Abstract
A new method of Differential Non-linearity (DNL) error correction for the successive approximation A/D converter (SAADC) is proposed.
The SAADC has very short conversion time and is very useful in various fields. But it has the largest DNL in comparison with any other type of ADC.
The DNL is the channel width irregularity of ADC, and is the most difficult error to deal with since it cannot be eliminated by adjustment.
The most famous research to equalize the channel width of SAADC is the Gatti's “Sliding Scale Method” proposed 28 years before approximately.
This new method of DNL error correction is based upon the peculiar feature of the SAADC.
First, the generation mechanism of DNL error is analyzed.
Second, a new method of DNL error correction is proposed.
Finally, the feasibility of this newly proposed method is confirmed through the computer simulations.
