A loopback-based INL test method for D/A and A/D converters employing a stimulus identification technique

We propose a new method for the integral nonlinearity (INL) and differential nonlinearity (DNL) testing of D/A-A/D converter pairs employing the recently developed stimulus identification method. This allows both converters to be measured independently but simultaneously without significant fault masking problems. Simulations show that the INL and DNL estimation errors for 12-b A/D and D/A converters are less than 0.5 least significant bit (LSB) units, and experimental tests give similar results.