Tag Archives: Personal Instrumentation

Goodbye, Podium: an Engineering Course Puts Theory Into Practice

The following was originally published 1 October 2012 in the Chronicle of Higher Ed.

I don’t do lectures anymore. Not in the usual sense. And I’ve never had so much fun teaching.

If I get an idea at home for my electronics-instrumentation class, I plug my Mobile Studio IOBoard—a small, inexpensive circuit board that allows students to do multiple electronics tasks without a lot of bulky equipment—into my laptop. I then build a circuit activity, record a lecture, add a paper-and-pencil exercise and an appropriate computer model, and I’m all done. I don’t have to wait until I get to the campus and find an open time in my lab. I can even ask a TA or a former student or a colleague at another university for feedback. The students can carry out their experiments anywhere, I can do my work anywhere, and I can get help from anyone because we all have the same set of simple, mobile learning tools.

Students get the same lectures I would give in person, but the focus is on doing things with the information rather than sitting passively and watching someone else demonstrate. When we meet for a two-hour session, they’ve already listened to the lecture, sketched out a circuit diagram, done some calculations. They’re ready to build and test a circuit at their desks, or may have done part of the activity at home. The recorded lectures become one more tool for the students to consult to help them through the experiments. One of my friends who teaches at a university in Utah won’t let students into her electromagnetic-theory class until they prove they’ve watched the lecture; they also have to bring proof that they’ve done the reading and some kind of homework.

The whole point is to use the class time well.

When students complete a lab experiment at home or in a staffed lab on campus, they come to class better able to explain what they’ve done and why they think the approach is correct, and to provide explanations or questions about any problems they encountered.

What is so cool is that the learning experience has all the key aspects of the complete engineering-design cycle—no matter where the students do the work. The combination of traditional paper-and-pencil calculations, simulation, and experimentation leading to a practical system model makes it possible for them to think and act much more like practicing engineers.

Here at Rensselaer Polytechnic Institute, we call this hands-on approach the Mobile Studio Project (mobilestudioproject.com). The concept grew out of some fantastic but hideously expensive studio classrooms (about $10,000 per seat) that RPI built in the 1990s to bring multiple engineering activities into one well-outfitted room. Each station had a full set of lab equipment, a desktop computer, and tables for taking lecture notes and doing hand calculations. There was a natural progression from introducing a topic and advancing to paper and pencil, simulation, and experiments, with breaks for group and one-on-one discussions. Maybe there was an hour of lecture or maybe 10 minutes, but after that the class would try something. More often than not, the class began with a demonstration or a hands-on activity. You’d build, you’d talk.

It was so much fun. I just loved it. We thought we’d ushered in a new way of teaching. But very few engineering schools adopted this model because it was so expensive and the studio classrooms held just 30 to 40 people. Our enrollments went up, and we had more students than we knew what to do with. The model simply was not scalable, even for us.

With the advent of laptops, we realized we didn’t need a special studio room. We could do all the activities except those that required access to lab equipment. We just had to figure out a way to add that capability to the students’ laptops. We tried a variety of existing options, mostly involving some kind of inexpensive data-acquisition board, but either they did not have the functionality we needed or they were much too expensive. And then we discovered we were at one of those magical crossroads where it became possible to imagine that every engineering student could be given his or her own personal mobile electronics laboratory.

What happened? A combination of better and cheaper electronics, strong leadership, and financial support from the National Science Foundation and industry led Rensselaer—with help from Howard University and the Rose-Hulman Institute of Technology—to develop the Mobile Studio.

The latest version of the Mobile Studio hardware costs about $150 per student—cheap enough that every engineering student gets his or her own board. (For information on acquiring the hardware, visit the project’s Web site.) So now we can take a studio approach in any decent classroom. More important, when students learn with Mobile Studio, their homework and test scores go up and learning improves, as documented by the University at Albany Evaluation Consortium, which provides independent assessment of research and pedagogy.

The most exciting results come from synthesis questions in which students are required, for example, to design a circuit with a specific functionality. Students who work with the Mobile Studio have significantly higher scores than those who do not.

Students can pursue their own ideas, build something, and then try it either just for their own satisfaction or, in my class, for more points. This style of teaching closely resembles the way engineers do their jobs and allows the students to achieve understanding based on what they do best.

