I think that I’ll do post-class notes on this blog on a regular basis, to have a record of what went on in the class (there are no class notes) and to reflect on what I might need to change in future offerings of the course.

The first part of class today was selling the parts kits to the students, and going over the full parts list with them, so that they could identify the parts and tools later on. I was a bit disappointed that I only sold 15 of the 20 parts kits, as there were 18 students enrolled in the course in the morning. One sent me e-mail during the day, saying that he had dropped, but the other 2 missing students were silent no-shows. This means I’m left with 5 of the $65.50 parts kits, and $327.50 was a little more than I had been willing to lose in order to teach the course. Anyone local want to buy a parts kit at cost? (List of parts ) I’m not willing to ship them—too many breakable things.

Hauling the parts up the hill with my big steel bike trailer was a bit of a workout also. Tomorrow I have less to haul up the hill (the thermoses and coffee cups for the thermistor lab), so I’ll probably use my lighter weight Burley trailer, which will also be easier to get through doors.

It seems that there is one student in the course who has had some electronics hobby experience, as he said that the cheap solder suckers I had bought did work well—I had mentioned their use as “erasers” for soldering mistakes, but said that I had used a different model, and had not tested the cheap ones included in the parts kit. Later on, the same student volunteered the most in the basic electronics review—at one point I had to ask specifically for a different person to contribute. I did get at least a third of the class, and maybe half the class, to ask questions or contribute answers today, so I think most are pretty engaged.

We found out that many of the screwdrivers in the cheap kits ($1.70 for the 11 screwdrivers, including shipping) had the rivets holding the tops on a little too tight, so that they did not swivel properly. I suggested trying a drop of oil, but I thing that simply turning the top relative to the shaft using a pair of pliers may be enough to loosen the rivets. I tried that with one of the leftover sets (which had 2 screwdrivers with tight tops) and it worked fine.

The technical material we covered I got mainly from Socratic questioning of the students about stuff they had in physics. We managed to get Ohm’s Law and Kirchhoff’s voltage and current laws onto the board, and they derived the output voltage of a voltage divider using them. I then explained that the way they would convert the variable resistance of the thermistor into a voltage is with a voltage divider: using the thermistor as one leg and a fixed resistor as the other leg.

Their prelab for tomorrow is to figure out which leg of the divider gets which component, and what resistance to use for a specified operating temperature T_{opt}. I didn’t give them a number for T_{opt}, saying that it would be specified in the lab tomorrow. I did remind them that what we were looking for was a fairly straight line for voltage as a function of temperature with as steep a slope as possible at T_{opt}, and reminded them of their calculus: the slope of the line is the derivative of the voltage with respect to temperature and to optimize the slope, we take its derivative with respect to temperature and set it equal to 0. I’m a little bothered that I did not think to rearrange the voltage divider formula into its clearest form: . Maybe I can give them that tomorrow as a reminder.

I wanted to start them out from the beginning with design exercises, and motivating the theory as a way to solve a design problem. So both the voltage divider and the optimization by taking two derivatives came up fairly naturally for this first lab.

I did give them two pieces of advice that might help (or not):

- They can use symbolic algebra programs like Mathematica (too expensive for me, but I think that the students have free access on some of the University computers) or Wolfram Alpha to do the calculus. After all, this is an engineering class, not a calculus class. I don’t care so much whether they can differentiate with paper and pencil, as whether they can use basic ideas of calculus to solve engineering problems.
- The answer, when they get it, will be very easy for us to check.

We’ll probably go over the prelab very quickly at the beginning of the lab, before showing them how to use the multimeters. I suspect that a couple of the students will need help with the calculus,but perhaps we can pair up students so that at least one student in each pair managed to do most of the prelab.

I hope I’ve not put too much stuff in the lab for these students to do—I don’t want to lose any more students this quarter.

On Friday, I plan to talk about other temperature sensors, so that they can discuss in the lab write-ups when thermistors would be an appropriate design choice.

Outside of class today, I tried seeing if we could do the sampling demo with just the Arduino data logger and a function generator (plus something to provide a DC offset, to keep the sine wave within the Arduino’s 0–5v limits). We certainly get aliasing as we go to higher frequencies, but the picture is not as clear as with the sampling board that has the ADC-DAC pair. For one thing, we can’t do dual traces, with one being sampled and the other being continuous. Without that visual cue, it is hard to see what is happening looking only at the sampled signal. So I’m still feeling a bit frustrated by the sampling lab—there are some important ideas there that the students need, but I’m having a hard time coming up with a design exercise (or even a sufficiently interesting demo) that will cement their understanding.

The DataLogger had a bug fix mid-day today, so the best working version is not the beta-5 version, but the current tip version. I’m wondering whether I should go around to all twelve machines tomorrow morning and update the software (I think that we installed beta-2 last week). It would take a while to do, mainly because booting up and logging into Windows is so slow. The actual transfers from a flash drive will go fairly fast. Maybe if I go to the lab at lunch time I can have everything set up by the time class starts at 2—oops, there is a departmental seminar 12–1:30, so I’d better go into the lab by 11 at the latest, and 10:30 would be better.

My co-instructor just e-mailed me a handout he uses in another class for explaining lab write-ups. I’ll try to adapt it to this class tonight.

## Leave a Reply