A 16.28 ppm/ C Temperature Coefficient, 0.5V Low-Voltage CMOS Voltage Reference with Curvature Compensation
summary
The gist
This paper presents a fully-integrated CMOS voltage reference designed in a 90 nm process node that achieves an excellent temperature coefficient, low line sensitivity, and power efficiency at a very
In short
This paper presents a CMOS voltage reference designed for low-voltage operation (0.5V) with excellent temperature stability across a wide range (-40°C to 130°C). The design uses mutual compensation of PTAT and CTAT currents, enhanced by curvature compensation, to achieve a temperature coefficient of 16.28 ppm/°C and low line sensitivity.
Key concepts
- PTAT Circuit
- The Proportional to Absolute Temperature (PTAT) circuit uses two MOSFETs operating in subthreshold. Its current is proportional to the absolute temperature, allowing the circuit to generate a voltage that remains stable even when the ambient temperature changes significantly. This improves power efficiency by mimicking BJT characteristics.
- CTAT Circuit
- The Complementary to Absolute Temperature (CTAT) architecture exploits multi-threshold MOS transistor characteristics. It generates a voltage component that is complementary to the absolute temperature change, helping to further stabilize the output reference voltage across varying temperatures.
- Curvature Compensation
- This technique uses an additional transistor placed in parallel with a resistor to exploit the exponential-like temperature dependence of leakage current. By adjusting this component, the design can extend its stable operating range and improve the overall accuracy of the temperature coefficient performance.
Terminology used across episodes
This episode discusses
- A 16.28 ppm/ C Temperature Coefficient, 0.5V Low-Voltage CMOS Voltage Reference with Curvature Compensation · Paper Radio
The paper
A 16.28 ppm/ C Temperature Coefficient, 0.5V Low-Voltage CMOS Voltage Reference with Curvature Compensation · Read on arXiv
Birla Institute of Technology and Science
This paper presents a fully-integrated CMOS voltage reference designed in a 90 nm process node using low voltage threshold (LVT) transistor models. The voltage reference leverages subthreshold operation and near-weak inversion characteristics, backed by an all-region MOSFET model. The proposed design achieves a very low operating supply voltage of 0.5 V and a remarkably low temperature coefficient of 16.28 ppm/°C through the mutual compensation of CTAT, PTAT, and curvature-correction currents, over a wide range from-40 °C to 130 °C. A stable reference voltage of 205 mV is generated with a line sensitivity of 1.65 %/V and a power supply rejection ratio (PSRR) of-50 dB at 10 kHz. The circuit achieves all these parameters while maintaining a good power efficiency, consuming only 0.67 μ W.
Transcript
Introduction to the show: ident: Robotics Radio. Generated commentary on the latest robotics and control papers.
Rosa: Today's paper: "A 16.28 ppm/ C Temperature Coefficient, 0.5V Low-Voltage CMOS Voltage Reference with Curvature Compensation".
Dev: This paper presents a fully-integrated CMOS voltage reference designed in a 90 nm process node that achieves an excellent temperature coefficient, low line sensitivity,
Rosa: First, who's behind it and why it matters.
Paper summary: Dev: So, looking at the "A sixteen point two eight ppm/ C Temperature Coefficient, 0 point 5V Low-Voltage CMOS Voltage Reference with Curvature Compensation," the authors' main achievement is delivering that specific temperature coefficient of sixteen point two eight ppm/°C while operating at such a low supply voltage of zero point five V <ref:2508.15729#pg0>.
Rosa: And the authors managed to do this by using a combination of PTAT, CTAT, and curvature-correction currents for mutual compensation across different MOSFETs <ref:2508.15729#pg0>. It seems like they’ve successfully engineered a solution that maintains remarkable stability over the entire range from-forty °C to one hundred thirty °C <ref:2508.15729#pg0>.
Taro: From an autonomy research standpoint, this means we have a reference voltage source that doesn't drift significantly when deployed in unpredictable external conditions, which is crucial for systems that need consistent power levels for their sensors and processing units <ref:2508.15729#pg0>.
