reach to 1000W/m².

Solar radiation energy differs greatly from zone to zone. Its worldwide distribution in 2017 is presented in the following image:

Figure 1.2. Distribution of direct solar radiation on the globe. [3]

Sun elevation and solar radiation spectrum

The insolation on Earth’s surface directly depends on the Sun’s apparent elevation relative to a terrestrial observer. Solar radiation is affected by the planet’s atmosphere through reflection, diffusion and absorption, an angle near the zenith (the angle formed by the horizontal plane at the observation point and vertical on this plane, passing through the center of the sun) as shown in figure 1.2. means that solar radiation will have a shorter trajectory through the atmosphere and, implicitly, a large quantity of energy will reach the surface of the planet. Also an angle near the horizontal implies that a lower level of radiation will penetrate because of the thickness in the atmosphere, which causes a diminished radiation flow.

Figure 1.3. The zenith angle and angles of penetration.

The atmospheric mass index (AM – air mass) indicates the total thickness ratio of the atmospheric layer that is penetrated by Sun light at a certain elevation and the thickness of the layer when the Sun is at zenith, at which point the coefficient is considered to be 1. Also at the outside extremity of the atmosphere, the same index is considered to be 0.

The expression of mass index is:

where the sun elevation angle γ is the point of observation.

Earth's atmosphere affects not only the intensity and, implicitly, the solar radiation energy as a whole, but more than that, affects its spectrum through its total elements. For example, the ozone layer filters most of the high energy frequencies, such as UV, X, gamma, and so on.

As long as the path through the atmospheric layers is longer, as in the case of sunset or at sunrise, the more the spectrum of radiation loses a large number of stripes of small wavelength. In its components only remaining the low energy, infrared frequencies.

The figure below illustrates the differences between the spectral distributions of direct solar radiation at the sea level and at the outer boundary of the Earth’s atmosphere.

Figure 1.4. The spectrum of direct solar radiation at sea level and at the outer limit of the earth's atmosphere. [4]

Measuring solar radiance

Solar radiance can be measured directly using pyranometers or photovoltaic sensors, or indirectly, by analyzing satellite images. The pyranometers are high precision sensors that measure solar radiation on a flat surface. Basically they are composed of two glass hollow domes, a black hyper-absorbent metal plate, thermo-sensitive elements positioned below the plate, and a white metal case.

The solar rays pass through the glass domes and fall perpendicular to the absorbing surface, heating it. Because warming directly depends on the degree of irradiation, the difference in temperature between the absorbent plate and the outside environment, more precisely the white housing, makes it possible to determine the solar radiation intensity.

Figure 1.5. CMP3 Pyranometer. [5]

As an alternative to pyranometers, much cheaper, but also less precise, are the photovoltaic sensors or irradiance sensor. These sensors have a lower accuracy compared to that of the pyranometers, because of their low spectral sensitivity.

Figure 1.6. Photovoltaic sensor for measuring solar radiation intensity. [6]

This type of sensor contains a photovoltaic cell, which generates an electric current, whose intensity is directly proportional to the incident radiation intensity. Because of limited sensitivity, the photovoltaic cell, is not affected by infrared waves, decreasing performance.

The solar cell and photovoltaic effect

Introduction

The photovoltaic effect was firstly observed in 1839 by French physicist Alexandre-Edmond Becquerel. The phenomenon consists on the appearance of an electrical voltage or an electric current in certain materials, in the moment they were exposed to sunlight.

The quantum mechanics progress has led to a better explanation of the photovoltaic effect. Starting from the notion of photons, the phenomenon has been described as the effect of detaching the electrons from the valence band, and their entry into the conduction band, followed by photon energy absorption by certain materials.

Figure 2.1. The band model for insulators, semiconductors and conductors. [7]

For semiconductor materials, unlike metals or electrical insulators, there is an energy difference between the valence of at least eight conductors, which is small enough that the absorption of a photon can displace an electron and places it in the strip line.

In photovoltaic cell construction for terrestrial applications, the cells require compatibility with the spectrum of radiation, which is different between the surface of the planet and the outer boundary of the atmosphere. The most widespread material is silicon, in the mono or poly-crystalline form. Pure silicon however, has few free electrons, insufficient to generate a useful electrical current.

In order to make the silicon efficient, it will be filled with materials from Groups III or V from the periodic table, that have one valence electron more or less than silicon, to increase the electrical conductivity. Some examples of substances are boron or phosphorus.

