Reliability HotWire

Issue 110, April 2010

Accelerated Degradation Analysis Using ALTA

[Editor's Note: This article has been updated since its original publication to reflect a more recent version of the software interface.]

Degradation analysis involves the measurement and extrapolation of degradation or performance data that can be directly related to the failures of a product. Many failure mechanisms can be directly linked to the degradation of the parts of a product, and degradation analysis allows the user to extrapolate to an expected failure time based on the measurements of degradation or performance over time. When accelerated testing techniques are used to reduce testing time, the analysis of the degradation measurements obtained at the accelerated stress levels can be extrapolated to normal use conditions. In this article, we will give an example that uses the accelerated degradation analysis tool in ALTA.

Consider one type of light-emitting diode (LED) in which the light intensity changes with the current stress and the time. A light intensity of 55 mcd is set as the failure criterion, or critical degradation. The manufacturer wants to know the B10 life at the usage current of 27 mA. However, the time schedule for the testing task is very tight. To reduce the testing time, the testing group decided to conduct an accelerated degradation test and use the accelerated degradation analysis tool in ALTA to analyze the data.

The group tested 10 samples with 5 samples at each of the 2 accelerated current stress levels, 35 mA and 40 mA. Table 1 shows the different light intensity measurements collected during the test at different currents.

They entered the collected data into ALTA's degradation analysis folio. For the extrapolation, they used an exponential model for the degradation curve and set the critical degradation to 55, as defined above. To analyze the extrapolated failure times, they chose the inverse power law as the life-stress model, and selected the Weibull distribution as the failure time distribution.  They also defined the normal use stress level of 27mA by clicking Set Use Stress in the control panel and entering the value in the Use Stress Level window.

Figure 1 shows the resulting 10 exponential curves that were used to extrapolate the failure times.

Figure 1: Degradation Curves in ALTA

After calculation, the control panel shows the parameter values for beta and K, as shown in Figure 2.

Figure 1: Analysis of Failure Times Extrapolated from the Degradation

To calculate the B10 life, they opened the Quick Calculation Pad and entered 27 for the stress value, and entered 10 for the BX% Life At value. The B10 life for a current of 27 mA is about 1176 hours, as shown in Figure 4.

Figure 4: Calculating the B10 Life using the Quick Calculation Pad

Conclusions

In this article, we discussed degradation analysis using ALTA. As you may know, Weibull++ also has a degradation analysis tool. The difference between these two degradation analyses is that in Weibull++, the analysis is conducted at only one stress level (the use stress level) and predictions are valid only for that stress level. In ALTA, the analysis is conducted at elevated stress levels, and the extrapolation from the elevated stress levels to the use stress level is based on the life-stress relationship. Product performance can be predicted either at the use stress level or at any level for which that life-stress relationship is applicable.

Copyright ® 2010-2014 ReliaSoft Corporation, ALL RIGHTS RESERVED