A Current Ratio Model for Defect Diagnosis using Quiescent Signal Analysis

I DDQ test has been used extensively both as defect reliability screen and as a defect diagnostic technique. However, the increase in subthreshold leakage currents in deep sub-micron technologies has reduced the effectiveness of I DDQ in these applications. Quiescent Signal Analysis (QSA) is a novel diagnostic technique that uses I DDQ measurements made at multiple supply pads on the Chip-Under-Test as a means of locating shorting defects in the layout. The use of multiple supply pads reduces the adverse effects of leakage current by scaling the total leakage current over multiple simultaneous measurements. In previous work, a resistance model for QSA was developed and demonstrated on a small circuit. In this paper, the weaknesses of the original QSA model are identified, in the context of a production power grid (PPG) and probe card model, and a new model is described. The new QSA algorithm predicts the position of the defect in the layout through the analysis of hyperbolic current ratio contours. SPICE simulation experiments are used to demonstrate the improved prediction accuracy of the new model on a portion of the PPG. I DDQ has been a mainstream supplemental testing method for defect detection for more than a decade with many companies. With the advent of deep submicron technologies , the use of single-threshold I DDQ technique results in unacceptable yield loss. Setting an absolute pass/fail threshold for I DDQ testing has become increasingly difficult due to the increasing subthreshold leakage currents [1]. Current signatures [2], delta-I DDQ [3] and ratio-I DDQ [4] have been proposed as a means for calibrating for these high subthreshold leakages. These techniques rely on a self-relative or differential analysis, in which the average I DDQ of each device is factored into the pass/fail threshold. However, these proposed forms of calibration are expected to become increasingly less effective over successive technology generations. An alternative calibration strategy that may have better scaling properties is to distribute the total leakage current across a set of measurements. This is accomplished by introducing probing hardware that allows access to the individual supply port I DDQ values. The method proposed in this work, called Quiescent Signal Analysis (QSA), is designed to exploit this type of leakage calibration for defect detection and as a means of providing information about the defect's location in the layout [5][6]. A resistance-based model for QSA was developed in previous works and simulation experiments were …