I built EEGFrontier, a compact open-source EEG acquisition board based on the ADS1299 and an RP2040.
The goal was to design a low-cost board that works with dry electrodes while exposing the full EEG signal chain — no abstractions, no closed firmware.
What surprised me most during this project were the practical issues that datasheets don’t really prepare you for: grounding (REF/BIAS), noise coupling from digital lines, routing constraints, and how small layout decisions drastically affect signal quality.
The repository includes full KiCad files, firmware, a BOM with cost references, and documentation images. This is a V1 board and already works, but I’m actively iterating on shielding and noise mitigation.
I’d really appreciate feedback from people with experience in EEG, biosignals, or analog front-end design — especially criticism.