Electronics Production

---

Goals for Week 8:

Document design rules & process of sending traces to the milling machine,

Do the group assignment,

Make and test a microcontroller development board that you designed

Group assignment page:

• Group assignment page

Developing the Xiao ESP32 C3 board

Edge cuts file

Traces file

---

Design Rules

These images show a line test used to determine the minimum trace width and spacing that can be reliably milled on a PCB using our current setup. The labels (.001, .010, .020) refer to the trace widths and spacings in inches. By comparing the digital design with the milled result, we can see that the .001" features are too fine and result in poor quality traces or bridging, while the .010" region shows improvement but still some imperfections. The .020" region mills cleanly and consistently, indicating it's within our tool’s reliable capability. Based on this test, our minimum trace width and spacing should be at least 0.010", though 0.012" or more is safer for consistent results. We can use this information to set design rules in our PCB software to avoid creating features too small for our mill to handle.

While 10 mm/s speed is standard, the tip of the endmill kept snapping off halfway through the job. To fix this, we changed the speed from 10 mm/s down to 6 mm/s for the traces.

Use the document below to understand how to export the board from KiCad all the way to uploading the files to VPanel for cutting. Use this link to access mods.

The first issue I encountered was one of the traces touching GND, resulting in a short circuit. I fixed this by carefully removing some of the copper trace with a x-acto knife.

Unfixed

Fixed

Next, I realized that 3 of the pins were having a short. This was because I made the traces much too close around the resistors for the bottom 4 pins. I fixed this by milling a new board (I probably could've used the technique above, but I thought milling a new board would be more time effective.) I made sure to make the traces around the resistors further apart.

New Board

The final issue I encountered was the fact that I had unnecessary 10 kOhm resistors that were messing up the inputs from my keypad. I was under the impression that all input devices needed 10 kOhm resistors for each pin. However, this is definitely not the case for keypads. I removed the unnecessary 10 kOhm resistors with a hot air soldering iron and tweezers, and my board was finally fully functional.