论文部分内容阅读
运用加速步进应力和恒定的应力与时间关系的试验技术证实了器件在使用状态的可靠性预测模型。该模型是从小信号二极管、小信号及功率晶体管、硅可控整流器和金属氧化物变阻器上获得的。每种器件都遵循可靠性预测的阿列尼乌斯模型。己证明本技术可用来确定:(1)小信号二极管在极高的加速试验状态与现场工作状态之间的加速因子;(2)高应力水平和使用状态之间的加速因子或倍增因子,该使用状态能用来预测小信号二极管在超时间工作时的性能。讨论了潮湿对器件可靠性的影响。功率器件在工作时比周围环境有较低的相对湿度(RH)。这种低的相对湿度对潮湿的激活机构起了抑制作用。叙述了用来淘汰有早期制作型缺陷的晶体管的高可靠筛选技术。这种筛选对于改进电子系统的可靠性是有效的和经济的。综合考虑若干项影响可靠性的加速因子技术,已在二极管上得到证实。同样的技术也适用于多数的器件可靠性预测。但加速因子不能外推到超过器件最大额定值的应力水平,在这种应力下将会产生新的失效模式,它不遵从原有的失效率与应力之间的关系。本论文中给出的失效率直线范围都终止于这些阈值极限。
The test technique of accelerating step stress and constant stress versus time confirms the reliability prediction model of the device in service. The model is derived from small signal diodes, small signal and power transistors, silicon controlled rectifiers and metal oxide varistors. Each device follows the Arrhenius model of reliability prediction. It has been demonstrated that this technique can be used to determine: (1) the acceleration factor of the small signal diode between the very high acceleration test condition and the field operating condition; (2) the acceleration factor or multiplication factor between the high stress level and the service condition The state of use can be used to predict the performance of small-signal diodes when operating over time. The effect of moisture on device reliability is discussed. The power device has a lower relative humidity (RH) than the ambient environment at work. This low relative humidity dampens damp activation mechanisms. Describes highly reliable screening techniques to eliminate transistors with early fabrication defects. This screening is effective and economical to improve the reliability of electronic systems. Considering a number of acceleration factor techniques that affect reliability has been demonstrated on the diode. The same technique is also suitable for most device reliability prediction. However, the acceleration factor can not be extrapolated to a stress level exceeding the device maximum rating. Under this stress, a new failure mode will be created which does not obey the original relationship between failure rate and stress. The linear range of failure rates given in this paper ends at these threshold limits.