Global Enterprise-Wide Test Management Software Solutions.  Actionable Data. Measurable Results.

Increased Yield and Yield Learning

Optimal+ solutions increase your yield by reclaiming devices wrongly labeled “bad” in both real-time and offline, as well as through early detection capabilities utilizing Optimal+-Detect. This allows organizations to track performance metrics more effectively and helps to ensure maximum values for yield, quality, utilization and throughput are achieved. Using Optimal+’s solutions, customers have realized yield increases of between 1% - 3% at wafer sort and final test.

Yield improvements are achieved primarily through the unique combination of:

  • A reliable and complete database of test data
  • Pro-active rules which catch problems as soon as they start to occur (real-time)
  • Additional rules executed offline on the database to compare yield across the test fleet
  • Early detection utilizing Optimal+-Detect to track, capture and detect extreme changes in your products' manufacturing, test or assembly
  • Powerful tools to analyze that data

Additional yield learning can be achieved by augmenting your test process on sampled dice or parts with additional tests.

Examples

Maximizing Yield through Probecard Performance Analysis [read more]

Finding the Issue

A high-level product-based report in OT-Portal showed that a product was showing lower than expected yield and higher than expected retest rates. 

Performing the Analysis

By drilling down to individual tester performance in OT-Portal, the engineer was able to identify a specific probe card, which was causing the issue within a matter of minutes.  The test house was notified and the probe card was removed for inspection.

Preventing Future Recurrences

A rule was created in OT-Rules to automatically catch lots exhibiting high site-to-site yield discrepancies.  The next time this problem occurred an email alert was sent immediately to the user so that the problem could be resolved straight away.

 

Maximizing Yield
Step 1 – Analyzing a probecard showed site 3 with consistently low yield

Site to Site Deviations
Step 2 – Creating a rule to catch site-to-site deviations for this product

Alert
Step 3 – Receiving an alert when the problem next occurs

Maximizing Yield through Probecard Performance Analysis [pdf, 471 KB]

 

Saving Yield by Managing Probe Cards [read more]

The Challenge

A large fabless company tests a number of different products at various test houses around the globe.  Each test house maintains its own supply of probe cards for each of the tested products.  Using OT-Portal, the fabless company detected that significant yield loss is attributed to poorly maintained probe cards.  The company wanted a mechanism to alert when a probe card was nearing the end of its life span or requiring preventative maintenance, before yield loss started to occur.

The Solution

Test data collected in the Optimal+ Database contains an accurate count of touchdowns and cleans performed by each probe card.  A “target” was defined in the Optimal+ database for each individual probe card specifying the number of touchdowns at which an alert was to be triggered.  When the alert was triggered, a request was sent to the supplier to perform preventative maintenance.  When the probe card was re-instated, the target was updated with the new value for the next maintenance cycle.
OT-Portal views were defined to display aging probe cards together with data from probe card related rules showing issues with probe cards.  In the first example, the dashboard shows rule alerts by probe card with a summary of the top “offender” probe cards that triggered the most alarms for yield/quality issues.  The second example contains information about each probe card with its performance and health status including total touchdown count, average lot yield by site, bin summary by site etc… It also shows the target spec for each probe card and its status in relation to that spec (the colored column.

SavingYield1

 

Saving Yield 2

 

Saving Yield by Managing Probe Cards [pdf, 422 KB]

 

Uncovering Geographic Issues at Final Test (Data Feed Backward) [Read More]

Finding the Issue

When a chip has an Electronic Chip ID (ECID) it is possible to discover wafer geography issues based on final test data.  Even though the parts passed at wafer sort, they failed at final test and the fallout was related to wafer geography and fab process issues.

The trigger for performing this analysis was a final test parametric test that had a higher than expected failure rate.  This was discovered using a failing test pareto chart in OT-Portal.

Performing the Analysis

In this example, an engineer used the wafer map reconstruction capabilities of OT-Portal to analyze the fallout of a parametric test performed at final test.  Wafer map reconstruction is an excellent tool for yield learning analysis. The wafer clearly shows that most of the failures occurred near the edge of the wafer.  The impact of this issue is that failures, which could be detected at wafer sort, were deferred to final test, causing costly and unnecessary packaging of bad devices.

This issue was resolved by created a screening test in the wafer sort process so that these problems were caught earlier.  The fab was notified so that improvements could be made to the wafer manufacturing process.

We call this useful capability “Data Feed Backward” – taking data from a later operation and using it for yield learning in earlier operations.

Data Feed

Uncovering Geographic Issues at Final Test (Data Feed Backward) [pdf, 279 KB]

 

Using Rules to Catch ATE Issues [Read more]

Finding the Issue

A by-8 probecard was failing on all sites in some touchdowns due to a tester issue.  This problem was detected automatically by an offline touchdown-monitoring rule defined in OT-Rules.  Standard yield monitoring mechanisms failed to discover the issue because wafer yield was still above the acceptable threshold.

Performing the Analysis

The user received an alert via email minutes after wafer probing was completed.  The user viewed the OT-Portal wafermap tool and informed the test house.  The wafer was re-probed, significant yield was reclaimed and the user was able to view the results of the retest in OT-Portal.  The tester was investigated to find the root-cause of the problem.

Wafermap

Using Rules to Catch ATE Issues [pdf, 595 KB]

 

end faq

 

 

Companies on the Highway...
AMD     Amkor     ASE     Broadcom     Global Foundries     KYEC     Marvell    Nvidia     Qualcomm     Samsung     SMIC     SPIL     ST     StatsChipPAC     TSMC     UMC     UTAC