No abstract

No abstract

No abstract

The goals of this paper are to discuss some issues involved in integrating simulation software into an undergraduate dynamics course, and to suggest an approach to this task. The software package in question is MSC -ADAMS ® , a package that is currently used worldwide in industry for advanced mechanical simulation. The course is MECH -310 (Mechanics III) and is currently being taught at Kettering University by the author.There is a real need in industry for mechanical engineers with CAE experience and the author finds that introducing computational tools into this course is an excellent way of initiating students into mechanical simulation. Also, through simulation of textbook examples, students can observe problem solutions for full intervals of time, in contrast to traditional methods, which in general allow only for "snap-shot" type solutions (i.e. restricted to an instant in time). This can lead to an improved understanding of the subjects being taught.The choice for a high-end simulation package is based on the principle of teaching a tool that can be used during the students' senior design courses and in their professional careers as well.Carrying out the integration of the software into the course can pose some challenging issues. For example, integrating software teaching and assignments into an already full course schedule. Here the author suggests a Web-based approach. Student feedback on the integration is included and an example assignment is also given in the paper.

A driving shaft coupled to a driven shaft by a universal joint is considered. The shafts are taken to be rigid and motion is restricted to one plane. The non-homogeneous differential equation of motion has time-dependent coefficients and both parametric and forced resonances can occur. Here the question of whether one can ‘‘drive’’ through the resonances using a driving angular velocity linear variation is addressed. Also, how long can one ‘‘dwell’’ at a potential resonance before actually encountering it is investigated. Numerical studies led to the following conclusions. For linearly increasing speed profiles no practically feasible sweep rates to avoid resonance build up were found for the forced motion resonances. For certain torque and damping values, parametric resonances are seen for slow angular velocity variations but they are not observed for practically feasible fast variations, thus raising the possibility that one can accelerate through them. For a trapezoidal speed input the dwell time is key in building up instabilities. As the dwell time increases larger response amplification is observed. For the case studied, it was shown that it is possible to drive through the instability if the dwell time is equal to or less than forty times the period of the parametric excitation.