A calibration method for non-binary successive approximation register analog-to-digital converters based on code density

Abstract

Successive Approximation Register Analog-to-Digital Converters (SAR ADCs) are significantly impacted by capacitor mismatch during the quantization of analog signals, which severely affects ADC accuracy. This paper builds upon code density calibration based on the binary algorithm to calibrate a non-binary redundant SAR ADC designed for 12-bits resolution at 187.5 MS/s. A comparative analysis of the two search algorithms before and after calibration was performed. Fifty spectrum analyses were conducted under capacitance mismatch levels of 0.5%, 1%, and 10%. The mean values of Signal-to-Noise-and-Distortion Ratio (SNDR), Spurious-free Dynamic Range (SFDR), and effective number of bits (ENOB) were calculated. Under capacitor mismatches of 0.5% and 1%, the code density method demonstrated effective calibration results for ADC using both search algorithms. However, under 10% mismatch, before calibration of binary algorithm: ENOB=4.58bits, SFDR=31.54dB, SNDR=26.45dB, after calibration: ENOB=6.71bits, SFDR=57.58dB, SNDR=49.19dB. Before calibration of non-binary search algorithm: ENOB=3.93bits, SFDR=28.4dB, SNDR=25.8dB, after calibration: ENOB=11.82bits, SFDR=88.24dB, SNDR=73.28dB. Therefore, it can be concluded that the code density calibration method is more effective for calibrating non-binary redundant SAR ADCs.