Keywords: anchorages, anchorage head, bearing plate, numerical modelling, laboratory experiments, integrity testing.

Ground anchorages are applications that are mainly used to support structures such as tunnels, mines and retaining walls etc. The components of anchorage systems are the protruding, free, fixed anchorage lengths and a bearing plate. The main areas of interest of the anchorages are the bond quality/characteristics of the fixed anchorage length and the anchorage head assembly characteristics. The anchorage head assembly consists of a bearing plate and stressing nut for the case of bolt anchorages and a bearing plate, barrel and wedges for the case of strands. Furthermore, the rock surface that the bearing plates are usually placed against may be different. It can vary from smooth concrete surfaces to the very rough surfaces found in tunnel walls and ceilings.

Research on ground anchorages began in Aberdeen in 1986 and went through several stages. Initially, field tests on rock bolts in tunnels in South Wales were undertaken and the outcomes of that research led to a new stage that involved the development of the non-destructive testing method for testing the integrity of ground anchorages called GRANIT (GRound ANchorage Integrity Testing). The non-destructive testing method developed has been improved by the development of a numerical model that gives a better insight into all the components of the anchorage system. This represents the current stage of this research project.

A lumped parameter model, developed to observe and understand the behaviour of ground anchorage systems when subjected to a dynamic load addresses the influence of the anchorage head assembly characteristics on the dynamic response. In order to find the properties of the anchorage head assembly required as input for the model, an anchorage head assembly was constructed. From the results from each of the anchorage components, obtained from the model it appeared that the most influential part of the anchorage system with respect to the detection of the load is the anchorage head assembly itself. In order to investigate how changes in the bearing plate characteristics influence the stiffness characteristics of the anchorage head assembly, different geometries of bearing plates were manufactured and placed on the anchorage head rigs. From the tests undertaken it is shown that the bearing plate geometry play a major role in determining the stiffness characteristics of the anchorage head. A typical load/displacement graph is shown to be highly non linear and is represented as a cubic curve in the model [1]. The GRANIT technique has implemented this finding and the changes in response frequency with post tension level are shown to be determined by the characteristics of the anchorage head. Furthermore, two types of anchorage heads have been recognised: Class I where the changes in response with load are present and determined by the flexibility of the anchorage head and Class II where the changes in the response with load are not detectable because of the high anchorage head stiffness [2].

From the initial laboratory research described, it was recognised that further tests on the anchorage head assembly can replicate better both the real mine or tunnel wall surfaces and the test procedures when GRANIT testing was undertaken. Firstly, since the usual GRANIT training procedure involves a load cell for measuring the post tension load and this therefore is part of the anchorage head assembly, a new set of tests was undertaken in order to see the changes between the anchorage head with the bearing plate only and with the addition of the load cell. Secondly, changes to the surface roughness of the concrete rigs that the tests were taken on previously are believed to represent more closely the real state of the rock surface found in mines and tunnels. The surface of the anchorage head rigs was chiselled in order to make the surface rough and the results from a series of tests with bearing plates of different geometry are presented. In order to measure the displacement of the anchorage head two LVDTs were placed, one closer to the bearing plate and the other farther from the anchorage head. From the results obtained from the experiments it is shown that the type of the anchorage head assembly i.e type of bearing plate used, the inclusion of load cell and the type of the surface the bearing plate is placed against do influence the load/displacement curve used in the model. This relationship will be used for obtaining the frequencies for each load level which are then implemented in the GRANIT system for load detection of the anchorages. This suggests that the more representative the curve that is chosen, for a particular anchorage type, the better the load diagnosis that is achieved.

1

Ivanovic A., Neilson R.D. and Rodger A.A. "Influence of Prestress on the Dynamic Response of Ground Anchorages", Journal of Geotechnical and Geoenvironmental Engineering, ASCE, 128(3), 237-249, 2002. doi:10.1061/(ASCE)1090-0241(2002)128:3(237)

2

Starkey A., Ivanovic A., Rodger A.A., Neilson, R.D. "Condition Monitoring of Ground Anchorages by Dynamic Impulses: GRANIT System" Meccanica 38:265-282, 2003. doi:10.1023/A:1022850520198

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £95 +P&P)