EGR Cooler Field Return Rate Evaluation Based on Product and Application Variation

The automotive industry drives some of the most stringent product requirements to ensure long product life and customer satisfaction. To demonstrate compliance with these requirements new and more accurate evaluation methods are needed. Thermal fatigue life in EGR coolers for heavy duty diesel applications have historically been a critical focus for engine OEMs. Being able to accurately evaluate product return rates due to thermal fatigue failures gives the OEM confidence that all end users will be satisfied, and allows program management to properly make fiscal decisions. Additionally, weight and cost optimization can be conducted with greater confidence. This is accomplished by accounting for product variation and application variation in thermal fatigue life evaluations. Including these variations requires a simplified numerical method to calculate product life, as tens of thousands of samples will be run through the analysis to represent real life random variation. First, time dependent component temperatures are calculated based on engine conditions. The component temperatures are used to determine strain, which is then used to evaluate thermal fatigue life. Appling this life calculation method to individual application duty cycles or all duty cycles allows field return schedules to be calculated. This method has been proven to accurately evaluate warranty return trends by studying a legacy EGR cooler, which has sufficient field experience. It was also found that specific application thermal fatigue life can be different than the thermal fatigue life of the entire fleet. This is a critical distinction as fleet life determines financial liability for the product, while specific application life is key to end user satisfaction. With a greater understanding of engine duty cycles, the physics of thermal fatigue, and with a new evaluation method it is possible to more accurately evaluate EGR cooler thermal life, which provides critical data for program management.