Department of Electrical and Electronic Engineering
BEng (Electronic) (University of Stellenbosch)
MEng (Electronic) (University of Stellenbosch)
PhD (Electronic) (University of Stellenbosch)
Courses given during the past few years
Microwave Electronics 813 (post graduate)
High Frequency Electronics (4th year)
Electronics (1st, 2nd and 3rd years)
Electronic Design (3rd and 4th years)
Computer Aided Network Analysis (post graduate)
Filter Theory (4th year)
Signals and Systems (2nd year)
Microwave Networks (post graduate)
Fields of Interest
Non linear circuit analysis
Satellite communication systems
First a summary of the names/topics and then a list of the same students with the thesis abstract added. Names of students are in order of degree or registration.
Cobus van der Merwe (MEng - degree in 2019)
Title: Magnetron Output Power Controller Used in the Application of Rooibos Tea Sterilisation
The objective of this project was to design a microwave sterilisation system that can be used to sterilise Rooibos tea. Rooibos tea is currently sterilised with a steam treatment, but this a˙ects the tea's colour and ˛avour. Mi-crowave irradiation was investigated as an alternative form of sterilisation.
Often microwave sterilisation tests are performed without the proper imple-mentation of microwave power control. Accurate microwave output power control was achieved by designing a custom magnetron power supply that uses a TRIAC to switch the mains voltage every half-cycle. A control system, that measures the applied and the re˛ected microwave power, was implemented to ensure a constant microwave power is delivered to the load. Calibrated mea-surements were taken of the system input power, microwave output power and applicator temperature. Adjustments were made to the microwave applicator to ensure that the dry tea did not burn.
Microbiological tests were performed on the treated samples for Salmonella, E.Coli/Coliforms and aerobic plate counts (TVC's). The South African stan-dard for microbial safety of Rooibos tea was used as reference. The microwave sterilisation system worked well for Salmonella and TVC's, but E.Coli/Coliforms proved more resilient. A de˝nite batching problem was identi˝ed with the sterilisation of raw farm products that contain di˙erent concentrations of mi-croorganisms. This study provides a good foundation for any further research on the microwave sterilisation of Rooibos tea, as well as a range of hardware solutions for various other microwave power control applications.
Anneke Stofberg (PhD.Eng - degree in 2018)
Title: Phase Tracking Electronically Variable Attenuators with Receiver Protection
This dissertation presents the development of a set of optimal phase tracking electronically variable attenuators. Secondly, a compact high power PIN diode limiter is developed and its minimum attainable resistance is extracted through high power measurements.
Close range reflections cause the receiver of a multi-channel digital beamforming radar to saturate. Controlled attenuation over time, implemented with electronically variable attenuators, is used to prevent receiver saturation (sensitivity time control). An electronically variable attenuator is placed in front of the first low noise amplifier in each channel; its insertion loss directly adds to the receiver’s noise figure.
A multi-channel digital beamforming radar receiver requires good phase tracking between its receiver channels to minimise direction of arrival estimation errors. The set of electronically variable attenuators used for sensitivity time control need to track in phase over the control range. In this dissertation, sensitivity analysis is used to identify a set of optimal phase tracking electronically variable attenuators. A root sum square error measure is derived from the multiple output sensitivities of an electronic network. The error measure gives the expected RMS phase error within a set of networks. Applying the error measure to several electronically variable attenuators over the control range, the cascaded parallel quarter-wave attenuator is identified as having optimal phase tracking within a set of attenuators over control range. The cascade parallel quarter-wave attenuator is developed further and optimised through the application of sensitivity analysis. The final attenuator has excellent attenuation flatness, attenuation range, phase tracking and a simple biasing scheme.
A multi-channel digital beamforming radar receiver also has to be protected against large signals. These large potentially damaging signals are either due to the radar’s own transmitted signal, or from other radars transmitting large amounts of power in the same frequency band. A receiver protector (e.g. a limiter) typically supplies this function.
In a multi-channel digital beamforming radar, a compact circuit based high power limiter has many advantages in terms of space and cost when it is compared to a waveguide limiter. The compact high power limiter developed in this dissertation consists of PIN diodes implemented on a multi-layer printed circuit board. The circuit is referred to as an active PIN-Schottky limiter. The maximum power handling capability of the active PIN-Schottky limiter is determined by the PIN diode at the limiter’s input. The minimum attainable resistance is not given by the manufacturers, so that the diode’s minimum attainable resistance can not be found from the datasheet information. It is difficult to estimate how much power is dissipated in the diode when a large signal is incident. Through a temperature controlled measurement, the PIN diode’s voltage decrease as a function of junction temperature increase is measured. By fitting the PIN diode’s measured and simulated junction temperature increase, it is possible to extract the resistance of the diode when large forward bias is applied. Once the resistance is known, the power dissipated in the PIN diode can be calculated for different operating conditions.
