Instrumentation for an Embedded Control Systems Design Course Incorporating the Digilent Electronics Explorer Board

Auburn University’s Electrical and Computer Engineering curricula include a junior-year, laboratory-intensive course on embedded control systems design. Teams of students design, implement, test, evaluate, and report on a microcontroller-based embedded control system that includes both analog and digital electronic elements. The laboratory infrastructure for this course has been updated around the Digilent, Inc. Electronics Explorer Board (EE Board), which is an integrated analog and digital circuit design station. In this paper, the authors present how the EE Board and associated PC-based WaveformsTM software meet the instrumentation needs of the course, and enhance the students’ abilities to effectively conduct and communicate their work. I. Overview of Course Objectives and Student Outcomes The course (ELEC 3050: Embedded Systems Design Lab) has these learning objectives: 1. Use modern engineering tools to design, implement, and test an embedded control system. Elements of the system include a user interface (keypad), analog signal conditioning (sensor), and power electronics (amplifier) to drive an actuator. System elements are integrated via a microcontroller module and its associated software. Students design, create and integrate hardware and software to simultaneously meet several performance specifications. Throughout the engineering processes, students regularly conduct experiments to measure and collect data about their system. Computer aided design tools include: (a) CodeWarrior integrated design environment (IDE) for HCS12 microcontroller system design, (b) Mathworks® MATLABTM and SIMULINKTM for data analysis, data visualization, and system level simulation, and (c) PSpice for electronic circuit simulation. Instruments used in experiments include: (a) the digital logic analyzer, (b) digital sampling oscilloscope, (c) digital volt/ohm meter (DVM), and (d) a waveform generator. 2. Practice life-long learning skills by performing their own research about unfamiliar technologies and engineering science concepts. 3. Periodically report engineering progress via memos, a midterm report, an oral presentation, and a final report. For all of these communication exercises, students perform data analysis and practice methods to concisely summarize data by designing figures and tables. Communication tools include the word processor, spreadsheet, graphics editor, and presentation software. Each student also archives his or her engineering work in a personal engineering notebook that is periodically reviewed. 4. Study how engineering practice relates to a professional code of ethics. Students examine peer-reviewed case studies, identify ethical dilemmas, and propose professional practice solutions. The following student outcomes are assessed during the semester: • Ability to design and realize an electronic system to meet performance constraints. • Ability to create experiments and draw meaningful conclusions from experimental data. • Ability to function as a member of an engineering team. • Awareness of professional ethics in engineering practice. • Ability to communicate effectively in both oral and written forms. The primary assessment objects are the students’ written reports and oral presentations. To effectively communicate their work outcomes, students need laboratory instrumentation that not only supports electronic measurements, but also records data and eases integration with the previously described analysis and reporting tools. In the next section, the authors present a typical course project having multiple options for analog and digital subsystems. These design options create several instrumentation needs that will be further described in Section III. II. A Typical Embedded System Project The system design problem is one that may be encountered in contemporary electrical engineering practice – the control of electric motor speed. A block diagram of the control system is shown in Fig. 1. A matrix keypad is the user interface (Fig. 2a). A dc motor with integrated ac tachometer is provided (Fig. 2b), and is modeled as part of the “plant” in Fig. 1. The plant output is the ac tachometer signal. An electronic circuit conditions the tachometer signal for processing by the digital control algorithm, which is part of the “controller” in the block diagram. The controller block includes the keypad interface logic, as well as conversions of analog-to-digital forms for feedback signals, and digital-to-analog forms for output signals. The controller’s analog output is amplified to drive the electric motor; the power amplifier is part of the “plant” block. In summary, the “plant” input and output are analog signals. Therefore, the signal conditioning input and “controller” output are also analog signals, but these signals are represented and realized in digital form within the microcontroller of the embedded controller. A typical set of system performance