This paper describes a compact resistor-based temperature sensor intended for the thermal monitoring of microprocessors and DRAMs. It consists of an RC poly-phase filter (PPF) that is read out by a frequency-locked loop (FLL) based on a dual zero-crossing (ZC) detection scheme. The sensor, fabricated in 65 nm CMOS, occupies 5800 μm 2 and achieves moderate accuracy (±1.2 °C (3σ) inaccuracy) over a wide temperature range (−50 to 105 °C) after a one-point trim. This is 2 better than previous compact resistor-based sensors. Operating from 0.85 to 1.3 V supplies, it consumes 32.5 μA and achieves 2.8 mK resolution in a 1-ms conversion time, which corresponds to a resolution FoM of 0.26 pJ·K 2 . Index Terms-CMOS temperature sensor, resistor-based temperature sensor, RC poly phase filter, frequency-locked loop (FLL), dual zero-crossing (ZC) detection, one-point trim.
This article presents an energy-efficient dual-RC frequency reference intended for wireless sensor nodes. It consists of a digital frequency-locked loop (FLL) in which the frequency of a digitally controlled oscillator (DCO) is locked to a temperature-independent phase shift derived from two different RC poly-phase filters (PPFs). Phase shifts with complementary temperature coefficients (TCs) are generated by using PPFs made from different resistor types (p-poly and silicided p-poly). The phase shift of each filter is determined by a zero-crossing (ZC) detector and then digitized by a digital phase-domain modulator (-M). The results are then combined in the digital domain via fixed polynomials to produce a temperature-independent phase shift. This highly digital architecture enables the use of a sub-1-V supply voltage and enhances energy and area efficiency. The 28-MHz frequency reference occupies 0.06 mm 2 in a 65-nm CMOS process. It achieves a period jitter of 7 ps (1σ ) and draws 142 μW from a 0.9-V supply, which corresponds to an energy consumption of 5 pJ/cycle. Furthermore, it achieves ±200 ppm inaccuracy from −40 • C to 85 • C after a two-point trim.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.