░  ░░░░  ░░        ░░        ░░  ░░░░  ░░░░░░░░   ░░░  ░░        ░░   ░░░  ░░        ░
▒  ▒  ▒  ▒▒  ▒▒▒▒▒▒▒▒  ▒▒▒▒▒▒▒▒  ▒▒▒  ▒▒▒▒▒▒▒▒▒    ▒▒  ▒▒▒▒▒  ▒▒▒▒▒    ▒▒  ▒▒  ▒▒▒▒▒▒▒
▓        ▓▓      ▓▓▓▓      ▓▓▓▓     ▓▓▓▓▓▓▓▓▓▓▓  ▓  ▓  ▓▓▓▓▓  ▓▓▓▓▓  ▓  ▓  ▓▓      ▓▓▓
█   ██   ██  ████████  ████████  ███  █████████  ██    █████  █████  ██    ██  ███████
█  ████  ██        ██        ██  ████  ████████  ███   ██        ██  ███   ██        █

OUTPUT

SCHEMATIC

Since I didn't end up using it last week, I wanted to do something with the thermistor this week, while also using a OLED screen for output. I wanted the temperature, converted from the thermistor's resistance level, to be displayed on the screen. At first I thought I'd have to use a wheatstone bridge circuit to accurately calculate the temperature but from most resources I read, apparently this wouldn't be necessary for a project where extreme accuracy was not so necessary, and that a simple resistor divider would suffice.

LAYOUT

This time I made sure to edit the footprint so that the number of pins was accurate to my esp32, as well as flipping it to account for it being on the back of the board. In retrospect I shouldve given more space between the esp32 footprint and the 4-pin header for the OLED screen, but it didn't actually seem to cause any issues. I also included pads for pullup resistors for the SCL and SDA lines since I read that sometimes the internal pull-ups on the esp32 pins are not strong enough for a stable connection, but upon testing the screen, it seems like I don't need to use them.

CARVING AND SOLDERING

The board cut almost perfectly on the first try, though there was one problem area. For some reason the carvera just gave up on cutting the power trace in the middle. I was faced with either recutting and risking the entire board failing due to bits being more worn down, or just trying to carve that section of the trace out myself with an x-acto knife. I chose to try my luck with the latter. It doesn't look very nice, but it doesn't short anymore. After this, soldering went fine.

TESTING

First, I wanted to test if the screen worked, since that was the main goal of this week (output) and I was a little nervous with how little space I gave it next to the esp32. Luckily it worked fine, and I think it's pretty cool how easy it is to use a OLED display through I²C!

Despite the success with the screen, I have unresolved issues with the thermistor. It displays an outrageously inaccurate temperature reading, which apparently can be the cause of poor soldering or an unstable voltage. I didn't get a chance to try to fix this with lab tools yet, so I plan on doing so today after class. My biggest suspect is of course the shoddily carved out power trace having low resistance to ground, because I mean, look at it. I hope that this is it and I can actually fix it by carving it out more, because it'd be really cool to have both input AND output working for once.