June 18th: Finished the Box Plot

I finally finalized my box plot for the PCB! It's somewhat heavily inspired by Aecert Robotics' design, but I removed the pressure sensors because I don't see the point in them. Also, I'm going to use the Bno055 as the gyro because the one Aecert uses is kind of bad :/. However, I want this design to be original - uniquely mine, so I'm going to base the power from better buck converters, have the main MCU be a Nucleo F446re, and use more affordable budget servos.

Total time spent: 4hrs

June 20th: Made first Schematic :D

The box plot makes wiring so easy! Passives like capacitors are annoying, though (but that's what ChatGPT is for). Good thing that datasheets give circuit diagrams, though! I'm having a hard time comprehending some parts of the circuits, like resistor dividers, but I'm slowly learning. Additionally, there are a bunch of ERC errors, and I don't think the nucleo footprint has nearly enough pins? Finding symbols/footprints was a massive pain, too. For example, for stuff like the passives, I kinda don't know what footprint to use because there are so many and I don't want to use the wrong one! Also, for things like the Bno055, all I could find on the internet were symbols of the chip itself, not the breakout board. I ended up having to make my own symbol.

Total time spent: 3hrs

June 25th: Finally fixed my schematic ugh -_-

I finally found the rest of the Nucleo footprint(s)! Turns out, there are 6 different parts. Although, I'm only using the two that have morpho pins XD. Also thought to clean up the space a little. The buck converter passive circuit was becoming confusing so I just wired it. The servo motor array is the bane of my existence - so annoying to rearrange. An annoying thing is, each pin has like 5 different purposes and I can't find any concrete this pin does specifically PWM/UART/I2C. Despite that, I CAN FINALLY START MAKING THE LAYOUT.

Total time spent: 5hrs

June 26th: Layout Designing go brrr (SO MANY ERRORS UGH)

The buck converters' passives are sooooo annoying. I have to place so many things so many times. I'm also starting to realize I might not have enough space (maybe it was a bad idea to use ChatGPT to find footprints). I'm going to use something like Digikey to find new footprints for everything. Also, the servo --> PWM connections are all over the place (rearranging them is going to be a painnnnnnnn). I genuinely don't think I have enough space for the PCA servo driver, so I might get rid of it. Also, I don't know how to route power because 15 amps is a lot and my trace calculator said it has to be 300 mils...

Total time spent: 3hrs

June 27th: Layout Designing go brrr - Take 2

I finally got a more optimized power system!!! Also, I found a way to fit everything on the board. Turns out that the footprints that ChatGPT gave me were wayyyy too big to fit. However, I do think that I'm placing things a little close together (in order to fit everything on the board) that there might be heat interference or smth. Routing is going to be fun!

Total time spent: 5hrs

June 30th: Layout Designing go brrr (fml)

I HATE DRC. IT TEARS APART MY HOPES AND DREAMS. Routing was fun while it lasted, but there. were. so. many. errors.............. I feel like I don't know how big I should make the buck converter traces. My calculator said that they need to be crazy wide (which cannot fit on the VERY small space that I have), but ChatGPT (that I kind of don't trust anymore) reassured me that the short length of the traces would negate damage or something.(Also, I realized I can use pours for the power :O)

Total time spent: 6hrs

July 1st: DRC PASSED!

After a lot of adjustments, all the DRC errors are now gonnnneeee!!!! I also learned about and implemented JLCPCB-specific design rules. Most of my errors consisted of: courtyard overlaps from components being close together, clearance issues of traces because the Kicad auto-route feature (I discovered) is based on a preset clearance that is just shy of the JLCPCB clearance, and some miscellaneous footprint/net errors. However, I'm still concerned on whether the servo voltage pours / vias will be able to carry the current. Prob a few tweaks left, but almost done!

Total time spent: 2hrs

July 7th: I'M DONE

I moved around the layout and increased the trace width of the buck converter circuits. I uploaded my schematic and designs onto several Kicad communities for final insight. Additionally, I also exported a 3d model of the PCB from kicad into OnShape CAD to check if it would physically fit. I also fully mapped out all of the necessary components in Digikey to the correct dimensions and requirements I need. I think I'm ready. I'm confident with the board and it's ready to be made!

Total time spent: ~2 hrs