 |
|
|
|
Resistors are notoriously sensitive to manufacturing "corners." A resistor on one wafer may have a significantly different base resistance than one on another. Consequently, resistor-based sensors typically require one- or two-point calibration to achieve high accuracy (e.g., error < ﾂｱ0.5ﾂｰC).
The fundamental principle involves measuring the voltage drop across these resistors when biased with a constant current or using them within a Wheatstone bridge configuration. Advantages over Traditional BJT Sensors Resistor-based Temperature Sensors in CMOS Tech...
These utilize the doped regions of the silicon substrate. They often exhibit a higher TCR but are more susceptible to noise and substrate interference. They often exhibit a higher TCR but are
At the heart of a resistor-based sensor is the Temperature Coefficient of Resistance (TCR). In CMOS processes, different materials offer varying thermal responses: where they trigger refresh rate adjustments
For decades, the "Proportional to Absolute Temperature" (PTAT) voltage generated by BJTs was the industry standard. However, resistor-based sensors offer several distinct advantages in the nanometer CMOS era:
Resistor-based sensors are now ubiquitous in , where they trigger refresh rate adjustments, and in IoT nodes , where power budgets are measured in microwatts. As we move toward 3nm processes and beyond, the focus is shifting toward "all-digital" temperature sensors that leverage the delay of resistive-capacitive (RC) networks, further blurring the line between analog sensing and digital processing.
High-ohmic polysilicon resistors can be fabricated in a smaller footprint than the multi-transistor arrays required for high-accuracy BJT sensing.