In the case of phosphorus, as only four of the five available electrons create bonds with the adjacent silicon atoms, the weak link of the fifth one can be broken easily, carrying that electron in the conduction layer.

This type of doping is called negative doping, because it adds an excess of negative charge electrons.

Figure 2.2. Phosphorous doped silicon. [8]

Similarly, for group III materials, such as boron, which have less valence electrons, there is a gap in the atomic bonds. This allows migration of the free electron from the silicon atom adjacent to the boron atom, filling that empty space.

This effect creates the appearance of another empty space, and leads to a “current of voids”. This type of doping is called positive doping, as it creates an excess of positive charge.

Figure 2.3. Boron doped silicon. [8]

A photovoltaic cell is essentially a p-n junction, formed by overlapping two layers of semiconductor material, one that is positively doped and the other negative doped. This juxtaposition generates a contact area that negative type loads cross over to fill the hollows in the positive doped layer, creating electron-hole pairs. The absorption of photons by such pairs leads to the breaking of the pairs, which forms free-charge carriers, thereby forming an electric current.

Figure 2.4. Photovoltaic cell – operating principle. [9]

Measurable parameters and I-U curve

The photoelectric conversion efficiency of cells is limited by multiple factors, such as cell reflection surface, thermodynamic efficiency, spectral distribution, efficacy in separating charge carriage pairs, or conduction of the material from which it is built. Also, any flaw in the structure of the material is reflected upon efficiency.

Because these factors are difficult to measure directly, substitutes are used to characterize cells such as the Maximum Power Point (MPP), the current-voltage characteristic (I-U curve), open circuit terminal voltage (VOC), short circuit current or filling factor.

Essentially, the I-U curve is characterized by the following three sizes:

a). The maximum power point – designated as the nominal power of a photovoltaic cell, MPP represents the point on the current-voltage curve in which it operates at maximum power. For MPP, both nominal current (IMPP) and nominal voltage (UMPP) are specified.

b). Short circuit current (ISC) – ISC is generally 5 to 15 percent under the current at the maximum power point. ISC varies, primarily, depending on cell technology, for monocrystalline cells around 3A. An important feature of short-circuit current for CPV systems is that the irradiation is linear-dependent. Thus, if the irradiance triples, so does the value of the current.

c). Open circuit voltage (UOC) – represents the voltage between the cells metallic terminals when it does not have a load connected. Just like ISC, UOC depends on the material from which the cell is made, in the case of crystalline silicon, reaching a value of about 0.5 – 0.6 V, and in the case of amorphous silicon ranging between 0.6 – 0.9 V.

Figure 2.5. Generic I-U curve of a photovoltaic cell. [10]

The fill factor (FF) is used to distinguish the quality of the solar cell. In the hypothesis of an ideal rectangular I-U curve, in which the maximum power point is obtained by multiplying the values of the short-circuit current and the open circuit voltage. The filling factor represents the deviation of the real characteristic from the ideal one.

From a geometric point of view, the fill factor represents the ratio between the surfaces formed by the two characteristics in the orthogonal axis system. The value of FF is always a number between 0 and 1 and has the following formula:

Figure 2.6. Graphical representation of the fill factor. [11]

Photovoltaic cells efficiency

The photovoltaic cell conversion efficiency represents the electricity flow in relation to the energy absorbed by the solar radiation incidence. In other words, a cells efficiency is, in simple terms, the proportion of the total solar energy absorbed by a cell, which was converted into electricity. This is calculated by dividing the peak electric power of the cell (expressed in W) to the incident radiation power density at the moment the maximum power is reached (expressed in W/m2) and at the cell surface (expressed in m2):

In photovoltaic cell production, due to the dependence of their electrical characteristics on temperature and radiation spectrum, the conversion efficiency is determined according to standard test conditions (STC). These include a temperature of 25°C with a tolerance of ± 2°C, irradiated incidence of 1000W/m2 and a spectral distribution of the radiation corresponding to the air mass index of AM 1.5.

The STC corresponds to a clear day in which the Sun has an elevation angle of 41.81° and the cell is oriented towards it at an angle of 37°. Under these circumstances, a cell with a surface area of 100 cm2 and a conversion efficiency rated at 20% will have an electrical power of 2W.

Figure 2.7. The conversion efficiency of the world's best photovoltaic cells between 1976 and 2018 [12]

In practice, the conditions set forth in the standard test conditions are rare, but the conversion efficiency in the technical documentation of the cells is always indicated in relation to the standard test conditions and represents the nominal efficiency:

where A represents the cell area.