Francois Schonken (PhD.Eng - degree in 2017)
Title: Modelling and mitigation of specular multipath interference in a dual-frequency phase-comparison FMCW system
Digital signal processing technology has improved greatly over the last two decades. Increased processing power, cheaper memory and higher sampling rates have enabled the application of Frequency-Modulated Continuous Wave (FMCW) radar to a myriad of new areas. FMCW offers a number of advantages, such as continuous coverage and low peak output power, making it an attractive technology for industrial and automotive applications.
Expansion into new application environments and the use of new signal processing algorithms have created a need for new multipath interference models. This study aims to fulfil that need through rigorous mathematical modelling of both the physical multipath environment and the two-dimensional Fast Fourier Transform (FFT) signal processing method, in the context of a Dual-Frequency Phase- Comparison FMCW radar sensor. It will be shown that specular reflections can have a profound effect on the amplitude and the phase of an FMCW radar’s base-band received signals. The multipath phase error interacts with the signal processing method, resulting in new and interesting effects. Furthermore, new mitigation methods will be proposed and critically evaluated by means of simulated and real-world measurements.
Adelaide Oberholzer (MSc.Eng - degree in 2017)
Title: Development of a continuous flow sterilisation system using microwaves
The focus of this project is to design a continuous flow system to be used for the sterilisation of biological growth media by using only microwaves. The output power of the magnetron has to be controlled in order to control the temperatures to which the media will be heated.
A study was done on different methods of microwave power control and it was found that anode current control is the most suited method. The developed method controls the anode current of the magnetron while keeping the cathode at its constant voltage. Two high voltage transformers used in domestic microwave ovens were connected to the same magnetron. This allowed the anode transformer’s supply voltage to be changed by using a triac control circuit. This changes the anode voltage, and so the current, while keeping the cathode constantly heated. Methods of relating the microwave power to a safely measurable voltage or current were also investigated. In the final design, a Hall effect current sensor was used to measure the current on the primary side of the anode transformer and this current was related to the output power of the magnetron.
Temperature sensors to be used for the inlet and exit fluid temperatures were also investigated and PT100 resistance temperature probes were used in the final design. A small peristaltic pump was used to pump the fluid through a PTFE coil inside the microwave cavity. PTFE was selected because it is not susceptible to microwaves and it is chemically non-reactive.
A study was done on the relationship between microwave power and the maximum temperatures reached by the fluid for different flow rates. This was used to develop a control system which was implemented using Matlab and two Arduinos as microprocessors.
The final system was moved to the Biochemistry Department to commence sterilisation tests. Because there are different types of microorganisms, it is important to test for different types as they may react differently to external stimuli.
For this study, both gram negative and gram positive bacteria were tested as well as yeast. The specific gram positive bacteria used was Micrococcus Luteus, strain: NCTC 8340; for gram negative bacteria Escherichia Coli, strain: DH5α containing a pGKCherry plasmid, and for yeast Saccharomyces Cerevisiae was used.
All tests were done in duplicate to confirm the results. The target exit temperature was 90 °C and the flow rate was 3.5 l/h. An initial test was done on M. Luteus with a concentration of 103 cells per ml. This batch was completely sterile and it was decided to increase the concentration to 106 cells per ml for all of the microorganisms tested. All three of these tests achieved sterility.
It was then decided to reduce the temperature to see if this had an effect on the results. These tests were done with M. Luteus at a concentration of 106 cells per ml. The temperatures selected were 70 °C, 50 °C and 37 °C. Only the 70 °C batch achieved sterility.
It is concluded from this project that continuous flow microwave sterilisation is possible and very effective.
Jason Bonarius (MSc.Eng - degree in 2016)
Title: Adaptive Noise and Interference Cancellation for Mobile Communications
In an arena of electronic warfare measures, to inhibit enemy operations as well as provide safety for allied forces from radio triggered explosive devices, interference and noise coupling degrade allied communications severely. In an attempt to allow for communications to occur uninterrupted while maintaining electronic warfare capabilities, a noise and interference cancellation system is to be investigated and developed, whereby wave superposition principles are to used to adaptively provide noise and interference cancellation.
