Agile Eye Solar Tracker

Abstract

In response to the growing demand for efficient solar energy solutions, our team presents an innovative solar tracker design. Focused on enhancing energy harnessing during peak hours, we explore an agile eye mechanism as a versatile and affordable alternative to existing mechanisms. This report outlines the mechatronics part of this project, covering the prototype of the product. We explore key project statements, State of the Art analysis, and design requirements. Subsequent sections delve into the embodiment design, sub-system design and an integration guide, offering a comprehensive overview of our sustainable energy solution.

The complete report can also be found inside the project repository.

Table of contents

1. Project motivation
2. Functional Analysis
3. State of the Art
4. State of the Art
5. Patent analysis
6. Conceptual Design
7. Problems and features
8. Concepts
9. Embodiment Design
10. Subsystems
11. Concepts
12. Choices
13. Design
14. Material
15. Manufacturing
16. Assembling
17. Design of sub-systems
18. Mechanical systems
19. Agile eye mechanism
20. Sliding system
21. Base and supports
22. Final CAD Design
23. Circuitry & Sensors
24. Sensor parsing
25. Button parsing
26. Commanding motors
27. Solar panels
28. Complete circuit
29. Software
30. Integration guide
31. Demo of project
32. Review
33. Sustainability
34. Bill of Materials
35. The team
36. Project repo

Supplies

For a detailed Bill of Materials: see the repository.

Step 1: Project Motivation

Present day, we all use the well-known fixed solar panels. From companies to individuals, more and more people are using this sustainable way of producing energy. But the problem with this technology is that, depending on the position of the sun, conventional panels do not always produce the maximum amount of energy. This is because the panels are in a fixed position. If the panel were able to follow the path of the sun, the same energy could be achieved using smaller panels. This is exactly what a solar tracker does.

That there is a real need for solar trackers can be proven by the following data. Looking at Figure 1 (Baert, 2023), the highest grid load in Belgium takes place between 6-12h AM. This is exactly the time when solar trackers can make a big difference in power as seen in Figure 2 (Al-Jumaili, Haglan, Mohammed, & Eesee, 2020).

Beside the previous need, there is the need for an affordable price. At this moment, solar trackers with panel cost between $1500-2000 (Clissitt, 2023). If there could be a cheaper alternative, this would answer this need.

There are multiple types of solar trackers, the most known with a single or double axis. It will not be possible to develop one of these types since they are patented. For that reason, another mechanism called agile eye will be used.

The product has been developed to be purchased specifically by private individuals. Therefore, an attempt will be made to reduce its cost.

This solar tracker is intended for use in everyday situations so, the owner of the product will be able to supply his own energy in a sustainable way and do it in a portable way that suits his needs. The idea is to develop a compact, affordable and easy-to-handle product. It will be made portable, which means that it could also be used on the go. On top of that, it could also be used at home, by attaching it to a roof or placing it in a garden.

Step 2: Functional Analysis

To make sure we have something to strive for and something to measure the finished product up against, we need to define the requirements of the product. This list of requirements can be seen in Table 1. The scale of importance is a number between 0 (unimportant) and 4 (very important), just like the scale of flexibility: F0 (not flexible at all) to F4 (flexible). Most notable in the requirements list are the items ‘Rotate angle’, ‘Tilting angle’, and ‘min velocity (tracking)’. If the first two are not met, we won’t be able to track the sun in all its orientations and if we don’t meet the minimal velocity while tracking the sun will outpace the tracker. The minimum velocity is calculated with an estimated min daytime of 6 hours. The 50dB noise limit is placed so a normal conversation (60dB) won’t be hindered by the functioning of the machine. The battery size is chosen to be able to charge a modern smart phone up to 2-3 times one and to give enough time for the tracker to find the sun to begin using the solar panel. All of this should be manageable constraints that will help to design the agile eye solar tracker.

Step 3: State of the Art

a.   State of the Art

The two main technologies used in solar trackers are single-axis and dual-axis. The single-axis solar tracker, as its name suggests, is only capable of orienting the panel around one axis of rotation. As can be seen in Figure 3 (Rooij), it can be either on the horizontal or the vertical axis.

This contrasts with the double axis solar tracker, which can combine two axes of rotation, as can be seen in Figure 4 (Niclas). This means greater precision and higher performance.