In addition to the dependence of irradiation, the crystalline cell efficiency is inversely proportional as the temperature increase. Thus, photovoltaic elements have maximum efficiency at low temperatures. The temperature coefficient is dependent on the material used, and in the case of crystalline silica it is around -0.0045 (-0.45%) per °C. Considering the power density of the incident radiation and its constant spectral distribution, the variation in cell efficiency is as follows:

𝛥 𝜂 ≅ −0.0045 × (𝑇 (𝑆𝑇𝐶) − 𝑇) × 𝜂 (𝑆𝑇𝐶)

Figure 2.8. Temperature dependence of I-U characteristic of photovoltaic cells [13]

For example, if the rated power under standard irradiation conditions and the spectral distribution of a cell is 200W and its temperature is raised to 45 ° C, the effective power reaches 182W. Conversely, if cell temperature is forced to drop to 5°C, effective power would reach 218W.

USED HARDWARE AND SOFTWARE

General aspects

Solar tracking is the most appropriate technology to enhance the electricity production of a PV system. To achieve a high degree of tracking accuracy, several approaches have been widely investigated. Generally, they can be classified as either open-loop tracking types based on solar movement mathematical models or closed-loop tracking types using sensor-based feedback controllers [14–16].

In the open-loop tracking approach, a tracking formula or control algorithm is used. Referring to the literature [17, 18], the azimuth and the elevation angles of the Sun were determined by solar movement models or algorithms at the given date, time and geographical information. The control algorithms were executed in a microprocessor controller [19,20]. In the closed-loop tracking approach, various active sensor devices, such as light dependent resistors (LDRs) [20, 21] were utilized to sense the Sun’s position and a feedback error signal was then generated to the control system to continuously receive the maximum solar radiation on the PV panel.

Solar tracking approaches can be implemented by using single-axis schemes [20], and dual-axis structures for higher accuracy systems. In general, the single-axis tracker with one degree of freedom follows the Sun’s movement from the east to west during a day while a dual-axis tracker also follows the elevation angle of the Sun. In recent years, there has been a growing volume of research concerned with dual-axis solar tracking systems.

Figure 3.1. Closed loop dual-axis sun tracking system.

However, in the existing research, most of them used two stepper motors [22, 23] or two DC motors [21, 24] to perform dual-axis solar tracking. With two tracking motors designs, two motors were mounted on perpendicular axes, and even aligned them in certain directions. In some cases, both motors could not move at the same time [16]. Furthermore, such systems always involve complex tracking strategies using microprocessor chips as a control platform.

Hardware used for this project

For the completion of this project, the following hardware has been used:

1 x Solar panel

2 x Servo motors

4 x Light Detecting Resistors (LDR)

4 x 10K Ohm Resistors

Jumper wires, Terminal blocks and Connector cables

Solar panel

For this particular low dimension project, the 10W mono CL-10 WM solar panel was chosen, because of its reduced dimensions, low weight of 1.10 Kg, due to the aluminum alloy frame, moderate amount of environmental protection and overall cheap price.

Figure 3.2. Solar module 10W mono CL-10 WM

Solar panel specifications:

With a rated power of 10 watts, the module is very suitable for any standard photovoltaic application. Suitable for grid-connected systems, caravans, camping, garden sheds off-grid systems. It is designed for applications in places without electricity and for small systems. The modules are precisely manufactured, equipped with anodized universal frame and hail-resistant safety glass. The cables can be connected on the back at the waterproof junction box (IP65).

Motors

Regarding the motors, this project uses S3003 Futaba Servo motors. These motors are perfect for this project because of their reduced weight of 37.2g, their capacity to move a desired payload, in this case the photovoltaic panel, the vertical axis arms, horizontal movement disc implementation and the motor’s self-weight.

One key feature is that the chosen servo has the ability to move at specific angles (0-180 degrees) in both directions, and a command wire, which makes it compatible with the used microcontroller (Arduino UNO), that will be presented further in the project.

Figure 3.3. S3003 Futaba Servo motor.

Detailed motor specifications are shown in the following datasheet:

Microcontroller

For the brain of this project an Arduino Uno has been used, because of its low cost, reduced size and perfect connectivity with the other used components in terms of the I/O pins.

The Arduino Uno is one of the most popular compact development boards on the market. The board is based around the removable Microchip ATmega328P microprocessor, which features 8-bit resolution and its own USB boot loader for reprogramming. With 32 KB of Flash memory, 2 KB of SRAM, and 1 KB of EEPROM, this little processor can carry out most any task you can ask of it.

