University of Malta
 

2014 Projects
UOM Main Page
 
 
 
Apply - Admissions 2016
Newspoint
Campus Map button
Facebook
Twitter

Final Year Projects Academic Year 2013/2014

An Induction Welder

Student: Christian Mizzi
Supervisor: Prof. Carmel Pulé

2014_A

Introduction
Induction heating/welding is used for many applications. It creates a high frequency magnetic field by using a coil, which induces high frequency eddy currents in any metal object present in its field. Since the heat energy is transferred very efficiently without any contact with the object, it is essential in applications where oxidisation has to be avoided. [1]

Project Objectives
The objective of this project is to investigate the operation of induction welding by designing and constructing an induction welding system capable of welding a pair of spectacles. This involves studying different topologies and range of frequencies.

Project Methodologies
The induction welder can be separated with three main sections which are; the resonant converter which is used to generate the high frequency magnetic field; the supply which uses an H-Bridge to generate a high frequency power source to drive the resonant converter; and the control circuitry used to control the output power and the H-Bridge operation through the gate drivers. The resonant converter makes it possible to change the power flowing by changing the frequency generated by the H-Bridge. The frequency is set by a feedback system, which monitors the input current and sets the frequency in order to match the current set by the user. [2] – [3]
  
Results and Achievements
The system was built by modifying a computer case, to fit the components inside of it. The system was tested gradually by using a DC supply and then with a variac, each time slowly raising the voltage to 240V. The results achieved matched the theory and the simulation results.
The setup was operated at 800W for 15 minutes and no malfunctions occurred. The mild steel bolt used as a load reached a maximum temperature of around 6000C which is enough to melt lead and braze aluminium. For a smaller coil, this amount of power is more than enough for brazing a pair of spectacles since the magnetic field would be concentrated in a smaller area.

References
[1] GH Induction Atmospheres. (2011, June 18). Introduction to Induction Heating [Online]. Available: http://www.gh-ia.com/induction_heating.html 
[2]. Sadiku, M. N. O. (2007). Elements of Electromagnetics (fourth ed). New York (USA)/Oxford (UK): Oxford University Press. p. 386.
[3]. Prof Ned. Mohan. Power Electronics: A First Course, New York, Wiley, October 18, 2011, pp-155-162.

   

Characterisation of Piezoelectric Materials for an Automotive Self-Powered Module

Student: Clare Camenzuli
Supervisor: Dr. Ing. Andrew Sammut
Co-Supervisor: Dr. Ing. Evan Dimech

2014_B

Introduction
Automotive sensors need a lot of harnesses to power them safely and reliably. In the automotive industry, it is desirable to eliminate these harnesses to make vehicles lighter and reduce carbon emissions. With the development of ultralow-power sensor nodes and new energy harvesting technology, it has become more feasible to power sensors from sources like vibrations, thus eliminating the use of harnesses.

Project Objectives
This project, which is being carried out in collaboration with Methode Electronics Malta Ltd., aims at investigating the use of piezoelectric sources to power a module that is to be placed in an automotive environment where vibrations are abundant. Piezoelectric sources are materials that generate a voltage when stress or strain, such as those present in vibrations, are applied to the material.
    
Project Methodologies
In order to establish the feasibility of such a self-powered module, the possibility of powering a system as shown in [1] through the use of piezoelectric materials had to be investigated.
This was done by acquiring several piezoelectric materials after consulting the most important piezoelectric concepts, constants and definitions, and performing several tests on them in order to characterise them. Through characterisation, the best combination of material, shape, thickness and area that yields the most power under certain environmental conditions established by Methode was determined.
Certain frequency, acceleration and temperature inputs were applied to all the piezoelectric materials according to the industrial requirements given by Methode. The inputs were applied through the use of a vibration jig whose frequency and acceleration can be controlled and monitored externally. The temperature was varied and controlled by using a temperature chamber. All equipment was available at Methode.

Results and Achievements
From the characterisation of piezoelectric materials, it was determined that the best combination of piezoelectric materials yielding the most power comes from a piezoelectric disc made of the material PZT-J, having a large thickness and area.
Further tests were then carried out on the six materials that gave the most power during testing. These six materials all exhibited most or all of the parameters that were characterised to yield the most power from a piezoelectric material. From these tests, the materials were shown to be reliable and durable over a prolonged period of 12 hours. The materials were subject to certain random inputs to further investigate their reliability within an automotive environment.
From these tests, a total power of 24.44µW and a current of 49.43µA were generated from one particular material in an environment resembling that found in the automotive sector, thus proving the fact that these materials’ energy can be harvested and used to power various sensor modules in a vibration-based environment. 