In addition to these types of solar trackers, the possibility of using an agile eye mechanism to orient the panels has been taken into account. This type of mechanism can be seen in Figure 5 (Rangaprasad, Bandyopadhyay, & Ghosal, 2013), this allows the movement of the panel in two or even three degrees of freedom. The difference between the two will be explained in detail later.

By way of analysis, a table has been drawn up to evaluate the possibilities that can be offered by each of these types of panels. Table 2 (Cooke, 2011) includes several parameters to be evaluated among the different products. In addition to the three types mentioned above, the use of fixed panels has been included in the analysis to obtain a more complete study.

b.   Patent analysis

As told before, the single and dual axis solar trackers are patented. The patents can be found below in Figure 6. On the left patent EP4002685A1, composed 25/05/2022 for the EU. In particular, the patent refers to a single-axis horizontal solar tracker being configured in such a way that the effects of the wind are reduced (Europe Patentnr. EP4002685A1, 2022).

The second patent is EP1998122A1, composed on 3/12/2008 for the EU. This one refers to a dual-axis solar tracker that tracks the sun both horizontal and vertical, characterised by a hollow tower that both supports and cools the head, preventing a drop in performance at high temperatures on hot days (Europe Patentnr. EP1998122A1, 2008). Because of these patents, it will be our choice to use another mechanism.

For the solar tracker, an agile eye mechanism will be used. In combination with a solar panel, this mechanism is not yet patented. Just like the dual-axis, the agile eye has a lot of freedom of movement. There exist versions with 2 degrees of freedom (Callegari & Carbonari, 2013) or with 3 degrees of freedom (Sadeqi, Bourgeois, Park, & Arzanpour, 2017) as showed in Figure 7. For this problem 2 could already be enough.

Step 4: Conceptual Design

a.   Problems and features

To start designing the agile eye solar tracker, all the problems to solve should be identified. With these problems in mind, features for the solar tracker should be chosen. These technical problems and features are listed in Table 3. Since it is a solar tracker, the first and obviously most important problem would be to follow the sun. This shall be done by using a solar tracker mechanism like single-axis, double-axis or agile eye.

To reduce the cost, weight and dimensions, there should be looked at the material choice. At first thought the prototype could be manufactured by MDF and PLA. This will be discussed in more detail later. The solar panels should be attached to the mechanism in a way that they are easily replaceable if they break. For this reason, a solar panel sliding mechanism shall be implemented.

The necessary torque should be calculated to know how many motors need to be used and what torque they need to have available. If the solar panels attach immediately on the tracker mechanism, they will bump into the mechanism if it turns. This means the solar panel sliding mechanism will have to be placed on top of a rod of certain length.

All these features can be implemented using different means. To choose between the means, it is recommended to compose a morphological chart (Table 4). In this table, the options that will be adopted have been marked in green. In the next section, the different concepts will be discussed to justify the choice of each mean.

b.   Concepts

In this section, the different concepts of means will be investigated. In the section about patents was concluded that there are patents on single and double axis solar trackers. To have complete freedom to operate, an agile eye mechanism will be used. As seen before in Figure 7, there are versions with 2 or 3 motors. To know which version should be used, let us look at a spherical coordinate system as shown in Figure 8. It can be quickly seen that it is only necessary to be able to change θ and φ. The radius of the mechanism will be constant. This means that 2 motors will be sufficient.

To choose the type of motor, the four options are compared in Table 5. Like this, it can be seen that the best options are stepper and servomotors, since these operate at a constant torque and have high precision. As both are valid for the prototype and due to the availability of stepper motors in FabLab, it has been decided to use them instead of servomotors.

A similar table can be constructed to determine which kind of sensor should be used; this can be seen in Table 6. It should be noted that the temperature sensor and camera are not included in this table. A temperature would be too unsensitive, while a camera is very expensive and collects unimportant data. Instead, two different possible kind of light intensity sensors are included: a photoresistor and a photodiode. While the photoresistor is the cheapest, the others have a higher sensitivity and gain and are active devices. Yet, a sun sensor would be very expensive (Quarktwin, 2023). For this reason, the solar tracker will use photodiodes. 

