Customer Solutions

Improving Physical Understanding through Laboratory Automation

Author(s):

Steven Orth, University of Wisconsin-Madison

Industry:

University/Education

Product:

Data Acquisition, GPIB & Instrument Control, LabVIEW

The Challenge:

Automating an undergraduate laboratory on electrical and electromechanical power conversion.

The Solution:

Using LabVIEW as the communication and user interface software, we created an icon-based, graphical programming environment referred to as "virtual instruments," which alludes to the strong link between the software and its GPIB, data acquisition, and analysis capabilities.

The undergraduate engineering laboratory is often the most important opportunity for students to observe and explore real-world applications of fundamental theories learned in the classroom. However, the actual learning experience is often compromised by the lack of time, effort, and equipment. Over the course of a two-year period, we examined ways of automating an undergraduate laboratory on electrical and electromechanical power conversion.

Laboratory Automation Goals
Most electrical engineering laboratories require instrumentation for generating excitation signals and measuring the results of the excitation. Instruments such as function generators, power supplies, digital multimeters, oscilloscopes, and spectrum analyzers commonly appear in these laboratories.

Learning proper instrument operation, interconnecting the experimental setup and the required instruments, and verifying the various connections are not trivial tasks. One major objective was to reduce the instrumentation learning time, so that most of the lab time was used for "core" material. A second major objective was to appropriately automate data acquisition, while a third was to expedite all forms of feedback and analysis so that the lab session results in a meaningful discussion of the experimental results. A final objective was to encourage discussion and preparation of report documents within the laboratory itself.

Laboratory Automation Equipment Requirements
To automate an electrical engineering teaching laboratory, basic equipment requirements focus on capturing input and output signals as well as on recording and analyzing data. Regardless of the specific laboratory work, most automated labs use a computer workstation with high-speed communication interfaces; general instrumentation with GPIB communication interfaces; A/D, D/A, and digital I/O hardware interface boards on the computer bus; software for numerical analysis, graphics, and word processing; software for bidirectional communication between hardware and the workstation; and a programming environment for providing custom graphical user interfaces.

The communication and user interface software used for our study was LabVIEW running on Apple Macintosh computers. Routines created in this icon-based, graphical programming environment are referred to as "virtual instruments," alluding to the strong link between the software and its GPIB, data acquisition, and analysis capabilities.

Setting Up the Lab
To improve the mechanics of conducting an experiment, we discovered that the laboratory automation software needed a consistent operating shell and user interface. In the first generation of the automation software, the operating shell varied from one experiment to the next. This was confusing not only for the students conducting the experiments, but also for the instructors. However, uniformity in the operating shell did not resolve all student concerns. In addition, we added consistency in operating the software used to conduct the experimental procedures to improve productivity. Another element that improved the mechanics of the experiment was to restrict user interaction with laboratory instruments to only those functions needed.

Improved Teaching Productivity
Lab automation substantially improved the teaching productivity of lab exercises by including analytical models in comparison to experimental results, online analytical models, online analysis of experimental vs. theoretical discrepancies, readily apparent means for online plotting (on screen), and readily apparent, rapid means for data recording to files.

Students are frequently called on to prepare laboratory reports simulating the experiment using a mathematical model and then comparing these results with the actual data collected in the lab. By incorporating this process as part of the experiment, we found that the theoretical model of the experiment can be become an important interactive tool during the experiment.

One of the most effective examples of this was found in an introductory DC machine experiment. Initially, the students determined the armature resistance and the back-emf constant of the motor. Using these values in the steady-state model, the students then applied the model to independently predict the armature voltage necessary to attain speeds at given loads. In effect, the students were playing the role of a cruise control in a fictional electric vehicle. Once the actual motor speed reached ±1 percent of the target speed, the load requirement was changed. Following this, the same set of operating points was achieved by closing a speed control loop on the motor. After completion, the students were invited to compare the armature voltages they computed with those resulting from the speed controller. This interactive use of the model within the experiment not only illustrates the usefulness of the model in a real-world setting, but also exposes the limitations of the model, which did not incorporate friction, drag, and other second-order effects.
Another means of employing the theoretical model during the experiment is by plotting the actual and the theoretical results simultaneously. This practice offers an excellent opportunity for the instructor to interact with students by asking the student to explain any differences. This was very successfully employed in both a magnetic actuator experiment and in a power electronics experiment in which a simple DC-DC buck converter circuit was investigated.

Conclusion
Introducing and integrating automation into an undergraduate laboratory on electrical and electromechanical power conversion improved student productivity. In addition, the interaction between instructor and student focused less on procedural matters, such as debugging the experimental setup, and more on the discussion of the key concepts covered in the experiment. The introduction of laboratory automation has become a powerful tool in emphasizing the understanding and exploration of the fundamental physical principles presented in electrical engineering teaching laboratories.

360532a1.pdf

View the entire user solution in Adobe Acrobat PDF format.