A proposed upgrade of PHY2302/2312, Electronics Laboratory for Physics Majors

John E. Furneaux
Professor of Physics and Astronomy

Here we propose to upgrade and modernize the 1-year electronics laboratory, PHY2302/2312 required for all majors in the Department of Physics and Astronomy. It is typically taken in the sophomore year. The rapid evolution of modern electronics and measurement techniques necessitates the revamping of this course to emphasize computer aided data acquisition, computer interfacing, and modern computer aided design and modeling. The use of industry standard software tools and high-end instrumentation is clearly indicated to give our graduates a competitive edge. This is currently cost effective because OU has purchased a site license for LabView, a $1,500 savings to this proposal. Further, the change of emphasis in the course from "cookbook" methods and rote learning to more active learning techniques and an emphasis on problem solving is needed. These improvements can be accomplished at relatively modest cost because the high-end instrumentation has been made available to the electronics laboratory by the P. I. and substantial educational discounts are available from vendors. Thus, a project that could easily cost at least $60,000 can be accomplished for less than $18,000 in funds from the College of Arts and Sciences with an additional $5,000 contributed by the Department of Physics and Astronomy.

1. Introduction.

In their sophomore year, majors in the Department of Physics and Astronomy take a required Electronics Laboratory Course that consists of 2 coupled 1-semester courses PHY2302 and PHY2312. Along with our majors, a few students from other departments take this course for a total enrollment of about 30 to 35 students per semester. The purpose of these courses is to give these students a practical introduction to electronics at a hands-on level. Traditionally, this has been a "cookbook" laboratory that performs a required textbook laboratory each week and an overall individual laboratory project each semester. In the last 10 years this course has been somewhat upgraded to include an introduction to data acquisition by computer and the use of microprocessor techniques. Projects are assembled on breadboards and are laboriously hand wired. Designs are prototyped, built, and tested, then redesigned, modified, and rebuilt. Modern electronics has left this paradigm. The rapid pace of development and the demand for reliability have lead to a number of software and hardware tools that have revolutionized electronics. Much of this technology is accessible to students. It is imperative that our students benefit from these technological advances and have a working introduction to modern electronic techniques. We feel that our students should be introduced to experimental techniques and tools, including instruments, hardware, and software, which serve as a practical introduction to modern electronics and provide a sound basis for the measurement of quantities of physical interest. There are three major interlocking areas which need to be addressed to meet these goals: upgrade the outdated equipment and software, revise the paradigm for designing and building electronics projects, revamp the method and emphasis of instruction. Each of these topics will be discussed below with an emphasis on the first because it is the direct subject of this proposal.

2. Equipment and Software Upgrade.

The laboratory is now equipped with outdated MAC II’s. They are not configured to acquire or analyze data and they are not networked. Thus, the most crucial upgrade is to these computers. There are 8 laboratory stations that each can accommodate a team of 2 or 3 students. We therefore propose to upgrade these 8 computers to Pentium class Windows NT machines. In general, computer aided data acquisition is not a particularly demanding application so that state-of-the-art machines are not required (approximate specifications are included in the budget). The requirements for modeling are somewhat more demanding but not overly so. For instance, Dell OptiPlex G1 computers in the minimum configuration are completely adequate. We therefore estimate that the required modeling, data acquisition, and analysis computers can be purchased for $10,000. In order to provide the required software efficiently and to allow modern methods of data transfer to be practiced, these computers need to be networked with a server. This server needs to be more capable. A Pentium II running at least 350 MHz with a 100 MHz backplane, 128 M of RAM, and a 10 Gbyte hard drive is needed. The server is also where the PC board layouts will be made demanding a large, 17" – 19", high-resolution monitor and a fast video card. Thus, it is estimated that the server will cost equivalent to a Dell Power Edge 1300 is estimated at $2,000.

Besides the networked computers, there is need for data acquisition equipment. In the modern laboratory electronic data is either taken directly with a computer using an analogue to digital converter interface card or it is taken indirectly with a special purpose instrument which is interfaced to the computer usually via the General Purpose Interface Buss (GPIB). It is therefore important for the students to become familiar with both methods of data acquisition. Therefore, 4 computers will be configured for each of the types of data acquisition. For more demanding special purpose experiments or lab projects the proposed GPIB interfaces are easily portable, hooking onto the USB serial interface of the computer. For data acquisition via the GPIB, "real" instruments are needed. Such instruments typically cost $3,000 to $5,000 each or more. We feel strongly enough about this upgrade of the Electronics Laboratory that we are making a number of general-purpose instruments available to the students from our own research laboratory. These instruments include 3 PAR lock-in amplifiers, 2 high-precision (5.5 to 6.5 digits) digital multi-meters (DMM’s), a 50 MHz digital oscilloscope, 3 programmable voltage sources, and a "universal source" programmable function generator. We are in the process of gaining access to other similar instruments from other laboratories within the Department.

