Search Results

A powered seat adjuster is a complex mass-produced assembly that is heavily optimized for low cost and light weight. The consequence is an inevitable degree of uncontrolled variation in components, subassemblies, and final product. Automakers are driving an exceptional focus on quality and the showroom experience of the car buyer is paramount. Therefore, any seat adjuster with the potential to not satisfy the customer's expectation is likely to be screened on the production line. This paper describes NVH metric design in the context of automated production line detection of seat adjuster defects. A key requirement of the production environment is that the metrics offer intuitive explanations of possible defects and are based on industry-standard formulations. The metric set is a hybrid of objective and subjective parameters with a focus on ensuring a robust sorting process that maximizes detection while minimizing the possibility of failing acceptable product.

A new method for extracting order related signal components from rotating machinery data is proposed. The notion of the Gabor expansion and the Gabor transform for time/frequency analysis are introduced and the relationship between them is described. Gabor Order Tracking (GOT) is offered as one possible masking strategy for extracting desired signal components from the time/frequency domain for reconstruction in the time domain. Practical examples are utilized to illustrate the proposed method in comparison to established techniques.

A new approach of estimating the modal parameters of operating piston engines is presented. The developed approach represents a combination of concepts from currently existing analyses such as the natural excitation technique (NExT), conditioned input analysis (CIA), and conditioned source analysis (CSA), and is hence termed “conditioned NExT analysis (CNA)”. NExT analysis can be employed to estimate modal parameters of structures in their naturally excited states. However, the existence of strong combustion induced periodic forcing makes the application of NExT analysis to operating engines difficult, if not impossible. CIA and CSA, built on concepts of partial and virtual coherence respectively, can effectively condition operating engine vibration data so as to remove any periodic energy associated with the process of combustion.