Figure 3.4. Arduino Uno development board. [25]

Sensors

For the sensors, this project contains four light detecting resistors also known as LDR’s. These are very common and they can be found in almost every electronic shop around every city.

They work by modifying their level of resistance based on how much light is hitting them. More light equals less resistance.

In this project they are implemented on the shadowing system, that helps properly track the movement of the sun. This will be presented further in the project.

Figure 3.5. Typical construction of a light detecting resistor. [26]

Software used for this project

Regarding the software used for making the project, the following programs have been used:

Arduino IDE

The Arduino IDE software is the main program for programming the Arduino board, it stands for Integrated Development Environment (IDE). It is compatible with a wide variety of operating systems, including Windows, Mac OSX, and Linux. It is an open-source, and can be downloaded from the main Arduino website [25].

The syntax, or the words and structure of the code are almost similar to C/C++ and Java.

The following image shows the UI of the Arduino IDE program, where the simple sketch “Sweep” is imported from the examples, Servo menu.

Adobe Photoshop

The Adobe Photoshop program is a design software that helps transform basic photos via adding various elements to it, creating concepts on a 2D scale and even 3D.

In this project Adobe Photoshop has been used to create explications about the theory used, add various highlighting elements to the photos taken on the final design, and create block diagrams, as well as coming with the final concept design.

It is a gateway of modifying real created elements, and adding some elements to pinpoint and highlight specific components in this design.

Adobe Photoshop was chosen because of my background in graphic design, which made the project progress speed up.

Microsoft Visio

Microsoft Visio is a program specialized in vector graphics, and mainly it is used for creating diagrams. It was used in this project to create the main block diagram that can be found in chapter III. PROJECT DESIGN, 1. Circuit diagram.

PROJECT DESIGN

Circuit diagram

Based on the chosen hardware, general knowledge, and intense studying, the following block diagram was designed.

Figure 4.1. Sun tracker block diagram.

Light detecting resistors

The top part of the circuit diagram (the four light detecting resistors) is responsible with data acquisition. In the presence of light the LDRs get a value between 0 and 1023 that depends on the brightness of the light source, value that is sent to the Arduino via the analog pins. The sensors are connected to the 5V pin on the Arduino, and to the ground pin (GND).

Figure 4.2. Connectivity of the LDRs to the Arduino.

Powering the Arduino and servo motors

The next part is about the powering and connectivity of the servo motors and the Arduino. The motors can be powered from 4.8V to 6V, and the Arduino can be powered with 5V. The project includes two buck converters connected to the battery, which lowers the voltage from 12V to 5V. Both servo motors are powered from one buck circuit, and the Arduino is powered from the second buck converter.

The servo motors commands are placed in digital pin 10, for the vertical servo, and in digital pin 9, for the horizontal servo.

The following block schematic shows the presented connectivity.

Figure 4.3. Powering and connectivity of the Arduino and servo motors.

Charging technique

In this last part the charging of the battery from the solar panel via the Maximum Power Point Tracker (MPPT) is shown.

Figure 4.4. Charging the battery from the solar panel using the MPPT.

In the case of sunny days, or cloudy days, the MPPT acts similar to a Buck-Boost circuit and lower the maximum voltage from the solar panel, which is around 23V, to 12V or raises the voltage in case it is under 12V, for an optimized charging technique.

Final circuit design

Figure 4.5. Final circuit diagram.

Code design

The code was written in the Arduino IDE software, and was designed by studying various tutorials over the internet, guides and scientific documents. [28]

Defining the servo motors

In order to use servo motors with the Arduino, the “servo.h” library must be included in the code.

// defining the horizontal servo

Servo horizontal; // horizontal servo

int servoh = 105; //middle point of the horizontal servo

int servohLimitHigh = 180; //highest value of the horizontal servo in terms of degrees

int servohLimitLow = 0; //lowest value of the horizontal servo in terms of degrees

// defining the vertical servo

Servo vertical; // vertical servo

int servov = 83; //middle point of the vertical servo

int servovLimitHigh =145; //highest value of the vertical servo in terms of degrees

int servovLimitLow =30; //lowest value of the vertical servo in terms of degrees

For the horizontal servo, values between 0 degrees and 180 degrees were implemented, so that it can properly track the sun.