Next for the data acquisition to be effective, software is needed. Here, we feel that it is useful for the students to be introduced to software which is commonly used in our research laboratories, other research laboratories, and industry. This software is LabView from National Instruments. It is very intuitive and user-friendly. Because National Instruments has a policy of supporting the use of its products, especially software, by academic institutions, a license for this software has already been purchased by OU, a savings of $1,500 for this proposal. National Instruments also maintains a library of application guides for this software, many which are applicable for its use in the teaching laboratory.

A second piece of software is needed to update the paradigm of electronics design, electronics modeling software. In modern electronics design, there is very little building of prototypes. Instead, sophisticated models are used to test the design. The standard engine for these models in the electronics industry is SPICE. We wish our students to be able to model their designs and to be able to use the latest electronic parts in these designs. SPICE gives this capability. Therefore, we wish our modeling software to utilize the SICE engine. A second capability that we need for this software is to produce printed circuit (PC) board layouts with output files appropriate for sending to a manufacturer to produce these boards. Such boards provide much more reliable circuits and more realistic and rewarding projects for the students. One possibility is that these projects could be aimed at improving the data acquisition and interfacing in the other laboratory courses taught in the Department. Thus even though this upgrade only directly effects our majors, it could have a wider impact on students in other Departments. The software that meets these demands and has similar discounts and support to LabView is Electronics Workbench. It has been used for a few years in the Electronics Laboratory but not very successfully because there was not a site license and the computers were too slow to produce meaningful results in a timely fashion. The PC board layout feature was not implemented either. A 10-copy site license is $1,500.

In the future we should be able to maintain the equipment requested here and to upgrade the facilities to keep them current via the use of the laboratory fee for this course. Software licenses should be able to be upgraded with modest fees. This is particularly true for LabView because OU has purchased a site license.

3. Revise the Electronics Design Paradigm.

The traditional paradigm for designing electronics, especially for physicists who are not experts at this task, is to find a "cookbook" design that approximately meets the desired application and then build it as a prototype on a breadboard with hand-soldered wires for connections. Then the design is tested and modified until it is "good enough for government work" or time has expired. Final products from such a process tend to be unreliable because they include many hand-soldered joints, and they do not incorporate the latest electronic components because cookbooks take time to be published and probably have been in the library for a number of years. This is the approach that is used for student learning and projects in PHY2302/2312. The more modern approach, which is much more cost effective and produces more effective and more reliable circuits, is to use some sort of electronics design and modeling software such as Electronics Workbench from Interactive Image Technologies, Ltd. With such a system, one may still begin with "cookbook" designs but they will probably be from a database of designs or from manufacturers "application notes" available on the Web. The designs are then modeled using the industry standard modeling engine SPICE. The effects of parameter and component changes including ""worst-case" scenarios can then be modeled. As a result of these tests the design can be modified to be more robust and effective. The newest components can be incorporated because SPICE models for components are routinely provided on the Web by manufacturers. Such models are easily incorporated into Electronics Workbench. Finally, because the model has been optimized through modeling, it is cost effective to transfer the design to a PC board for the production of even a single circuit. This is time effective also because Electronics Workbench has a PC board Layout utility that will automatically generate the appropriate artwork for the production of the circuit. The output of this utility is the appropriate files that a company can use to produce the desired PC board. The electronics technician in the Department assures me that special rates from such companies are available to academic institutions especially for classroom use. We therefore plan to use such facilities to produce professional, robust projects for the students. This has a number of laudatory consequences. The students produce a project that works and that looks and feels like it was professionally produced. They experience the process in a way that is much closer to actual practice than previously. Finally, because the results of the projects will be considerably more robust, projects that will continue to have an impact on the Department and other students can be contemplated. Last semester, I was in charge of the laboratory associated with the freshman physics course. There were many comments about the primitiveness of the computer usage in this lab and the poor quality of the transducers. I believe that many of these things could be effectively upgraded and improved using Electronics Laboratory projects. This is planned.

In the area of digital circuit design and practice, there has been even a greater revolution. Now complex circuits with many chips are rarely built. Instead special purpose circuits are produced via programmable logic devices of varying complexity. The resulting designs are then produced in one-of-a-kind chips or are "burned" into an alterable chip. All that the design consists of is a logical connection between inputs and outputs. Systems for producing designs for such a paradigm are available and in fact generally free to academic institutions. The production of the final products is also very reasonable because the companies are very interested in generating relationships with universities. One of the leading companies, Vantis an offshoot of Advanced Micro-Devices, has provided free software and special manufacturing offers to us for this purpose. We will therefore alter the teaching of digital electronics to emphasize the use of the simplest components, that are useful for simple interface functions and to get an understanding of the properties of digital components, and the design of more complex functions that are appropriate for special purpose chips. We hope to have the digital project include some production of special purpose chips to perform special functions appropriate for the level and interest of the students.

