11/26/2025 - Did some planning

A RTL-SDR will be used to receive the ADS-B packets, which are sent to the MCU

I plan to have the MCU:

• process the packets
• broadcast them to Flightradar24 (free business account!!)
• display the nearest plane's info: airline, type, altitude, heading, v-speed, ground speed, etc.
• potentially handle packet feed over WiFi/USB from another receiver for future use with an external antenna

BUT

DIY ADS-B receivers are usually done using a Raspberry Pi, and are well documented. However, in my opinion, Pi boards are unreasonably expensive. I looked for an alternative, and I discovered that HP made extremely cheap thin PCs like the T520, which I will be using:

Only £6.49 on eBay(!) AND the RAM and SSD are replaceable!
I do spot the odd power plug, so that may raise a problem to solve later...

update 7 dec 25: you just need a resistor over the "smart" power pin of the motherboard (see here)

I plan to run some flavour of Linux on it, together with a USB to GPIO breakout to connect to the display's 16 pin connector:

11/28/2025 - Time to make an FPGA board

I've decided to make my own FPGA board to get the precise timings needed to drive the display that I've chosen. It will need precise timing with clock pulses, so an FPGA is perfect. I'll be doing it around a ICE40UPK5 chip in EasyEDA.

Why not get an FPGA board? It's a great challenge for me, I can customise the form factor, and only have the things I need on it.

11/30/2025 6 PM - Completed the FPGA board schematic

I've now completed the schematic for the FPGA board. I used the iCE40 UltraPlus Breakout Board documentation as a reference while making my board. I've used jumper connectors as it used, which I find to be a great way to have selectable modes.

I have a 64Mb (8MB) flash chip interfaced with it, along with a FTDI chip to program either of them directly.

Now it's time to do some PCB routing...

oh... a name you ask?
I'm calling it Salvador (yk... iCE and all that 😅)

11/30/2025 11 PM - PCB placement done

Now, I've finished doing PCB part placement. I changed the headers from 2, 1x24pin headers to 2, 2x12 headers, since my placement is quite compact.

There are M3 mounting holes on each corner.

Time to enter PCB routing hell... wish me luck

12/2/2025 - Finished the FPGA board

As soon as I started routing, I realised my placement was just wishful thinking :(

I rearranged parts as I went, connecting to the internal 3v3 and GND planes where needed. The signal traces are fully on the top and bottom layers (yes the back looks a little weird)

It's about 49x43mm, so pretty small. I mainly used 0402 sized caps and resistors to save space and make routing a little easier.

12/7/2025 3 PM - Updated PCB board for UART over USB

I didn't consider communication directly over USB via UART. The FTDI chip has integrated UART: it's actually the main use of the chip in other applications.

I broke out the TXD and RXD lines on Channel B on the chip, and assigned them to pins on the headers and on the iCE40.

Also changed the font + added an easter egg, so now the PCB is complete and ready for fab.

12/7/2025 5 PM - Sorted out a BOM of electronic parts

Parts for 1 board sourced from LCSC: total $34.39 with shipping (exported BOM in gh repo)

PCBs sourced from JLCPCB: total $3.30 with shipping (holy cow that's cheap)

Other parts:

Ebay:
HP T520 - £6.49 + P&P
+ Power supply - £9.00

Total: £19.04 (~$25.50)

Aliexpress:
P2 128x64 LED matrix display - £21.99
SDR/ADS-B Receiver - £17.89
5V 5A adaptor - £5.29
24k resistor - £0.93
270k resistor - £0.94

Total: £48.63 (~$64.89)

grand total: $128.08

12/27/2025 9 PM - Created an enclosure

I created an enclosure that's large enough, and easy to make with sheets of 9mm MDF.

It requires three sizes of board:

• 247x275mm x2
• 129x257mm x3 (2 cut down by 10mm crosswise)
• 76x129mm x1 (width can be variable. Instead of ordering a sheet of this size, it ends up being cheaper getting one more of the above.)

MDF can be glued together with PVA glue, and I will drill holes for the cables and airflow.

12/27/2025 10 PM - Finished the initial firmware

Finally got back to writing firmware I started earlier. I have written it in SystemVerilog, which is a new language to me, and it successfully synthesises.

Currently, it should show a test frame of all borders white, split into quarters with a cross. If this works, I can start going about livestreaming the frames, and playing with PWM for a larger colour range, but only after I have the FPGA and display to test with, due to the mig lack of docs online.