For the vertical servo, values between 30 degrees and 145 degrees were chosen, so that the vertical movement does not flip the solar panel over, and while analyzing how the sun moves in the sky, even at sunrise or sunset the light source will be above a certain point, so the panel does not need to be orientated perpendicular to the ground.

Servo and LDR pin connections

The light detecting resistors are connected to the analog pins from 0 to 3 on the Arduino, and the servo motors are connected to digital pins 9 and 10 on the Arduino.

// LDR pin connections

// name = analogpin;

int ldrlt = 1; //LDR top left

int ldrrt = 0; //LDR top right

int ldrld = 3; //LDR down left

int ldrrd = 2; //LDR down right

void setup()

{

Serial.begin(9600);

// servo connections

// name.attacht(pin);

horizontal.attach(9);

vertical.attach(10);

horizontal.write(180);

vertical.write(45);

delay(3000);

}

Sensor readings and tolerance

In this section the program reads the sensors data, and creates a tolerance based on these readings.

void loop()

{

int lt = analogRead(ldrld); // top left

int rt = analogRead(ldrrd); // top right

int ld = analogRead(ldrlt); // down left

int rd = analogRead(ldrrt); // down right

int dtime = analogRead(4)/20; // read potentiometers

int tol = analogRead(ldrld+ldrrd+ldrlt+ldrrt)/4;

Calculation algorithm

This algorithm is the main part of the program that is linked with analyzing data and transmitting it to the servo motors in order to properly track the position of a light source.

int avd = (lt + rt) / 2; // average value down

int avt = (ld + rd) / 2; // average value top

int avr = (lt + ld) / 2; // average value right

int avl = (rt + rd) / 2; // average value left

int dvert = avt – avd; // check the difference of up and down

int dhoriz = avl – avr;// check the difference of left and right

//the calculation algorithm

if (-1*tol > dvert || dvert > tol) // check if the difference is in the tolerance else change vertical angle

{

if (avt > avd)

{

servov = ++servov;

if (servov > servovLimitHigh)

{

servov = servovLimitHigh;

}

}

else if (avt < avd)

{

servov= –servov;

if (servov < servovLimitLow)

{

servov = servovLimitLow;

}

}

vertical.write(servov);

}

if (-1*tol > dhoriz || dhoriz > tol) // check if the difference is in the tolerance else change horizontal angle

{

if (avl > avr)

{

servoh = –servoh;

if (servoh < servohLimitLow)

{

servoh = servohLimitLow;

}

}

else if (avl < avr)

{

servoh = ++servoh;

if (servoh > servohLimitHigh)

{

servoh = servohLimitHigh;

}

}

else if (avl = avr)

{

// nothing

}

horizontal.write(servoh);

}

Building process

The project design will be structured based on the chronological order of the building process. The logic of creating the project was going from bottom to top. After choosing and ordering the right devices for the sun tracker, some measurements were taken regarding the final weight and dimensions of the project.

It was made in a personal workshop using a wide variety of equipment and devices such as:

drilling machinery

boring machinery

vise

steel cutting devices and machinery

wood cutting devices and machinery

measuring devices (calipers, roulette, tape measure)

welding machinery

wide variety of grinders

clippers and screwdrivers

As a remark, and a key point of this project is that all the equipment used for making the casing of the project, the mechanical moving parts such as the horizontal disk, the little moving wheels under it, as well as the vertical shaft, the spacers, the casing reinforcements and the resting wooden plates are scavenged from household items and devices. This scavenging process made the final costs of the project drop with around 70%.

The stability plate and the casing

The first thought was that the project will have some moving parts that may create some difficulties regarding the stability and overall equilibrium of the device. The sun tracker is made to be transported in different types of outdoor places, and indoor places (near a window, or on a balcony) and be placed on different types of materials. The point is it needs to sit in a stable manner, so as a resting plate I used a wide circular piece of wooden material.

Figure 5.1. Created wooden stability plate

Wood was used because it’s a softer material, it doesn’t slide on different materials that much, and would not scratch or deal damage to the surface it is left on.

On top of the wooden resting plate the casing that protects the circuit part was designed. The casing has a box shape, with the exteriors made from some plastic material, which provides an easy way of mounting different pieces of circuit elements, due to the low material thickness, and protects the circuits inside from short circuit, in the case of rain and as well as insects or other dangerous hazards.

Figure 5.2. Protective exterior casing.

