The effect of derating on component selection and product design
Kunwar Bahadur Singh, R. K. Saket, S. Chatterji
Abstract
Kunwar Bahadur Singh, R. K. Saket, S. Chatterji
Abstract
Derating affects heavily the component selection and the thermal design of products. Therefore, it would be expected that this procedure is well motivated and that the de-rated temperature is selected so that the component after derating meets the reliability requirements, but on the other hand, the derating should not add too large safety margins. Too large safety margins may easily result in unnecessary and expensive cooling arrangements. In this paper, the effect of using linear derating guidelines on lifetime is discussed and an alternative derating approach, that takes better into account the temperature dependency of the lifetime, is introduced. The effect of parameter uncertainties on lifetime prediction is also explained. Finally, a new derating procedure that uses a physical lifetime model and compensates for the parameter uncertainty is proposed.
OpenAlex reports 2 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Derating affects heavily the component selection and the thermal design of products. Therefore, it would be expected that this procedure is well motivated and that the de-rated temperature is selected so that the component after derating meets the reliability requirements, but on the other hand, the derating should not add too large safety margins. Too large safety margins may easily result in unnecessary and expensive cooling arrangements. In this paper, the effect of using linear derating guidelines on lifetime is discussed and an alternative derating approach, that takes better into account the temperature dependency of the lifetime, is introduced. The effect of parameter uncertainties on lifetime prediction is also explained. Finally, a new derating procedure that uses a physical lifetime model and compensates for the parameter uncertainty is proposed.
Key concepts: Derating, Component (thermodynamics), Reliability engineering, Reliability (semiconductor), Selection (genetic algorithm), Dependency (UML), Computer science, Engineering