PAPERS
Development of Non-loosening Rail Fastening System with Leaf Spring Clip Applicable to Existing PC Sleeper
2025 Volume 66 Issue 3 Pages 177-182
Details
2025 Volume 66 Issue 3 Pages 177-182
Rail fastening systems using bolts and leaf spring clips are widely used in Japan. This rail fastening system requires regular maintenance to prevent the bolts from loosening. To eliminate the need for retightening bolts, some railway companies are replacing rail fastening systems using leaf spring clips with boltless rail fastening systems using round bar clips. This replacement involves replacing the existing PC sleepers with another type of sleepers for round bar clips. Therefore, some railway companies find it difficult to introduce boltless rail fastening systems using round bar spring clips due to construction costs and labor. In response to this problem, we have developed non-loosening rail fastening systems using leaf spring clips that are compatible with existing PC sleepers.
Rail fastening systems currently used in Japan generally use a method in which rail clips are deflected by bolts to fasten the rails. This rail fastening system uses leaf spring clips made from processed spring steel plates. Figure 1 shows an example of this type of rail fastening system, specifically the commonly used 5N-type rail fastening system (for 3-type PC sleepers and JIS 50kgN-rails applied to conventional lines) in Japan. This rail fastening system requires periodic monitoring of bolt loosening to maintain the rail clamping force. If loosening is detected, retightening must be performed based on indicators such as tightening torque.
In contrast, rail fastening systems that do not require bolt retightening are referred to as boltless rail fastening systems. Figure 2 shows an example of a boltless rail fastening system, specifically the e2009 type which is widely used in Japan. This rail fastening system utilizes wire spring clips made from processed spring steel. The rails are fastened by inserting the wire spring clips into the shoulder embedded in the PC sleepers by a specific tool. Therefore, replacing the rail fastening system with bolts and leaf spring clips with the boltless rail fastening system eliminates the need to retighten the bolts. However, this requires the existing PC sleepers to be replaced with PC sleepers compatible with wire spring clips. As a result, some railway operators face difficulties in introducing boltless rail fastening systems from the viewpoint of the associated costs. Therefore, in this study, we developed a non-loosening rail fastening system using leaf spring clips that can utilize the structure of the existing PC sleepers without removal [1, 2].
The 5N-type rail fastening system mentioned above was selected as the application for the boltless rail fastening system. The requirements for examining the structure of this rail fastening system are as follows:
● The structure of existing PC sleepers can be utilized
● Easy installation on existing PC sleepers
● No loosening of the components after rail fastening
2.2 Structural proposalWe devised the structure shown in Fig. 3, based on the total number of components, ease of manufacturing, and material strength. The components consist of leaf spring clips, bolts and washers. This structure is designed to fasten the rail by adhering bolts to the plugs embedded in the PC sleeper, placing a leaf spring clip on top of the sleeper, and inserting a washer between the bolt and the leaf spring clip.
Based on the structure devised in the previous section, the leaf spring clip, bolt and washer were prototyped, as shown in Fig. 4. The cross section of the leaf spring clip was designed identical to that of the leaf spring clip consisting of the 5N-type rail fastening system. The circular hole in the leaf spring clip was made larger than the bolt head so that it can be fitted after the bolt adheres to the PC sleeper. Additionally, the tip of the upper spring section of the leaf spring clip was shortened to prevent it from contacting the rail. This ensured that the tip of the upper and lower spring sections make contact after rail fastening. The washer was shaped in a U-configuration to be inserted between the bolt and the leaf spring clip. The bolt was designed with stepped sections of different diameters, not to be inserted above the stepped section. This ensures that the distance from the underhead of the bolt to the top of the sleeper remains constant. The distance from the underhead of the bolt to the stepped section of the bolt was set at 51 mm ± 0.5 mm, including the permissible dimensional tolerance during manufacturing.
Figure 5 shows a diagram of the rail fastening process. Each step in the figure is specifically as follows:
1) Bend the leaf spring clip to make the upper and lower spring sections contact.
2) Place the washer in the gap between the bolt and the leaf spring clip.
3) Release the bending force applied to the leaf spring clip and allow it to clamp the washer between the bolt and the spring, thereby fastening the rail.
Following the process, it was confirmed that the rail could be fastened as shown in Fig. 6 without applying significant load on the bolt.
It is a concern that the rail clamping force of the developed rail fastening system may be less than that of the 5N-type rail fastening system. Therefore, this section examines the rail clamping force of the developed rail fastening system.
