2009Unpublished venueRequires access

Design of an 8-bit pipelined ADC with lower than 0.5 LSB DNL and INL without calibration

El-Sayed Eid, Hassan Eldib

Open publisher page 4 citations

Abstract

This paper describes the design of an 8-bit fully differential pipelined analog-to-digital converter (ADC). The design methodology employed in this work follows a technique of allocating appropriate error budgets to the various ADC errors such that the maximum differential nonlinearity (DNL) error is less than 0.5 least significant bits (LSB). Simulation results show that the ADC maximum DNL errors are +0.3/-0.45 LSB, and maximum integral nonlinearity (INL) errors are +0.3/-0.35 LSB. These low figures of nonlinearity errors are achieved without applying any calibration techniques. Simulation also shows an ADC power consumption of only 1.32 mW for a 1.2 V supply, an effective number of bits (ENOB) of 7.94 bits and a figure of merit (FOM) of 2.63 pJ. The pipelined ADC is designed in a 0.13 ¿m CMOS process, resulting in a very small active area of 161 ¿m × 139 ¿m.

About this research paper

What this paper is about

This paper describes the design of an 8-bit fully differential pipelined analog-to-digital converter (ADC). The design methodology employed in this work follows a technique of allocating appropriate error budgets to the various ADC errors such that the maximum differential nonlinearity (DNL) error is less than 0.5 least significant bits (LSB). Simulation results show that the ADC maximum DNL errors are +0.3/-0.45 LSB, and maximum integral nonlinearity (INL) errors are +0.3/-0.35 LSB. These low figures of nonlinearity errors are achieved without applying any calibration techniques. Simulation also shows an ADC power consumption of only 1.32 mW for a 1.2 V supply, an effective number of bits (ENOB) of 7.94 bits and a figure of merit (FOM) of 2.63 pJ. The pipelined ADC is designed in a 0.13 ¿m CMOS process, resulting in a very small active area of 161 ¿m × 139 ¿m.

Why it matters

OpenAlex reports 4 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

This paper describes the design of an 8-bit fully differential pipelined analog-to-digital converter (ADC). The design methodology employed in this work follows a technique of allocating appropriate error budgets to the various ADC errors such that the maximum differential nonlinearity (DNL) error is less than 0.5 least significant bits (LSB). Simulation results show that the ADC maximum DNL errors are +0.3/-0.45 LSB, and maximum integral nonlinearity (INL) errors are +0.3/-0.35 LSB. These low figures of nonlinearity errors are achieved without applying any calibration techniques. Simulation also shows an ADC power consumption of only 1.32 mW for a 1.2 V supply, an effective number of bits (ENOB) of 7.94 bits and a figure of merit (FOM) of 2.63 pJ. The pipelined ADC is designed in a 0.13 ¿m CMOS process, resulting in a very small active area of 161 ¿m × 139 ¿m.

Key concepts: Differential nonlinearity, Integral nonlinearity, Effective number of bits, Least significant bit, Figure of merit, Calibration, CMOS, Computer science

Related papers

Back to paper searchBrowse research topicsOriginal source
Design of an 8-bit pipelined ADC with lower than 0.5 LSB DNL and INL without calibration — Research Paper | ScholarLens