The Fall quarter is about to start, and I don’t have my web sites set up with all the homework assignments yet, but I’ve been thinking more about my Winter and Spring courses than my Fall ones.
This Winter I’ll be trying to run a new freshman design seminar for bioengineering majors. This is intended to be a 2-unit course (that is, about 60 hours total work) and to have a lot of engineering design thinking. The basic idea of the course is to have students design and build something relevant to their major, and to get them used to thinking of themselves as engineers, rather than as students who memorize stuff from books or as hands in a bio lab, who behave like low-cost robots to do whatever protocols they are told to implement.
The problem, of course, is that I can’t count on freshmen having learned anything useful yet, at least not universally. Most of the students will have had little or no physics, many will have had little chemistry or biology, most will have had no computer programming, and essentially all will have had no electronics. Collectively, there may be a fair amount of learning in the room, but individually, not so much.
This argues for doing a group project (the traditional way to get something done with large numbers of untrained people). The danger, as with many group projects, is that one student might end up doing all the work, because that is more efficient than parceling out the work to less competent team members. With only 60 hours per student (including class time, reading, lab safety training, and any other activities), I can’t have them picking large projects that are too big for one person to handle alone, which is the usual engineering school approach to ensuring that group projects are really group projects.
The course is not a required one, so I won’t have to deal with reluctant students who don’t want to be there, which removes one problem, but it also means that I have to come up with a course that excites the students, and makes them want to put in the effort to accomplish the project. That means coming up with project ideas that are feasible for freshman at an only moderately selective school (we accept about 50% of applicants) with limited time, but that are exciting to do. A tough challenge!
In June, I mentioned 2 possible projects:
- A low-cost optical density meter for continuous monitoring of OD 600 (with an LED light source) or OD 650 (with a narrower-spectrum laser-diode light source) of cultures on a shaker table.
- A pulse oximeter for measuring blood oxygenation.
Over the summer, I played a bit with the pulse-oximeter idea, and I think it may be a bit too difficult for students with no analog electronics and no mechanical design skills—a pulse monitor using a bright IR or red LED shining through a finger seems feasible though.
I thought about some other projects students could work on, mostly on the theme of do-it-yourself lab equipment for home or cash-strapped high-school labs:
- 3-color colorimeter. This consists of an RGB LED, a disposable cuvette, a phototransistor, and a microcontroller with analog-to-digital conversion. The mechanical design is straightforward, the electronics trivial, and the programming not too difficult. The resulting device would be useful in a number of AP chem labs.
- Conductivity meter. I’ve posted about this idea already. It also has trivial electronics and fairly simple programming. There would need to be a bit more background on the idea of polarizable and non-polarizable electrodes, and most of the design difficulty is in mechanical design—designing a small probe that is consistently wet by the solution and can be handled by the user.
- Spectrophotometer. Initially I thought that a spectrometer was a bit too ambitious for the freshman design seminar, but I’ve seen a couple of crude open-source designs (PublicLab design and Scheeline cellphone design) that could serve as starting points for a usable, low-quality spectrometer. The biggest difficulty here is in getting the spectrum into digital form—the open-source designs use an external camera and a lot of existing software to finesse this problem. I don’t particularly like the idea of using a DVD as a diffraction grating (the lines are spiral, rather than straight), especially since linear diffraction gratings cost under $1 each (down to 30¢ each in lots of 200). I looked up the track pitches of various optical disks:
CD 1.6±0.1 µm 3.95GB DVD-R 0.8 µm 4.7GB DVD-R 0.74 µm BluRay 0.32 µm
and they are comparable with the 1µm pitch of the cheap linear diffraction gratings. The finer the pitch of the diffraction grating, the wider the resulting spectrum.
Note that 4 of the 5 projects involve optoelectronics, and that all of them require programming (unless we take already written programs, like that available with the online spectrophotometer projects). Mechanical design for eliminating stray light paths and making sure that the optical path is consistent is important on all 4 projects, and the conductivity meter has similar mechanical design problems (for ensuring that the electrical paths between the electrodes remain consistent.