Bernard Meyer (MSc.Eng - degree in 2015)
Title: Comparing baseband and intermediate frequency FMCW receivers
Frequency modulated continuous wave (FMCW) radars have a large range of applications in the civilian and military sectors, including aircraft navigation (radio altimeters), sub-surface examination, weather monitoring, and many more remote sensing applications. In general receivers have limited dynamic range. Therefore, FMCW radar receivers can saturate due to large reflections from close range targets or poor isolation between the radar system’s transmit and receive paths.
It is hypothesised that certain radar receiver topologies are more suited to coping with saturation. Therefore, the intermediate frequency (IF) and the commonly-used baseband (or zero-IF) FMCW radar receiver topologies are compared in terms of component non-linearity. Receiver saturation generates unwanted signals that can result in the detection of false targets. The study proposes that an IF receiver, under specific operating con-ditions, can suppress unwanted harmonics and intermodulation products (IMPs) much more e˙ectively than an equivalent baseband receiver.
A detailed IF receiver design is presented that includes a discussion on image rejection, signal quantization, anti-aliasing, and sensitivity frequency control. It is found that the IF receiver is immune to the detection of false targets caused by poor isolation between the demodulator’s local oscillator and radio frequency ports. A noise analysis shows that the IF receiver’s noise performance is potentially greater than that of an equivalent baseband receiver due to pink noise that has a 1/f power spectral density, which decreases with an increase in operating frequency.
The non-linear behaviour of the IF receiver is investigated by analysing the circuit components that have the highest risk of saturation in the receiver chain. The magnitude of harmonics and IMPs are determined using an algebraic approximation of a non-linear model presented in the study. The model is used to predict under which conditions false targets will be detected. It is found that the zero-IF receiver is susceptible to distortions that are suppressible by an IF receiver. Though the application has limitations with regards to the received signal characteristics, the IF receiver has better or equivalent performance when compared to a baseband receiver.
Kobus Myburgh (MSc.Eng - degree in 2015)
Title: The instantaneous frequency measurement (IFM) receiver
The instantaneous frequency measurement (IFM) receiver is normally a subunit used in electronic warfare receivers. Its purpose is to measure the frequency of pulse and continuous wave signals over a very wide bandwidth and power dynamic range. It should be noted that the band of interest can range from 500 MHz to 40 GHz. This is an extremely wide bandwidth. These receivers have a very high resolution (typically 1 MHz) over this very wide bandwidth and the error of these receivers is typically a few megahertz. The IFM receiver should reveal the presence of signals. It is not generally known in advance whether signals are present, even if they are, nor their frequency and other characteristics.
An inherent disadvantage of all traditional IFM receivers is that overlapping pulses within the same power range corrupt the discriminator outputs to the extent that frequency measurements appear random. This is particularly bothersome, as high duty cycle and continuous wave emitters reduce the frequency measurements to little value.
This thesis entails the design and development of an IFM receiver (co-pulse IFM receiver) that can distinguish overlapping pulses within the same power range.
Different acquisition receivers were researched to make sure that the IFM receiver is still the best option, taking bandwidth, frequency resolution/accuracy and throughput time (latency) into consideration. The traditional IFM receiver was researched and simulated in MATLAB.
While researching the traditional IFM receiver, literature was discovered that identified Prony’s method as a solution to distinguish overlapping pulses (simultaneous signals). Literature proposing Prony’s method that had been submitted from as early as 1989 was found. However, no evidence of the use of the method in current commercial IFM receivers was found, indicating that the implementation of Prony’s method is the difficult part. Prony’s method was simulated in MATLAB, then implemented, simulated and tested in hardware.
The results of the hardware implementation were documented and demonstrated a solution to distinguish overlapping pulses.
Johann Botha (MSc.Eng - degree in 2014)
Title: Design of an RF Ion Thruster
Recent years have seen a decline in the rate of space exploration due to the ineÿciency of chemical rockets. Therefore alternative fuel eÿcient propulsion methods are being sought to enable cost e˙ective deep space exploration. The high fuel eÿciency of electric thrusters enable a spacecraft to travel further, faster and cheaper than any other propul-sion technology available. Thus electric propulsion has become the propulsion of choice for scientists and engineers. A typically electric thruster contains some sort of electrode to ionise the propellant. Although this is feasible for short space missions, it becomes im-practical for more ambitious space missions as electrodes erode over time. The alternative is to ionise the propellant using electromagnetic fields, which eliminates lifespan issues associated with electrode based thrusters.
In order to examine methods of improving the lifespan and performance of electric thrusters, this thesis aimed to study the method of microwave discharge ionisation for an electric thruster. This includes the design of an RF Ion Thruster with extraction and acceleration grids to generate thrust. A 600 W 2.45 GHz magnetron (obtained from a conventional microwave oven), coupled to circular TM010 resonant cavity, was used to ionise neutral argon gas. The process of electron cyclotron resonance (ECR) was used to ensure the eÿcient ionisation of a high density plasma. The thrust was achieved with a three-grid system biased at high voltages to accelerate positively charged argon ions to high exhaust velocities.
