Browse Prior Art Database

Adaptive In-Test Tool Yield Performance Monitor

IP.com Disclosure Number: IPCOM000176338D
Original Publication Date: 2008-Nov-13
Included in the Prior Art Database: 2008-Nov-13
Document File: 2 page(s) / 123K

Publishing Venue

IBM

Abstract

Disclosed is method for adaptively implementing a yield performance monitor during wafer level or package level test. Benefits include, setting accurate test hardware cleaning cycles, reduced product retest, improved product yield, and test platform independent yield monitoring.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 54% of the total text.

Page 1 of 2

Adaptive In-Test Tool Yield Performance Monitor

Background:
In an ideal wafer level or package level production test environment,

incurred as a condition of the product test equipment is considered to be negligible. However, when testing products in high volumes, observed product yields degrade as the number of tested

products increases. This trend is easily explained by a buildup of foreign material on the

product's front end hardware (FEH) contact pins originating from the devices under test (DUTs).

Degrading electrical contact with the DUT increases the number of false parametric, AC, and ultimately open test fails as the resistance in the electrical test path increases.

    Poor electrical contact with the DUTs therefore requires retesting products leading to significantly increased test cost, reduced product FEH lifetime, unintended product scrap of good devices, reduced product quality, and reduced test equipment capacity.

General Description:

    The disclosed method is an adaptive yield monitor implemented directly from the product test program. The product test program continuously monitors yield variations over a discreet historical sample of devices when compared to the overall product population yield. When the discreet sample yield deviates from the population yield by a static limit, the product test

program delivers a signal to the test equipment to halt testing. A test equipment operator or an

automated cleaning routine resolves the test hardware issue.

Typical Embodiment:

Data structures created in the test program track the product population yield and the

product discreet yield. After testing each device, the population yield and the discreet yield over

the previous n number of devices is calculated. If the discreet yield falls below the population by a set percentage, the product test program sends a signal to the test equipment to suspend testing or to start a cleaning cycle. Once the operation is cleared...