Once students could do labs at home, the new technology suddenly opened up dimensions we hadn’t thought of before. Courses that never had lab experiments have them now. For example, mechanical- and civil-engineering majors learn circuits through minilabs that might last 20 minutes. Students can now be asked to do homework involving hardware. They can also tinker at their own projects.

As I said, if I get an idea at home, I just set up my Mobile Studio, build the circuit, and see what happens. I don’t have to wait for the classroom. This is the direction in which engineering education is going. New modes of delivery made possible by an ever increasing array of products will make the present way we teach unrecognizable. I might never need to stand behind a podium again.

Advertisements

A Dialog on Mobile Studio

The following was originally published in the Fall 2012 issue of ECE Source and co-authored by Mohamed Chouikha of Howard University

Can you give us a little background on Mobile Studio? What is its history? What are its cost and capabilities?

The Mobile Studio is a small, inexpensive hardware platform for use in a home, classroom or remote environment. When coupled with the Mobile Studio Desktop software, the system duplicates a large amount of the hardware often used to teach electronics intensive courses in ECE and other STEM disciplines (e.g. scopes, function generators, spectrum analyzers, etc.). The goal is to enable hands-on exploration of STEM education principles, devices, and systems that have historically been restricted to expensive laboratory facilities.

ms2

The Mobile Studio was the brainchild of Don Millard, who presently serves as a Program Director in the Division of Undergraduate Education at NSF. In 1999, he was at RPI, where he was looking for a way to make Studio Pedagogy work more effectively and much more affordably. Studio instruction, developed at RPI primarily in the 1990s and used in essentially all of the core electrical and computer engineering courses (for which RPI’s ECSE Dept. received the ECEDHA Innovative Program Award in 2001), was found to be a very good way to deliver engineering education, especially in ECE programs, and attracted a steady stream of visitors all of whom went away hoping they could implement something similar. However, with very few exceptions, none were successful because the costs were so high. In round numbers, the facilities necessary to provide lectures, paper and pencil problem solving, numerical simulation and traditional experiments all in the same room, cost about $10k per seat, which is just not practical.

Don’s vision for a new, inexpensive studio for teaching electronics was based on replacing the very expensive standard set of instruments found on a typical lab bench (scope, power supplies, function generators, multi-meters, etc.). He hoped that someone was selling something he could use for this purpose, but nothing he found met his needs. His next step was to design and build a small board that could duplicate the functionality he needed. With the help of Analog Devices and Doug Mercer (an ADI fellow who graduated from RPI in ’77), an amazing RPI student (Jason Coutermarsh, who now works for ADI), funding from NSF and Hewlett-Packard, and the help and support of a growing, but small number of true believers (including the authors of this piece), he went through several designs, with varying degrees of success, until what is called the RED2 board became generally available in 2008. Earlier designs (including RED and BLUE) showed that his idea works very well, but were, as a colleague at Rose-Hulman has said, not quite ready for prime time. The RED2 board had all the necessary functionality required and the robust design to survive regular usage by undergrads. Information on all three boards, along with the software necessary to run them, etc. can be found at mobilestudioproject.com. All three boards were designed by RPI personnel (Don and Jason, primarily, who also arranged for and supervised their manufacture). The cost of each was about the same as a textbook or about $150.

How difficult is it to adapt existing experiments to Mobile Studio? How much effort is required for a trial run? Are there any limitations in developing experiments vs. standard equipment? Are students able to use standard equipment after learning on Mobile Studio?

Because the Mobile Studio platform contains most of the instruments necessary for standard electronics experiments, adapting existing labs generally only requires a different set of instructions because the wires have to go to different places. It is also the case that the instruments available through Mobile Studio do not have the bandwidth or dynamic range of more expensive, standard instruments, so some experiments must be modified to work at lower power or under 200kHz. This is generally not a big issue because most basic circuits and electronics are best taught in the audio range anyway. It is possible to modify the experiments in an existing course, over the term of a semester, without the need for special outside work. This has been done at several universities, including, most recently, UW-Madison. A short description of the innovations they are pursuing in engineering education are described in their annual report which includes a short section on Mobile Studio.

