Matthias Eberlein, Harald Pretl,
"Thermal Sensor and Reference Circuits Based on a Time-Controlled Bias of pn-Junctions in FinFET Technology"
: Harpe, P., Makinwa, K.A., Baschirotto, A. (eds) Analog Circuits for Machine Learning, Current/Voltage/Temperature Sensors, and High-speed Communication, Springer, Cham, Seite(n) 191-207, 2022
Original Titel:
Thermal Sensor and Reference Circuits Based on a Time-Controlled Bias of pn-Junctions in FinFET Technology
Sprache des Titels:
Englisch
Original Buchtitel:
Harpe, P., Makinwa, K.A., Baschirotto, A. (eds) Analog Circuits for Machine Learning, Current/Voltage/Temperature Sensors, and High-speed Communication
Original Kurzfassung:
This research introduces a new concept to generate PTAT and CTAT voltages precisely through switched-capacitor operation. In replacement of the classical BJT, the active bulk diode is utilized and forward-biased by a charge-pump. During capacitor discharge, the respective pn-junction voltages are sampled at different time points and further combined by charge-sharing techniques. A bandgap reference with this architecture shows an untrimmed accuracy of ±0.73% (3?), consuming only 21 nA in a 16 nm FinFET process. Similarly, a thermal sensor was implemented with an 8-bit SAR readout, which achieves a precision of ?2 °C without calibration on 2500 ?m2 silicon area. The simple structures feature intrinsic supply rejection down to 0.85 V and a digital-alike operation.