References
[1] MIDE. (2013). Piezoelectric Energy Harvesters. [on-line]. Available: http://www.mide.com/pdfs/Volture_Datasheet_001.pdf

 

Design of an Embedded Differential GPS System

Student: Darren Cachia
Supervisor: Dr. Ing. Andrew Sammut

2014_C

Introduction
Global Positioning System (GPS) is a satellite based navigation system that provides a position and velocity solution with accuracy of tens of meters. Differential GPS (DGPS) was developed in order to improve this accuracy. DGPS uses fixed reference stations that transmit corrections to user receivers so as to obtain a more accurate position.

Project Objectives
In this project, the hardware and software for both the reference and user receivers had to be designed to obtain a functional embedded DGPS system. The DGPS algorithms had to be based on a previous dissertation [1] in which these were implemented and tested on MATLAB.

Project Methodologies
The hardware was designed by first choosing the main components which were the microcontroller and GPS module, and then designing the rest of the hardware to ensure compatibility. An important design choice was the selection of a data link between the two boards which was chosen to be internet, with the reference station using Ethernet and the user receiver using GPRS. Additionally Power over Ethernet was used for the reference station power supply. On a software level the main design issues were the implementation of the off-chip interfaces and GPS algorithms. The DGPS algorithms from previous work had to be adopted and optimized for use on the microcontroller chosen.
  
Results and Achievements
The hardware for the reference station and user receiver was designed but due to time constraints only the reference station was assembled. This was successfully tested and full hardware functionality was achieved. Figure 1 shows the finished reference station. In order to test the DGPS system development boards were used as the receiver station.
Figure 2 shows the positional error of the user receiver over an 8-hour period. The blue result shows the position error without the DGPS corrections and the red and green plots show the position error with two different differential correction techniques. An improvement of 82% in the mean offset and 68.7% in the standard deviation were obtained when comparing the blue to the green plot, these were comparable to previous work [1].


References
[1] N.Ebejer, GNSS Local Area Augmentation System, B.Eng. dissertation: University of Malta, Malta, 2012.

   

To study the performance of a magnetic amplifier

Student: Dion Scerri
Supervisor: Prof. Carmel Pulé

2014_D

Introduction
The saturable core reactor is a predecessor of the magnetic amplifier, a technology first reported in the early 1900s that found most of its uses in radio and television applications. Various researchers, mostly in Canada, America and Germany made major contributions to almost half a century in developing different applications incorporating the use of a magnetic amplifier.

Project Objectives
The magnetic amplifier is today considered an old and lost technology since most of its applications and advantages got replaced and superseded by the introduction of semiconductor devices as early as the 1950s. Documentation and information about its working principles and characteristics to different applications is almost inaccessible or very limited considering the period it was published, the WWI WWII era. This is where it found its major electrical applications within the military. This is practically, the main reason of such unpublished restricted material.
The objective of this project is to demonstrate and document the characteristics of the main components of a magnetic amplifier and lay foundations to the possibility of further additional research associated to it.

Project Methodologies
The project was tackled from basics since no direct approach could be found as to how to build or what approach to take to set up even the basic working model of a magnetic amplifier, that is, a device that could amplify a given input signal within its operational parameters.
Principles of magnetism and electromagnetic induction were revised and reported with the objective of developing a deeper insight of the principles of operation of the magnetic amplifier. Design was not possible and thus off-the-shelf equipment had to be purchased and tested one by one, in different circuit configurations and parameters, for the first indicative results. This involved EI type transformers of various (below 100VA) ratings and resistive loads.

Results and Achievements
The magnetic amplifier set up was primarily focused to target a common application that may still be found in industry today. This is the power regulation of highly resistive loads such as stage and theater lighting control.  Of course, this project was based on a much smaller power scale than the one that would be required in practice/ industry.
A working model of a lighting control magnetic amplifier was achieved under different circuit configurations and a respective experimental study was reported.