specification is as follows: • The desired speed must be selected from a 16-button matrix keypad. • The system must demonstrate nine different non-zero speeds. • The sampling period for the system must be 10 ms +/1 μs. • The measured steady-state speed of the motor must be within 5% of the desired speed, whether the motor is under load or not. The test load will be a single turn of string around the motor pulley, with a 15 g washer weighting the free end. Fig. 1. Block diagram of a typical embedded system project • Extra credit: In addition to the above requirements, the closed-loop step response settling time is one-half the open-loop settling time. Overshoot must be less than 10%. The step change will be from 40% to 50% of motor full speed, without external load. Teams of two students design and implement subsystems to perform several functions: detect and interpret user input from a keypad, measure the motor speed (sensing), execute a digital control algorithm, convert the algorithm’s digital output to an analog form (controller output), and amplify the analog signal to drive the dc motor (power electronics). Many of these functions can be realized by either analog or digital means; these various options are described below. II.A. Options for analog subsystems Analog signal conditioning options exist for both the feedback signal and the controller output. For feedback, the tachometer produces an ac voltage whose amplitude and frequency are proportional to the motor shaft speed. Several options exist to condition the tachometer signal. A frequency-to-voltage converter can produce an analog voltage proportional to the ac frequency. Alternatively, an envelope detector can estimate the ac signal amplitude. Either of these analog signals can be converted to digital form by the analog-to-digital converter that is included on the microcontroller chip. The controller output can be a voltage proportional to the calculated control algorithm value; a digital-to-analog converter can be used to perform the conversion. Alternatively, the digital control value can modulate the duty cycle of a pulse-width-modulation (PWM) signal. Using the microcontroller, students study two on-chip methods to produce an accurate PWM signal: employ the programmable timer, or use the PWM generator. II.B. Options for digital subsystems The tachometer ac signal can also be conditioned to digital logic levels, which opens additional signal processing options for the feedback signal. Students design a comparator circuit that (a) (b) Fig. 2. Embedded system hardware: (a) matrix keypad, (b) dc motor with integrated ac tachometer. produces a square wave synchronized to the tachometer ac signal. The square wave has the same frequency as the tachometer signal, but the amplitude is constrained to logic levels (0-5 V). The period of the square wave can be measured by the timer subsystems within the microcontroller. The user interface is a 16-button matrix keypad. Students design hardware and software to detect and decode button presses. Periodic scanning can be used to monitor button presses, or the keypad can be used to generate interrupts that are serviced by the microcontroller. In summary, the student teams study several analog and digital signal options that can be used to meet requirements of the design problem. Measurement of analog and digital signals is a common thread throughout these studies. The laboratory infrastructure needs to support the students are described in the next section. III. Laboratory Infrastructure Needs A variety of measurement instruments are essential in the course: 1. Volt/ohm meter: Typical tasks include verification of dc power supply voltages, and measurements of component values. 2. Logic analyzer: Tasks include verification of keypad interface signals, and analyzing signals from the timer and counter exercises. Signal bandwidths can be up to 1 MHz. 3. Oscilloscope: Typical signals measured include the ac signal from the tachometer, the PWM signals generated by the microcontroller, and the output of the power amplifier. The bandwidth of these signals is usually below 1 MHz. Signal amplitudes in correctly operating circuits are typically within the range of +/– 30 V. The switching power amplifier has the potential to generate much higher voltage spikes because the motor winding is an inductive load. These voltage spikes are clamped in a properly designed circuit, but the oscilloscope must be sufficiently robust to be employed in improperly designed circuits. 4. Signal generator: Both sinusoidal and square wave signals may be used as test signals throughout the semester. Logic level signals may be used to test timer and counter subsystems. Generated signal bandwidths are usually below 1 MHz. 5. Power supplies: Provide electrical energy to all hardware circuits. Both positive and negative supply voltages are typically used. Power supplies are needed for 5 Vdc logic circuits, comparator and operational amplifier circuits, and dc motor drive electronics (up to 1 A curre