Results yielded the success of the designed electromagnetic based thruster, measuring approximatively 1.78 mN of thrust with a specific impulse of Isp = 3786 seconds. The ECR process produced a high plasma density with a plasma absorption rate of approx-imately 77% of the total input microwave power. The final results obtained were found to match the predicted results extremely well and resembled results found in literature. This demonstrates the eÿciency of the RF ion thruster that was designed in this project and the future use in space exploration activities. However, future research needs to be undertaken on a controlled feedback system that will ensure optimal operating conditions for maximum performance. In addition, the method of grid-less acceleration needs to be studied to achieve maximum thrust and specific impulse.
Anneke Stofberg (MSc.Eng - degree in 2014)
Title: IQ Reflected Power Canceller for an FMCW radar
Large close range environmental reflections or poor isolation between the transmit and receive paths of an FMCW radar can overload the receiver. The In phase and Quadrature phase (IQ) Reflected Power Canceller (RPC) provides a solution to the problem by cancelling any close range reflections. In this study a procedure to optimise the design of an RPC is developed and the performance limits of a practical RPC is investigated in depth.
There are four focus areas in the evaluation and design of the IQ Reflected Power Canceller. First, an analysis was performed on a theoretical IQ Reflected Power Can-celler, which provided insight into how the system functioned and made it possible to identify practical application issues that would arise during the design.
The next focus area was the IQ Reflected Power Canceller’s dynamic range. Equations, based on the power and noise characteristics of each component in the canceller, were derived. From these equations, a system, with an optimised dynamic range, could be developed.
Next, the IQ Reflected Power Canceller’s feedback loop stability was investigated. The canceller is an active negative feedback control system but, in order to obtain the negative feedback, the feedback signal has to be phase shifted by 180 degrees to the phase of the input signal. An analysis of the canceller’s RF phase contribution resulted in an equation that can be used to manage the nett RF phase in the feedback loop.
The evaluation model of the IQ Reflected Power Canceller produced favourable results. The tests performed on the system included measuring the level of cancella-tion that can be achieved, whether the dynamic range corresponds to the predicted values and the amount of RF phase error that can be introduced in the feedback path while maintaining a stable system. The IQ Reflected Power Canceller was found to perform well in the evaluation. It provided a cancellation of more than 45 dB for close range reflections and the canceller remained stable across a wide range of RF centre frequencies (1 GHz). This means that the FMCW radar’s frequency modulation bandwidth will not be limited because of the IQ Reflected Power Canceller. The evaluation clearly showed that the modulator in the feedback loop is the critical element that determines the dynamic range of the radar with an RPC.
Erik-Jan Moes (MSc.Eng - degree in 2008)
Title: A Low Noise PLL-Based Frequency Synthesiser For X-Band Radar
This thesis discusses the design, development and measured results of a phase-locked loop based frequency synthesiser for X-band Doppler radar. The objective is to obtain phase noise comparable or lower than that typically achieved with direct analogue frequency synthesis techniques. To meet this objective, a theoretical study of the noise contributions of individual components of the synthesiser and their effect on the total phase noise within and outside the loop bandwidth of the PLL is performed. The effect of different phase margins on the closed-loop frequency response of the PLL, and hence the total phase noise, is investigated. Based on the results, an optimal phase-frequency detector reference frequency, loop bandwidth, adequate phase margin, and suitable components are chosen for optimal phase noise performance. The total phase noise at the output of the synthesiser is calculated and it is shown that the phase noise specification can be met.
A significant part of this thesis is devoted to the design, modelling and characterisation of a frequency multiplier, as well as to a combline and interdigital bandpass filter required for the frequency synthesiser. In the first case, a piecewise linear circuit model is used to model the behaviour of the nonlinear multiplier circuit. Fourier theory is used to calculate the large-signal driving point input and output impedances of the nonlinear circuit, enabling the computation of the circuit elements for the input and output matching networks. The measured response of the frequency multiplier under various different operating conditions is presented and discussed. The design of the microwave bandpass filters is based on the theory of coupling and external quality factors. To aid in the verification and optimisation of the design, a software simulation tool is used. The presented S-parameter measurements of the filters show how well the theory matches with what is obtained in practice.