The following equation expresses the relationship between the rail clamping force and the rail creep resistance. [3]:
| (1) |
Where γ is the rail creep resistance per set of rail fastening systems (kN), P is the clamping force per leaf spring clip (kN), μ1 is the coefficient of friction between rails and leaf spring clip, and μ2 is the coefficient of friction between rail and rail pad.
The rail creep resistance of the rail fastening system for ballast tracks must generally exceed the standard values for longitudinal ballast resistance, which range from 6 to 10 kN/m [3]. Therefore, in order to achieve a rail creep resistance greater than 10 kN/m, the required rail clamping force per leaf spring clip P is estimated to be 3.56 kN, calculated from equation (1) with a sleeper spacing of 641 mm, μ1 = 0.25 [4], μ2 = 0.65 [4].
A linear elastic FEM analysis was conducted using NX Nastran ver. 10.0 to determine the rail clamping force of the developed rail fastening system. Figure 7 shows the analysis model of the leaf spring clip. In this model, the bolt axial force is assumed to act vertically downwards at the center of the upper spring section of the leaf spring clip. As a reaction force, the rail clamping force is generated at the tip of the lower spring section, and the remaining reaction force is generated at the rear end of the lower spring section. The clearance at the tip of the leaf spring clip before fastening is set to 9 mm and the contact between the upper and lower spring sections is not considered. The deflection of the leaf spring clip is the value at the tip of the upper spring section and at the point of application of the bolt axial force.
Figure 8 shows the relationship between the deflection of the leaf spring clip and the rail clamping force obtained from the FEM analysis. When the upper and lower spring sections make contact, the rail clamping force is 4.08 kN, which exceeds the required rail clamping force of 3.56 kN mentioned above. Additionally, at the rail clamping force of 4.08 kN, the deflection at the point of application of the bolt axial force is 3.97 mm, and at the required rail clamping force of 3.56 kN, the deflection is 3.46 mm. In this case, the difference in deflection at the point of application of the bolt axial force is 0.51 mm. On the other hand, the distance from the underhead of the bolt to the stepped section of the bolt is set with a tolerance of ± 0.5 mm for the permissible dimensional tolerance when manufacturing. Since the deflection at the point of application of the bolt axial force in the leaf spring clip is within a ± 0.5 mm tolerance, when deflected to contact the upper and lower spring sections, it can be concluded that the design ensures that the required rail clamping force is achieved.
To evaluate the performance of the prototype, we conducted a tip spring constant test and a two-directional loading test to verify the safety of the rail fastening system against fatigue failure. In addition to these tests, we performed a creep resistance test.
3.1 Tip spring constant testA tip spring constant test was conducted to assess the rail clamping force, and to determine the tip spring constant, which represents the vertical spring constant at the tip of the leaf spring clip. Figures 9 and 10 show the setup of the tip spring constant test, and the test results, respectively. In this test, the actuator was fixed to the test rail head, and the rail pad was removed by lifting the rail. Subsequently, repeated vertical load was applied. The sign of the rail displacement was defined as positive for upward movement and negative for downward movement. The tip spring constant and rail clamping force were determined based on the relationship between the load and rail displacement during rail lifting. As a result, the tip spring constant converted to per leaf spring clip was 1.0 MN/m. In addition, the rail clamping force converted to per leaf spring clip was 4.10 kN, which was generally in agreement with the results of the FEM analysis conducted in Section 2.5.
As well as the tip spring constant and rail clamping force obtained from the test in section 3.1, based on the design loads and track conditions shown in Table 1, using the loading conditions for the two-directional load test were calculated [5] as shown in Table 2, the test was conducted. In this test, the bolts were fixed to the PC sleepers using an epoxy resin-based adhesive.
