fix(examples): Migrate buzzer tone() to hardware PWM Timer.

[nedseb]

Description

Several examples use a software tone() function that generates a square wave by toggling a GPIO pin in a blocking loop. This blocks the CPU during the entire tone duration, preventing other operations (reading sensors, polling buttons) from running concurrently.

Affected examples

File	Current approach	Impact
lib/mcp23009e/examples/dpad_piano.py	sleep_us() loop	CPU blocked during notes — buttons unresponsive while playing
lib/lis2mdl/examples/metal_detector.py	Hardware PWM (Timer)	Already fixed in PR #251

Solution

Replace the software tone() with hardware PWM using machine.Timer:

from machine import Pin, Timer

buzzer_tim = Timer(1, freq=1000)
buzzer_ch = buzzer_tim.channel(4, Timer.PWM, pin=Pin("SPEAKER"))
buzzer_ch.pulse_width_percent(0)

def tone(freq, duration_ms=0):
    buzzer_tim.freq(freq)
    buzzer_ch.pulse_width_percent(50)
    if duration_ms:
        sleep_ms(duration_ms)
        buzzer_ch.pulse_width_percent(0)

def no_tone():
    buzzer_ch.pulse_width_percent(0)

The SPEAKER pin is PA11 (TIM1_CH4 on STM32WB55).

Prerequisite

Validate that hardware PWM works on the STeaMi board (Timer 1, channel 4, PA11). This was implemented in metal_detector.py (PR #251) but needs hardware testing by @Charly-sketch.

Tasks

• Validate PWM tone on hardware with metal_detector.py
• Migrate dpad_piano.py to use hardware PWM Timer
• Check if other examples use software tone (future PRs)

Related

• PR examples(lis2mdl): Replace test scripts with practical examples. #251 — lis2mdl examples (metal_detector already migrated)
• PR examples: Add practical examples for mcp23009e driver. #247 — mcp23009e examples (dpad_piano to migrate)

[semantic-release-updater]

🎉 This issue has been resolved in version 0.1.3 🎉

The release is available on GitHub release

Your semantic-release bot 📦🚀