To select how many of these photodiodes should be used, multiple layouts should be considered. This can be seen in Figure 9. If one sensor would be used, the tracker should take values on different positions and calculate the optimal position by itself. This should be done periodically to keep tracking the sun. This is a tedious process. A similar problem occurs when two sensors are used, yet only in one direction. The best option would be to use four sensors and compute the light differences between all of them to know in which direction to move.

The solar panels can be attached to the solar tracker using three different options: Bolts, glue, or a special designed sliding mechanism (see Figure 10). A comparison can be seen in Table 7. It would be an advantage if the solar panels would be easily replaceable if they break. For this reason, it is recommended to use multiple smaller ones instead of one large one. The replacement would not be possible if they are glued together. The sliding mechanism has a higher cost, but can protect the panels more than the bolts, while there are no holes needed in the panels. It can be concluded that a self-designed sliding mechanism for multiple small solar panels is the best option.

When looking at the battery-options, the most comfortable to use because they are available at the FabLab, are power supplies. They give the required energy and are the most suitable option. So, in this case this will be the way to supply the prototype.

Talking about how to implement the code of the prototype, the most suitable and easy to use is the Arduino Uno.

Step 5: Embodiment Design

This section covers all the concepts and choices made during the project. One part is dedicated to all the concepts the group came up with. Another is dedicated to the choices made based on the concepts introduced. The end of the chapter is dedicated to the choices made for the materials and assembly of the entire prototype.

a.   Subsystems

In Figure 11 is the block diagram that summarizes how the whole prototype interacts to orient itself towards the sun.

The aim of the mechanism is to move the solar panels, and therefore the mechanism, using motors. The direction of rotation is determined by the Arduino, which receives signals from the sensors.

To know in which direction the motors should turn, it is possible to determine this by simple addition and subtraction of sensors 2 by 2 as can be seen in Figure 12.

b.   Concepts

In the previous chapter, the morphological means were determined. Yet for some means, there remain some options of how to implement them: the 2DOF system, sliding system, 4 sensor layout and handles all have some different embodiment concepts. In this chapter, those will be discussed to select the best concept (based on the requirements determined in Table 1).

Starting with the 2DOF agile eye, there can be 2 options for the singular arm as seen in Figure 13: An L-arm (Callegari & Carbonari, 2013) and a U-arm. It can easily be seen that the U-arm will induce a higher stability.

The next concepts pertain the implementation of the sliding system. They can be seen in Figure 14. Originally, a large rod was chosen. It could also be tried with a much smaller rod. Of course, this will highly limit the possible movements of the tracker. Secondary, the choice between a sliding- or placing mechanism must be made.

Third, there are the concepts of sensor placement with many choices as seen in Figure 15. The sensors could be placed in the corners, in the middle and on the side. Besides, they could also be placed under an angle (here 45° is taken as an example). It should be noted that that the first and fourth options lower the precision, while the second takes away a lot of room for the solar panels.

c.   Choices

To determine which embodiment concepts should be chosen, there can be looked at Table 8. The agile eye with a U-arm will be preferred over the one with an L-arm, since the first would be more stable and rigid. The stability also implies that the system will be more precise. The sliding system was chosen mainly for its ease of maintenance. It can also be seen that a small rod is not a good option since it decreases the minimum tilting angle a lot. The sensors will be placed on the side of the sliding system, unrotated. This way the solar panels do not get covered (like with the middle placement), the precision stays high (which is not the case with the corner placement) and the calculations are simpler than with the rotated placement.

d.   Design

Now that the concepts are fully constrained, the designing can begin. Using Autodesk Inventor Professional 2024, the different parts of the solar tracker were fully designed. In this subchapter, the most significant points of this design will be described. For more details, the technical drawings of all components can be found in the project repository. The assembled subparts can be seen in Figure 16.

