Development of embedded bending member to model dowel action

William G. Davids*, George M. Turkiyyah

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

20 Scopus citations

Abstract

The development and implementation of a general, three-dimensional (3D) embedded bending member suitable for modeling dowel action is presented. The development is motivated by the desire to efficiently model rigid concrete pavement slabs with doweled joints using solid brick elements without requiring the mesh nodes to coincide with the dowel element nodes. All derivations are done with respect to an arbitrarily oriented coordinate system. The stiffness matrix of the embedded dowel element is expressed as a transformation of the unembedded dowel element stiffness matrix, eliminating the need for special integration techniques. Consideration is given to axial debonding of the dowel, as well as the nonlinear case where a gap exists between the dowel and surrounding material. Details of the object-oriented implementation of the element and its incorporation in a general, nonlinear solution strategy are presented. Convergence and parametric studies were performed using a realistic model of two rigid, concrete pavement slabs subjected to static axle loading near the joint. The results of these studies demonstrate the applicability of the element as well as indicating the potential for significant detrimental effects on doweled joint performance when gaps exist between the dowels and the pavements slabs.

Original languageEnglish (US)
Pages (from-to)1312-1320
Number of pages9
JournalJournal of Structural Engineering
Volume123
Issue number10
DOIs
StatePublished - Oct 1997
Externally publishedYes

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Building and Construction
  • General Materials Science
  • Mechanics of Materials
  • Mechanical Engineering

Fingerprint

Dive into the research topics of 'Development of embedded bending member to model dowel action'. Together they form a unique fingerprint.

Cite this