The plastic walls of the exterior are mounted on four reinforcement steel pillars using screws. The pillars not only have the role of keeping the exterior plastic walls together, but they also act as a resting point for the upper part of the solar tracker. This is important because if they would not support the weight of the upper project, the entire tracker would collapse.

As for the top of the box, a piece of stout wood was used, again to provide protection, but mainly for the fact that is strong and can sustain the weight of the top part of the project.

Figure 5.3. Jointing system for the exterior walls and one of the four reinforcement steel pillars.

Problems encountered:

When cutting the plastic pieces, the equipment friction on the material caused temperatures to rise and melt some of the pieces that were eventually replaced and handled with grater care.

The steel pillars were hollow, making the welding process more difficult and requiring more skill and precision.

The horizontal disks

The horizontal disks are the main guiding system of the horizontal axis. They are mainly made of ribbed aluminum, because it is a strong, resistant and also light weight material. The disks are in number of two, which will be explained up next.

The first disk has the role of merging the lower part of the project design with the upper part. This is made by attaching four stability pillars, using screws and nuts, on the previously built box. The stability pillars dimensions were taken accordingly to the servo motor dimensions. The size of the disks is about 30 cm in diameter.

The first disk also has the role in holding the horizontal servo motor in place, using screws and the original servo mounting system.

Figure 5.4. Top and bottom view of the first horizontal disk.

The second horizontal disk is the one that transfers the horizontal movement of the project design from the horizontal servo motor to the top part of the project, in order to accurately pinpoint the position of the sun.

Because the top part of the project lies on the horizontal servo that is mounted on the first horizontal disk, some little wheels were adapted to reduce that pressure on the motor.

Figure 5.5. The second horizontal disk, the support wheels and servo mounting spot.

The adapted support wheels have a huge impact in how the horizontal movement is made, due to the position they were placed in. The numbers of wheels used in this design are three, and they are placed in a triangular manner, so that the weight of the top part is equally distributed.

To eliminate some of the friction, they had to be oiled up. The oil markings can be seen in figure 3.6. in the top view of the first horizontal disk. They were mounted using screws. The wheels were scavenged from an old car radio.

Figure 5.6. One of the three adapted support wheels.

Problems encountered

Because the first design of the project used micro servo motors, which were of reduced size, and considerably power, the horizontal movement was impaired. After finding the second types of motors, the problem was that the space between the first disk and the top of the box was insufficient, so the new motors could not fit. The problem was solved by changing the dimensions of the four stability pillars.

Another problem was encountered in the gap between the two horizontal disks. The new motor had different proportions and the support wheels were not touching the surface of the first disk. The problem was solved by carefully measuring the motor mounting, enlarging the space that the motor was and placing it a little bit under so that the wheels can perfectly sit on the first plate.

The vertical arms

The vertical arms are the parts responsible for the vertical movement via the servo motor mounted there. They are in number of two and are made from the same ribbed aluminum as the horizontal disks. Because the aluminum used is ribbed gives them an extra point of toughness, so that they don’t bend under the movement stress. Their height and width were calculated based on the solar panel size, and motor size.

They are mounted on the top horizontal disk via two pairs of nuts and screws

Figure 5.7. The left and the right vertical arms.

It can be seen from figure 3.9. the servo is mounted on the left arm, and on the right arm, an bronze muff has been mounted. The role of the bronze muff is to reduce the friction in the movement of the shaft that was attached on the right part of the solar panel, and make everything go smooth. The bronze muff was scavenged from an old broken motor.

Figure 5.8. Scavenged bronze muff and shaft

Problems encountered

Because the motors had to be replaced, and the new ones were bigger, the left vertical arm was reconstructed in such manner that the new motor could fit.

The sensor cross

The method used makes use of the formation of shadows by change in position of the sun to pinpoint its current position. Generally four light detecting resistors (LDR) are placed closed together separated by black plastic walls forming a cross, with one LDR sensor in each section. The readings obtained from the four LDR sensors are compared to determine the relative luminance and hence find the position of sun relative to the panel.

Figure 5.9. The built sensor cross.

Problems encountered

The first created sensor cross was in reduced dimensions. The light shining on the sensors was improperly distributed, so the tracking was not precise. This was fixed by properly creating a bigger cross made from another type of material, and adding little protection to the outside of the sensors.

IMPLEMENTATION

Final application

In this chapter the hardware and software used and project design have been implemented head to head, resulting in the final application.

Figure 6.1. Final version of the Solar Tracker.