The motors are attached to a base plate. In the bottom of this plate is a standardised screw hole that can used to mount the whole mechanism on top of any tripod. Using the motors, the mounting plate in the middle of the agile eye will be able to turn. On top of this plate, the sliding mechanism will be mounted. Buttons (hinge levers) will also be used to prevent the "agile eye" from colliding with the various supports. They are used to determine the maximum angular amplitude of our device. Here, it's 45°. Two buttons will be used, one for each arm as we can see in Figure 16.

e.   Material

For the prototype, basic, but strong materials were chosen: MDF, PLA and PVC. The MDF is a good choice since it is a strong material, its price is low and the process to manufacture something out of it is very fast. This constitutes the main parts of the system (the base, the supports for the motors and the limit switches and the sliding mechanism). The PLA is used for the mechanism itself: the arms of the agile eye and the plates to hold the rod. These were made in PLA, since they were more complex and thus 3D-printed. The rod must be strong and light to not overload the motors. The easiest way to manufacture such a rod is thus to cut it from an already existing PVC-pipe.

f.   Manufacturing

In order to use those materials for the prototype, some machines need to be used. The manufacturing processes must be cheap and fast to suit if a piece breaks or does not have the proper dimensions. Thanks to a comparison of different manufacturing processes, we can choose the laser cutter machine for the MDF-plates and the 3D printer for the PLA (see Table 9). Other manufacturing processes, like a saw are not good processes since they have bad accuracy (which is important to make the assembling). This while once a CAD-model has been made, the 3D printing and laser cutting can be done very fast.

g. Assembling

The prototype has been designed to make the assembling easy between each part. The Table 10 shows the methodology to choose the most suitable assembling mode. The prototype must be cheap, which means the assembling mode should be cheap. Regarding the table, it can be seen that screwing is the most suitable since it is easy if a part needs to be replaced. The supports are designed as shown in Figure 17. There are spaces to put screws and nuts to fix the parts.

Step 6: Design of Sub-systems

a.   Mechanical systems

After seeing the different parts of the prototype and talking about the materials and the type of manufacturing chosen, let’s now discuss in more detail the different subsystems that correspond to the mechanical part of the prototype. The different mechanical subassemblies of which the prototype is composed are the agile eye mechanism, the sliding system for the solar panels and the base of the assembly that includes the supports for the motors and the switches.

i.    Agile eye mechanism

In order to develop the proposed mechanism, it was necessary to have a series of requirements that the design must meet. On the one hand, the correct movement of the mechanism must be guaranteed, verifying that there is complete movement and that parts do not collide with each other. In addition, the use of material must be optimized, making the design as compact as possible.

With reference to other agile eye mechanisms, the different components of the mechanism were developed. As can be seen in the Figure 18, it is mainly composed of three arms and a mounting plate.

As previously discussed in the embodiment design section, it has been decided to use a U-shaped arm to hold the mounting plate in a more stable way. This arm will be attached to one of the two motors. This arm has been defined as ‘Arm1’ and is shown in the Figure 19.

The other motor is connected to one of the L-arms, which in turn is connected to the other L-arm by means of a screw that allows the rotation of both on its axis. This last arm will be anchored to the mounting plate. In this way the movement will be transmitted from the motor to the first arm and the first arm will make the other arm move, taking also into account the position of the U-arm. The arm attached to the mounting plate has been defined as ‘Arm 2’ (Figure 20) and the one which is connected to the motor is ‘Arm 3’ (Figure 21).

The mounting plate (Figure 22) is designed in such a way that the arms can be attached to it by means of screws that allow the arms to rotate on their axis. In addition, it has four holes in its base to anchor the bar to which the sliding system would be attached.

The screws that were used to attach each of the arms to the mounting plate have been chosen of a large size to act as rigid axes. These screws are M10 and allow the parts to rotate around their own axis. In Figure 23 can be seen how they are connected.

For the manufacturing of the complete agile eye mechanism it was decided to use 3D printing, as discussed in the previous section. This decision is mainly based on the curved shapes of each of these parts.

For printing the mounting plate there was no problem, as the entire part could be printed at once. On the other hand, to print each of the arms, each of the pieces had to be split in two and then glued together with a specialized PLA glue. For better understanding of the solution taken, Figure 24 shows one of the arms designed to guarantee a correct printing.

ii.    Sliding system

A number of requirements were taken into account when designing the assembly in which the solar panels are located. On one hand, the system should be easy to assemble. In addition, the panels must have a protection against damage. That is why it was decided to use the sliding mechanism. An overview of the complete assembly is shown in Figure 25.

Besides, other requirements must be met, such as not colliding with the agile eye mechanism. For this reason a rod has been added to lift the assembly, giving greater freedom of movement. Another requirement that this assembly must meet is that it must have a place for the sensors and that they must be arranged in such a way that the quadrants are well defined and the light they receive is not the same in order to correctly orient the mechanism.

