Resolving ESD-Induced Charging Failures in a Market- Deployed Wearable Sports Device

Project Background

The client’s product was already on the market when a critical issue surfaced.
The device is a portable sports performance monitoring unit, worn by athletes during training and competition to measure movement and acceleration data for later performance analysis.

During normal use, it is exposed to constant movement, frequent handling, and repeated attachment and removal from the charging dock. Between training sessions, reliable charging is essential not only to replenish the battery but also to enable consistent data access and transfer for post-activity analysis.

Because the device is expected to be charged frequently and used intensively, stable and predictable docking behavior is a core functional requirement. In this case, that requirement was not met, because when placed on the charging dock, the device performed unpredictable power behavior and would randomly power on and off, directly undermining its usability and reliability.

At the time the issue appeared, the devices were already deployed to end users, leading to growing customer frustration caused by unreliable charging behavior and product returns.

Our team was engaged because of prior experience developing a similar class of portable, docked devices, including charging and docking behavior. This background allowed us to quickly recognize architectural patterns and identify potential weak points that were overlooked during early product development.

 

 

The Problem

When placed on the charging dock, the device exhibited unpredictable power behavior:

  • Random power-on and power-off cycles
  • Intermittent charging failures
  • Occasional unresponsiveness while docked

As a result, data could not be reliably retrieved, and in some cases, the device became unusable until manually reset.This represented a critical issue, affecting both end users and the client’s business operations.

Our Contribution

Initial Investigation

The initial assumption was a fault in the charging dock itself. Our team performed a detailed analysis of:

  • Charger hardware
  • Electrical characteristics
  • Dock-to-device interface behavior

However, extensive testing showed that:

  • The charging dock was functioning correctly
  • The issue could not be consistently reproduced under standard lab conditions

This combination made the problem particularly challenging, since the failure appeared intermittently in real-world usage but resisted deterministic reproduction during conventional testing.

Reproducing the Issue

Through controlled experimentation and targeted stress testing, we discovered that the malfunction is reliably triggered by electrostatic discharge (ESD) effects.

This was particularly relevant given the wearable nature of the device. Athletes wearing the device during intensive movement naturally generate electrostatic charge through friction with clothing and motion.

In practical terms:

  • Electrostatic charge accumulation caused unstable behavior on certain external input pins
  • Once electrically charged, the device could:
  • Fail to charge correctly when docked
  • Enter undefined microcontroller states
  • Exhibit erratic power cycling

This also explained why the issue was difficult to reproduce in the lab, yet appeared in real usage.

Root Cause Analysis

After the client granted us access to the codebase, we conducted a code review and architecture review and identified a critical gap.

An external signal pin from the docking station was connected directly to the microcontroller. There were no TVS diodes or RC filters on all external signals. Additionally, there was no adequate firmware-level protection or signal conditioning.

As a result:

  • contact bouncing and ESD-induced instability were not handled
  • false signal transitions propagated directly to the MCU
  • the system entered undefined and unpredictable states
The Solution: Firmware-Level Signal Stabilization

Because the devices were already in the field, hardware changes (such as adding TVS diodes or redesigning the input circuitry) were not an option. The solution had to be implemented entirely in firmware.

Firmware Architecture Improvements

We designed and implemented a dedicated software module responsible for:

  • Input signal stabilization
  • Protection against contact bouncing
  • Filtering false transitions caused by ESD effects

The solution was built around:

  • Interrupt-driven detection of signal changes
  • Activation of a timing counter on interrupt
  • Continuous monitoring of the pin state over a defined stability window
  • Event processing only if the signal remained stable for the required duration

A validation window of approximately 80–100 ms, with a total observation period of around 400 ms, was used to confirm signal stability before triggering any system action.

This ensured that only valid, stable events were allowed to influence system behavior.

User Control Button Issue Identified During Testing

While validating the charging fix, additional testing revealed a related issue with the device’s user control button:

  • Short and long presses were inconsistently interpreted
  • Long presses sometimes triggered short-press actions

The same stabilization and debouncing module was reused to address this issue.

This unified approach:

  • Stabilized button behavior
  • Eliminated false triggers
  • Ensured consistent interpretation of user input
Implementation Details
  • Firmware development was done using Zephyr RTOS, consistent with the existing codebase
  • A modular library was implemented and integrated into the main application
  • The client’s original firmware was updated without disrupting existing functionality
  • Unit tests were added to validate the new module
  • Extensive functional testing confirmed stable behavior under previously problematic conditions

Results

After deployment of the updated firmware:

  • Random power cycling during charging was eliminated
  • Charging behavior became stable and predictable
  • User button behavior was fully corrected
  • Reliable data access and retrieval were restored

Most importantly, the device regained full reliability without requiring hardware recalls or redesigns, which was a critical outcome for a product already on the market.