ms1

An example of the efficacy, ease and immediate results of the Mobile Studio approach is a trial session done several years ago at Howard University. Using a make-shift studio space, a few fold-up tables/chairs and a wireless network, the set-up took only 30 minutes and by 8:00 am the session started. Twenty students (self-organized in teams of two) participated in this activity. After a 30 minute overview of the activity (incorporating the “Filters CAD” module) with a demonstration of a working circuit (using an electric guitar as the input signal), students were given 90 minutes to construct and test their designs. All of the teams created protoboard versions of the circuit and tested them with the instrumentation, while 6 of 10 successfully demonstrated a functional circuit. The participating faculty, students and administrators were so impressed and excited with this result that the following semester the Mobile Studio was formally introduced in the junior electronics course as part of EE curriculum at Howard.

The equipment necessary for the trial run at Howard, with the exception of the guitar, fits in a standard carry-on travel bag. This makes it possible to offer workshops and outreach activities almost anywhere, as long as the participants have laptops. We have given workshops, run K-12 programs, etc. in many different countries with minimal difficulties because everything is so small. There are now universities in Africa (e.g. in Cameroon and Ethiopia), RET programs serving teachers in Native American schools (e.g. through the CIAN ERC at Arizona), community colleges, etc. all using Mobile Studio because it is so simple to create the experience anywhere and anytime. Our colleague from Morgan State – Yacob Astatke – has been particularly active in improving engineering education in his home country of Ethiopia.

ms3

One of us (KC) recently received the following question from Israel.

I came across info about the Mobile Studio h/w and s/w and went through the stuff in your site and some tutorials on YouTube. I like the idea of working with Mobile Studio which seems like an available inexpensive solution. However, I got the impression that from a student’s point of view there is not much difference between the MS and using simulation such as Matlab (that students are currently using in a communication class). After all, students are not using a real scope, function generator, or a spectrum analyzer, but rather a s/w interface which may seem similar to a Matlab simulator. Dr. Avi Silbiger, Jerusalem College of Technology.

You have done a nice job of asking one of the questions we get fairly often, but in a well-defined context. Restating your question somewhat – if students see the same information through a software interface, how is the experience really different from a simulator? In fact, there are a few very minor aspects of using the Mobile Studio that differ little from a good simulation. For example, it is possible to use one of the Arbitrary Waveform Generators to reconstruct a square wave from its harmonics and play it back directly through connections on the board to display the results on the oscilloscope. One can listen to the signals using the audio output to obtain a sense of what harmonics mean. Clearly, this can be done using Matlab. Nothing is physically connected to the board … no physical components are used … etc. However, very, very little of what we do with Mobile Studio is anything like this.

Before I get into the real differences, I will address your statement that ‘after all, students are not using a real scope, function generator or a spectrum analyzer…” As I explain to my students and everyone else who I talk to about Mobile Studio, what we have really is a collection of real instruments … just not ones in boxes with the usual knobs and displays. NI, especially, and Matlab, to a lesser extent, also struggle with the misconception that a small piece of hardware connected to a computer is less of an instrument than the big old scopes, etc. that we have used forever. I think all of us suffer somewhat because of the unfortunate name that NI gave to their control programs – Virtual Instruments or VI’s. In some of the documentation I give to my students, I carefully draw boxes around each part of the board so that they see it as a very compact collection of boxes and not just a board with lots of connections. I also point out that modern instruments are really configured largely the same way except that they are selfcontained rather than share a single computer. For example, in my radar lab, I have high frequency spectrum analyzers, network analyzers, oscilloscopes, etc., all of which I can operate remotely from my laptop because they all are Windows boxes. They are computers that look like instruments.

OK, so much for philosophy … What is the student experience like (addressing the rest of your sentence)? First, we do exactly the same experiments we did with standard instruments (mostly Agilent, with the cost of a single station greater than $10,000). We build all the basic op-amp configurations and measure the input and output voltages on the breadboards. Students get all the same experiences they have with real experiments, including dealing with noise, poor connections, power limitations, etc. All of the materials for my course are available at http://ei-rpi.org. Once the students master the use of Mobile Studio, teaching them to use a standard oscilloscope is very simple. In fact, colleagues at Rose-Hulman have found that students learn the concept of scope measurements much more quickly with Mobile Studio than with standard equipment.

