Santa Cruz Mini Maker Faire 2017

Today I spent about 10 hours on the 2017 Santa Cruz Mini Maker Faire.  The hours for the Faire were 10–5, but I spent some time setting up and tearing down afterwards, so I left the house around 8:30 a.m. and had the bike trailer unpacked and everything back in the house by about 6:30 p.m.  I figure that I spent only about 10 hours earlier on setup for this Faire: applying for the Faire, setting out all the displays and testing them at home, preparing new blurbs for my book and blog, making table signs telling people how to use the interactive parts of the display, blogging about the Faire, and doing load-in last night.  That is a lot less than last year, as I was able to reuse a lot of the design from last year.

Here is the table display I ended up with:

The bare corner at the front left was reserved for the students in my freshman design course who were coming to display their muscle-controlled robot arm, but they decided to set up in back (you can see one of the lead students in the background).

I had four interactive displays (from left to right):

• A pair of function generators and an oscilloscope showing Lissajous figures.  I changed this from last year, as I did not use the FG085 function generator this year, but one of the function generators from the Analog Discovery 2.  I still used the Elenco FG500, despite the very low quality of its waveforms, because it has a knob that is easy for kids to turn, and is easy to reset if they mess it up (unlike the jamming buttons on the FG085).  I did not use the second function generator on the Analog Discovery 2, as I did not want kids playing with just a software interface (and a rather complicated one at that).  It might even be worthwhile for me to build a simple audio sine-wave oscillator with a single big knob over the summer, so that I can have something for kids to play with that is fairly robust and that can’t be easily set into a weird state.  I could even do two, just for Lissajous figures, though having one fixed oscillator worked well this time.  I had the Analog Discovery 2 oscilloscope showing on the laptop next to the old Kikusui CRT oscilloscope, showing both the waveforms and the XY plot, so that I could explain to adults what was happening with the Lissajous figures and about the differences between classic oscilloscopes and modern USB-based ones.
  A lot of people asked me about the Analog Discovery 2, which I was very enthusiastic about—Digilent should be giving me a commission! (They aren’t, although I’m sure I’m responsible for at least half a dozen sales for them, and a lot more if we go ahead with our plan to use them in place of bench equipment in my class next year.)
• In front of the laptop showing the Lissajous figures, I had a standalone optical pulse monitor using the log-transimpedance amplifier and the TFT LCD display.  Using the log-transimpedance amplifier worked well, as did using a lego brick to block light to the sides and back of the phototransistor.  A lot of people have trouble holding their hands still enough to get good readings (particularly children), so it would be good to have some sort of clip instead of resting a finger over the phototransistor.  I’ve tried making clips in the past, but I’m not good at mechanical design, and I’ve always ended up with either a clamp that is too tight (cutting off circulation and getting no reading) or too loose (falling off).  Ideally, I’d want a pressure between systolic and diastolic pressure, so about 12kPa (90mmHg).  People did like the use of Lego as a support, though—it provided a familiar element in the strange world of electronics.
• To the right of the pulse monitor was a pressure sensor.  I had a mechanical gauge and the electronic sensor both connected to a piece of soft silicone tubing taped to the table top.  Kids pressed on the tubing to get an increase in pressure, visible on the gauge (about 20–60 mmHg) and on graph PteroDAQ was running on the little laptop (which we refer to as the “Barbie” laptop, because of its color and small size).  I explained to kids that the tubing was like the tubing stretched across roads sometimes to count cars, with a pressure sensor that recorded each pulse as wheels compressed the tubing.  (For some of the old-timers, I reminded them of when gas stations used to use a similar system.)
  PteroDAQ worked well for this setup, running all day at 20 samples per second without a glitch.  The only problem was occasional display sleep from the laptop, fixable by touching the touch pad.
• At the far right end of the table, I had a phototransistor which kids could shadow with their hands, with the result visible on another channel on PteroDAQ.  This was a last-minute change, as I was getting very unreliable results from the capacitive touch sensor when I tested it out last night.  The capacitive touch sensor worked fine at my house, but in the kindergarten room at Gateway I has a different electrical environment, and it would not work unless I grounded myself. Rather than fuss with the touch sensor, I made a new table sign and put in a light sensor instead.
  I might want to experiment this summer with different ways of making touch plates—trying to get one that doesn’t rely on the toucher being grounded.  My initial thought is that if I have two conductors that are not too close together, but which would both be close to a finger if the touch plate is touched, then I may be able to get more reliable sensing.  I could try some wire-and-tape prototypes and maybe make PC boards with different conductor layouts.  (OSH Park‘s pricing scheme would be good for such tiny boards).

