The IEC Exchange

 

January 2022
 

Featured Article

Kenneth A. Connor

Professor Emeritus, Electrical, Computer, and Systems Engineering

Rensselaer Polytechnic Institute

Member, Board of Directors, Inclusive Engineering Consortium

IEC lens: A Perspective on how IEC can work for you. 

A common theme in problem solving these days – especially in business – is trying different ways of looking at problems. Some people build on the idea of mindset while others emphasize lenses. Some do both. Here, I would like to suggest that a good way to look at how IEC can work for you is to consider the IEC lens. Often, we are overwhelmed by many problems that clearly require more resources to address than are available locally. We know we would like to do a full update on a key undergraduate or we would like to offer a new course in a hot topic like AI or we have a technical research idea but no facilities or colleagues to work on it. What IEC offers is the opportunity to address problems-issues-opportunities by joining forces with colleagues from similar Minority Serving Institutions, practicing engineers from industry, and now, thanks to the addition of several very strong research universities, who have joined us as affiliate IEC members, new potential colleagues from Predominantly White Institutions. That is the IEC lens which has us look at a new enterprise and start thinking about what kind of a team we need to make it happen. 

Engineers like to think concretely, so here are some examples of activities that IEC can facilitate.

Elective Courses – More often than not, there is not enough bandwidth left in smaller, teaching-focused departments to offer elective courses that address the big issues of the day. Just as often, there are too few students interested in each topic to justify the time and energy to develop and deliver a new course. I have been thinking about this recently in the context of solid-state lighting because I have been involved in a couple of meetings sponsored by DOE on Diversity, Equity, Inclusion and Respect in the lighting industry. As most of us know, lighting was once a key part of nearly every electrical engineering program. I had some fun back in the mid-80s writing a history of electrical engineering at RPI (‘The Advent of Electrical Engineering at Rensselaer: 1900-1940, K. A. Connor and M. F. Walker, IEEE Transaction on Education, Vol E-27, No. 4, November 1984, 226-230), which was first offered in 1906. The first undergraduate curriculum included Lighting Theory in the senior year. Lighting left our programs generally after World War 2, because the electrical issues associated with it were no longer of general interest to EEs. Only a small minority of our graduates worked in the lighting industry. Lighting became the responsibility of designers, architects, civil and architectural engineers. 

With the advent of solid-state lighting, with its almost unlimited ability to impact smart systems, lighting is beginning to come back to EE. However, the need for lighting engineers is still rather small compared to the overall production of graduates. Assume that 200-300 electrical or computer engineers each year should go into the lighting industry. That averages out to about 1 graduate per ECE department. If students could take a lighting course or two, they would be much better prepared to succeed or even to consider such a career. The only way to provide such courses to everyone would be online. A small group of faculty, plus a few practicing lighting professionals could make this happen. 

There are two small experiments going on right now to help IEC schools offer substantive lighting design courses. One has been developed by Bob Davis from Pacific Northwest Labs and is offered at Tennessee State in Architectural Engineering. Another is offered in Architecture at Morgan State by Ed Bartholomew, a lighting designer who recently started his own consulting firm in Cambridge, MA. The course is co-taught by Greg Guarnaccia of DoublEdge Designs in Baltimore. Both offerings could be modified to be more multi-disciplinary and deliverable online. 

Lighting is clearly only one example of an elective subject that very few schools could muster the resources to teach if it is treated as a local, rather than a global issue. Another technical area that is close to my heart is plasma engineering, particularly plasma science for nuclear fusion research. Plasma engineering has also played a big role in lighting. Again, there are rarely enough students anywhere to justify teaching the sequence of courses necessary to train someone for a successful career. At RPI, I often taught a plasma course for 3-6 students, always as an overload. It worked well for the students, but I know a much better course could be created by several faculty working together. 

I did two experiments to look for other approaches. In one, I collected the full course documentation from friends at other schools and shared that information with my students. I was not teaching a course collectively, but I at least could provide multiple approaches to introducing various topics so students could find the one that worked the best for them. The other approach was more like what IEC can enable. A friend at UC-Davis, and a former RPI undergrad, wanted to teach the intro to plasmas course for grad students, but could not find enough students. When we were chatting about what could be done, we realized that he could teach my students and his if he delivered his lectures remotely. This was before the internet made this really easy, but RPI had some outstanding facilities for distance learning. We got permission to use one of the distance learning classrooms, my friend taught our students a great intro course, I graded the homework and quizzes for my students, and everyone came out a winner. What we did not do was successfully sell this idea to colleagues at other schools. It remains an opportunity available to anyone who wants to create the right kind of course. 

