The normal stress of the constant normal stiffness direct shear apparatus is applied by weights through a lever, and the normal stiffness is controlled by a normal spring connected to the lever. Different normal stiffness boundary conditions can be achieved by replacing springs with different stiffnesses. During shearing, the shear stress at the interface is controlled by a micro-motor in the system, with a maximum shear stress of 10kN. The measurement and control system can simultaneously measure and output the interfacial shear displacement and normal displacement in real-time. Data can be directly collected to the control computer through the data acquisition system. The working principle of the direct shear apparatus during the shearing process is shown in Figure 5a. By measuring the stiffness of soils with different properties as required by the experiment, different stiffness values can be controlled for the experiment. The maximum and minimum stiffness values obtained are 1100kPa/mm and 200kPa/mm, respectively. Two intermediate values are then selected, and all springs are uniformly processed using 65Mn carbon steel (Figure 5b). The internal effective dimensions of the shear box are 100mm × 100mm × 50mm, with both the upper and lower shear boxes having a height of 25mm. The sample consists of a 100mm × 100mm × 25mm soil layer on top and a 100mm × 100mm × 25mm mortar cube below. The sample is tightly attached to the shear box to reduce shear stress errors during preloading. A schematic diagram of the force applied to the sample is shown in Figure 5c.。
FIGURE 5 (a) Direct shear instrument (b) Spring selection (c) Specimen stress diagram

2.5 Testing Process

Boundary conditions are crucial factors that influence the mechanical response of soil. During the shearing process, changes in the normal stress applied to the interface due to volume changes in the shear zone can lead to alterations in shear stress. Existing research has shown that constant normal stiffness (CNS) conditions are closer to actual engineering conditions compared to constant normal stress conditions38. Under constant normal stiffness conditions, the change in normal stress at the interface is denoted as,and the change in normal displacement at the interface is denoted as , the normal stiffness at this time is represented by k =constant.
The process of interfacial shear testing includes the following steps: (1) Calculate the weight of the soil sample required for a compaction rate of 95%. Install the prepared mortar block into the mold, and then fill the soil sample into the mold evenly in three layers. Fabricate the interfacial shear sample using the layered compaction method and wrap it in plastic film to prevent moisture loss during the freezing process. (2) Place the interfacial shear sample in a -20℃ freezer for rapid freezing for 12 hours to prevent moisture migration during the freezing process. (3) After rapid freezing, remove the mold from the solidified sample, wrap it in a new plastic film, and transfer it to a constant temperature freezer. Maintain the sample at the target experimental temperature for 24 hours. (4) Adjust the environmental temperature and instruments for the test, ensure the sealing of the freezer to minimize any potential errors, install the sample on the shear apparatus, and begin the interfacial shear test. (5) After the shear test is completed, take photos of the sheared interface and record the test details.
The experiment was conducted using a rapid shear mode with a shear rate of 0.8mm/min. The shearing duration was set to 15 minutes, resulting in a shear displacement of 12mm. The experimental conditions considered during the test included various normal stiffnesses (200kPa/mm、500 kPa/mm、800kPa/mm、1100 kPa/mm),normal stress(40kPa、80kPa、120kPa、160kPa),water content of soil(10%、13%、16%)and testing temperature(20℃、-2℃、-4℃、-6℃).