| Design wheel load | Load A (kN) | 98 |
| Load B (kN) | 86 | |
| Design lateral load | Load A (kN) | 30 |
| Load B (kN) | 15 | |
| Sleeper span (mm) | 641 | |
| Vertical coefficient of rail pad (MN/m) | 110 | |
| Vertical coefficient under sleeper (MN/m) | 40 | |
| Lateral coefficient of rail fastening system (MN/m) | 90 | |
| Test load PA | Max. value (kN) | 34.2 |
| Min. value (kN) | 10.0 | |
| Loading angle θA (°) | 58.7 | |
| Test load PB | Max. value (kN) | 27.7 |
| Min. value (kN) | 10.0 | |
| Loading angle θB (°) | 67.1 | |
| Height of load application h (mm) | 100 | |
 | ||
Figure 11 shows the setup of the two-directional load tests. First, a static load test was conducted using two actuators to apply static alternating loads to the rail from inside and outside the track gauge to measure the leaf spring clip stress and the lateral rail head displacement. Then, a dynamic load test was performed to check for any abnormalities in the components when the two actuators applied alternating dynamic loads to the rail. Figure 12 shows leaf spring clip stress obtained from the static load test, plotted on the fatigue limit diagram of SUP9 steel. The leaf spring clip stress was within the Goodman line for 105 cycles and yield limit line, so there was no problem to solve. Additionally, the displacement of the lateral rail head was approximately 0.5 mm, which was less than the design limit of 7.0 mm for conventional railway lines.
The dynamic load test was conducted for 1 million cycles at a loading frequency of 5.5 Hz. As a result, no loosening or displacement of the washers was observed after the test, and no abnormalities were found in the components.
3.3 Rail creep resistance testFigure 13 shows the setup of the rail creep resistance test. In this test, the PC sleeper was fixed to the test machine table, and the rail was loaded twice in its longitudinal direction to measure the rail creep resistance. Figure 14 shows the test results. The rail creep resistance averaged 7.78 kN per rail fastening system over two loadings. Converted to a sleeper spacing of 641 mm, this is equivalent to 12.1 kN/m, exceeding the ballast longitudinal resistance of 10 kN/m.
To implement the proposed rail fastening method, a tool with a mechanism for bending the leaf spring clip is required. For the purpose, two types of installation tools were developed, a manual installation tool and a mechanical installation tool as shown in Fig. 15. The manual installation method is a tool that bends the leaf spring clip by hand without the use of electric machinery. This method utilizes the p rinciple of leverage, where the reaction force is taken from the rail, and the leaf spring clip is bended at the tip of the arm by tilting the arm. The mechanical installation method is a tool that bends the leaf spring clip using an electric wrench or similar device. It operates by taking the reaction force through two plates attached to the bottom of the tool and placed against the rail's underside. The hexagonal section at the top of the tool is connected to a standard maintenance wrench or similar tool and rotated to bend the leaf spring clip. Using these two types of installation tools, we conducted a workability evaluation. As a result, both tools were confirmed to effectively bend the leaf spring clip, allowing the upper spring section to be tightly pressed against the lower spring section. In addition, it was verified that the fastening washer could be installed properly in the gap between the bolt and the leaf spring clip in this state.
A test installation was conducted on an operational line to verify the performance of the developed rail fastening system under actual installation conditions. A straight section of a conventional railway line was selected as the installation site. Figure 16 shows the installation process using the manual installation tool. The developed rail fastening systems were installed on five sleepers, with one sleeper skipped between each installation, totalling ten sets of rail fastening systems. The installation was carried out following the rail fastening process described in Section 2.4 using the manual installation tool. As shown in Fig. 17, it was confirmed that the rail fastening system could be successfully installed on the operational line without any issues.
A non-loosening rail fastening system with leaf spring clips was developed that does not require the replacement of PC sleepers and utilizes their existing structure. Performance tests confirmed that the fastening system possesses the necessary performance for application on actual tracks. In addition to the fastening system, two types of installation tools were developed to facilitate rail fastening, and their workability was verified.
In this rail fastening system, even with the smooth shape of the upper surface of the washer and the bolt neck, no detachment or movement of the washer was observed during a dynamic load test. However, for mass production, a spherical surface processing is planned for the washer and bolt to ensure that they do not prevent detachment or movement. Additionally, after the test installation on the operational line, ongoing monitoring will be conducted, including durability testing of the adhesive, and further improvements will be made as necessary.
In conducting the field installation tests for this research, we received significant cooperation from the West Japan Railway Company. We would like to express our sincere gratitude to them here.
|
Daiki YAMAOKA Researcher, Track Structures and Components Laboratory, Track Technology Division Research Areas: Continuous Welded Rail, Rail Fastening System |
|
Tadashi DESHIMARU, Ph.D. Senior Chief Researcher, Head of Rail Maintenance and Welding Laboratory, Track Technology Division Research Areas: Railway Rails, Rail Fastening Systems |
|
Shingo TAMAGAWA, Ph.D. Senior Researcher, Track Structures and Components Laboratory, Track Technology Division Research Areas: Continuous Welded Rail, Rail Fastening System |
