You hear the laments in so many places: “Analog design is dead”, or “They're not teaching analog circuit design anymore”, or “Students all want to write apps, not do analog circuit design.”
Perhaps there is some truth to these claims, but there's also a lot that's mistaken about them. How do I know this? Because for the fifth year, I had the privilege and honor of being a judge for the Texas Instruments Analog Design Contest, which culminates with the awarding of the Engibous prize, named after former TI chairman and CEO Thomas Engibous. (The contest press release is here.) [Disclosure—judges receive an honorarium, as well as the honor of judging.]
The top three prizewinners received well-deserved cash awards, while the ten finalist team members attended a three-day event in Dallas to present their projects and meet TI experts. (Due to a schedule conflict, I could not attend, unfortunately).
The rules are fairly modest. Each design requires full documentation, of course, and must use a few TI parts (but not exclusively). Since the specific parts to be used are not called out, the teams have a pretty large field from which to choose. This requirement is really not limiting or burdensome. The teams have to build the project, or at least try to. Not every finalist project makes it to completion within the time limit. Welcome to reality, I say.
As you'd expect from an analog-centric design environment, the projects were not merely software applications to help you find the nearest ethnic restaurant, or GUIs with lots of flash and color. Each project interfaced with real-world signals and equipment, including motors, ECG signals, test signals, and much more. Just looking at the titles of the finalist projects gives you a clear sense of the diversity and creativity involved in these projects:
- an ECG Demonstration board
- an AMSAT maximum power-point tracker
- an improved Tesla coil
- a tablet-assisted mortar aiming/firing unit
- a disposable, low power blood glucose meter
- a wireless-instrumentation network for test aircraft
- a food-safety assessment device
- an automated pill dispenser
- a low-power bird-call recorder
- an automatic guitar-tuning system with motorized tuning heads
All the finalist projects were genuinely impressive, in many ways. Each entry was judged on originality of design, creativity of design, level of engineering analysis (typically 50-100 pages, plus code listings), and quality of written description of how each TI analog IC or processor benefited the overall design. Students designed and built circuit boards, monitored signals, debugged circuitry and code, and did all the things any project team would do (except get the design ready for volume manufacturing).
The winner was a somewhat outside-the-box entry (good!): Adam Munich, a freshman from Rochester Institute of Technology, whose project focused on an improved Tesla coil. His design featured a 10 kVA IGBT H-bridge, which switches at the primary current zero crossing point. No doubt about it, there's nothing casual about managing this level of voltage and power. It is not for the meek or foolish.
The contest will be running again next year, and teams are already getting their projects planned and started, I am sure. As in the past years, I look forward to their creativity, and to discussions of the issues, problems, challenges, and solutions. I look forward, perhaps most of all, to the sheer enthusiasm and energy which come through via their reports.