Advising – Teaching-focused MSIs are known for the excellent relationships created between faculty and students. Students generally get better help and advice from their professors than is the case at large, research-focused PWIs. There are still many frustrations in this area. One big challenge is helping students build a better understanding of what an electrical and/or computer engineer is and, especially, what their job will be like when they graduate. Having a much better connection to practicing engineers will help both faculty and students. I imagine that in your discussions with students there are often questions that you cannot easily answer about the nature of the workplace at likely employers. If you see an opportunity to help students by getting the answers to these questions, IEC can help you develop and deliver new tools, communication channels, etc. to enhance your network and the networks your students are trying to build to help them navigate the pathway to a successful career. I think one such opportunity is to build connections with Employee Resource Groups (also known as Affinity Groups) that now exist in most companies. In breakout discussions during our recent EquiTECH meetings, I learned that ERGs are used to help recruit students, so some baby steps have been made. Now we need to help get people talking and work out more elaborate approaches to connecting students, faculty and practicing engineers into a much more robust network.

Developing New Learning Opportunities for Students – One of my strengths as a teacher came from my desire to play with the ideas I presented in my class. I liked to create activities that made it possible for students to also tinker. Usually, when I came up with something, I offered the activity as an extra credit opportunity and then as an optional project before implementing it as a regular assignment. That way, I could figure out whether my idea had some big holes in it (they usually did) that might end up wasting a lot of student time for little learning gain. Someday I will write up all of my ideas because most of them still work fine (some are too connected to obsolete technology like disposable flash cameras, which I used for their easy high voltage charging capabilities). 

What I discovered through the years was that I could greatly speed up the development process and achieve a much better result if I engaged some practicing engineers and colleagues from other universities who taught similar courses. Insights from practicing engineers are always quite useful. I was inspired to do this, in part, by a great experience I had when I had my first opportunity to teach using studio pedagogy. (‘A Studio Format for Innovative Pedagogy in Circuits and Electronics,’ E. W. Maby, A. B. Carlson, K. A. Connor, W. C. Jennings, P. M. Schoch, Frontiers in Education, 1997 Proceedings Vol. 3) Ed Maby, who was my colleague at the time, had developed a studio version of the Intro to Electronics course at RPI. I taught a section in its second offering. Ed had lots of great ideas for student-focused, hands-on learning activities, most of which were at least a bit impractical. However, we were very fortunate to have a really talented TA who would do Ed’s experiments while Ed was giving the short lecture that introduced each studio session and figured out how to make them work. That TA was Jerry Lopato, who was hired by National Instruments when he graduated and has had a great career there. 

I have shared the following with engineers from both Analog Devices and Tektronix, but I thought it would also be fun to get IEC people involved with an idea I have been playing with for a couple of months. Please feel free to try it yourself. As you will see, it requires very little in the way of resources. This is the kind of an exercise that IEC can help with.

The experiment I have been playing with is exceptionally simple. I am using an Analog Devices ADALM1000 (aka M1K) and the solar cell from a dollar store solar pathway light. I disconnect the cell from its circuit, but I leave it in its package. I then connect the wires from the cell to channel A and ground of the M1K. The signal observed by the scope and the spectrum analyzer (using ALICE Desktop software) look something like as shown in the figure.  

A picture containing text, electronics, computer, indoor

Description automatically generated

My purpose in doing this experiment is to provide an example that is easy for students to reproduce an experiment where they can readily identify the components of the data and, thus, be able to tell the story that is found in the data. Getting students to do more than just turn in their data without comment can be a challenge. Specifically asking them to tell a story seems to help.

 

The solar cell will detect time-varying light signals well into the audio range. The cell is not very large so its capacitance is small. The purpose here is to observe the light coming from various sources in my office. In my office, I have three LED light bulbs (2 from CREE and 1 from Philips, purchased at a local Home Depot) that are in very simple switched fixtures (no dimming). If you decide to do this, I suggest using dimmable bulbs because they are higher quality, rather than the cheapest bulbs available. However, any bulbs are fine. I also have an LED light in a ceiling fan that is controlled by a dimmer. The light signal from the dimmer-driven fixture is likely to be more complex than from a bulb that is not dimmed. There are three laptops (Thinkpads) and a 32" Samsung TV. The laptop displays do not produce a lot of light, but the TV does. Finally, there are two windows so that there is also the possibility of sunlight during the day. Sunlight has an easy signature since it is just a constant DC level, at least when clouds are not passing overhead. The time-varying signal observed with the M1K looks like a DC voltage (about 0.6V) with some significant, non-sinusoidal ripple at about 120Hz. (6 periods in 50ms) 

A screenshot of a computer

Description automatically generated with medium confidence

At this point, it is possible for students to turn the various sources on and off to identify the impact each has on this signal. This will not be particularly illuminating because the individual contributions so not differ a great deal. In ECE, one of our real strengths comes from approaching a problem simultaneously in both the time and frequency domains. This is another set of lenses whose views offer quite different information quality. Sometimes there are start differences in time dependence and sometimes differences are clearer in frequency. 

