This repo consists of a small bit of code written and data collected while helping my son run his science experiment "What brand of AA battery lasts longest?"

We will use a simple resistive load to exercise several samples of several brands of AA battery. The selection of the load was primarly based on time, i.e., getting all the different brands tested within a week. The datasheet for Duracell's AA's shows that at 10 ohms it akes about 20 hours to train below 1V, and 24 ohms is about 50 hours. It was easiest to home-build a 20 ohm resistor, so I went with that to land somewhere in between.

1. Plug the Raspberry Pi into the the USB port of the host computer.

2. Select 3 units of a given brand of battery and insert them into the battery holders.

3. Load battery-monitor.py into the Raspberry Pi (FIXME: Say how) and start it running.

4. Allow data to collect until all battery readings are consistently below 1V. This will take a day or so.

5. Stop the monitoring program and save the collected readings to a file.

6. Remove the spent batteries and store them together.

7. Repeat for the next brand.

• Raspberry Pi Pico RP2040 ($5)

• Half-size breadboard ($5)

• 15 100-ohm resistors. I soldered the legs of 5 100-ohm resistors together in parallel to form a hacky 20 ohm resistor (FIXME: link to article about parallel resistors), and made 3 of these bundles. It would have been much better to buy some nice high wattage, low ohm resistors but time was tight and I had the 100-ohm resistors on hand. ($0)

• 3 Duracell alkaline AA batteries

• 3 Energizer alkaline AA batteries

• 3 Amazon Basics alkaline AA batteries

• 3 Safeway alkaline AA batteries

This is what the output of battery-monitor.py looks like:

Minutes,Time,random0,random1,random2
0,2022/01/30 13:35:02,1.245878,1.227297,1.425544
5,2022/01/30 13:40:02,1.176539,1.148341,1.373175
10,2022/01/30 13:45:02,1.06455,1.02905,1.264408
15,2022/01/30 13:50:02,1.011325,0.9863492,1.223268
20,2022/01/30 13:55:02,0.9976287,0.9734584,1.210378
25,2022/01/30 14:00:02,0.9766811,0.9500938,1.187819
30,2022/01/30 14:05:02,0.9637903,0.9404257,1.180568

• The connections on the breadboard were very sensitive to touch and measurements were affected sometimes. After the first few rounds of experimentation I ended up soldering some of the connections better but there is still room for improvement.

• Voltage readings at ADC2 are consistently higher than the others. I haven't investigated this.

• It's too hard to get the batteries out of the older. It needs one of those ribbons that lays under the batteries that you can use to pull them up. And the holder needs to be mounted to something the user easily can apply force to.