Field drawing test of fully bonded BFRP anchor under cyclic loading

2.1. Geological survey of the test area

The site pull-out test area is located near Hujiagou, Longquan Temple Town, Yongdeng County, Lanzhou City, in the DK76+0344.5-BK76+343.40 subgrade slope protection anchorage Engineering Department from Zhongchuan Airport to Wuwei section of the new railway from Lanzhou to Zhangye Third and fourth lines. The engineering geological conditions of the site are as follows: the stratigraphic lithology is dominated by the Upper Pleistocene aeolian loess of the Quaternary system (12-20 m thick, light yellow, mainly silty, slightly dense and slightly wet, Grade II common soil, ${\sigma }_0=$120 kPa) and the underlying Upper Neogene Upper Neogene mudstone (brown-red, mainly clay minerals, poor diagenetic and soft, grade IV soft stone, ${\sigma }_0=$350 kPa) [11-13]. It belongs to the Qilian uplift belt in the Qilian fold system, with stable strata, undeveloped fault structure and simple geological structure. There is no bad geology. The special soil is eolian sandy loess with self-weight collapsion (grade Ⅳ, 12-20 m thick), soft soil with slightly dense state and weak expansive mudstone. Surface water is not developed, and no groundwater is found in the depth of exploration drilling.

2.2. Test materials and equipment

The experimental materials mainly used 5 full-threaded BFRP bolts, with lengths of 2.8 m, 3.3 m, 3.8 m, 4.3 m and 4.8 m, diameters of 32 mm, densities of 2.05 g/cm³, tensile strength of 834 MPa, elastic modulus of 45 GPa, elongation at break of 1.89 % and shear strength of 255 MPa.

The test equipment includes multiple systems such as hole forming, grouting, loading, monitoring and laptop computers, as shown in Fig. 1. The pneumatic anchor drill is selected for drilling, and the compressed air is used as the power to ensure the drilling accuracy and efficiency. The loading system is based on a manual oil compression core jack with a rated tonnage of 65 tons and a stroke of 18 cm. It operates in cooperation with ZB4-500 electric oil pump, Φ50 and Φ32 anchors, reaction pier and steel cushion plate [14-15]. A dial indicator with a range of 50 mm and an accuracy of 0.01 mm for displacement monitoring; Drawing load is collected by Changzhou Jinmu 100Φ80 vibrating string anchor cable dynamometer (JTM-V1800) and supporting YT-RG-01 reading meter; The strain acquisition relies on DH3816N static strain test system and BX120-20AA resistance strain gauge to meet the needs of complex conditions and multiple measuring points.

2.3. Test scheme design

2.3.1. Prepare the anchor bolt

This test takes BFRP anchor rod as the key object. After selecting different lengths of bolt according to the test design, prepare bolt according to the following steps: First, the steel sleeve with a length of 0.8 m, an outer diameter of 50 mm and a wall thickness of 5 mm was centralized and filled with 1:1 epoxy resin and curing agent with filling agent. Then, auxiliary fixtures were prepared, the measuring points were marked and sanded, strain gauges were affixed, and the strain gauges were waterproof with latex coating. An 8-meter wire was prepared, the DH3816N static strain test system was connected and waterproof treatment was done, and the centring bracket was installed. The production process is illustrated in Fig. 2.

Fig. 1Field drawing test system

Fig. 2Bolt preparation flow chart

2.3.3. Loading test

After leveling the drilling slope, the reaction pier, steel backing plate 1, core jack, anchor cable dynamometer, steel backing plate 2, anchor, fixture and dial indicator are installed in sequence. The dial indicator is fixed on the self-made scaffold by magnetic base. The device schematic diagram is shown in Fig. 3.

This BFRP anchor pulling test followed relevant specifications, adopted cyclic loading, and controlled loading and unloading load and speed by manual oil pressure gauge (adding 0.07 MPa/s and unloading 0.14 MPa/s). The load is divided into multiple cycles. After the initial load is stabilized for more than 1 minute and the first cycle load is stabilized for more than 10 minutes, the load is unloaded back to the initial load for more than 1 minute. Then load step by step, stabilize the pressure, unload back to the initial load. The test shall be terminated until the anchoring system is damaged or the judgment condition is reached. If the criterion is not met, the test shall be continued on the basis of the maximum failure load multiplied by 1.2 times and 1.5 times respectively. Failure judgment basis: the end displacement of bolt is not convergent, the anchor solid is pulled out or the bolt is pulled out; The displacement increment of the load in the latter stage exceeds that of the upper stage by 2 times.

Fig. 3Schematic diagram of BFRP bolt drawing test device

3.1. Failure mode analysis of BFRP anchoring system

This experiment explores the failure form and mechanism of BFRP anchor system under cyclic pulling load, and its apparent changes are observed and recorded during the whole process of pulling test. The initial load is small and the response of the anchoring system is weak. The “squeaking” sound comes out when the load increases, which is inferred that the interface between bolt and slurry is subjected to shearing stress. With the change of load and holding load, the five groups of anchors appear the second interface failure, accompanied by local soil collapse at the hole, mud expansion near the anchor hole and inclusion crushing and slip-off, etc. In the test, the anchorage system is mainly the second interface sliding failure.

3.2. Analysis of the relationship between cyclic load and displacement of BFRP anchor

The displacement curve of BFRP anchor rod under cyclic load presents two significant features: First, the slip displacement of each anchor rod gradually increases with the increase of cyclic load, and the displacement increment of the latter cycle is larger than that of the previous cycle, and the increment difference of the first two cycles near failure is particularly prominent; Second, the maximum slip displacement point of most bolts (4.8 m, 4.3 m, 3.8 m) lags behind the maximum cyclic load point. In the early stage of unloading, the bolts still slip out due to large residual load, and the displacement rebound occurs at low load, indicating that the bonding force still exists after unloading. Although the displacements of the 3.3 m and 2.8 m bolts are not synchronized with the maximum cyclic load, their displacements continue to increase from unloading to the end, and show a linear and rapid growth trend after the maximum cyclic load.

3.3. Comparative analysis of bolt dynamometer load and design cyclic load

In this test, the design cyclic load is applied to BFRP bolt by jacks, and the feasibility of the operation scheme is verified by the monitoring data of the connected bolt dynamometer. The B1 bolt is taken as an example. As can be seen from Fig. 4, the monitoring load is positively correlated with the phased changes of the design cycle load, but due to uncertainties in practical operation, the monitoring load is difficult to match the theoretical value. However, the time-history curves of the two are consistent and the values are similar, indicating that the experimental operation is highly feasible. Looking at Fig. 5, except for the slightly larger peak load of the bolt dynamometer in the second cycle, the rest are small, indicating that there is a stress loss in the force transmission, but the overall two are close, indicating that the test effect is good and the operation plan is feasible.

Fig. 4Comparison of time history curve between monitoring load and design cyclic load by bolt dynamometer

Fig. 5Comparison of peak curve of bolt dynamometer and design cyclic load