Design of an Indoor Localisation System

Student: Johann Cassar
Supervisor: Ing. Brian Zammit 

2014_E

Introduction
Indoor Wireless Sensor Networks (WSNs) have attracted a lot of interest due to their vast and flexible applications. A WSN consists of numerous nodes whereby each node is connected to a sensor. Localisation is a very important aspect of such networks since the knowledge of a sensor’s location is critical in order to process information originating from this sensor. Through a minimum network of three fixed location beacons a roaming node can be localised in the region of operation.

Project Objectives
The main targets of this project are to gain knowledge on different techniques of indoor localisation and tracking while also designing and implementing a system capable of exhibiting localisation information over an adequate coverage of a predetermined floor area. The design of specific algorithms in order to handle accuracy issues due to typical inaccurate distance measurements is an essential part of the project.

Project Methodologies
Through a network of three ultrasonic receiver node sensors and an ultrasonic transmitter sensor node on a moving robotic platform, three line-of-sight distances are obtained. The receiver nodes are placed physically at the vertices of an equilateral triangle with 1.5m sides and attached to the ceiling of the room facing downwards. Each receiver sends wirelessly time-of-flight information to the central PC where through data fusion the concept of Trilateration is applied as seen in Figure 1. The intersection of the three circles results in the location of the unknown node. A Least Mean Squares (LMS) algorithmic approach is implemented in the application of this concept to minimize errors. Improvement of accuracy is attained through the use of the Kalman Filter. This recursive algorithm is ideal for this application where noisy data needs to be filtered out.
  
Results and Achievements
Initial experiments were done for fixed position localisation over a region covering a 3m x 3m area and later on the robot was set to follow a series of both straight line and circular trajectories.  Two main Kalman filter implementations were tested out and compared. The Odometric Kalman Filter uses the physical dynamics of the robot as a model and also the wheel encoder readings of the robot to drive the model. The Kinematic Kalman Filter models the robot as a moving target with no use of odometric data. Through the obtained results the Odometric Kalman Filter was deemed to give better results.  The Least Mean Squares Algorithm results for fixed position localisation cover an area of 5cm x 5cm for a set of repeated distance measurements. With a properly calibrated Odometric Kalman Filter this region was reduced to 2cm x 2cm. Depicted in Figure 2 is a trajectory example showing the improvement achieved through the use of the Kalman filter as the robot is moving in a slightly curved trajectory.


Low Noise Amplifier for Square Kilometre Array

Student: Jonathan Camilleri
Supervisor: Dr Ing. Owen Casha  
Co-Supervisor: Dr. Kris Żarb Adami

2014_F

Introduction
The Square Kilometre Array (SKA) is an international effort to build a next generation radio telescope having a total antenna surface area of 1 km2, in the South African and Australian deserts. The objectives of the project are to see deeper into the origins of space, understand dark matter and also find signs of extraterrestrial life amongst others [1].

Project Objectives
The objective of this project was to design a low noise amplifier (LNA) to cater for the needs of the Square Kilometre Array Low frequency range, and provide low power, minimal noise amplification to the low frequency signals picked up by the antenna array. 

Project Methodologies
First, a topology for the LNA was chosen amongst several options. The differential cascode capacitively cross-coupled common gate (CCC-CG) topology was chosen for the core LNA, since it can provide adequate amplificatio nwith limited noise figure and low power demand [2]. The main circuit parameters were mathematically modelled for an indepth understanding of the circuit whilst facilitating the design [2]. A transistor was then chosen to ensure the best results possible [3]. The transistor small-signal model was then derived and values for its parasitics were obtained [4][5]. The first design of the LNA was to be with discrete components on a printed circuit board. However, due to several design limitations, the entire LNA was shifted to AMS 0.35µm CMOS integrated technology. The design was then optimised further until the desired specifications were achieved.
  
Results and Achievements
The final LNA was designed, consisting of the core LNA, followed by a differential output buffer stage, and the current mirrors providing the necessary biasing current for the transistors.
The final simulation results shot that the LNA exhibits a broadband response, from 50 MHz to 300 MHz, whislt demanding a power consumption of just 15.22 mW from a power supply of just 1.5 V. An input reflection loss of -15.78 dB and an average gain of 25.62 dB was achieved with a minimal noise figure of 1.183 dB.
Future work may include further improvement of the amplifier’s forward gain, the actual layout of the integrated circuit and actual manufacturing and testing.