As can be seen on the front command panel of the Solar Tracker, 3 switches have been used. The first switch turns the main application on and off, the red LED will signal if the Arduino is powered on, the second switch turns the vertical orientation on and off, and the last switch turns the horizontal movement on and off.

The point of having individual switches for the servo motors is useful in many ways. If at some point during construction of the project something would have went wrong, the problem could be narrowed down by switching on and off the motors. Another useful thing is that the system can work on a single axis. For example, if the Solar Tracker is placed and oriented by a human operator facing east, the horizontal tracking can be switched off, and the system will track the position of the sun only using the vertical axis, going from east to west.

A mini voltmeter is added and connected directly on the solar panel, for the purpose of not consuming the battery, and because it requires 3V to work, it will stay off when the solar panel is not in the presence of solar radiation. The mini voltmeter will show the voltage of the solar panel in real time.

The command panel also contains a USB port, mainly designed for charging smart phones or other similar devices. It has fast charge capabilities.

Figure 6.2. USB port charging an smart phone.

Figure 6.3. Final application circuit design.

Calibration

After the project has been built, the circuit was designed based on the build, and the code has been implemented, the entire project had to be calibrated in order to work.

The calibration was made using some mechanical knowledge, and using the Serial Monitor from the Arduino IDE program.

First the range of motion of both the servo motors was studied, in order to find a middle point, so that when the project is started, it can go either left or right, up and down within some boundaries, without damaging itself.

Using the serial monitor, the degrees at which the motors work could be easily monitored, and calibrated.

Figure 6.4. Motor and sensor data form the Serial Monitor.

The figure above shows data from the LDRs, as well as servo motor data. Some real time degrees of the vertical and horizontal servo are shown, the average top, bottom, left and right values from the sensors in real time, the time at which the LDRs send data to the Arduino, in this case 34ms, and the tolerance that was chosen through constant experiments.

The serial monitor code for LDR, time and tolerance looks like this:

//serial monitor display for sensors

Serial.print(avt);

Serial.print(" AVE TOP ");

Serial.print(avd);

Serial.print(" AVE BOT ");

Serial.print(avl);

Serial.print(" AVE LEFT ");

Serial.print(avr);

Serial.print(" AVE RIGHT ");

Serial.print(dtime);

Serial.print(" TIME ");

Serial.print(tol);

Serial.println(" TOL ");

The servo motors degrees code:

//serial monitor display for servos

Serial.print(servov);

Serial.print(" V ");

Serial.print(servoh);

Serial.print(" H ");

Project performance

The sun tracker performs perfect under indoor testing, the sun light was simulated artificially using different types of lighting such as flashlights of different power, room lighting, and lighting coming from the windows.

Outdoor testing was almost perfect, the sun tracker was orientating perpendicular to the best light source he had. But because in the outdoor testing period, meteorological factors were not optimum, 80% of the days were cloudy and rainy, further testing should be done

As a final performance of the project, it works in almost 85-95% of the time, depending on random lighting factors.

CONCLUSIONS

A dual axis solar tracker system is an efficient method to track the sun and obtain maximum solar radiation at any time of the day, regardless of where the device is placed. It requires almost no knowledge to be deployed, but mainly, it needs to be placed somewhere in the presence of the sun, and needs to be powered on. Solar energy is becoming one of the most efficient types of alternative energy, because is highly available, and does not cost too much to implement a harnessing device.

One key element of this project was how to build something useful for day to day activities using household items, and the scavenging process.

A financial analysis will be provided on all the items and devices used in the project. As a remark, everything will be averaged, because prices differ on the market.

Financial analysis

It will be structured based on items and devices found laying around my house, or scavenged, and items and devices I had to buy.

Household/Scavenged items and devices:

Wooden plate made in a specific stratified manner( 8mm thickness) – 100 LEI

Ribbed aluminum plate (2mm thickness) – 200 LEI

Plastic plate (2mm thickness) – 30 LEI

Steel pipe (3 cm diameter) – 20 LEI

2x Buck circuits (one from a car charger, one unknown but functional) – 40 LEI

12V Battery (scavenged from an broken UPS) – 200 LEI

TOTAL = 590 LEI

Purchased items and devices:

Solar panel – 70 LEI

2x Servo motors – 60 LEI

Arduino UNO – 30 LEI

Wires (red and black) 5m – 30 LEI

Arduino jumper wires – 10 LEI

Mini voltmeter – 10 LEI

TOTAL = 210 LEI

Total cost of the project is around 800 LEI, and by using household items and scavenging, the total costs have been reduced by 73,75%.