The measured spectral and phase noise response of various components comprising the synthesiser, are discussed. These measurements provide insight into the response of individual components under different operating conditions and show the behaviour of important subsystems of the synthesiser. The thesis culminates in the presentation of the measured phase noise of the complete synthesiser. It is shown how well the measured phase noise correlates with the calculated phase noise. In addition, the measured spectral content and transient behaviour of the synthesiser are investigated and discussed. High power spurious components at some output frequencies are identified. and reduced. The feasibility of using the developed prototype phase-locked loop based frequency synthesiser for coherent X-band Doppler radar is discussed and demonstrated.
David Smith (MSc.Eng - degree in 2006)
Title: Satellite data transmission by menas of a multi-channel system
The aim of this thesis is to find an affordable and upradeable manner of increasing the data rate of current satellite systems. The compression, modulation and amplification techniques and telecommunication regulations restrict the data rate, which are bypassed by designing a parallel channel configuration. In order to test this solution a system is developed that is based on the existing standards of DVB-S and MPEG-2. The combination of these standards protect the data from interference, package the data for ease of storage, modulate the data to more effectively radiate the signal and shape the spectrum to adhere to telecommunication regulation. The spacing between the channels is reduced and the transmitted signal is amplified to inject interference into the system. The effect of this interferece on the receiver is shown by transmitting image data and comparing the received image with the original.
Wynand de Wit (MSc.Eng - degree in 2006)
Title: Linearization of an FM-CW 94 GHz millimeter wave radar
The topic of millimeter wave radar systems is introduced. These radars are used in a wide range of applications in both the aviation and automotive field due to the resolution advantages which MMV systems have above their counterparts. MMV components are studied and characterised to improve on existing linearisation technique. Different possible linearisation techniques are discussed and compared to choose the best possible technique for this application. This technique was developed and implemented in the existing system.
Pieter Winter (MSc.Eng - degree in 2005)
Title: Position Control of a Mobile Robot
Position calculation of mobile objects has challenged engineers and designers for years and is still continuing to do so. There are many solutions available today. Probably the best known and most widely used outdoor system today is the Global Positioning System (GPS). There are very little systems available for indoor use. An absolute positioning system was developed for this thesis. It uses a combination of ultrasonic and Radio Frequency (RF) communications to calculate a position fix in doors. Radar techniques were used to ensure robustness and reliability even in noisy environments. A small mobile robot was designed and built to test and illustrate the use of the system.
Thinus Neethling (MSc.Eng - degree in 2004)
Title: A Broadband Microwave Limiting Amplifier
Limiting amplifiers are employed in electronic warfare (EW) systems requiring a high measure of amplitude control. These EW systems employ sensitive signal processing components that are unable to accept the full dynamic range of input signals the system must face. The limiting amplifier, however, offers the unique capability of reducing the received signal spectrum to a suitable dynamic range.
Typical application of the limiting amplifier is in the instantaneous frequency measurement receiver (IFM) where the limiting amplifier allows the receiver to accurately measure pulsed signals over a wide input dynamic range. The aim of this study is the design and analysis of a broadband limiting amplifier. Focus is placed on the design of a so-called “backbone” limiting amplifier (BLA) which forms an integral part of a proposed modular design approach for realizing a design with improved input dynamic range. The designed BLA is discussed in this thesis while insight is given as to the intricacies associated with its mechanism of operation. Over its 45 dB (-40 to 5 dBm) input dynamic range, the designed 2-18 GHz limiting amplifier offers a typical saturated output power of 7.5 dBm while harmonic suppression of better than 8.6 dBc is achieved.
The BLA design was based on an existing limiting amplifier design, the so-called “baseline” limiting amplifier, employing alternating amplifiers and attenuators. Evaluation of the “baseline” limiting amplifier design allowed for formulation of a design hypothesis for realizing the BLA design. Physical measurements on the BLA were then used to scrutinize and validate the formulated design hypothesis.
The requirements for realizing the BLA design were the establishment of a thorough radio frequency (RF) amplifier design capability, and understanding of the nonlinear phenomena associated with the RF amplifier and the utilization and control thereof within the limiting amplifier. Different RF amplifier designs that were done are discussed in this thesis, while it is shown how they were used to further investigate important design considerations for application in the BLA design. The computer-aided design packages namely MultiMatch and Microwave Office (MWO) were successfully used in realizing the desired broadband RF amplifier designs and the eventual BLA design.
Quintin Engelbrecht (MSc.Eng - degree in 2004)
Title: The Design of an FMCW Proximiry Radar
The aim of this thesis is the analysis and design of a broadband PIN diode switch
. Pin diode switches are gaining popularity in RF and microwave applications today. This is due to their excellent switching and power handling capabilities, reliability, low cost and small size. An analysis and design procedure for broadband PIN diode switch design, using the series, shunt and compound topologies respectively, is presented.
