./nkardas
Back to projects
ElectronicsSchool ProjectsMarch 2024

Training Guitar

Guitar prototype with capacitive sensors and LED matrix to help beginners. Real-time detection of pressure on strings and frets.

Training Guitar

Training Guitar - Capacitive Sensors

Overview

Design of a training guitar for beginners using capacitive sensors to detect pressure on strings and frets. Real-time visual feedback via LED matrix to help novice guitarists position their fingers correctly.

Problem Statement

Learning Challenge

Learning guitar, while facilitated by online tutorials (YouTube), remains difficult for beginners:

  • Pressure on strings must be correct to obtain the right sound
  • A beginner without musical ear cannot easily detect if pressure is insufficient
  • Notes sound wrong but the error is difficult to identify

Question

How to help a beginner know if they are pressing the guitar strings correctly?

Innovative Solution

Principle: Capacitive Sensors

Guitar strings and frets are metallic → use as 1-electrode capacitive sensors:

  • Positive electrode: String or fret (5V applied)
  • Negative electrode: Guitarist's finger (ground)
  • Contact modifies capacitance detectably

By combining string + fret information, we can identify which note is played.

Visual Display

4×2 LED Matrix (4 strings × 2 frets):

  • Each LED represents a fretboard position
  • LED on = string AND fret correctly pressed
  • Advantage: Precisely locates error (vs OLED screen showing only note name)

Theoretical Validation

COMSOL Multiphysics Simulation

1. Guitar strings (2D)

  • Modeling: Rectangle 0.5mm × 1m (nickel-plated steel)
  • Parameter: Finger-string distance (0-20mm)
  • Results:
    • Contact: 5 nF
    • No contact: ~10 pF
    • → ×500 detectable variation ✅

2. Guitar frets

  • Modeling: Rectangle 20mm × 30mm + string + finger
  • Parameter: String/finger assembly ↔ fret distance
  • Results:
    • Contact: 3 nF
    • No contact: ~10 pF
    • → Detectable variation ✅

Prototype Architecture

Main Components

1. Sensors

  • 4 guitar strings (shortened)
  • 2 frets (metal pieces)
  • = 6 capacitive electrodes

2. Conditioning Circuits

  • 6 analog circuits (1 per electrode)
  • Capacitance → square signal conversion

3. Microcontroller

  • Nucleo64 L152RE (STM32)
  • Timers for frequency measurement
  • GPIO for LED control

4. 4×2 LED Matrix

  • Real-time display
  • 100Ω resistors

5. Integration

  • Custom interface PCB (Kicad)
  • 3D-printed enclosure

Conditioning Circuit

Operating Principle

Objective: Convert capacitance variation into square signal whose frequency reflects capacitance.

Architecture (3 Op-Amps):

  1. Voltage divider bridge

    • 5V → 2.5V (2× 10kΩ resistors)
    • Follower for stability

  2. Voltage/Current converter

    • Constant charge current on electrode

  3. Hysteresis comparator

    • Triangular signal → square signal (0-5V)
    • Thresholds: 0V and 5V

Mathematical relation:

C = 1 / (R₆ × F)

where F = square signal frequency

Simulation Validation (PartQuest)

  • Finger-electrode distance variation
  • Conforming square and triangular signals obtained
  • Resistor calibration validated ✅

Implementation

  • Electrical schematic under Kicad
  • PCB without vias for simplification
  • 6 circuits fabricated and tested

STM32 Programming

Nucleo L152RE Configuration

Timers (Input Capture Mode with interrupts):

  • TIM5 (32-bit, high precision): 2 channels
  • TIM2/TIM3 (16-bit): Additional channels
  • Interrupt on rising edge of square signal

GPIO:

  • 8 outputs for 4×2 LED matrix control

Detection Algorithm

1. Capacitance measurement

  • Frequency calculation in interrupt callback
  • Averaging to smooth values
  • Period → capacitance conversion in while loop

2. Press detection

  • Threshold: 30 pF (between 10pF idle and 3-5nF contact)
  • Double condition for LED:
    • Fret touched (row)
    • AND String touched (column)

Technical Limitations

  • Only TIM5 high precision timer (32-bit)
  • TIM2/TIM3 less precise (16-bit) → sometimes unstable values
  • Sufficient for functional prototype

Technologies Used

Electronics: STM32 L476RG, TLC555 (oscillator), 4×2 LED matrix

Sensors: Capacitive sensors (strings and frets as electrodes)

Programming: Embedded C, timers, GPIO

Simulation: COMSOL Multiphysics (capacitive validation)

Future Prospects

Technical Improvements

Full guitar:

  • More powerful microcontroller (more precise timers)
  • Bluetooth module for smartphone transmission
  • Mobile app displaying virtual fretboard
  • Discreet integration in guitar body

Future Applications

1. Playing correction Detect accidentally touched neighboring strings that alter sound.

2. Score tracking App comparing in real-time what is played vs displayed score.

3. Automatic score generation Recording played notes → automatic score creation (solo memorization).

Learnings

Multidisciplinary Skills

Electronics:

  • Analog conditioning circuit design
  • Op-Amp and comparator mastery
  • PCB design and fabrication

Embedded Programming:

  • Timers in Input Capture mode
  • Interrupt management
  • Real-time signal processing

Physics:

  • 1-electrode capacitive sensors
  • Multiphysics simulations (COMSOL)
  • Theoretical validation before prototyping

Prototyping:

  • 3D printing for enclosure
  • Mechanical/electronic integration
  • Complete system testing and validation