EMG PCB

EMG PCB Project

A custom wearable electromyography (EMG) acquisition system built around the Texas Instruments ADS1299-4 analog front end and the Seeed Studio XIAO nRF52840.

The project includes:

• A custom 4-layer KiCad PCB
• Arduino firmware for ADS1299 configuration and sampling
• Bluetooth Low Energy (BLE) data streaming
• A Python receiver for live plotting, diagnostics, and CSV recording
• Project-local KiCad symbols, footprints, and 3D models

Project status: Active prototype development. The hardware, firmware, BLE protocol, and signal-processing workflow are still being tested and refined.

Project Overview

The system measures small differential biopotential signals using the ADS1299-4, transfers samples to the XIAO nRF52840 over SPI, and sends the data wirelessly to a computer over BLE.

Skin electrodes
      │
      ▼
Input protection and filtering
      │
      ▼
ADS1299-4 analog front end
      │  SPI + DRDY
      ▼
XIAO nRF52840
      │  Bluetooth Low Energy
      ▼
Python receiver
      ├── Live plot
      ├── Packet diagnostics
      └── CSV recording

Add image: photo of the assembled PCB and battery-powered

Current Capabilities

• Reads the ADS1299 device ID successfully
• Configures the ADS1299 through SPI
• Supports internal test-signal and input-short testing
• Samples at 250 samples per second
• Currently focuses on ADS1299 Channel 4
• Decodes signed 24-bit two's-complement ADC samples
• Streams custom BLE notifications from the XIAO nRF52840
• Receives and plots samples in Python
• Optionally records received samples to CSV
• Tracks malformed, missing, duplicate, and out-of-order packets
• Includes project-local KiCad libraries for portability

Current development work includes:

• Reducing electrode-motion artifacts
• Investigating clipping during movement
• Improving the flexed-muscle signal-to-noise ratio
• real-time data processing
• Comparing operation with and without ADS1299 BIAS drive
• Refining packet reliability and receiver-side validation
• Developing a future Android application

Repository Structure

EMG_PCB_PROJECT/
├── firmware/
│   ├── emg.ble/
│   │   └── emg.ble.ino
│   └── test_ads_mcu_communication_preinstal/
│       └── test_ads_mcu_communication_preinstal.ino
│
├── python/
│   └── emg_project/
│       ├── emg_ble_receiver.py
│       ├── requirements.txt
│       └── .venv/                  
│
├── hardware/
│   └── kicad/
│       └── EMG_PCB/
│           ├── EMG_PCB.kicad_pro
│           ├── EMG_PCB.kicad_sch
│           ├── ADS1299_Section.kicad_sch
│           ├── MCU_Section.kicad_sch
│           ├── Power_Supplies.kicad_sch
│           ├── EMG_PCB.kicad_pcb
│           ├── fp-lib-table
│           ├── sym-lib-table

Hardware

Main Components

Main Power Rails

The PCB uses the following rails:

• +2.5 V
• -2.5 V
• 3.3 V
• Li-ion battery input
• USB VBUS for development and charging

The ADS1299 analog supply and reference arrangement are implemented on the custom board. The XIAO nRF52840 provides the embedded processing and BLE connection.

KiCad Design

Open the complete project using:

hardware/kicad/EMG_PCB/EMG_PCB.kicad_pro

Do not normally open one of the child schematic files directly.

Schematic Hierarchy

The root schematic is:

EMG_PCB.kicad_sch

It references these hierarchical child sheets:

ADS1299_Section.kicad_sch
MCU_Section.kicad_sch
Power_Supplies.kicad_sch

The child .kicad_sch files are not duplicates. Each file stores one hierarchical page used by the root schematic.

Project-Local Libraries

Custom KiCad resources are stored in:

hardware/kicad/EMG_PCB/lib/

This includes:

• Custom symbols
• Custom footprints
• STEP/STP 3D models
• Project library tables

Keeping these resources in the repository makes the project easier to open on another computer without losing custom parts.

Manufacturing Files

The EMG_PCB_Manufacture folder contains generated Gerber and drill files. Snapshot of a specific manufactured board revision.

Firmware

The firmware is currently developed in the Arduino IDE for the XIAO nRF52840; however, switching over to platformer.io and Cursor for serious development.

Current Acquisition Configuration

The active configuration has included:

• 250 SPS output data rate
• Channel 4 enabled for acquisition
• Channels 1–3 powered down during single-channel testing
• 24-bit signed conversion data
• Internal square-wave test mode
• Input-short noise testing
• Optional normal electrode mode
• Experimental operation with ADS1299 BIAS disabled or enabled

Register settings may change as testing continues. The firmware source should be treated as the authoritative current configuration.

BLE Data Receiver

The Python receiver is located at:

python/emg_project/emg_ble_receiver.py

It scans for the BLE device:

Nykin-EMG

The receiver decodes the custom BLE packet format implemented by the firmware and can:

• Plot samples live
• Print samples
• Record data to CSV
• Run for a fixed duration
• Operate without plotting
• Report packet and protocol statistics

Development Roadmap

• Design custom ADS1299/XIAO PCB
• Assemble initial prototype
• Verify main power rails
• Read ADS1299 device ID
• Capture internal test signal
• Perform input-short noise testing
• Stream samples over BLE
• Plot and save data in Python
• Finalize safe electrode-mode register configuration
• Improve motion-artifact rejection
• Validate voltage scaling and gain calculations
• Add repeatable electrode-placement documentation
• Add digital filtering and signal-quality metrics
• Develop Android receiver application
• Complete a revised PCB based on prototype findings

Author Nykin Leskiw

Developed as a personal electrical engineering and embedded-systems project