In order to do a proper switch design, accurate practical models for the components are required. Parameter extraction therefore formed an important consideration for this study. A parameter extraction procedure is presented, which enables the designer to very accurately extract the required models for the components in the environment they operate in. The designer can then do a proper design to ensure that the switch response when measured, closely corresponds to that simulated.
A compound configuration switch was designed, built and measured to confirm the validity of the design procedure. The results illustrate that if the extracted models of the components are integrated into the design, the measured and simulated response compare remarkably well.
Valpré Kellerman (MSc.Eng - degree in 2004)
Title: The Design of an FMCW Proximiry Radar
The preliminary development of a FM radio receiver module is discussed. An existing narrowband system operating between 48MHz and 50MHz will be replaced. Digital components were investigated, compared and used with analogue techniques to build a more flexible two-way radio receiver system. A direct digital Synthesizer was considered as a replacement for the current synthesized phased lock loop oscillator and much attention was given to the local oscillator and mixer design, characteristics and measurement procedures.
A detailed study of receiver systems was undertaken to determine the specifications needed for every receiver component to achieve satisfactory receiver performance in the end. Receiver characteristics as well as receiver measurement procedures are defined. A software tool was developed to aid the design process, establishing computationally whether the receiver specifications are met prior to the final design.
The complete design process, from fundamental specifications through to the developed final receiver module is discussed. A modular design approach was used to guarantee easy manufacturing, substitution and testing, This approach comprises the break-down of the receiver into well defined components that are each matched to 50Ω. The separate components of the system were designed, measured and characterized to make it possible to replace only a single component instead of the entire system when a part becomes redundant.
Robert Kellerman (MSc.Eng - degree in 2004)
Title: The Design of an FMCW Proximiry Radar
The development of a small, cost effective short range FM-CW radar as well as the design and optimization of a double helix antenna is discussed. The FM-CW radar operates at S-band (2450-2550 MHz) and is limited at a maximum transmitted power of 5 dBm. The system is required to detect reflections from objects up to a 30m range. This was made possible largely by the effective high gain antenna design and improved system isolation between the transmitted and received signals.
The complete design process from fundamental requirements, through various simulations and system analysis to the final electronic circuit is discussed. The design was also built and measured in the laboratory and initial field measurements were taken. The following aspects were considered:
- Relation between the beat frequency and various FM-CW radar system parameters based on the theory, preliminary measurements and simulations.
- Radar component assessment and design.
- Design of a low cost Lange coupler that isolates the system’s transmit and receive signals, which operates as a 90o hybrid and is used to feed a duel-fed circularly polarized patch antenna.
- A circularly polarized patch antenna and double helix antenna were designed, built and measured in an anechoic chamber. The antennas were implemented in the final radar system and compared on the basis of their contribution to the system’s performance.
- Integrating the final radar system on a circular 60mm diameter PCB with a double helix antenna mounted on top.
The final radar system is able to measure range up to 40m and meets all the design requirements. The signal processing of the measured beat frequency, however, can be taken further.
Jacques Malan (MSc.Eng - degree in 2004)
Title: Microwave Data Transmission by means of an Optimal Bandwidth Multichannel QPSK System
The increase in the complexity and functionality of new satellites leads to a growing demand for a digital communication system operating at a much higher bit rate than before. The main purpose of this thesis is to address precisely this problem.
The theory as well as a practical high speed system are custom-developed. The goal of this thesis is not to reinvent the wheel, but to use off-the-shelf products as far as possible. This system demonstrates that a very high data rate is not out of reach for a developer with limited funds.
The secondary goal is to understand the operation of this communication system. The emphasis is on the inaccuracies in the system. A design tool s created to act as a vehicle for understanding the influence of a single variable on the performance of the entire system.
Johann Gerber (MSc.Eng - degree in 2002)
Title: A Load Independant Microwave Source
In the past few years a need for consulting on microwave heating has surfaced. Since most specialised microwave components are imported, this adds a hefty price tag in developing the infrastructure for such consultation. This thesis looks at the design of low cost alternatives for use in the laboratory. A microwave source output with variable output power is often required for consultation. The calculation of product costs may also be simplified if the available microwave power is constant despite load changes. The magnetron as a load was first investigated and a high voltage SMPS was then designed to control the output of the magnetron according to operator / consulting requirements. To ensure adequate feedback of system output and load matching, a dual directional coupler has been designed and implemented. Since the amount of microwave power required has a direct impact on the initial capital expenditure, costs need to be kept low by optimising the system. An impedance matching unit was designed to match the magnetron with the load, saving energy and reducing system costs. Through automation, the system will strive for the optimum load condition without operator intervention.