Below is a description of the different parts that make up this assembly. In addition, we will discuss how each one is manufactured and its function.

The base of the mechanism is composed of four parts. The first is a base (Figure 26) on which the panels are supported. This has elongated cutouts in the location of the panel wires so that they can slide in. It also has countersunk holes to allow the piece to be joined to the rod.

On top of this plate would be the second one (Figure 28), which will act as a wall for the panels. This piece will have an open side to be able to insert the panels. To close the open part there is another small elongated plate that will act as a closing device (Figure 27).

The last part is another plate that is placed on top (Figure 29). This plate has a slightly smaller recess than the size of the panels. This way, the panels will not be able to come out of the top and they will be protected. In addition, this plate has a square-shaped protrusion on which the sensors will be placed.

All of the plates discussed above are made from 3 mm MDF and have screws to fasten them together.

Talking about the sensors, as mentioned before, each one will be placed in a quadrant. To separate them from each other, a cross (Figure 30) has been designed to separate the light reaching each sensor. The pieces that make up the cross were made using 3 mm MDF and then glued together with wood glue to ensure that they will stay together.

The last part of the assembly would be the rod. As mentioned above, its function is to elevate the entire assembly so that it does not interfere with the movement of the agile eye. For this purpose, a PVC rod (Figure 31) has been used, selected for being light, and some 3D printed bases (Figure 32). These bases have holes to screw one of them to the sliding mechanism and the other to the mounting plate, so that both subassemblies are connected.

To help understand how the whole system works, Figure 33 shows how the panels would be inserted into the assembly.

iii.    Base and supports

To finalize the whole assembly, it has been necessary to design a base on which all the parts will be placed. This base has a series of holes so that the necessary supports that hold the whole assembly can be fixed to it. The only requirement that this subassembly must meet is that it is strong enough to support the rest of the structure.

The parts that form this assembly are the base, the supports of the motors and the supports of the switches. The base consists of a rectangular-shaped MDF board, which has a series of holes that allow the supports to be anchored to it by bolts.

On the other hand, the motor supports have also been manufactured by laser cutting in 3 mm MDF panels. Initially, a design was created that was only connected to each other by means of slits in the wood, so that the parts fit together. It quickly became apparent that the rigidity of the assembly was not sufficient. Thus, a new design was made in which the different pieces would be joined together with M3 bolts, as can be seen in the Figure 34.

Considering the good result of the design in MDF with bolt connection, it was decided to use the same technique for the support of the switches, which are used to mark the origin of coordinates of the agile eye. This structure, similar to that of the motors but of a smaller size, can be seen in the Figure 35.

iv.    Final CAD Design

Finally, Figure 36 shows the assembly of the complete prototype. In the figure it can be seen how the three sub-assemblies that have been explained throughout this section are included.

b.   Circuitry & sensors

To make the solar tracker function, several different circuits need to be made to meet the previously mentioned requirements. In total 4 sub circuits need to be made for these functionalities:

1.     Sensor parsing

2.     Button parsing

3.     Commanding motors for moving the agile eye mechanism.

4.     Solar panels

These sub circuits come together to communicate to an Arduino Uno. This integration will be discussed at the end of this section. For each of these sub circuits, components need to be selected. The selected components are discussed with the design of the sub circuits. Each of these sub systems was tested on its own before combining them for the final circuit.

i.    Sensor parsing

The prototype has been fitted as previously discussed with photodiodes. The selected photodiodes are the BPW20RF photodiodes from ‘VISHAY’. This diode was selected due to its sensitivity range which matches the illuminance of the sun: On a dark day a minimum of 100 lux, on a sunny day a maximum of 100 000 lux as can be seen in Figure 38. Another consideration was the sensitivity relative to the angle of the incoming light. The chosen diode performs well in this regard as seen on Figure 37.