I also had my laptop displaying my book; some quarter-page blurbs with URLs for my book, PteroDAQ, and this blog; my 20-LED strobe; my desk lamp; and a PanaVise displaying one of the amplifier prototyping boards.

I’d like to think of a more exciting project for kids to play with next year—perhaps something I could build over the summer.  Readers, any suggestions?

In addition to my display, some of the freshmen from my freshman design seminar class demonstrated their EMG-controlled robot arm (which uses the MeArm kit):

The students built a MeArm from a kit, then programmed a Teensy board to respond to muscle signals amplified by amplifiers designed by other students in the class. The combined project had two channels: one for controlling the forward-backward position of the arm (using the biceps), the other for controlling the gripper (using muscles in the forearm). With practice, people could pick up a light object with the robot arm.

The scheduling of the Mini Maker Faire was not ideal this year, as it conflicted with the Tech Challenge, Santa Cruz County Math contest, the California Invention Convention, and the Gem and Mineral Show, all of which draw from the same audience as the Mini Maker Faire.

The Faire seemed to be reasonably well attended (rather slow for the first hour and half, but picking up considerably in the afternoon).  There was plenty of room for more exhibitors, so I think that organizers need to do a bit more outreach to encourage people to apply.  It would probably help if they were quicker responding to applicants (it took them over three months to respond to my application, and then only after I nudged them).

Some obvious holes in the lineup: The Museum of Art and History did not have a display, but I saw Nina Simons there, and she said that MAH definitely plans to do it next year, but the Abbott Square renovation is taking up all their time this year.   The fashionTEENS fashion show was April 21, just over a week ago, so it would have been good to get some of them to show their fashions again: either on mannequins or as a mini-show on the stage.  It might be good to get some of Santa Cruz’s luthiers or fine woodworkers to show—we have a lot of top-notch ones, and many do show stuff at Open Studios. The only displays from UCSC were mine and the Formula Slug electric race car team.

Of the local fab labs, Cabrillo College Fab Lab and Idea Fab Labs were present, but The Fábrica and the Bike Church were not.  I thought that Cabrillo did a great job of exhibiting, but Idea Fab Labs was a little too static—only the sand table was interactive.

It might be good to have Zun Zun present their Basura Batucada show (entirely on instruments made from recycled materials) and have a booth on making such instruments.  It might be hard to get Zun Zun to volunteer, but they used to be very cheap to hire (I hired them to give a show at my son’s kindergarten class 15 years ago—they were very cheap then, but I don’t know what their prices are now).

One problem my wife noted was the lack of signs on the outside of classrooms, so that people would know what was inside.  The tiny signs that the Faire provided (I think—I didn’t get one) were too small to be of any use.  It may be enough to tell makers to bring a poster-sized sign to mount.  I had my cloth banner behind my table, but a lot of the displays were hard to identify.  Instructions or information mounted on tables would also have been good—again these would have to be provided by the makers.  I did not see people carrying maps this year—they can also be helpful in getting people to find things that were tucked away in odd corners.  Not many people made it back to the second kindergarten room where FabMo and the Lace Museum activities were.

Update 2017 May 1: It turns out that there were some things I missed at the Faire.  The principal of Gateway sent me email:

… we did have 4 of the Fashion Teens exhibit their creations on the stage at 11:30—might be cool to have them put those on mannequins and have a booth next year. Also we had two more UCSC projects—Jim Whitehead and the Generative Art Studio, and Project AWEsome from the School of Engineering. We would LOVE to have more UCSC-related projects …

 

Revised pressure sensor lab went very well

Today I ran a revised version of the pressure sensor lab (see
Pressure-sensor lab went well, Class-D lab revision didn’t work, Blood pressure monitor, Blood pressure lab, and Blood pressure lab part 2 for descriptions of the old labs).

The revised lab included both blood pressure cuffs and breath pressure using the simplified breath pressure apparatus of Simplified breath pressure apparatus:

The ½” elbow is small enough that I can put my lips around the opening, which would have been a bit difficult with the 1″ tee.

To make the apparatus, the students had to drill 2mm holes in PVC elbows, so I packed up my drill press last night and hauled up the hill in my bike trailer this morning.  For those unfamiliar with Santa Cruz, that is a 3-mile ride with a fairly steady 4% slope, resulting in a climb of about 715′ (218m).  Needless to say, I went slower than usual uphill!  There is a drill press only about 150′ from the lab the students were working in, but the bureaucracy for getting the students access to the drill press is incredible (I tried, and failed, to get a dozen students access last quarter). So it was easier for me to haul my own drill press up the hill on my bike than to deal with the dysfunctional bureaucracy at UCSC to use the drill press supposedly there for student use.