4. Revamp the Method of Instruction.

Because electronics technology is progressing and changing so quickly, it is necessary for our students to be able to keep abreast of these developments independent of specific devices and models. This requires a higher level of learning in the Electronics Laboratory. I feel that it also needs a change in emphasis from specific electronic models to hands on measuring. I therefore plan to rewrite the laboratories and reorganize the presentation to take this into consideration. Labs will be more exploratory and problem solving in nature with an emphasis on measuring and getting the best possible physical data within the constraints of time an money. We will use measurements of the characteristics of electronic components and systems as the examples. This re-enforces the learning of the general properties of the electronics. In the first few sessions I wish to introduce a set of tools which the students can use to explore and measure electronic parameters. Then they will be able to use these tools to generate knowledge and further tools/instruments. My goal is for the students to have a comfortable working knowledge of electronics and its capabilities at the end of the course. I also hope that they will gain some expertise in solving problems of "How do I measure X?". Because there are many software packages to introduce to the students and because I am contemplating considerable change to the labs, I feel that it is imperative that the course materials be tested and developed with some real students. Thus, I have included 2 undergraduate student assistants and one part time graduate student assistant as novice developers and "guinea pigs" to insure that the resulting laboratories are accessible to students and that the expected outcomes in fact reflect some portion of the real outcomes. They will help to develop materials to make the laboratories accessible to their participants. We plan to put the course on the Web so some of the results and materials may be accessible to others within the university and academic community.

5. Conclusion.

Here we propose an ambitious undertaking, not only to upgrade the computer and instrumentation facilities of PHY2302/2312, Electronics Laboratory for Physics Majors, but also to alter the organization, emphasis, and presentation of the course to reflect modern electronics technology and practice. I feel that as a result of this project, our students will benefit greatly. In particular, I feel that they will be more competitive in the workplace and that they will be able to produce more meaningful and rewarding Junior Lab and Capstone projects.

1. Computers. There are 8 positions for experimental groups to work with 2-3 students in each group; thus the request for 8 computers. Dell OptiPlex G1 computers with the minimum configuration from the OU Computer Store would be adequate. The list academic price for these is $1,500 each. For the server a Dell Power Edge 1300 with the appropriate software license and hard drive upgrade is $2,075. With a bundle purchase, the estimated discount gives a total for computers of $12,000. There are now available only MAC II’s which are so slow they make modeling untenable.
2. Computer interfaces for electronic measurement and control applications. These pieces of hardware are the very latest from National Instruments and include an educational discount of 10%. They will remain appropriate for the students for at least 5 years. The GPIB interface will continue to be important for the foreseeable future.
3. This is the newest state-of-the-art software. LabView is the most popular and universally used data acquisition and analysis software. National Instruments also supports educational uses of its products with application notes for classroom use and a 70% discount on software products. For modeling, an important consideration was that the software use the electronic industry’s standard modeling engine, SPICE, so that models for the newest components could be accessed and realistic responses could be obtained. Also the ability to produce printed circuit board layouts and the digital files for their commercial production was a high priority. Electronics Workbench from Interactive Image Technologies Ltd. meets these goals superbly. This company also supports the educational use of its products with substantial discounts, 75% for a site license of 10, and considerable support via extra material produced especially for the use in laboratory courses.
4. Personnel Support. It is important that the course materials be tested and the proposed laboratories be proven. Assistance is also needed in preparing the appropriate Web based materials to make the proposed course as interactive and user-friendly as possible. Therefore, the use of one part-time graduate students and 2 undergraduates who have just completed the PHY2302/2312 sequence is proposed. The graduate student will help prepare course materials and the Web Site. The undergraduates will evaluate and test the materials and give feedback to insure that the materials and the course are at an appropriate level and pace. They will also assist the graduate student and myself with the preparation of materials. This experience will give all the students valuable insight into the preparation of laboratory course materials and solidify their preparation in the important area of experimental measurement using electronics. All of the proposed students are from groups that are underrepresented in physics and engineering.
5. A substantial portion of the resources that are necessary to upgrade this course is in the form of modern high-tech instrumentation. It is vitally important that our students are exposed to modern instrumentation that is representative of that used in research, development, and industry. A representative selection of such equipment will be made available to this laboratory course so that the students will get such exposure. This exposure will not only give the students an introduction to the "real world" but also it will give them a good introduction needed for their further laboratory experience at OU, the Junior level advanced Laboratory and the Capstone Experience.