The ‘signal’ created by this sensor needs to be converted to 0-5 Volts to be able to communicate to the Arduino. For this reason, a sensor parsing sub-circuit needs to be designed and constructed. This circuit will need to be copied 4 times to facilitate the sensor array discussed before. 2 types of sub circuits where considered:

1.     Trans conductance amplifier coupled with an inverting amplifier.

2.     Voltage divider.

In the first type the circuit (per sensor) utilizes 1 opamp, 1 resistor, +12V, -12V and 10V for the trans conductance amplifier as seen in Figure 40. For the inverting amplifier to go from -5-0V to 0-5V, an extra opamp and 2 resistors are necessary. Together this makes: 2 opamp’s, 3 resistors, +12V, -12V and 10V. This is a lot of components and complexity in comparison to the second type which needs 1 resistor and 5V. For this reason, we utilize the second type of sub circuit. The resistor value was chosen as 1MΩ to limit the current draw per sensor. This gives us the final sensor parsing sub-circuit seen in Figure 39.

ii.   Button parsing

Buttons are sensitive to bouncing. Bouncing is a phenomenon that happens when a button closes/opens. The switch contacts ‘bounce’ against each other giving multiple impulses/ false positives upon closing/opening due to being spring-loaded. To fix this, a debouncing circuit was utilized. This circuit is a RC low pass filter with a capacitance of 0,1uF and a 220kΩ resistor as seen in Figure 41. With this circuit the output is 5V when the switch is closed and 0V when the switch is opened. The used switch can be seen on Figure 42.

iii.   Commanding motors

As discussed before, the solar tracker prototype will utilize 2 stepper motors of type Nema 17(Figure 43) with its properties in Table 11.

The Nema 17 motor was selected due to its availability, current limit per phase and holding torque. The expected min holding torque needed for normal operation was estimated with following calculations:

m = estimated weight of solar panels and rod system = 0,303kg

L_rod = 14 cm

g = 9,81 m/s^2

The estimation of the weight is done by calculating the volume of the used MDF (density of 730 kg/m^3) and adding the weight of the solar panels.

T_min holding = m*g*L_rod = 0,42Nm

The minimum holding torque was calculated by simplifying the rod and agile eye system to a force on a lever where the length of the lever is the length of the rod.

Communication and commanding the motor is done via a stepper driver. For this the A4988 stepper driver was chosen (Figure 45). The driver allows us to command a motor with a max current per phase of 2A and is compatible with the Arduino Uno which is utilized as logic brain of the tracker. For this the minimum viable circuit will be used (Figure 44). For more accurate positioning the MSi pins can be utilized for smaller steps, but this isn’t needed for our application. The current limit of the driver was also set to 1.68A utilizing the provided potentiometer on the driver.

iv.    Solar Panels

The 4 solar panels in use are the ‘Seeed Studio 0.5W Kit’ solar panels. These solar panels were put in series for a max voltage of 25.6V. To be able to monitor the output of the solar array, a voltage divider was utilized to step down this voltage to a range of 0-5V.

v.    Complete circuit

After putting all the sub circuits together, we get the following circuit diagram (Figure 47). The circuit consists of the components listed in Table 12. The input voltage for the stepper drivers was provided with a 100uF capacitor to eliminate voltage spikes. See below for a summary of the circuit.

• Power
• Arduino Uno provided with +12V and GND
• Stepper driver A4988 provided with +12V, +5V and GND
• Sensor sub circuit provided with +5V and GND
• Inputs/outputs Arduino Uno
• Solar panels analog 0-5V pin A0
• Sensor input analog 0-5V pins A1-4
• Stepper drivers pins 1-5
• Limit switches pins 6 and 12

c.   Software

The program to operate the solar tracker runs on an Arduino Uno. This means the program was written in a variant of C++ . The explanation of the code will be done utilizing flow charts. The program was written so the robot works in 3 steps upon startup as seen in Figure 48.

The initialization of the 2 motors must be done because we work with stepper motors without encoders. This means that upon startup we don’t know the orientation of the motors. To initialise the orientation 2 limit switches were utilized, one for each motor. This was done with the steps shown in Figure 49.

After finding the switch the motor resets the angle saved in the memory of the Arduino. Then it turns to the opposite limit (max angle) to check if the arm is unobstructed and then turns to a starting position in the middle of the working area by turning to half of the max angle. After doing this for both of the motors as seen in Figure 48 the program starts the ‘Acquiring and tracking the sun’ step. This step is visualized in Figure 50. This diagam is repeated until the tracker is shut off.

The flow diagram goes as follows: The sensors (photodiodes) get read and Dir_1 and 2 get calculated. These values determine if the motor needs to turn anti- of clockwise for motor 1 and 2 respectively. This calculation is done by intelligently utilizing the layout of the sensors as seen in Figure 51.