rhit

It is not just that the students have a real hardware experience in the classroom, essentially identical to what we got in the past with standard scopes, etc., what is really powerful is that each student is given a full set of tools similar to those used by practicing engineers. I discuss this in a document I give them called Differentiators and Integrators (available on the course website http://ei-rpi.org) in which I address the tools required for an engineering design cycle which includes both simulations and experiments. Students get to work with the same kind of tools they will have on the job as they learn to be engineers. Even more important, they own these tools and can use them anytime and anywhere. Thus, the learning they do based on hardware experiments is not limited to a classroom they only see 2 or 3 hours per week, but can be continued at home and with friends. Note that instructors also have the freedom to try ideas out at home in little more space than it takes for one’s computer. Check out the article I just wrote for the Chronicle of Higher Ed on this topic. In the end, this approach is very attractive if one believes in the value of hands-on experiments, not just simulations.

There are other programs similar to ours that we are now collaborating with. Kathleen Meehan at Virginia Tech, for example, offers Lab-In-A-Box classes that never really meet in the traditional sense. Students do all of their labs at home and then come in to demonstrate their results. They use different hardware and, ironically (given your questions), used a Matlab interface for their measurements. Thus, it really can look like a simulation on first glance, but then you see the actual experiments with all real-world characteristics included and you see that the students are having a complete learning experience.

I understand that Mobile Studio boards are no longer available but Digilent has a new product that can easily replace it. Are there any other options available if one is interested in pursuing this approach to engineering education?

We have indeed stopped manufacturing Mobile Studio hardware. However, there are still several hundred boards which can be purchased by interested parties. A better choice, though, for long-term program development, is one of the similar commercial products. The Analog Discovery board from Digilent, developed in collaboration with Analog Devices, provides the same functionality as Mobile Studio with better specs and a lower price. This seems to be the curse of the industries our graduates work in – everything must simultaneously get better and cheaper. Analog has completely revamped their excellent university program around Digilent’s board. There is an excellent video on the program on their website. The Virginia Tech Lab-In-A-Box project is now using the Digilent board. The NI myDAQ is also an excellent choice for what we are now calling Mobile Learning Platforms. Bonnie Ferri’s TESSEL Center at Georgia Tech has had great success in adding hands-on content to existing ECE courses using myDAQ. There are other products on the market, but these two provide the most extensive support infrastructure and are or are becoming widely used.

Mobile Studio, TESSEL and Lab-In-A-Box have all been supported by NSF through the CCLI (now TUES) program. A collaboration of the participants in these programs (The Center for Mobile Hands-On STEM) presently has support from TUES. The universities involved are RPI, Virginia Tech, Georgia Tech, Howard, Morgan State, and Rose-Hulman with assessment provided by the University at Albany. Several other universities and community colleges are actively applying the pedagogy being developed at the new center. 

Blended Learning for Circuits and Electronics

The following was originally published in the May 2014 issue of ECE Source 

As I near the end of my 40th year as a professor of electrical and computer engineering, I remain excited about teaching electronics to engineering students. It would be natural to expect at least a little burnout at this advanced point in my career but I find I am having more fun than ever because we now have some amazing new tools available and, through a fortuitous series of recent experiences, I am meeting more and more remarkable teachers from ECE departments throughout the US, Canada and the world who want to fundamentally change the way our students learn about and with electronics.

My experience teaching circuits and electronics appears not to be typical. A large fraction of ECE students learn these critical subjects from faculty who treat the assignment as a chore maybe because their only direct interest and background comes from their undergraduate years. When I interviewed faculty at an outstanding research university recently (call it UXY), I was told that most of the faculty in basic circuits were from the communications group. This coincides with my own experience as an undergrad in the 1960s when I had a terrible experience in my Intro to Circuits course at Wisconsin. Full disclosure – I was assigned to the 7:45AM section three days a week for this theory only course, so I lacked a strong incentive to attend all class meetings. I did not have my Circuits Lab until a later semester. The students at UXY are better off in that they usually take their lab in the same term, although the schedules are not coordinated. When I took Linear Systems from the same instructor a couple years later, I was pleasantly surprised to discover that he was a really good teacher and thoroughly enjoyed the course. Unfortunately, giving students a positive experience in circuits is critical to building interest in ECE. In another recent interview, a colleague at a West Coast school whose experience teaching circuits is similar to mine, sees basic circuits as the breaking point where students either start loving EE or running away from it.