I measured the frequency content with the ALICE Spectrum Analyzer. As can be seen in the photo of the experiment, both the Spectrum Analyzer and the Oscilloscope can be displayed at the same time.  The largest signal is at 120Hz, which is consistent with the observed ripple frequency in the scope image. The yellow arrows mark this signal and its harmonics. Most of the harmonics come from the ceiling fan LED, which was operating at its maximum output so the signal still looks roughly sinusoidal. When it is dimmed, the signal looks nothing like a sinusoid. The ripple on the light from the CREE and Philips light bulbs is much more sinusoidal so its spectrum has much smaller harmonics. The green arrow marks the 60Hz electrical noise. The red arrows mark the signals from the Samsung TV. The various signals were identified by turning the sources on and off. The 60Hz noise is what was left when the solar cell was covered up by a dark cloth. The 60Hz noise was larger when I covered the cell with my hand because my body capacitively coupled the noise into the circuit. The signal from the TV is the most interesting because it has no obvious relationship to the 60Hz power signal. Both the ceiling fan light and the TV can be controlled by remotes, which makes switching them on and off quite simple. 

A screenshot of a computer

Description automatically generated with medium confidence

I recorded a simple video of this experiment showing the differences in the signals from each source: https://youtu.be/iAph2TDyvvM Please feel free to add comments under the video.

As part of experiments like this, I usually compare the results to what I can do with my Tektronix scope (TBS 1072B) and then mention and/or have the students explore some of the issues to consider when more capable instruments are used. In this case, I reproduced the FFT spectrum analyzer measurement using my Tek scope and the OpenChoice Desktop software to capture the signal. Because the experiment uses a lot of wires and alligator clip leads, it is quite simple to have both measurements made simultaneously.  Both setups use a Flattop FFT Window because it is an option they have in common.  The only special thing I did with the Tek scope was to reduce its bandwidth to 20MHz. That did reduce the noisy stuff at the bottom of the image. The peaks show up at the same locations as the M1K although they are less stable. They are also a lot narrower. I added an RC lowpass filter that reduced the noisy stuff a bit more, selecting the components so that the peaks were not really changed much. Reducing the high frequency bandwidth suggests that most of the differences observed between the two measurements are due to aliasing, but I have not confirmed that directly. 

Graphical user interface

Description automatically generated with medium confidence

If, as I sometimes have done in class, students are asked to do the measurements with both types of devices, it is good to provide them with some helpful information. There is an excellent document, written for the TBS1000B scope) from Tektronix that addresses using spectral analysis and time-domain measurements to solve troubleshooting problems that is particularly helpful. (https://download.tek.com/document/40W_16563_4_HR.pdf

Connecting with ECEDHA – At IEC, we share the same outstanding support infrastructure as ECEDHA and use the resulting close relationship to the advantage of both organizations. We are open to establishing similar connections with other groups and are beginning discussions to see how we can also mutually benefit one another. 

There are two ECEDHA activities that I am involved in as a representative of IEC that are likely to be of interest to you. The first is Project RECET, which is building a repository of materials to make it easier to teach ECE classes remotely. We have some funding from NSF and will shortly be rolling out a new website that will make it easier to use and help further develop the content we have collected. Watch for an announcement on when the new site will be up and running. RECET in many ways owes its existence to the Experiment Centric Pedagogy project that brought so many of us together several years ago and that also motivated the establishment of IEC. Everyone in IEC should be paying close attention. 

The second activity is the ECEDHA annual meeting that will be held in New Orleans this year, for which I am helping to organize a session on ECE Education. The purpose of this session is to introduce the ECE community to non-traditional approaches to educational delivery by having as many of our colleagues as possible present some short soapbox talks on ideas that are likely to force us out of our comfort zones while offering the potential for significant improvements in ECE education. You are all invited to contact me if you have ideas that you would like to share. We plan to have people make short videos that will be available for everyone to watch as a way of promoting a productive dialog before, during and after the meeting in New Orleans. Watch for additional information on this opportunity both in weekly emails and on the IEC and ECEDHA websites. Note that, in addition to the in-person meeting in New Orleans, there will be an online meeting on 8 February that will kick off the discussion of ECE Education and provide additional information on the session in New Orleans and other activities after the annual meeting. I look forward to hearing from everyone with good, provocative ideas. 

 

Ken Connor

Madison, Wisconsin

25 January 2022

 

 

 

 


IEC Corporate Members 

   
 
   
 

  

 

Pathways Supporters

  1.