References

Adarsh, S., Anand, A., & Singla, J. (2015). Increasing the Efficiency of a Pv System Using Dual, (2), 37–41.

Shea, J. J. (2002). Clean electricity from photovoltaics [Book Review]. IEEE Electrical Insulation Magazine (Vol. 18). http://doi.org/10.1109/MEI.2002.1019908

Distribution of direct solar radiation on the globe. https://solargis.com/

The spectrum of direct solar radiation at sea level and at the outer limit of the earth's atmosphere. https://commons.wikimedia.org/wiki/File:Solar_Spectrum.png

CMP3 Pyranometer. http://www.omniinstruments.co.uk/cmp3-pyranometer.html

Photovoltaic sensor for measuring solar radiation intensity. https://www.indiamart.com/proddetail/solar-irradiance-sensor-13315671062.html

The band model for insulators, semiconductors and conductors. https://www.halbleiter.org/en/fundamentals/conductors-insulators-semiconductors/

Phosphorous doped silicon, Boron doped silicon http://solarjourneyusa.com/bandgaps.php

Photovoltaic cell – operating principle. https://global.kyocera.com/prdct/solar/spirit/about_solar/cell.html

Generic I-U curve of a photovoltaic cell. http://solarprofessional.com/articles/design-installation/field-applications-for-i-v-curve-tracers#.WyvSRKczaUm

Graphical representation of the fill factor. http://acse.pub.ro/wp-content/uploads/2014/08/licenta.pdf

The conversion efficiency of the world's best photovoltaic cells between 1976 and 2018. https://www.nrel.gov/pv/national-center-for-photovoltaics.html

Temperature dependence of I-U characteristic of photovoltaic cells. http://www.pvresources.com/en/solarcells/solarcells.php

Nayak,S.R.; Pradhan, C.R. Solar tracking application. IOSR J. Eng. 2012, 2, 1278–1281.

Lee, C.Y.; Chou, P.C.; Chiang, C.M.; Lin, C.F. Sun tracking systems: A review. Sensors 2009, 9, 3875–3890.

Rubio, F.R.; Ortega, M.G.; Gordillo, F.; Lopez-Martinez, M. Application of new control strategy for sun tracking. Energ. Conv. Manage. 2007, 48, 2174–2184.

Blanco-Muriel, M.; Alarcon-Padilla, D.C.; Lopez-Moratalla, T.; Lara-Coira, M. Computing the solar vector. Solar Energy 2001, 70, 431–441.

Reda, I.; Andreas, A. Solar position algorithm for solar radiation applications. Solar Energy 2004, 76, 577–589.

Abdallah, S.; Nijmeh, S. Two axes sun tracking system with PLC control. Energ. Conv. Manag. 2004, 45, 1931–1939.

Saravanan, C.; Panneerselvam, M.A.; Christopher, I.W. A novel low cost automatic solar tracking system. Int. J. Comput. Appl. 2011, 31, 62–67.

Barsoum, N. Fabrication of dual-axis solar tracking controller project. Intel. Control Autom. 2011, 2, 57–68.

Dasgupta, S.; Suwandi, F.W.; Sahoo, S.K.; Panda, S.K. Dual Axis Sun Tracking with PV Cell as the Sensor, Utilizing Hybrid Electrical Characteristics of the Cell to Determine Insolation. In Proceedings of 2010 IEEE International Conference on Sustainable Energy Technologies, Kandy, Sri Lanka, 6–9 December 2010; pp. 1–5.

Yan, Z.; Jiaxing, Z. Application of Fuzzy Logic Control Approach in a Microcontroller-Based Sun Tracking System. In Proceedings of 2010 WASE International Conference on Information Engineering, Beidaihe, China, 14–15 August 2010; pp. 161–164.

Yan, Z.; Jiaxing, Z. Application of Fuzzy Logic Control Approach in a Microcontroller-Based Sun Tracking System. In Proceedings of 2010 WASE International Conference on Information Engineering, Beidaihe, China, 14–15 August 2010; pp. 161–164.

Arduino Uno development board. http://beliketechies.blogspot.com/

Typical construction of a light detecting resistor. https://www.mouser.com/ds/2/737/photocells-932884.pdf

https://www.arduino.cc/en/Main/Software?

Documentation about the code. http://www.instructables.com/id/Simple-Dual-Axis-Solar-Tracker/

General ideas. http://acse.pub.ro/wp-content/uploads/2014/08/licenta.pdf