Characterization of mechanical systems with real joints and flexible links
The field of multibody simulations has seen great strides in recent years. Most multibody simulations are carried out under the assumption that the joints involved are ideal. However many real joints are not ideal; they possess some clearance between their components. The existence of these clearances allows for the freeflight motion of some of the joint’s constituent components. This motion is usually followed by an impact leading to load amplification and chaotic motion. The effects of impact-induced load amplification could be compounded when the system also contain members or links with some degree of flexibility. This study is an attempt to observe and characterize systems of this nature. The issue of characterizing these systems is addressed incrementally. Initially, a test rig with a clearance or real joint and rigid links is used to gather data on the characteristics of a real system with rigid links and a real joint for various system conditions. The resulting data is analyzed and used to construct a validated analytical model of the same system. This model is then used as a test bed to study the contact and hence the impact process occurring within the joint. This approach allows for various contact models to be evaluated and for a suitable one to be selected or formulated. Once this is done, flexibility is introduced into the study by modifying the test rig to include flexible bodies, and the new data is analyzed for any possible effects brought on by body flexibility. As expected and seen in other previous studies, joint clearance does indeed cause load amplification. Various existing impact models for this joint were studied and none were found to give satisfactory results, so a modified formulation was made and found to produce significantly better results. It was noted that during the impact process, damping, even in very small amounts, plays an important role and may not be ignored safely as is sometimes done. Flexibility in links is found to have an influence on the system’s responses depending on the system configuration, but by and large, it appears to have a deleterious effect. The phrase "flexible body” seems to imply a one-to-one correlation between the stiffness of a body under quasi static conditions and its dynamics in a working mechanism. It would seem from this study that such an assumption is not always true and perhaps a better term would be “real body."
Thesis (Ph.D.)--Wichita State University, College of Engineering, Dept. of Mechanical Engineering
Includes bibliographic references (leaves 97-101)