What is Quality?
Quality is a measurement of how close your product conforms to customer requirements and expectations and also a measure of the probability that your product will perform as it’s supposed to perform under specific conditions for a stated time-period.
What is reliability?
Reliability represents the wear out over time of a product. High reliability means that the devices have a very low probability of failing within a few years. Most semiconductor devices have lifetimes over many years. However, you cannot wait years to analyse the performance of a device; you have to increase the applied stress to simulate performance later in the life cycle of the device. These accelerate fail mechanisms, help with root cause analysis, and help semiconductor companies take actions to prevent the failures later in the life of the product. yieldHUB supports the study of these added stresses such as applied in HTOL testing.
Quality and Reliability in yieldHUB
The data generated in testing semiconductors that is available in yieldHUB will help capture the quality and reliability of the products and devices in question.
Some of the tools within yieldHUB that promote quality and reliability are as follows:
Drift Analysis (for burn-in, HTOL and life-test, supporting these key reliability processes)
Sensitivity analysis (e.g. which tests are sensitive to a program or material change?)
Program change detection (how has the program changed between revisions, e.g. any tests missing?)
AEC Parts Average Testing (for automotive)
Good Die in Bad Neighborhood (popular and not just for automotive)
Lot Norm (Lot Anomaly) detection for Aerospace
The analysis of High-temperature operating life (HTOL) reliability testing or other forms of Drift Testing is supported in yieldHUB. The "Per Die Trend" chart brings you into a comprehensive analysis of the drift of Part ID between sets of conditions such as hours and changed power supply setting which is common in HTOL testing. You can also see how the population drifts and the drifts relative to the original values and test limits.