Practical Voltage References for Precision Instrumentation Applications
Integrated voltage references are essential building blocks of precision instrumentation and mixed-signal systems as they set their absolute accuracy, stability, and reliability. This talk describes the evolution of such references from architecture development and circuit design to silicon validation and production release, with an emphasis on achieving excellent DC performance. The talk will begin with a discussion of precision reference topologies and their journey from whiteboard concepts to first silicon. Then, the focus will shift to the longer (and often more challenging) journey from first silicon to released product: characterization, qualification, and high-volume production testing, all of which are necessary to deliver reliable performance in the field at scale.
Click here to view the Author’s Bio
A Fully Integrated CMOS GMR Biosensing Platform with On-Chip Magnetic Excitation
CMOS technology continues to expand its role in biological sensing, enabling integrated platforms that push detection toward smaller scales and higher throughput. Giant magnetoresistive (GMR) sensors are especially promising for point-of-care diagnostics due to their CMOS compatibility and negligible magnetic background, but their μΩ-level signals on kΩ baselines impose extreme analog front-end requirements and have traditionally relied on external components (e.g., sensors, electromagnets, etc.). This work presents the first monolithic CMOS GMR biosensing system-on-chip integrating a 32-pixel array, localized on-chip magnetic excitation, a lock-in analog front-end, an incremental ΔΣ ADC, and full digital control. The chopper-stabilized front-end (<10 nV/√Hz) and incremental ΔΣ converter achieve 120 nT rms input-referred noise, demonstrating a compact, self-contained platform for high-sensitivity multiplexed biosensing.
Click here to view the Author’s Bio
High-Performance TMR Interfaces: Techniques and Insights
Tunnel magnetoresistance (TMR) sensors convert magnetic signals into resistance changes, enabling high-resolution measurements for contactless current sensing, magnetic field monitoring, and sensor interfaces. Their combination of high sensitivity, CMOS compatibility, and low power consumption makes TMR sensors attractive for both industrial and emerging applications. However, TMR sensors suffer from limited bandwidth, temperature sensitivity, and stress-induced drift, which complicate precision readout and long-term stability. This talk reviews circuit- and system-level strategies to tackle these issues and presents two illustrative examples. The first one demonstrates bandwidth extension techniques that enable wideband, ripple-free readout. The second focuses on stability compensation methods to mitigate temperature- and stress-induced variations, highlighting practical approaches to improve both speed and reliability in TMR-based sensing systems.
Click here to view the Author’s Bio
Powerless Contactless Rotation Counting Using GMR Nanowire Technology
We present a novel multi-turn sensor architecture based on a Giant Magnetoresistance (GMR) nanowire, enabling powerless, contactless rotation counting and precise angle measurement in a highly miniaturized form factor. The first part of the talk will explore the underlying physics of the sensor, detailing how the nanoscale wire facilitates contactless and energy-free rotation counting through the generation and propagation of magnetic domain walls. The second part will address the sensor readout methodology, introducing an innovative approach that significantly reduces the number of electrical connections required between the GMR sensor and the ASIC while enabling the readout of a large array of resistors. Together, these advancements establish the GMR nanowire multi-turn sensor as a compelling solution for next-generation applications in robotics, automotive systems, and industrial automation.
Click here to view the Author’s Bio
Hybrid Magnetic Field Sensors
For quite some time, the combination of on-chip sense coils and Hall-effect sensors has been explored in academic research, showing promising potential by breaking the traditional noise–speed tradeoff. For actual product designs the combination brings some extra challenges, of which some will be highlighted in this keynote.
Click here to view the Author’s Bio
Looking beyond CMOS to improve analog performance of precision sensor interfaces
Most of the world’s precision sensor interfaces are realized in CMOS processes, due to their wide availability, lower cost, and ability to co-integrate the front-end circuitry into a larger system. Yet, in stand-alone applications where both versatility and performance are required, bipolar processes continue to offer advantages in terms of 1/f noise, speed, distortion, input impedance and matching. This presentation will give an over-view of instrumentation amplifier architectures with a focus on bipolar-based precision sensor interfaces. Optimized devices, such as super-beta bipolar transistors and precision JFETs will also be discussed, as well as precision circuits based on bipolar transistors. Finally, some insights will be given into the challenges of producing bipolar-based amplifiers in high-volume.
Click here to view the Author’s Bio
Advances towards Efficient High-Resolution Incremental ADCs
High-resolution, high-efficiency data converters are dominated by noise-shaping and oversampling architectures, which excel when continuous-time operation and averaging are allowed. However, in applications where true Nyquist-rate conversion is required, such as single-shot conversion, multiplexing, or time-interleaving, neither oversampling nor noise-shaping can be used. This is due to the very concepts that allow them to combine efficiency with performance: they introduce memory into the signal path, thereby preventing sample-to-sample operation. This talk presents approaches to get around this, i.e. to combine Nyquist rate conversion with high power efficiency through innovative architecture and circuit design with a focus on incremental delta-sigma ADCs.
Click here to view the Author’s Bio
Capacitively-Biased BJT Based Temperature Sensors in CMOS Process
Most CMOS temperature sensors are based on BJTs, which can be used to generate base-emitter voltages with well-defined temperature dependencies. However, due to headroom constraints, such sensors typically cannot operate from sub-1V supplies, limiting their use in advanced process nodes. To address these shortcomings, the capacitively biased diode (CBD) technique was recently proposed, in which a pre-charged capacitor is discharged via a diode-connected BJT. This requires very little headroom (∼150mV) and simultaneously samples the BJTs’ base-emitter voltage. The latter can then be applied to a ∆Σ ADC to generate a digital representation of temperature. In this talk, I will present the results of recent research on CBD-based temperature sensors. Prototypes in process nodes ranging from 180nm to 22nm will be presented.
Click here to view the Author’s Bio