When calculating the turning direction of motor 1: Dir 1 is calculated by subtracting the sum of values of sensors 1 and 3 from the sum of values of sensors 2 and 4. This means that when Dir 1 is bigger than 0 the average brightness on the side of sensors 2 and 4 is bigger than the other side and thus motor 1 should turn to point in that direction. An analog calculation can be done for motor 2 (Dir 2). These values are then used as seen in Figure 50.

When interacting with the motors the A4988 stepper driver was used. This stepper driver makes turning the stepper motors easy. When a rising edge is detected on the step pin of the driver the motor will take a step in the direction given by the direction pin of the driver. An example of one step in code is given in Figure 52. Depending on the value of the variable 'Delay' the motor wil turn faster of slower when looped.

The program was written so that the code can run independently of the computer. This means that upon startup (connecting to power or pressing the reset button on the arduino) the program follows the flow diagram shown in Figure 48. The final code can be viewed in the project repository.

Step 7: Integration Guide

The Agile Eye Solar Tracker mechanism dynamically aligns solar panels with the position of the sun to efficiently harvest solar energy. This integration guide provides step-by-step instructions for the seamless integration of the Agile Eye mechanism into solar energy systems.

Step 1 - The material used:

The components and tools utilized in this project include the Agile Eye Solar Tracker parts (see repository), Arduino Uno Microcontroller, 2 Nema 17 motors, 2 stepper drivers (A4988), PLA for 3D printing, MDF for laser cutting, wires, hinge levers, screws and a breadboard. The integration process comprises mechanical assembly, motor integration, electrical wiring and Arduino programming.

Step 2 – Print the different parts:

The different design parts are made by the CAD Inventor. The main parts of the prototype are made thanks to laser cutting and assembled with screws. Those parts are designed to take the screws into account. The arms of the tracker are printed with a 3D-printer. Print the arms with the 3D-printer and the other parts with the laser cutting machine.

Step 3 – The wiring of the system:

Connect each motor to a A4988 driver and connect each voltage supply of the driver to the power supply of the Arduino, for a clearer view there can be looked at Figure 53 The ground must be connected to the ground of the power supply. Then connect the phases of each motor to the pins 1B, 1A, 2A and 2b of the driver. Pay attention to the colours of the wires leaving from the motor! For the limit switches (Figure 54), identify the 3 ports: NO, NC and Common. These ports help you to know how to connect them. Putting the button in the left (like in the figure) the left ports is the ground following by the NO and the NC. Link the NO with a pin in your Arduino (here pins 12 and 6) and link the common port with the supply voltage (5V). Then you connect the solar panels to an output pin of the Arduino and finally, you connect each sensor to an output pin also and the 5V pin.

Step 4 – Assemble the plates with the base:

Take the parts from the laser cutting machine and assemble them with screws. The plates with the base must fit well. Screw the motors with the dedicated holes in the mount plates. For the switches, fit the 3 ports with the holes of the little plate as can be seen on Figure 55. This plate is screwed with the other plates and the base like the other parts.

Step 5 – Assemble the arms and the solar panel system:

Take the arms and put each arm on a shaft of a motor. Put bolts in the sides of the arms and connect each side of the U arm with a bar which holds the sensor plate. Fix the sensor plate with the MDF sensor plate. For the solar panel system put the solar panel in the sliding system and screw the two parts of this system. Screw the rod with the sliding system. Then, fix the wiring for sensors in the above plate of the sliding system.

How can the solar tracker follow the sun light?

The sensors catch the light and a computation of the sum of the light intensity between 2 by 2 sensors allows to know in which direction the solar panels must turn. The information is given to the motors and the mechanism tracks the sun until a limit angle. This angle is reached when the rod is in contact with a switch (push button) which stops the rotation.

Step 8: Demo of Project

The voice-covered demonstration of the project can also be found in the project repository.

Step 9: Review

Finally, the prototype operates but is not perfect, some features must be fixed. The prototype does not work with the sliding system and the rod for the solar panels because the motors are not powerful enough. This can be fixed by replacing the actual motors with more powerful motors. To get a more sustainable system we can use the electricity produced by the solar panels to supply the system instead of a power supply. A solution is to connect the wires of the solar panels in the pins of Arduino (5V and Ground) and provide a battery to store excess energy.

