Laboratory pull-out tests on fully grouted rock bolts and cable bolts: Results and lessons learned

Abstract: 

Laboratory pull-out tests were conducted on the following rock bolts and cable bolts: steel rebars, smooth steel bars, fiberglass reinforced polymer threaded bolts, flexible cable bolts, IR5/IN special cable bolts and Mini-cage cable bolts. The diameter of the tested bolts was between 16 mm and 26 mm. The bolts were grouted in a sandstone sample using resin or cement grouts. The tests were conducted under either constant radial stiffness or constant confining pressure boundary conditions applied on the outer surface of the rock sample. In most tests, the rate of displacement was about 0.02 mm/s. The tests were performed using a pull-out bench that allows testing a wide range of parameters. This paper provides an extensive database of laboratory pull-out test results and confirms the influence of the confining pressure and the embedment length on the pull-out response (rock bolts and cable bolts). It also highlights the sensitivity of the results to the operating conditions and to the behaviour of the sample as a whole, which cannot be neglected when the test results are used to assess the bolt-grout or the grout-rock interface.

Introduction 

Fully grouted rock bolts and cable bolts are two reinforcement techniques widely used in civil and mining engineering. These support systems combine efficiency, flexibility, ease of installation and low cost (Stillborg, 1994; Fine, 1998). Due to these assets, they are extensively used in the underground to improve safety along roadways and large openings.

In general terms, a rock bolt or cable bolt consists of a bar inserted in a borehole that is drilled into a soil or rock mass and anchored to it by means of a fixture (Windsor, 1992; Windsor and Thompson, 1996). Fully grouted bolts comprise four elements: the bar, the surrounding ground, the internal fixture to the borehole wall and the external fixture to the excavation surface. The main characteristic of fully grouted bolts is that they only provide support action if the surrounding ground tries to deform; thus, they are passive reinforcement systems (Tincelin and Fine, 1991).

Worldwide experience suggests that failure of fully grouted bolts most likely occurs at the bolt-grout interface, by means of a debonding process that starts if the axial force on the bar exceeds a critical value, and then propagates along the interface (Goris, 1990; Hyett et al., 1992, 1995; Kaiser et al., 1992; Stillborg, 1994; Li and Stillborg, 1999; Moosavi et al., 2005). Analytical solutions for the debonding process were proposed recently (e.g. Li and Stillborg, 1999; Ren et al., 2010; Blanco-Martín et al., 2011). However, these solutions do not account for the interface normal behaviour explicitly. With the support of the European Commission’s Research Fund for Coal and Steel (RFCS), a new pull-out bench was designed and calibrated in the context of the PROSAFECOAL programme (Papamichalis et al., 2010) to gain more insight into the response of fully grouted bolts (axial and normal directions). This bench, described in Blanco-Martín (2012) and Blanco-Martín et al. (2013, 2016), allows testing several bolts, and investigating the influence of a wide variety of parameters, such as the confining pressure, the embedment length, the roughness of the borehole wall or the thickness of the grout annulus. Additionally, failure at the bolt-grout or the grout-rock interface can be studied. Blanco-Martín et al. (2013) suggested a procedure to assess the response of the bolt-grout interface from experimental results and theoretical considerations, and proposed a semi-empirical formulation of the interface behaviour (axial and normal directions) for resin-grouted steel rebars and fiberglass reinforced polymer (FRP) rock bolts.

As mineral resources are decreasing in Europe, mining companies are searching deeper underground to meet customers’ needs and maintain their activities. At large depth, stresses are higher and support systems must be intensified. In this context, a new laboratory pull-out set-up has been conducted within the framework of the RFCS AMSSTED research programme (Hadj-Hassen et al., 2015). In this set-up, a large range of bolt types has been tested, and attention has been focused on the influence of the confining pressure and the embedment length, since it has been previously shown that these parameters have a strong effect of the pullout response (Benmokrane et al., 1995; Hyett et al., 1995; Moosavi et al., 2005; Blanco-Martín et al., 2013). Moreover, the execution of the tests has demonstrated that pull-out results are very sensitive to the operating conditions and to the response of the sample as a whole (for instance, damage of the rock sample markedly affects the measured pull-out response). Fifty-two tests on rock bolts and thirty-two tests on cable bolts have been carried out, and the main findings are presented here.

