Discussion. Model Tests on Masonry Arches.

Abstract

Institute of Higher EducationThe Author's conclusions concur with my own work on 1 m, 1.5 m, 3 m, 5 m, and 6 m span bridge test"-21 which have resulted in the idealization shown in Fig. 17. It is from this theoretical model that the assessing engineer may start and thus determine the relative significance of each component. This allows the different modes of failure to be studied. If the contribution by the spandrel walls to barrel stability is high (which in most cases it will be), transverse distribution becomes significant. This has been observed during the monitoring of a two-span stone bridge during the passage of an abnormal load.22 The bridge showed no signs of distress longitudinally, but transversely, local to the wheel loading, tensile strains were recorded which were commensurate with crack formation. As it was likely that transversely there was no precompression (unlike longitudinally where compression is guaranteed), it followed that the recorded tensile strains did cause cracking. Fig. 18 shows the transverse strain and deflexion profiles. This raises the question as to whether or not the adoption of a full width KEL as the standard loading case is restrictive, as it increases the likelihood of a classical fourhinge mechanism or a three-hinge snap-through failure depending on the spandrel wall stiffening and barrel backfill. It was reassuring, therefore, to see confirmation of this concern by the Authors' use of a half-width loading case. Other types of loading may cause local distress or even a mode of failure not hitherto considered. Certain types of arch bridge are more likely to be susceptible to non-standard modes of failure, e.g. skew, elliptical or multi-ring brickwork arch bridges. It has been shown21F22 that point loads can cause local distress to an arch.47. The real value of model testing lies in the prediction of the mode of behaviour and the load-carrying capacity of full-scale structures. The Authors' analysis considered only one of the variables, i.e. vertical loading. A more extensive analysis using the parameters identified in Fig. 17 results in a more complex situation. Assuming the densities of the prototype and model are the same for each of the constituents, it follows that, using the Authors' notation, Pp/Pm for the various parameters is related t0.L; (for the vertical dead load), L j (for the horizontal longitudinal soil pressure), L: (for dead load of the spandrel wall), L; (for cohesion of backfill with the spandrel wall), L: (for frictional resistance of the backfill with the spandrel wall due to self-weight) and Lf (for frictional resistance of the backfill with the spandrel wall due to the applied loading). Additionally, as the Authors correctly explored, there is the effect of relative strength.48. It follows that any extrapolation of the results for application at full scale should be treated with care as the scaling factors are not the same for each parameter.49. In spite of the doubts raised, it is interesting that the correlation between model and full-scale behavi...

Numerical investigation of arches in brick-masonry bridges

“…At present, these structures need to be assessed considering current safety requirements and loading conditions. Past laboratory and in-situ tests (Hendry, Davies, & Royles, 1985;Pippard, Tranter, & Chitty, 1936;Pippard & Ashby, 1939;Pippard & Chitty, 1941;Royles & Hendry, 1991;Wang, 2004;Melbourne et al, 2007) showed that the structural response of masonry arches is very complex and characterised by different failure modes. Thanks to the development of computational techniques for nonlinear structural analysis, various numerical strategies for assessing arches in masonry bridges have been proposed in recent past (Nobile & Bartolomeo, 2015;Sarhosis, et al, 2016).…”

Progressive cracking of masonry arch bridges

“…While the beneficial effects of the spandrel walls on the ultimate load capacity has been demonstrated (Royles and Hendry, 1991;Melbourne et al, 1997) it has also been recommended that the integrity of the spandrel walls with the arch barrel should not be relied upon for the purposes of ultimate capacity assessment (Page, 1993;.…”

Mesoscale partitioned analysis of brick-masonry arches

“…Thus, over the past few decades a growing interest has developed in studying their behaviour up to collapse and assessing their ultimate load capacity. Significant laboratory and in-situ tests were carried out [3][4][5][6][7][8][9] and different analytical and numerical descriptions were proposed [10][11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26][27][28]. Current modelling strategies for arches in masonry bridges are mainly based upon the use of limit analysis concepts [20][21][22], the finite element method (FEM) [13,24,25,26,27], the discrete element method [23] or discontinuous modelling techniques [19].…”