Marius van der Merwe (MSc.Eng - degree in 2002)
Title: Harmonic Mixer Analysis and Design
Harmonic mixers are capable of extended frequency operation by mixing with a harmonic of the LO (local oscillator) signal, eliminating the need for a high frequency, high power LO. Their output spectra also have certain characteristics that make them ideal for a variety of applications. The operation of the harmonic mixer is investigated and the mixer is analyzed using an extension of the classic mixer theory. The synthesis of harmonic mixers is also investigated and a design procedure is proposed for the design and realization of a variety of harmonic mixers. This design procedure is evaluated with the design and realization of two harmonic mixers, one in X-band and the other in S-band. Measurements suggest that the procedure is successful for the specific applications.
Willem Koen (MSc.Eng - degree in 2001)
Title: Design and Characterization of Diode Detectors
The use of diode detectors for power measurement and AM demodulation is not a new subject. The design and characterization of optimum diode detectors can, however, always be improved. In this thesis a step-by-step design procedure is presented for the design of various diode detectors. The design itself is not a complex task, but the success of the results depends on the characterization of both the detector and the diode. A method to extract optimized diode models from DC and small signal measurements are presented. Analyzing the detector circuit with the harmonic balance technique does the characterization of the detector. These results can be used to create an error-correcting algorithm to compensate for the non-linear behaviour of diode detectors and to improve the dynamic range and sensitivity of the design.
Coenrad Fourie (MSc.Eng - degree in 2001) (co-supervisor with Prof. Willem Perold)
Title: A 10 GHz ADC in Hybrid SUperconducting Logic
Rapid Single Flux Quantum (RSFQ) logic cells are discussed, and new cells developed. The expected yield of every cell is computed through a Monte Carlo analysis, and where necessary these cells are optimized for use in a complex system. A mathematical study of the Josephson junction and SQUIDs (Superconducting Quantum Interference devices) as switching elements precede a discussion on the operation of RSFQ and COSL (Complementary Output Switching Logic.) These logic families are implemented in low temperature niobium technology, and require liquid helium cooling. A 10 GHz over-sampling delta modulating analogue-to-digital converter is then designed and constructed using RSFQ and COSL building blocks in a hybrid configuration. The design emphasis is on devising ways to test the operation of RSFQ with limited equipment. Yield analysis procedures on the complex system are discussed, followed by a detailed discussion on the circuit layout and layout problems. Software routines are developed to calculate the required dimensions of layout structures.
Nico Geldenhuys (MSc.Eng - degree in 2001)
Title: Microwave Frequency Discriminators
The topic of microwave frequency discriminators is introduced. Various applications and implementation techniques of these devices are given and discussed. This thesis may serve as reference foundation for future research on this topic. Discriminators are developed at both S- and X-band, with high linearity and good sensitivity. The systems used to implement these devices are developed as well. The S-band discriminator is used in a direct FM demodulation system, which works well.
Werner Steyn (M. Eng. - degree in 1998) (co-supervisor with Prof. Petrie Meyer)
Title: X-Band Receiver Front End Optimised for Hight Tmeperature Superconductors
The recent international progress made in the fabrication of HTS components and systems for microwave applications created the need to fabricate and verify the performance of these components locally. An investigation into the design and manufacture of an X-band receiver front end with HTS technology was launched with this report forming stage two of a long-term research project.
In this report a room temperature X-band receiver front end is presented. All the components of the receiver are designed and manufactured with transition to a future cryogenic X-band receiver in mind. While this will result in a more meaningful comparison between the two systems, it also gives invaluable insight into the design procedures and problems of the eventual HTS system.
Ingolf Meier (PhD - started 1997)
Title: Interference Patterns Synthesis for Microwave Dielectric Heating
One of the most difficult problems in microwave dielectric heating is the generation and control of field and heating patterns. A technique allowing the synthesis of different, pre-determinable heating patterns by interference modulation is proposed.
The proposed concept may be described by the term 'interference modulation'. Interference modulation is a technique that enables particular patterns, called features, to be obtained by signals from several sources interfering with each other. Known features may be selected by modulating the relative phases of the signals. Weights are assigned to these features, which may be combined over time to form a new heating pattern. Phase changes may then be used to switch to specific, known features, with weights that will determine the contribution of each feature to the desired overall pattern.
In the practical implementation described, magnetron tubes are the sources. Each of these narrow-band high-power sources was injection locked to a low power control signal. The control signals are derived from a reference source and their phase is set to select a corresponding feature.
