Minideck Pcb
total hours spent: 43 hours.
July 8, 2025
time spent: 2 hours
got started with the power stuff today because that felt like a good anchor
opened up a blank KiCad sheet and just kinda sat there before committing to the first buck converter
threw in a couple AP62300s, added some random caps that looked about right
did basic power rails like +3.3VA and +5V, mostly by guessing current draw
honestly spent more time figuring out resistor values than placing parts
it's still super rough, but at least the boardâs not completely empty anymore
definitely not thinking about routing yet
| what happened | image |
|---|---|
| got the power rails and buck converters roughly figured out |
July 9, 2025
time spent: 1.5 hours
dove headfirst into HDMI today and instantly regretted it
the connector footprint was cursed and I had to double-check every diff pair
wired up the HDMI chip and some tiny filter caps for signal integrity
had to decide whether to bother with hotplug detection and i just YOLOâd it in
tried to keep traces as short as possible even though I havenât routed anything
also added pull-ups and random ESD protection that I may or may not actually populate
got through it alive, but just barely
| what happened | image |
|---|---|
| threw in the hdmi stuff, wasnât sure if i was wiring it right tbh |
July 11, 2025
time spent: 2.5 hours
today was CM4 Lite day, and man this thing has a crazy number of pins
literally had to scroll through three different PDFs to not mess up the IO mapping
figured out where to connect PCIe and eMMC signals even though Iâm not using either
started drawing out the main headers and labeling them so I wouldnât cry later
threw in level shifters and testpoints just to be safe
| what happened | image |
|---|---|
| added the cm4 lite footprint and some essentials around it |
July 12, 2025
time spent: 1 hour
today was about the esp32 and the USB hub, two chips I really like but hate wiring
added the esp32 footprint and spent too long assigning IOs
for the usb hub, I was constantly bouncing between the datasheet and KiCad
wired up USB upstream/downstream lines and crossed my fingers on signal integrity
threw in some debug headers and left room for ESD stuff I havenât decided on yet
also started routing logic enable signals to give me more control later
kinda messy but functionally complete
| what happened | image |
|---|---|
| hooked up the esp and usb hub chip, still unsure about esd protection here |
July 14, 2025
time spent: 1.5 hours
this was cleanup day, had to declutter a lot of overlapping nets
focused on GPIO and voltage domains because things were starting to get scary
double checked voltage levels everywhere and pulled back any accidental 5V hits
renamed nets to stuff like USBD+ and PWRBTN so I donât get lost later
added a few jumpers where I needed config flexibility
| what happened | image |
|---|---|
| the gpio, ethernet, sd signals etc., are getting crowded | |
July 16, 2025
time spent: 2 hours
felt like expanding the UI side today, so buttons and switches got added
dropped in an IO expander to control all the random toggles I want
added RGB LED, backlight control, and the mandatory cyberpunk toggle
also threw in debug pads and labeled everything just in case I forget stuff later
double checked how the esp talks to the neopixel
| what happened | image |
|---|---|
| added a bunch of buttons, switches, and gpio expands |
July 18, 2025
time spent: 1 hour
focused entirely on the battery and power switch logic today
added charging ICs, battery monitor, and some MOSFET magic for switching
realized I forgot to add thermistor protection, so threw that in last minute
checked current sense resistor values like 5 times because battery fires arenât cool
rewired the main power path so everythingâs a bit safer
design is power-heavy now but kinda elegant
will probably still debug it with a fire extinguisher nearby
| what happened | image |
|---|---|
| threw in the power switch logic and battery protection circuit | |
July 20, 2025
time spent: 3 hours
long session today, mostly organizing the whole schematic properly
split things into hierarchy sheets because the top level was getting out of control
reassigned power domains and labeled them to make layout cleaner later
caught a bunch of dumb mistakes like misnamed nets and flipped diodes
also started writing notes inside schematic blocks to help future-me
deleted some sketchy looking wires and redrew for clarity
design now feels like an actual engineering thing and not a spaghetti nightmare
| what happened | image |
|---|---|
| general cleanup and better power domain separation | |
July 21, 2025
time spent: 2 hours
this one was about the screen and making sure it wonât fry
hooked up the LCD interface and added ESD resistors on all the touch lines
tried to keep traces short even though Iâm still in schematic mode
double checked the displayâs backlight, touchscreen, and data lines
added a debug header just to probe signals later if things donât work
organized the connector pins to make layout easier
this part was clean.
