Athol J. Carr. BE(Hons)(Civil), MSE, Ph.D.(Calif). F.IPENZ, M.ASCE
Athol J. Carr graduated from the University of Canterbury with a Bachelor in Engineering (First Class Honours) (Civil) in May 1964. After working as a design engineer in New Zealand he went to the University of California, Berkeley, and completed a Master of Science in Engineering (in Structural Engineering) in May 1966 and a Doctor of Philosophy (in Structural Engineering) in May 1967. Both degrees were under the supervision of Professor Ray W. Clough.
Dr. Carr joined the Department of Civil Engineering at the University of Canterbury in January 1968. Here he has pursued research and teaching interests in structural engineering, with an emphasis on computer based methods of analysis, structural dynamics and in finite element analyses. Since the early 1970s he has been actively involved in earthquake engineering, non-linear dynamic analyses and in software engineering. After 43 years teaching at the University of Canterbury he retired in 2010 as Professor Emeritus at the University of Canterbury.
He has given his course on advanced dynamic analyses at the The Rose School in Pavia Italy, in 2005, 2008, 201 and 2015.
He has developed the RUAUMOKO dynamic analysis computer program suite over the past 40 years and which is now used in over 130 universities, building research institutes and consultancies around the world for research, teaching and design analyses. Research work on new analysis and damping models is still taking place with graduate students at the University of Canterbury as well as directly within Carr Research.
Improved Models for Damping in Inelastic Seismic Analyses.
Athol J. Carr
Professor Emeritus, University of Canterbury, Christchurch, NZ
For several decades the modelling of damping in non-linear dynamic analyses has used a suggestion made by Lord Rayleigh in the 1880s. Rayleigh recognised that the damping phenomenon was probably hysteretic but as stated in his book he used a viscous damping model for mathematical convenience. Somehow with the passing 125 years engineers have generally not adopted more realistic models. Most current analysis methods still use Rayleigh’s suggested expedient of a viscous damping model (which now bears his name). The Rayleigh damping model assumes that the damping is viscous which is physically difficult to justify, and it implies very high damping for rigid-body modes which may be a problem for base-isolation studies, but also implies very high damping ratios for the higher modes of free-vibration. This high damping in the higher modes was shown to cause severe accuracy problems in the late 1970s, and although attempts have been made to mitigate these problems, Rayleigh damping still appears to hold sway even in the latest engineering text books. Recent work has developed a series of new damping models which provide a damping associated with each of the members in the structure, which means that different levels of damping (viscous, or non-viscous,) can be associated with different parts of the structural system. This cannot be achieved in analyses where damping is ascribed to the structure as a whole. These new damping models are shown to be able to represent the decay phenomenon observed in the free-vibration of engineering structures and yet give more realistic representations of where the dissipation actually occurs within the members which make up the structure.