Calculation and measurement showed that reliable locking occurs with a control signal power of at least 3% of the magnetron's emitted power.
Measurements of patterns were carried out with materials formed into sheets and blocks. Some were chemically prepared to reveal the overall heating pattern. The observed patters, simulations and field measurements concur, thus validating the concept and operation of the proposed topology.
Diana Viljoen (M. Sc. Degree in 1997) (Co-supervisor with MM Muller)
Title: Temperature Control for Food Research
The aims of this study were to test a conventional electric oven with an electronic temperature controller [ETC], to develop and test a computerized temperature control system [CTCS] in two similar ovens [oven A and oven B] and to determine the effect of heating on time/ temperature measurements and product quality characteristics in cakes and biscuits as determined by instrumental measurements and analytical sensory analysis. With the CTCS the aim was also to establish the significance of temperature control to within 1° C the effect of the presence of additional moisture on the environment and the product in the oven cavity, as well as the efficiency of two temperature control systems [four thermocouples versus eight thermocouples].
The ETC is less consistent than the CTCS within replications with regard to average oven temperature, moisture-loss and Hunter L-values.
The CTCS with eight thermocouples provides a more valid indication of the environment in the oven cavity and the effect of heat on products. No significant differences were fund in average oven temperature between oven A and oven B, as well as within treatment replications. However, there is a significant difference between oven A and oven B with regard to the dissipation of moisture. This is also shown in results with regard to moisture-loss and colour measurements. Cakes baked in oven B show significantly more moisture-loss [approximately 1 g] than those baked in oven A. The two ovens dissipate moisture in different ways and average oven temperature is significantly correlated with moisture-loss, which is possibly the reason for this phenomenon. Hunter L-values are also significantly correlated with average oven temperature and moisture-loss. No significant differences were fund between oven A and oven B with regard to height of cakes and texture of cakes and biscuits. In most cases, regression is linear with regard to selected time/temperature measurements and product quality characteristics as oven temperature increases by 1° in oven A and oven B. These results illustrates the efficiency of the CTCS.
The measurements conducted to determine product quality characteristics are valid with reference to their applicability in this study. The effect of temperature on the colour and moisture-loss of products is effectively illustrated by these measurements. However, height and texture measurements do not provide a clear indication of the effect of the small differences in temperature, i.e. 1°C, between treatments.
The CTCS significantly improves the reproducibility of the quality characteristics of baked products and therefore provides more consistent and repeatable results which, in turn, improves the reliability of research findings.
Pieter K Siebrits (M. Eng. - degree in 1997)
Title: Adaptive Control for Microwave Heating
The developments over the past few years in the microwave and dielectric heating field has led to this research project. Research identified the need for developing a controller that could accurately control the temperature of a wide variety of materials. The electronic and chemical engineering departments started working in close collaboration in this regard.
A robust adaptive controller was developed and compared to a classic Proportional-plus-Integral controller commonly found in industry. Both pulsed and continuous methods were used for varying microwave power. On evaluating the adaptive controller excellent experimental results were obtained.
To control the power and calculate the power efficiency, accurate calibration techniques must be implemented. Therefore different calibration techniques such as the through-reflect-line- approach and a predefined calibration kit were investigated.
Francois Visser (M. Eng. - degree in 1997)
Title: A Two-Element UHF Adaptive Receiver Array.
This document discusses the design, assembly and test results of a two-element adaptive receiver array. The system is intended for use as a ground station to link with the SUNSAT micro satellite once it is in orbit. A UHF narrowband receiver that was originally developed for SUNSAT's UHF system is used to demodulate the received signal.
The array optimizes received signal strength by adjusting the phase shift and attenuation in each of the antenna element channels. A received signal strength indicator [RSSI] output is provided by the receiver and is digitally sampled by a computer to be used as feedback signal.
A number of building blocks have been developed for the adaptive system. The phase shifters and attenuators are digitally controlled by computer via a central bus. A hybrid power combiner adds the signal power in the two channels together and feeds it to the UHF receiver to be demodulated.
Measurements that were performed on the system show that it has promising potential as a signal tracker. When a single transmitter is placed at a certain angle to the array, the received signal strength has a well-defined peak when the array is steered in that direction. Performance of the array also agrees well with the theory. A method was devised to determine mismatches between channels and to account for their effects with post-processing of the measured data.
A simple search algorithm was written to automatically adjust the array parameters for maximum received signal strength. Its performance is satisfactory.
Since each channel is designed to be identical, the modules may be duplicated exactly if the array is to be expanded to incorporate more elements.