BQ25672: ILIM_HIZ: Recommended circuit for robust, compliant 100mA POR limit

Tool/software:

Hello TI team,

I am designing a self-powered USB device using a charger from the BQ25672 family (with an ILIM_HIZ pin). My goal is to ensure a USB-compliant 100mA input current limit at Power-On-Reset (POR), especially for a cold start when no battery is present.

My understanding is that the ILIM_HIZ pin is the correct way to set this hardware default before the host MCU boots and takes control via I²C. However, I'm struggling to find a robust circuit solution due to two main challenges:

1. REGN Tolerance vs. Narrow Voltage Window: The datasheet indicates that to set the 100mA limit, the V_ILIM_HIZ must be below 1.08V. However, the REGN LDO, suggested as a pull-up source, has a very wide tolerance (e.g., 4.6V to 5.2V). A worst-case tolerance analysis with 1% resistors shows that a standard resistor divider from REGN cannot reliably generate a voltage that stays within the required window, especially considering the startup threshold.

2. Startup Threshold Ambiguity: The datasheet states the charger stops switching if V_ILIM_HIZ falls below 0.75V, but resumes switching if it becomes higher than 1.0V. For a guaranteed POR startup, must we ensure that V_ILIM_HIZ is always > 1.0V, even under worst-case conditions (min V_REGN and resistor tolerances)? This would make the valid design window for a 100mA limit extremely narrow and difficult to hit reliably (1.0V to 1.08V).

A circuit using an external shunt regulator (like a TLV431) or a Zener diode to generate a stable ~1.05V seems like a robust solution, but it feels more like a workaround than an intended design pattern.

Therefore, my question is:

What is the official, TI-recommended circuit design to reliably set the 100mA POR limit using the ILIM_HIZ pin, considering the wide tolerance of REGN and the startup voltage threshold?

Thank you for your clarification.