Automated Characterization Of Transient Threshold Voltage Shifts In MOSFETs

One of the most important parameters in the operation of the metal-oxide semiconductor field effect transistors (MOSFETs) is the threshold voltage.  It establishes when the transistor begins conducting. This voltage in both, N- and P-channel devices, changes when transistors are subjected to pulsing signals and varies with pulse amplitude and frequency.  Methods for characterizing such shifts in discrete devices have been studied and practical circuits for quantifying the change have been proposed.  However, proposed solutions require manual calibration to take measurement.  Manual calibration processes are time consuming and thus, they are not cost effective for production line testing. The objective of this project is to devise ways of reducing the calibration time, enabling the automation of the characterization process. Several solutions were explored, but discarded due to noise generation. We propose to create a computer algorithm to provide for a software-based virtual calibration mechanism. We hypothesize that a software solution shall eliminate the necessity of manual calibration without affecting the accuracy of measurements. Using LabVIEW as the implementation platform, allowed us to deploy machine controlled testing procedures that can be integrated into semiconductor device production lines.  Preliminary results obtained through this method allowed us to validate our hypothesis. Measurements of the threshold voltage shifts and relaxation times, in both packaged and wafer level devices show that our virtual calibration process introduces negligible errors with respect to the manual calibration.  Moreover, test time is reduced in several orders of magnitude while providing a procedure amenable for automated execution.