While it is generally accepted that labs are critical to providing the best possible learning environment for circuits and electronics students, little has changed in my four decades except that the instruments at most colleges are better and are interfaced with computers for control and data acquisition. They may be a little prettier than they used to be, but lab facilities are still expensive, limited access, often windowless with utilitarian desks and nearly all have the same set of standard instruments. Except for at a few schools that built expensive studio classrooms where all forms of content delivery (lecture, computer lab, experimental lab, problem sessions, recitations …) are possible in any length or combination in any class period, hands-on, hardware-based learning activities are only possible in these standard labs.

Easily the best thing about being an ECE professor is that we get to help equip energetic, bright young people with the skills and knowledge to change the world. Sometimes, when we are lucky, these changes directly impact what we can do in the classroom. Examples of the educational tools generated by the creativity of our graduates in circuits and electronics include National Instruments’ myDAQ, Digilent’s Analog Discovery, Syscomp’s CircuitGear, with others joining their ranks almost daily. For links to most products in this market, see http://hibp.ecse.rpi.edu/~connor/Mobile%20Studio/. What clearly distinguishes these products from traditional bench-instruments is their very low cost (somewhere near the price of a technical textbook) and their multiple functionality (scope, function generator, power supply, spectrum analyzer, logic analyzer …) that provides most of what is needed for analog and digital courses. They are truly the results of the relentless quest of our graduates for ever more capable and cheaper products.

It would be easier to talk about this on-going revolution in engineering education if there was a commonly accepted name for these new tools. At the recent ECEDHA meeting in Napa, Sam Fuller of Analog Devices called them Personal Instrumentation Devices. Others I have talked to have called them Hand Held Instruments. At RPI, where one of the earliest such devices – The Mobile Studio – originated, we choose to call them Mobile Learning Platforms. If you want to find information online, maybe the best terms to use for your search are USB Scopes, even though the functionality goes way beyond just simple voltage measurement and PCs are being replaced by tablets and phones.

None of the names mentioned is really fully descriptive or understandable and the overall market is changing so fast that the focus on the hardware probably limits discourse. Fortunately, as educators, we are better off keeping our sights on the pedagogy the new tools make possible. In the center that combines Mobile Studio (RPI, Howard, Rose-Hulman, Morgan State) with TESSAL (Georgia Tech) and Lab-In-A-Box (Virginia Tech) projects, we use Mobile Hands-On STEM or MOHS Pedagogy in honor of the whimsical units used previously for conductance. I will mention at least one other term below, but, in the remainder of this piece, I will use MOHS Pedagogy to mean the blending of inexpensive, hands-on, experiment-focused instruction with other traditional (e.g. lectures, paper-and-pencil problem solving, recitations) and, especially, newer (e.g. flipped classroom, problem-based learning, active learning) modalities.

One of the secrets for staying young, at least in spirit, is to embrace opportunities that go well outside one’s comfort zone. My most recent chance to do something with the potential to be really embarrassing was when my team (with Fred Berry of MSOE and Peter Lea of Bowdoin) was chosen for an NSF pilot program to see if I-Corps activities and training could work for educational research as well as it does for traditional technical research. This effort, with support also from Intel, took nine three-member teams through a very intense schedule of multi-day workshops and weekly online classes and meetings, both beginning and ending in DC. Probably the most demanding and rewarding part of the process was the development of hypotheses organized into a business model canvas and then getting out of the building to interview at least 100 potential customers for our ideas. The interests of the groups ranged from transition programs for veterans interested in engineering careers to teaching programming fundamentals to concept inventories. All of us were convinced we had great ideas, but none had ever worked so hard to define things from the customer point-of-view.

What did we learn from this experience? Some of the real nuggets we received were not totally new, but our overall approach went through quite a significant change. We learned that the teaching of circuits and electronics labs is largely driven by the nature of the available facilities. Most schools have moderately sized labs serving 10-25 students, requiring multiple sections scheduled throughout the week, often into the evening. Most labs are written so that nearly all students can finish them in the allotted time, although many also have some open shop time for students to catch up, if necessary. Most students do what they can during their class time and then make the most of the experience in their reports. A faculty member from a large East Coast community college extended her work with her students as long as scheduling permitted and then continued to work with some of them using the lab set up she maintains in her office. We talked to quite a few such dedicated and passionate faculty who do whatever is necessary to make sure their students learn the material. They are frustrated because they cannot take care of everyone. We also came to appreciate that we were taking a similar facilities focus in our work by starting from the hardware and not pedagogy. We have these cool new tools and were thinking of how they change what we can do, rather than starting from what we need or want to accomplish and then seeing what the news tools can make possible.

