Week 04

electronics design


Board Design

I started by working in KiCad, but ran into issues importing libraries and exporting. Eagle seemed to have better documentation (both in terms of HTM(A)A classmate support and online discussions), so I eventually switched to using it.

First, I layed out the schematic in Eagle. I got the basics down first, and then added an LED to pin 11 (with a resistor on the ground side) and a button to pin 5. I didn’t realize it at the time, but since I didn’t include a pull-up resistor with the button, I would need to rely on the ATTiny’s internal pull-up resistor. Here’s how things looked after completing the schematic:


Next, it was time to turn the ratsnest into the board layout. It took a few iterations to get a configuration where things worked out without requiring the other side of the board or jumper wires.



The milling process itself was smooth, but after the job was complete I realized that I hadn’t set up the design rules properly in Eagle. Some of the traces were too close, and the 164” endmill didn’t get between them (see the left side of the ATTiny, and the FTDI header).

pcb traces too close

Sure enough, I had set the minimum distance between traces to 0.01mm in Eagle, but looking back to Week 3 when we characterized the endmill, it should have been set to 0.020mm. I updated the design rules, and the warnings were exactly where they were expected:

pcb traces too close in eagle

I went back to fix them, and also realized that I needed to update the grid size to .5mm rather than .25mm so that the lines could snap more easily to where I wanted them. Then, I milled again:

final mill

Things I did wrong:

  • Set up design rules in Eagle that were too lax
  • Forgot to update the grid size for easier layouts


Next up: soldering. This went relatively smoothy, although the resonator was one of the tricker components to solder. I was sure to get the proper orientation for the ATTiny by looking at the board in Eagle and checking where the little circle was (top left). I neglected to check the orientation on the LED, which caused some problems when I went to test the board. The (nearly invisible) green line should go on the ground side.

soldered board

I tested connections with the multimeter, and didn’t find anything unexpected, so I moved onto programming.

Things I did wrong:

  • Flipped the orientation of the LED


Next, I used the USBTiny to program the board on the lab’s computer. Because of issues with the Ubuntu virtual machine, we were running Arduino on Windows. First, I ran the blink sample file to confirm that I had identified the pin correctly in code.

The LED didn’t light up, although I could see the blinking pattern when I used the multimeter to try to diagnose, at which point I learned that my LED was backwards. It was a good excuse to try out the heat gun to remove the LED from the board, before resoldering in the correct orientation. When I plugged the board back in, it blinked.

Next, I tried programming functionality in the button….

Things I did wrong were:

  • Identified the wrong pins for the LED and button in code.
  • Didn’t turn on the pull-up resistor in the ATTiny