| what happened | image |
|---|---|
| display interface is finally somewhat sane | |
July 22, 2025
time spent: 1.5 hours
wrapped up the final round of cleanup and review today
triple-checked pinouts, footprints, and power rails for errors
added fiducials, test points, and random extras like boot jumpers
also finalized the USB-C and HDMI protection stuff
July 23, 2025
time spent: 4 hours
worked on deciding the dimensions for the pcb borders and made sure that it fit my case dimensions that i have planned, also many components i placed on the board, just the process of deciding the border length took a lot of time, started routing the pcb and i finished abtu 20% of it, i am thinking of importing it to easy eda pro and autorouting it but i want the source files so ill keep it
july 24,2025
time spent: 3 hours
today i finished routing the display pcb, and started routing the main pcb, it did take a lot of retries since i had to replace parts and i finally found a suitable arrangement, routig all the parts in the main pcb was pretty hectic, ive finished like 70% of it, i've got school stuff to do now so imma take the night off
July 26, 2025
time spent: 2 hours
started work on the display PCB today and honestly this one's way calmer than the main board
got the basic stuff in first power input, the FPC connector for the screen, and some random passives
had to rotate the layout like 3 times to make the FPC orientation make sense
added pull-up resistors.
| what happened | image |
|---|---|
| finished the display PCB design, itâs looking clean and minimal |
July 28, 2025
time spent: 3 hours
big day finished up the main PCB layout and itâs giving boss battle energy
revisited every sheet and double checked for footprint mismatches or missing caps
replaced some jank passives from earlier with proper parts and cleaned up nets
went full sent on layout had to rip up the power plane.
| what happened | image |
|---|---|
| final layout of the main PCB, all the chaos finally connected |
july 29,2025
time spent: 3 hours
finished doing the firmware, got nothing to say about this one.
July 29, 2025 ( in the night )
time spent: 2 hours
Did more work on the case for the display PCB today. Most of the time went into figuring out the outer dimensions so the board actually fits in clean. Kept adjusting a few mm here and there to match the rough size I had in my head. Took way longer than it shouldâve. Once that was sorted, I started adding the cutouts and screw holes. made sure the ports had breathing room, and triple-checked all the hole spacing. Donât wanna be that guy who forgets to leave space for the USB port. Didnât bother adding fillets or rounded corners yet just getting the basic shape right for now. Might tweak the aesthetics later once I test fit the first cut. But yeah, itâs finally starting to look like a proper case.
| what happened | image |
|---------------|-------|
| sneak peaks | |
| sneak peaks |
|
July 30, 2025( early morning )
time spent 4 hours Started designing the bottom half of the display case today. The goal was to keep it simple but still make sure everything lines up clean. Took a bit of time messing with dimensionsâmainly double checking hole positions it matched the PCB layout. Rounded off the corners to make it less boxy and added a small notch for cable clearance later if I need it. The cutouts for the screen area and the mounts came out nice in the 3D preview.
Honestly just getting the case to sit right with the PCB was half the battle, but it looks solid now. Ready to move on to the top half next.
| what happened | image |
|---------------|-------|
| sneak peaks | |
| sneak peaks |
|
July 30, 2025( night)
time spent 2 hours
I started working on the top half of the case. This part was kind of annoying, to be honest. I had to figure out the exact cutout for the display and ensure it sat flush without pressing too hard on the screen. It took a few tries to get the depth right. I kept toggling between the model and the actual display dimensions just to be sure.Then came the USB port clearance, which was a real hassle. Tiny shifts made it either not fit or too loose. I finally added a small beveled area around it so it wouldnât scrape while plugging things in. I also had to make sure the mounting holes lined up perfectly with the bottom half. I used offsets and several measuring tools in Fusion to get it right. I added a little lip around the edges to help with alignment when snapping it on, and I created some space inside for routing wires from the display. Itâs not the prettiest lid, but it gets the job done. I might refine the outer surface later to make it look cleaner. For now, Iâm just focused on function.
| what happened | image |
|---------------|-------|
| sneak peaks | |
| sneak peaks |
|
July 31,2025
time spent 2 hours
spent most of the evening just modeling the top half of the display case. it honestly took longer than expected lining everything up with the pcb was kind of a pain. had to double check the display cutout, standoff placement, and make sure nothing would crash into the components later. i played around with the screen window quite a bit⌠didnât want it to look bulky, but also didnât want it too snug and risk it scraping the screen when assembled. added a few guiding walls inside so the pcb can sort of âslot inâ when itâs time. should help with alignment too. tweaked tolerances here and there, especially around the screen opening. might still need adjustments after the first print, but visually itâs looking nice. not the most fun part of the build, but weâre getting there.
| what happened | image |
|---|---|
| sneak peaks | |
| sneak peaks |