References
[1] “The SKA Project”, https://www.skatelescope.org/project/
[2] W. Zhuo, Z.Li, S. Shekhar, S. Embabi, J. Pineda de Gyvez, D.J. Allstot, E. Sanchez-Sinencio. “A Capacitor Cross-Coupled Common-Gate Low-Noise Amplifier”. IEEE Transactions on Circuits and Systems II, Vol. 52, No. 12, 2005. 
[3] “ATF-54143 - Low Noise Enhancement Mode Pseudomorphic HEMT in a Surface Moune Plastic Package - Datasheet”. Avago Technologies. 2012.
[4] Agilent De-embedding and Embedding S-Parameter Networks Using a Vector Network Analyzer. Application Note 1364-1. Agilent Technologies. 2004.
[5] L. T. Wurtz. “GaAsFET and HEMT Small-Signal Parameter Extraction from Measured S-Parameters”. IEEE Transactions on Instrumentation and Measurement, Vol. 43, No. 4. 1994.

Design of a Pedestrian Tracker in GPS Degraded Environment

Student: Jonathan Camilleri
Supervisor: Dr. Ing. Andrew Sammut 

2014_G

Introduction
The concept of positional tracking has nowadays become a widely accepted and dispersed practice.  These tracking systems usually make use of GPS. Nevertheless, there are circumstances where GPS signals are either completely blocked or severely degraded.  Such locations include indoor facilities, underground areas, and highly urban locations.

Project Objectives
This project aims at identifying the best GPS-alternative technology that can be used in these scenarios.  The necessary equipment, together with the ideal sensor location, are to be identified.  A series of algorithms then need to be developed such that 2-D positional information can be extracted from the acquisitioned sensor output readings.

Project Methodologies
First, several GPS-alternative technologies were researched and compared. Both the advantages and setbacks of each were critically analysed. Inertial tracking systems stood out as the sensing technology which are best suited for pedestrian tracking.
Additional research was carried out in order to compare a number of viable Inertial Measurement Units (IMUs) that emerged as possible candidates. Theoretical comparisons as well as experimental tests were carried out in an attempt to identify the most suitable sensor package, which was found to be the VN-100 Rugged.
By considering the nature of the data available from the chosen IMU, it was determined that the best location to place the inertial sensors was the flat face of the foot.
Gravity cancellation algorithms and axes-mapping transformations were derived and implemented so as to obtain Earth-referenced velocity signals.
Both static period and dynamic period compensation techniques needed to be implemented such that the velocity drift experienced could be dealt with and its effect limited to a minimum.  A number of data fitting methods together with different moving-window sizes were considered. It was concluded that an unsmoothed cubic spline with a moving-window of twenty steps was the compensation technique that gave the best performance.

Results and Achievements
From the results obtained during testing, the system was seen to successfully maintain a 1.4% average error for normal walking conditions and a maximum of 5% average error over time and over a range of paces, gaits, directions and distances.
The developed algorithms were seen to be able to track a wide variety of non-linear 2-D walking paths, travelled by a number of different pedestrians, with satisfactory accuracy.  Therefore, it can be concluded that this IMU system can be used as an effective means of motion tracking in situations where GPS cannot be used.

 

A Multi-Channel Industry Grade pH Probe Tester:
Interface Design and Characterization

Student: Nathalie Cauchi
Supervisor: Ing. Marc Azzopardi
Co-Supervisor: Dr. Ing. Andrew Sammut

2014_H

Introduction
In the chemical industry aqueous solutions are classified into acidic or alkaline solutions depending on their hydrogen ion concentration commonly known as pH.  pH measurements are necessary in the pharmaceutical fields, food sciences, environmental research and water treatment. An electrochemical cell, known as a pH electrode characterized by a very high impedance is used to obtain the pH measurement.  The pH electrode is made up of a glass and reference electrode separated by an electrolyte across which a potential is generated that is directly proportional to the pH.  The high impedance of the electrode requires special interfacing techniques for correct measurements to be obtained. The measurement is temperature dependent necessitating temperature compensation. 

Project Objectives
The aim of the project is to design an electronic hardware system that is capable of measuring the pH potential with high levels of accuracy and precision. The electronic hardware is to be interfaced with a LabVIEW based controller and data handler; and will form part of an Industrial grade pH probe tester for ProMinent Fluid Controls Ltd.  The whole system design is shown in Figure 1.  A series of experiments are to be carried out in order to characterize and evaluate the performance of the designed system.  The final design is a multilayer board with all the necessary noise suppression techniques to obtain a precise measuring instrument.

