STM32WB: HCI_HARDWARE_ERROR_EVENT (0x02 – BlueCore Time Overrun) Observed on Some Devices

Hi everyone,

I am encountering a HCI_HARDWARE_ERROR_EVENT with code 0x02 (BlueCore time overrun error) on STM32WB devices.

Key observations:

• Issue is observed only on some units, while others operate normally
• All devices use the same firmware, BLE stack, and FUS versions
• Affected units show unstable BLE behavior (disconnects or degraded performance)

Validation performed:

• Tested with STM32 reference example (BLE_p2pServer)
• The issue is still reproducible
• PCB layout and hardware design have been reviewed and validated
• No application-specific changes involved during testing

Questions:

1. Root cause
   • What typically triggers a BlueCore time overrun at the controller level?
   • Does this point more toward hardware timing issues (clock accuracy, RF timing), or internal controller or stack conditions?
2. Unit variation
   • Has anyone observed this error occurring only on a subset of devices?
   • Could this be related to component tolerances (HSE or LSE oscillator, RF matching network, power stability, etc.)?
3. Failure diagnosis
   • What is the best way to identify the source of this issue?
   • Any recommended approach to isolate whether it is:
     • Clock-related
     • Power-related
     • RF-related
     • etc.
4. Mitigation or fixes
   • Are there improvements in newer BLE stack or FUS versions related to this?
   • Any configuration or system-level practices to avoid timing overruns?

Summary:

• Reproducible under reference BLE application
• Design and layout validated
• Observed only on some hardware units
• Likely related to device-specific or marginal conditions

Please let me know if any additional details, logs, or configuration information are required.

Any insights or similar experiences would be greatly appreciated.

Thanks!