I explained to each pair of students how to use a drill press, including basic safety precautions, and had them drill a 2mm diameter hole in their PVC elbows.  There were no problems with this, and I plan to do the same for the lab in future.

Each pair of students designed an instrumentation amplifier with an INA126P chip as a first stage and an op-amp as a second stage, wired it up on breadboards, checked the calibration, and recorded both breath pressure and blood cuff pressure.  A few students used extra time to play around with some toys I brought in: a hand vacuum pump, a Lego pneumatics pump, and an aquarium air pump.  One group even tried using the pressure sensor as a microphone, using a loudspeaker with a 300Hz sine wave for input (the pressure sensor could detect the 300Hz input without problems, though I suspect that it was not registering the full pressure fluctuation, as I think that the sensor has about a 200Hz bandwidth).

Most groups were done with this week’s lab in the 3 hours of today’s lab, so Thursday’s lab will consist mostly of students doing make-up work to redo old labs, with a few finishing up this week’s lab.  I expect to spend most of Thursday’s lab time grading design reports (I’m about 2 days behind—I got the design reports that were turned in a week ago done and returned yesterday, and I got the prelabs turned in yesterday done and returned today, but I haven’t started yet on the reports turned in last Friday, nor the stack of redone work turned in last week).

The instrumentation-amp lab went surprisingly well this year, despite adding the drill press.  I think that the big advantages over previous years are that they did not have to solder the inst amps this year and that they had already done a multi-stage amplifier for the microphone amp.

I think that I should rewrite the book to introduce multi-stage amplifiers as the default design (since every amplifier they do in the course is now multi-stage), and talk about how to choose the gain for each stage in general, before getting into individual labs.

One minor problem in lab today—students didn’t have the short pieces of tubing to connect up their breath-pressure apparatus.  This turned out to be my fault—I hadn’t included them on the parts list for this year!  Luckily the BELS staff had some pieces leftover from last year, and I had about 20 feet of my own tubing in the box of stuff I’d packed for the lab, so we had enough for everyone to get 6″.

 

Simplified breath pressure apparatus

In Breath pressure measurements, I described my first attempt at using PVC to make apparatus for breath-pressure measurements, with a 1″ PVC tee to press against my face:

One leg of the tee holds the 2mm air leak, the stem of the tee holds the barbed fitting for connecting to the pressure sensor, and the other leg of the tee is for putting around my mouth.

I suggested

Perhaps I should try again with just a 1/2″ female threaded tee—that may be cheap enough that every student can have their own, and only the barbed fittings (which get no flow through them) would be shared.  Students wouldn’t even need to buy their own—I could have a stock of 50 of them, and wash them in a dishwasher after the lab. 

Today, when I went to the hardware store, I did not find any ½” tees with two female threads, so I simplified the design further, eliminating the screw-in plug and just drilling a 5/64″ (2mm) diameter hole into a ½” elbow (one side slip, the other female pipe thread):

The ½” elbow is small enough that I can put my lips around the opening, which would have been a bit difficult with the 1″ tee.

I was able to get similar expiratory breath pressure measurements with either apparatus, but I had trouble getting a good seal for inspiratory measurements with the ½” elbow—I kept getting leaks at the corners of my mouth. None of my measurements today (with either apparatus) got up to the pressures I observed yesterday.  I’ll practice with it a bit more—maybe adding a short length of ½” PVC that I could put a little further into my mouth.  If I can get it to work consistently, it is certainly a cheap enough solution for every student to have their own—79¢ each for the elbows at the hardware store, but only 20¢ from PVC fittings online.

I was not able to figure out where I had bought the 3/16″ barbed fittings from, but I found some for 62¢ each (in 10s) from Cole-Parmer.  At that price, I might even have the students each buy their own, with only the pressure sensors themselves being reused.

Breath pressure measurements

I was never very happy with the breath pressure measurements we made in the applied electronics class, because blowing into a little 3/16″ hose did not correspond well with the measurements reported in the literature.  For one thing, the professional measurements are not into a closed tube, but into a small chamber that has a 2mm diameter hole as an air leak. My fix for this last year was to ignore breath pressure and concentrate on blood-pressure cuffs instead, which worked pretty well. But breath pressure should be easy to measure also, so this week I started looking at making cheap apparatus for doing breath pressure properly.