Dev: I think the implication here is that we can design more efficient, low-power control systems where the reference voltage doesn't become a major source of error due to environmental temperature fluctuations or low supply voltages <ref:2508.15729#pg0>.
Rosa: That efficiency combined with stability opens up avenues for deploying these types of reference circuits in edge devices that need to operate reliably in varied conditions, which is what I'm looking at for field robotics <ref:2508.15729#pg0>.
Taro: The real impact could be on creating more resilient autonomous agents capable of surviving and operating effectively outside of highly controlled laboratory settings <ref:2508.15729#pg0>.
Dev: We also have to consider the power efficiency aspect, which they report at zero point six seven µW at zero point five V, suggesting that these low-voltage applications can be very power-conscious while still maintaining good performance <ref:2508.15729#pg0>.
Rosa: So, in simple terms, this paper presents a highly stable voltage reference that runs on very little power and stays accurate across a wide temperature range, which is what makes it relevant for robust field applications <ref:2508.15729#pg0>.
Conclusion: Rosa: So, to wrap up this discussion on "A sixteen point two eight ppm/ C Temperature Coefficient, zero point 5V Low-Voltage CMOS Voltage Reference with Curvature Compensation," we've seen how they managed to pack this level of precision into such a compact design running on just half a volt <ref:2508.15729#pg2>.
Dev: Exactly; the key takeaway is that achieving such high stability at these low supply voltages isn't just about squeezing the voltage down; it’s about intelligently compensating for the temperature changes that usually wreck those circuits.
Taro: And from an autonomy standpoint, having a reference voltage that maintains this accuracy across extreme temperatures means our navigation systems or sensor calibrations won't suddenly lose sync when we go into deep cold or intense heat.
Rosa: Right, and thinking about real-world deployment, does this stability translate to long operational time for a field robot before we need to recalibrate?
Dev: We’ve looked at the power consumption figures, and they report keeping that reference voltage within one percent of its target even when the operating environment swings wildly from minus forty degrees Celsius up to one hundred thirty.
Taro: If the system can handle those thermal stresses without significant drift, it opens up possibilities for autonomous agents operating in environments we haven't fully mapped yet.
Rosa: It really feels like we’re getting a solid foundation for more reliable hardware that doesn't need constant on-site adjustments, which is what field robotics demands.
Dev: The circuit design itself relies on careful mutual compensation between the PTAT and CTAT elements to counteract thermal drift, which is a neat engineering feat in CMOS technology.
Taro: What I find particularly interesting is how this approach addresses the uncertainty that comes with unpredictable external conditions in autonomous systems.
Rosa: It makes me wonder if we can use this kind of robust reference across multiple subsystems on a single platform for even better overall reliability.
More episodes
- 2610.11768-Narrow and Deep: An Ontology Tower as the Knowledge of an LLM Agent for an Industrial Equipment System
- 2610.11904-Large-Scale Partition-Based RIS Beamforming For Uplink RIS-Equipped Multi-User Systems: Asymptotic Analysis
- 2610.11885-Redefining fuel poverty: Introducing the temporal equity framework (TEF)
- 2610.11900-Reach-Stabilize Control of Control-Affine Systems with Unknown Affine Parameters
- 2610.11964-From Asymptotic to Designer-Assigned-Time Control: A Review of Stability Notions, Design Mechanisms, and Controller Architectures
- 2610.12226-Stabilization of Unidirectional First-Order PDE-ODE Coupled Systems with Boundary and Distributed Input Delays
- 2610.12028-Policy Synthesis for Finite Populations of MDP Agents under Aggregate Reach-Avoid Chance Constraints
- 2610.12103-Predefined-Time Integral Reinforcement Learning for Unknown Nonlinear Systems via Inverse-Optimal Design
- 2610.12110-Adaptive dynamic programming using Lyapunov function constraints
- 2610.12324-Convex Safety Filtering via Spectral Selection for Nonconvex Safe Sets