This paper is organized as follows. First, we describe the experimental bench used at laboratory-scale and the set-up designed to prevent unscrewing when testing cable bolts. Then, we present the samples preparation procedure, as well as the main characteristics of the bolts, the grouting materials and the rock type used to prepare the samples. Later, the main results obtained for rock bolts are presented, followed by the results for cable bolts. For a given bolt type and dimensions, our results compare well with past investigations (Benmokrane et al., 1995; Hyett et al., 1995; Moosavi et al., 2005; Ivanovic and Neilson, 2009). The experimental data presented here extend the available database of pull-out test results, and can be used both as a technical reference under the specified conditions, and as a means of comparison between model predictions (which include operating conditions, and sample components and behaviour) and laboratory-scale data.

Conclusions 

This paper presents pull-out test results performed at laboratory-scale on three types of rock bolts and three types of cable bolts. Each bolt has been tested using different embedment lengths and confining pressures. Resin and cement have been used as grouting materials, and in all cases, the bolts have been grouted into a sandstone sample. The tests have been performed using a pull-out bench that allows testing a wide range of parameters. Constant confinement or constant radial stiffness has been used as boundary conditions. The study presented has been performed in static conditions, and its main goal is to investigate debonding at the boltgrout interface.

Experimental results show that interface adhesion, friction and mechanical interlock (threaded bars) contribute to the bolt-grout bond. The comparison between pull-out results of smooth and threaded bolts clearly shows that the bolt profile plays an important role: not only the maximum load is greater for threaded bolts, but also the post-peak phase shows a more gradual force decrease. Moreover, the bolt profile is reflected in the oscillations measured in the post-peak phase, whose periodicity matches the indentations of the tested bar.

Results for HA25 and FRP threaded bolts show overall similar trends, with a stiff, quasi-linear pre-peak phase, followed by a drop of stiffness, a maximum force and finally a post-peak phase in which the axial force decreases towards a residual value, much lower than the peak force. Depending on the confining pressure, oscillations on the load occur according to the bolt profile; these oscillations are less clear for high pressures, for which the grout between the bolt asperities is sheared. The operating conditions in the radial direction affect both the radial and the axial responses. In the current bench configuration, the advantage of using constant radial stiffness conditions is that the confining pressure variation can be used to estimate the radial displacement. Results obtained for cable bolts are also consistent and show similar trends to rock bolts, with the axial load-displacement relationship showing four different phases, and the bolt profile and geometry being observed in the post-peak phase. However, the results obtained for rock bolts and cable bolts show important differences in the post-peak phase, with a much steeper load decrease in the case of rock bolts. Two likely reasons to this difference are the bar type (solid against initially twisted wires) and the bolt profile (geometry of indentations).

Regarding the Mini-cage cable bolts, their profile induced severe damage to the rock sample. In the case of the IR5/IN bolts, pull-out results were very sensitive to the configuration at the free end. These results, together with the radial fracturing that often occurs at low confining pressures (rock bolts and cable bolts), highlight the difficulty of assessing the bolt-grout interface behaviour from raw pull-out data, and also the importance of addressing the behaviour of the sample as a whole before focusing on the interface. In order to study the interface response accurately, the bench configuration should be kept as simple as possible to reduce uncertainty in the measurements and the processes underneath (fracturing, rock response, bench calibration, etc.).

The results presented are consistent with previous pull-out test results on similar rock bolts and cable bolts, and provide an extensive database of laboratory-scale results, focusing on the influence of the confining pressure and the embedment length on the pull-out response of fully grouted bolts. These data can be used both as a technical reference under the specified conditions, and as a means of comparison between model predictions (which include operating conditions, and sample components and behaviour) and laboratory-scale data.