It can be concluded that the assignment is successfully completed: An agile eye solar tracker was built. However, it would be nice if it also would be able to carry the solar panels themselves and immediately harvest a part of their produced energy.

Step 10: Sustainability

In the pursuit of making a solar tracker, one immediately thinks of sustainability. The solar tracker itself is of course already a key-player within eco-friendly power, yet what about the material and production?

The prototype was designed with sustainability in mind. Most of the connections are done with bolts and nuts, which means they are easily disassembled. That way, when something breaks or should be redesigned, this part can just simply be disassembled without throwing away any of the other parts.

Together with this, a lot of parts are made of PLA. This is a type of plastic that can be recycled and reused for 3D-printing. Yet the parts out of MDF are non-recyclable. These could be replaced by other more reusable materials, but that way you would lose the high speed of the manufacturing process.

The life span is currently the most limited because of the electronic components. To protect these and thus improve the lifespan, it is recommended that in the future these should be in a protective box.

Step 11: Bill of Materials

An exact Bill of Materials of the prototype can be found in the project repository, but a concluded version can be seen in Table 13. To calculate the volumes of MDF and PLA, the CAD model of the solar tracker has been used. With these in mind, an estimation of the complete price for the materials of the prototype could be made. This estimation is €108,71. It should be noted that these are the new prices of components such as steppers, drivers, switches… During the mechatronics sessions, it was possible to get cheaper (already used) versions. This means that this price is an overestimation.

Step 12: The Team

Ramses Jacquet

My name is Ramses Jacquet, and I am a master student Electromechanical engineering robotics in the BruFacE program at the VUB and ULB. I did my bachelors degree in ‘werktuigkunde’ at the VUB. For this project I helped with the CAD design, wrote the code for the project (reading the sensors, commanding the motors, and implementing the switches), designed and made the circuits used in the final robot, helped in selecting the sensor’s, motors. I enjoyed the project immensely, most of all the design of the circuits and the testing and integration of the Arduino code.

Arthur Lots

I am a master student Electromechanical engineering robotics in the BruFacE program. I did my bachelors degree at the VUB. During this project, I mainly focussed on the mechanical design, manufacturing and assembly of the solar tracker. Together with that, I tried to keep a general overview and management over the group. This included taking the lead during meetings, communication with professors and assistants, assigning tasks to everyone and keeping an eye on the deadlines. Since this was the first time I did this, it was a very challenging task, yet one I learned a lot from.

María Talavera

My name is María and I am a Double Degree student of the Master in Electromechanical Engineering of the BRUFACE program. I have studied the Bachelor and the first year of the Master at ‘Universidad Politécnica de Madrid’ and this year I decided to come to Brussels to continue my studies.

In this project most of the time I have dedicated to the mechanical part. I mainly have focused on the realization of drawings and sketches, CAD design, manufacturing and assembly. What I have enjoyed most about this project has been the design part, since it is a field that interests me a lot. Also having the experience of meeting new colleagues, working with them and learn new things from them.

Hassan Ballouk

I am a master student of electromechanical engineering robotics in the BruFacE program. On our project, the solar planner is the most beneficial experience for me. I learn a lot of things, including the programming of Arduino and how to connect the Arduino with circuit connection electronics. I learn how to assemble parts of different components through threading, and my friends in our group are very helpful and helped me with many things in my work on our group. I work in Arduino coding with my help of my freinds and final testing.

Yassine Bouhjar

My name is Yassine and I did my bachelor at the “École Polytechnique de Bruxelles”. During my third year, I took the biomedical option and then switched to mechatronics.

In this project, I mainly worked on the Arduino part. In particular, to make the link between the motors and the sensors. What I liked was the fact that we had the freedom to choose and implement the project as we wished. The fact that we were all working on different tasks and needed each other was what I appreciated most.

Zakaria Zouaoui

I am Zakaria and I am in the master's degree in electromechanical engineering in robotics at ULB. During this project I worked in the design of the plates for switches, in the manufacturing process and the study of the sensors. I helped for the Arduino connection for the switches. What I liked is the learning of lots of things like team working, project management and developed my skills in electronics.

Step 13: Project Repo

All references, complete report, code, BOM, project files and voice-spoken project demo can be found on the project repository:

Project repository