This paper on "Practical Contact Modelling: Examples, Oddities and Verification Techniques" was presented at the NAFEMS World Congress on The Evolution of Product Simulation From Established Methods to Virtual Testing & Prototyping - 24-28 April 2001, The Grand Hotel, Lake Como, Italy.

Abstract

At the 2000 NAFEMS AGM in Stratford-upon-Avon, UK the author was asked to present a paper detailing the main problem areas that designers were aware of when faced with Finite Element Analysis. Talking to design engineers attending the "Basics of Stress Analysis" and the "Fundamentals of FEA" (two NAFEMS-accredited courses run by the author) the following problem areas were identified:

• Need for rapid model building from CAD - cleaning and mesh quality problems
• Overuse of constraints and tendency to automatically pair them up "against" the loads
• Convergence/accuracy - tendency to trust the first results produced without questionAbsolute accuracy/failure criteria - significance of "elastic" stresses
• Results assessment and interrogation - tendency to rush the results interrogation phase.

Within the problem areas identified, by far the most widespread problem was that of faithful representation of constraints and boundary conditions. A large majority of those questioned seemed to want to apply loads "at one end of the structure" and reciprocal constraints "at the other end of the structure". The understanding being that one was opposite to the other. During class examples/questions there was a tendency for attendees to follow this approach rigorously and, in many cases, the constraints applied would assist the structure in carrying the loads. This is a dangerous approach as the stresses predicted are not conservative and the structure is NOT made to "work" as hard as possible. Including contact can overcome the sensitivity of constraints, and nowadays because setting it up is fairly straightforward, the user can realistically include contact in practically every job that is undertaken.
The aim of this presentation, therefore, is to highlight practical guidance in contact modelling. Guidance that can lead to increased efficiency and reliability in such problems. A number of examples are shown and these are based on the problems encountered during actual projects carried out over the last year or so. Some of the examples are set out to highlight the various "oddities" that can arise in contact modelling and suggestions to avoid these situations are given. This work should be of interest to a wide range of analysts as well as the code developers because some of the "oddities" appear to be code-specific.
One of the main objectives of the presentation is to highlight the post-processing checks that should be carried out in order to verify that the contact analysis h~ been completed in a satisfactory fashion. This paper seeks to provide practical advice and guidance in order that such problems can be completed more efficiently while ensuring that the results are correct and reasonable.