The effectiveness of the operation of digital control systems, traffic safety and diagnostics depends to a large extent on the accuracy of the analog-to-digital con- version. Methodical errors of the quantization transformation, time sampling and the method of restoration of the input signal. This article is devoted to the analysis SHAPE \* MERGEFORMAT of the influence of sampling and storage devices of voltage-code converters on the accuracy of analog-to-digital conversion. Models of transformation with instanta- neous counts and taking into account sampling and storage devices are considered, errors of transformation of random stationary analog signals for these models are determined. The difference in error estimates for models with instantaneous read- ings and sample-and-hold circuits (SHC) makes it possible to determine the influ- ence of SHC on the accuracy of the conversion. Taking into account that to increase noise immunity of the conversion, averaging of a given number of digital samples within the step of time sampling and tracking digital averaging is used, the cor- responding expressions for estimating errors on models taking into account digital averaging at instantaneous counts and taking into account sampling and storage devices are obtained. An estimate has been introduced that allows one to determine the effect of sampling and storage devices on the conversion error. A methodology for calculating this estimate has been developed. Examples of the transformation of a random signal with a constant spectral power density in the frequency band from 0 to ω max and with a power spectral density decreasing with increasing frequency are given. A technique has been developed that allows analyzing the effect of sampling and storage devices on the analog-to-digital conversion error of various stationary random signals.