Project Methodologies
In order to achieve a design having Industrial grade performance a review of the published literature on the interfacing of pH electrodes with analogue circuitry is carried out.  The second phase of the projects involves developing the hardware and software design for the system. This is then verified by the means of developing and testing a prototype board. Error characterization is carried out to evaluate the performance.  The third phase of the project is the design and development of a multi-layer printed circuit board, PCB that is able to measure up to ten pH electrodes concurrently.
The final phase of the project is the complete interfacing of the electronic hardware with the LabVIEW controller. The system is then fully tested and digitally calibrated.  

  
Results and Achievements
In order to achieve a design having Industrial grade performance a review of the published literature on the interfacing of pH electrodes with analogue circuitry is carried out.  The second phase of the projects involves developing the hardware and software design for the system. This is then verified by the means of developing and testing a prototype board. Error characterization is carried out to evaluate the performance.  The third phase of the project is the design and development of a multi-layer printed circuit board, PCB that is able to measure up to ten pH electrodes concurrently.
The final phase of the project is the complete interfacing of the electronic hardware with the LabVIEW controller. The system is then fully tA prototype board consisting of the pH and temperature electrode interfacing circuitry along with the digital circuitry was first built on a dual layer PCB. The analogue circuitry was tested and is seen to achieve a tolerance of 0.015%. The measured analogue signals are converted to their digital equivalent value with 15-bit resolution.  The microcontroller is capable of receiving commands from LabVIEW and perform the required commands and data collection. Data is transmitted to LabVIEW on request.
Once the design was verified, the final multi-channel PCB was created. The PCB is modular and standalone allowing for future expansion. Two redundant channels are present such that in case of a fault, production is not effected. All the necessary shielding and noise suppression techniques have been implemented to ensure that noise ingress is negligible.


A Multi-Channel Industry Grade pH-Probe Tester:
Data Acquisition, Reporting and HMI Considerations

Student: Nathalie Cauchi
Supervisor: Dr. Ing. Andrew Sammut
Co-Supervisor: Ing. Marc Azzopardi

2014_I

Introduction
The notion of ‘sourness’ is quantified by a measurement called the ‘pH’ value [1] which is a measurement that quantifies how acidic or alkaline a solution is. The chemical definition of pH is that it is a measure of the hydrogen ion and hydroxide ion concentration in a solution [2][3]. ProMinent Fluid Controls Ltd is one of the market leaders, producing close to sixty thousand probes annually. As part of the manufacturing process, ProMinent perform thorough testing and certification on each individual probe in order to confirm that this is working according to specification. In order to do this, an efficient, reliable and accurate testing system capable of testing multiple probes at once is required.

Project Objectives
The aim of this dissertation was to design and implement a multi-channel industrial testing station for pH probes that takes into account the process automation of the testing procedure as well as being scalable, reliable and accurate. Moreover, this project forms half of the integrated system that is composed of a hardware project and a software project. The purpose of this project is to develop the software element of the system. This system will be used by testing operators as well as the management personnel. Thus, this project is also aimed at designing an adequate human-machine interface, HMI, through which any user can control the overall process. Eventually, all of the testing results will be stored in a suitably designed database.

Project Methodologies
The first task involved studying the current testing station and production line, to obtain a good understanding of the operating principle, manufacturing and testing processes of pH probes. Through this analysis together with the predefined requirements, a high level design was developed. This was then segmented into modules concerning four main modes of operation. Production Mode is used to test a batch of probes as part of the production process. Engineering Mode was developed to allow the authorised users to visualize the response of the probes on a real-time graph. Calibration Mode was designed to calibrate the system and Setup Mode was designed to adjust the operative system parameters.
  
Results and Achievements
On completion, each unit was tested for functionality and eventually, these units were integrated to obtain the final product. On verifying that the system is doing what is required, a set of evaluation interviews were carried out to identify how much the system is in-line with the company requirements. All candidates adapted easily to the system while providing valid suggestions for improvement for both the functionality and HMI of the system.

References
[1] F. J. Kohlmann, “What is pH and How is it Measured? A Technical Book for Industry,” Hach Company, U.S.A, 2003
[2] E. K. Springer, “pH Measurement Guide,” Hamilton: The Measure of Excellence, Nevada, USA, 2006.
[3] Dr. A. Bier, “Electrochemistry Theory and Practice,” Hach Company, Loveland, Colorado, 2010.

Calendar
 
 
Last Updated: 24 October 2014

Log In back to UoM Homepage