We also learned (or confirmed in this case) that the facilities used for circuits and electronics labs were unexciting. Another topic discussed at length at the Napa ECEDHA meeting was how to recruit and retain a more diverse and larger ECE student body. Everyone in the room loves being an electrical or computer engineer and struggles to understand why so few young people want to join us. If we really think being in the ECE community is so great, why aren’t we building hands-on, active learning environments and highlighting them as showplaces to excite our future brothers and sisters? This idea partly came from some interviews with admissions personnel who immediately appreciated the potential to build on the nascent ECE interests of students who attend our pre-college summer programs by using MOHS tools to enable them to experience real ECE. Summer programs, like a lot of K-12 outreach, tend to emphasize mechanical and structural engineering (even those with a strong robotics flavor) over ECE and CS, which tend to make the enrollment imbalance in engineering even worse. One admissions officer even suggested buying a mobile device and a parts kit for each future ECE student to encourage them to tinker and provide them with a set of tools to use in their future courses. The return on a couple hundred dollar investment should be quite good.

During the present year, both before and after I-Corps, I have also gotten to work with an amazing group of schools – the HBCUs with ECE programs – on a new project using Experimental Centric Pedagogy (the alternative to MOHS I mentioned above) to address their recruitment and retention issues by providing a richer learning experience for their students. With leadership from the two schools already using MOHS(Howard and Morgan State), thirteen institutions are now building a model program using mobile instruments that should benefit the entire ECE community.  The lessons learned from our I-Corps experience have definitely provided a valuable contribution to this effort.

Another I-Corps lesson is a corollary to the cool learning approach that attracts new students. It is better if our students get to solve problems like real engineers if they are to develop the skills and confidence to become electrical and/or computer engineers. Real engineers do not sit around just solving academic problems, like I did as a sophomore. They do simulations and experiments with the goal of creating a workable systems model that can be used to produce the products their customers need. In recent years, with the availability of laptops, we have been able to add simulation activities to our circuits and electronics courses. Now with USB-based mobile instruments, our students are able to also do as much experimentation as necessary in their courses, no matter where and when they meet. It is no longer necessary to wait for a turn in a lab. They can also be given experimental homework. In fact, their entire lab experience (following the Lab-In-A-Box model) can be done at home. We can all give our student a learning experience that completely blends all the approaches taken by working engineers in ways that were never before possible.

I offer an example of what we can now do from the Electronic Instrumentation course I offer for large numbers of engineering students outside of ECE at RPI.  I asked my students to design a hardware switch debouncer using a 555 timer chip. I let them use any design they find online. To test out their design, I had them first simulate it with PSpice using an idealized switch signal with some bouncing. None of the designs they can find online will work with this sequence unless they change some components. Then I created the same sequence as a csv file and used this file with the Custom feature of the Analog Discovery Arbitrary Waveform Generator, so that the identical input could be used to test their hardware implementation. The students were able to test their design and see essentially identical responses from both simulation and experiment. I did this because I wanted the students to know they could simulate exactly what they are seeing experimentally, which validates the simulation. Then they can use any information easily obtained from the simulation to characterize their experiment. For example, I ask them to find two designs that use different average power. There is no direct way to measure power with standard instruments, but with PSpice it is trivial. This project involved roughly equal parts traditional hand calculations, simulation and experimentation and produced a working system.

I would like to end with a challenge – should we go as far as possible to eliminate traditional labs in ECE but rather incorporate experimentation as part of all standard undergrad courses? I would argue that traditional labs should indeed go and the use of lab facilities change focus to providing high performance or specialized measurement capabilities as a complement to the MOHS approach. We should change from labs being an add-on experience to becoming an integral part of a fully blended learning experience for our students. All the types of new pedagogy mentioned above (especially flipped classrooms) can then be incorporated based on the interests and capabilities of the students and faculty and not on what one can find in the lab.