My first attempt just stuck a barbed tee into the hose. Since the opening of the tee is about 2mm, this should give the pressure vs. air flow curve we need.  This cheap approach sort of worked, but I was not able to get very high pressures recorded—much less than what the literature suggests should be average for a man of my age.  I’m pretty sure my lungs are in better than average condition, both for volume and pressure, so I don’t think this simple setup measured breath pressure well. I don’t know whether the problem is the mouth-to-tubing fit, pressure loss along the tubing, or the tee not really behaving like a 2mm orifice—my understanding of fluid flow in tubes and through orifices is rather weak.

In any event, I decided that the solution was to make apparatus that looked more like the equipment shown in the literature.  I happened to have a few barbed fittings with 1/2″ pipe threads, so I went down to the hardware store to get some PVC parts: a 1″ tee, a couple of bushings to reduce the 1″ slip to 1/2″ female threads, and a 1/2″ male-thread plug.  I drilled a 5/64″ hole (2mm) in the plug and assembled my breath-pressure apparatus:

One leg of the tee holds the 2mm air leak, the stem of the tee holds the barbed fitting for connecting to the pressure sensor, and the other leg of the tee is for putting around my mouth.

The screw-in plugs allow replacing the hole, in case I want to experiment with different size air leaks. The 1″ circle of the tee provides an adequate seal around my mouth, but is not particularly comfortable. The parts cost about $5, so the apparatus is cheap enough.

With this device I managed to get breath pressure measurements comparable to what was reported in the literature, though it took some practice to get high pressures—normal breathing emphasizes high volume, not high pressure.

Plot of high-pressure exhalations followed by high-pressure inhalations (five attempts). I moved the apparatus away from my mouth between breaths, so these are not full breath cycles. I don’t know exactly what the noise at the end of the magenta trace is—probably jostling a loose connection.

So the breath-pressure apparatus works, but I’ll need to think more about whether to build a dozen of these for the course. There are questions about cleaning the apparatus between users, for example—I don’t want to be spreading cold viruses among my students! Is there a way to make the apparatus more comfortable to use and get a better seal around the mouth.  Commercial peak-flow meters use disposable cardboard or plastic mouthpieces that cost about 35¢ each (in 100s from Amazon), and redesigning the apparatus to use them might be worthwhile—but even then the recommendation is to clean the equipment between users, unless one-way (exhalation-only) mouthpieces are used.  The standard usage (based on pictures on the web) appears to be to have a mouthpiece small enough to go into the mouth, so that the lips seal around it—my current design does not have that.

Perhaps I should try again with just a 1/2″ female threaded tee—that may be cheap enough that every student can have their own, and only the barbed fittings (which get no flow through them) would be shared.  Students wouldn’t even need to buy their own—I could have a stock of 50 of them, and wash them in a dishwasher after the lab.  I’ll have to see how well that works.

Blood pressure lab part 2

I spent long hours in the lab yesterday (10 a.m.–6:30 p.m.), because students were not as far along as I had thought by the end of Tuesday. I stay in the lab until all the students are finished, even if that is a couple of hours past when the lab should end. A lot of students were still debugging their breadboarded instrumentation-amp circuits 2 hours into the lab session, when I had thought that they would be soldering after at most an hour more breadboarding.

Debugging on both the breadboards and the soldered boards took a lot of time, but the problems were pretty much the standard ones:

• power not connected
• wire missing
• wire to wrong location
• power supply hooked up backwards to one of the chips
• bent pin on chip not making contact with breadboard

A couple of groups asked me for debugging help but did not have a drawn schematic to work from.  As I had said earlier in the quarter, I would not help them debug unless they had a schematic to work from.  (I think in both cases the problem was that they had connected up one wire incorrectly, but without a correct schematic to work from, there was no way that they could detect the error.)  Some students have been a bit sloppy about trying to work out of their heads rather than putting things down on paper, and this sloppiness is beginning to be a problem for them.  It has mainly been the weaker students who have been being this sloppy, so perhaps it has been more the case that they have been trying to work out of other people’s heads—copying bits and pieces from other students rather than working things out themselves and then implementing it.

I won’t be helping students debug unless they have shown some effort at making their designs debuggable (like having a clean schematic before they wire things up).

I’m not 100% sure, but I think that all but one group got a soldered instrumentation amplifier working and a blood-pressure measurement run done with it.  The one group that didn’t get that far did have a breadboarded amplifier working.

There were some students missing due to illness, so I’ll probably have to arrange some time next week for make-up lab times.  I’m not sure when that can be.