Using reduced-order models in D/A converter testing

Static models of D/A converters can efficiently be used for testing purposes. For high-resolution converters nonparametric nonlinearity tests are not practically feasible for production testing due to the long test time. Modeling the device error response using a limited number of model parameters, the minimal number of required test points is significantly decreased. As a consequence, it becomes possible to estimate the error characteristics of the device from a few test points. In the paper, methods to select model parameters are discussed. The results are verified by the processed measurement data of 100 D/A converters.

[1]  J. Schoukens,et al.  Static nonlinearity testing of D/A converters , 2001, IMTC 2001. Proceedings of the 18th IEEE Instrumentation and Measurement Technology Conference. Rediscovering Measurement in the Age of Informatics (Cat. No.01CH 37188).

[2]  Gerard N. Stenbakken,et al.  : Modeling and Test Point Selection for Data Converter Testing , 1985, ITC.

[3]  Gene H. Golub,et al.  Matrix computations , 1983 .

[4]  Gerard N. Stenbakken,et al.  LINEAR ERROR MODELING OF ANALOG AND MIXED-SIGNAL DEVICES , 1991, 1991, Proceedings. International Test Conference.

[5]  Michael Peter Kennedy,et al.  A rigorous exposition of the LEMMA method for analog and mixed-signal testing , 1999, IEEE Trans. Instrum. Meas..

[6]  G. Stenbakken,et al.  Developing linear error models for analog devices , 1993, 1993 IEEE Instrumentation and Measurement Technology Conference.

[7]  Yves Rolain,et al.  Static nonlinearity testing of digital-to-analog converters , 2001, IEEE Trans. Instrum. Meas..