Search CINTEC.COM:

Advanced Search

Table of Contents :: Introduction :: Objectives :: Pop Bottle Bridge :: Test Arrangements :: Numerical Model :: Results :: Discussion :: Conclusions :: Appendices

Report No. B1660A/V16/R03
May 2004
Cintec International
Cintec House
11 Goldtops
NEWPORT
NP20 4PH

5. NUMERICAL MODEL

The numerical model that was developed by Gifford for the Special Assessment of Pop Bottle Bridge(3) has also been used to predict displacements and strains for comparison with the test results. Several minor modifications were made as follows.

  1. The four transverse cracks were explicitly represented in the modelled barrel by including a frictional cohesionless joint in the barrel through all rings at the crack locations. These cracks at approximately the 1/5 span, have been represented as being closed under permanent loads, prior to the application of the live load.
     
  2. In place of the codified live load cases used for assessment each of the 28 load cases used in the tests have been modelled by applying 11.5 tonne axles.
     
  3. The rules in BD 21 for distributing wheel loads transversely through the surfacing and fill have not been used. Instead axle loads have been applied over a 2.5m wide strip of barrel.

The extra strength and stiffness attributable to the effects of transverse load distribution, the effects of spandrel wall stiffening and the influence of skew have not been allowed for in the analysis. The analysis is based on a conservative two dimension plain strain representation and, therefore, is expected to give upper bound predictions for displacement and strain.

Figure 5.1 shows the DE model of Pop Bottle Bridge and the modelled transverse cracks.

 
Finite Element Mesh
DE Mesh of Crack
Discrete Element Mesh
Figure 5.1 DE Model of Pop Bottle Bridge used to Predict Test Responses
Cintec International

Cintec House
11 Goldtops
NEWPORT
NP20 4PH

Gifford And Partners Ltd

Carlton House
Ringwood Road
Woodlands
Southampton
SO40 7HT