Fast ADC testing by spectral and histogram analysis

The integral nonlinearity (INL) of analog to digital converters (ADCs) can be described by a behavioral error model expressed as one dimensional image in the code domain. This image consists of low and high code frequency components which allow describing the ADC performance with a small number of parameters. This paper presents new methods for low code frequency and high code frequency testing. The identification of the low code frequency components is performed by multi-harmonic sine fitting in the time domain. The high code frequency components are estimated in the statistical domain by a narrow band histogram test using a triangular stimulus signal. The performance of the proposed method is assessed for various ADC devices.