Browse Prior Art Database

Method of Using Parametric Data to Screen Modules in Final Test to Safeguard Product Quality

IP.com Disclosure Number: IPCOM000247050D
Publication Date: 2016-Jul-29
Document File: 3 page(s) / 71K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a method of testing integrated circuits at the final test area. The method is comprised of steps to check whether the lot is maverick, compare the parametric distribution of the good modules from a non-maverick lot(s), and rescreen the maverick lot with a test program using the analyzed predetermined test limit.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 3

Title

Method of Using Parametric Data to Screen Modules in Final Test to Safeguard Product Quality

Abstract

Disclosed is a method of testing integrated circuits at the final test area. The method is comprised of steps to check whether the lot is maverick, compare the parametric distribution of the good modules from a non-maverick lot(s), and rescreen the maverick lot with a test program using the analyzed predetermined test limit.

Problem

Semiconductor final tests often encounter maverick production lots after testing. Once a maverick occurs, it takes a lot of effort to determine the root cause of the problem, which eventually puts the lots on-hold. In addition, the lot might be sent to scrap or undergo burn-in and other stress test in order to not sacrifice product quality and reliability.

Solution/Novel Contribution

The novel contribution is a method of understanding and analyzing the parametric data in order to salvage the lot and prevent it from becoming scrap, without jeopardizing the quality of the good modules to be shipped to the end customer. This process translates to business revenue and reduces the holding time of maverick lots.

Method/Process

Besides the difference on the process flows, one key differences between the current process and the novel method is the length of time to "Hold the lot". The process on record (POR), might take months due to the stress test involved. For the new ID process flow, it might only take a few days.


Page 02 of 3

Figure 1: Comparison of process flows

To implement the process in a preferred embodiment:

Step 1: Test a lot in Final Test (See ID Process flowchart in Figure 1). Once the lot encounters a maverick due to parametric fails, the method proceeds to analyze the parametric data to check the distribution of the readings.

Step 2: Compare the maverick lot MV LOT 1 to non-maverick lots LOT 1 to LOT 5 of the same...