Khor , Wooi Kin
(2017)
Fault Isolation with ‘X’ Filter for Bogus Signals and Intensive Scan Cell Sequence Validation.
Masters thesis, Universiti Sains Malaysia.
Abstract
There are some concerns in silicon data collection by using the Design-For-Test (DFT). Bogus signal which carries ‘x’ value in simulation, results from the complex logic synthesis and power-up floating state can often mislead the fault isolation process with invalid failing condition. Besides, scan cells within the scan chain architecture is also having mismatch value in between the simulation data and silicon data due to the non-ideal mapping file passed down from the designer team. Hence, it is important to develop an integrated tool that can filter all the bogus signal online and to validate the correlation between silicon data and simulation data with minimum coverage of 90%. Data from actual Intel 6th generation microprocessor with 14 nm process technology, Skylake is imported to ensure that the application of this thesis in the current industry market. Necessary tools such as the “Differentiate and Display” feature to ease the analysis of data, the AND-logic operation to filter the bogus signal and X-OR logic operation to handle the inverted characteristic of signals are developed throughout the thesis. Results show that the developed integrated filter of bogus signals is successful and the minimum coverage of validation tool is 96.5%. Actual failure analysis case from industry is imported and the difference with and without the developed tools are compared. Inconclusive optical test result from the sample is obtained without the implementation of tools. On the other hand, defect of short circuit between the via and the metal line is found after the implementation of the developed tools. It is concluded that this thesis has achieved all the objectives set.
Actions (login required)
 |
View Item |
![[img]](http://eprints.usm.my/style/images/fileicons/application_pdf.png)
