Laboratory experimental of ballast–bituminous–latex–roving (Ballbilar) layer for conventional rail track structure

1 Introduction

The construction of a reliable, large-capacity, safe railway network is one of the many infrastructure projects prioritized by Indonesia’s Ministry of Transportation in an effort to stimulate local economies across the country. High operating speed and large train axle loads are the primary factors that determine the reliability, capacity, and safety of the railway network. The performance of the railway line structure, which is strong and resistant to traffic loads and other environmental factors, greatly influences both of these variables.

The upper and lower structures are the two main components of a typical railway track. Ballast, sub-ballast, and enhanced subgrade make up the structure of the lower part of the railway line. The ballast structure on railway lines is a layer composed of crushed stone aggregate and is located in areas experiencing the greatest concentration of stress due to train traffic. Ballast performance is commonly compromised due to damage, necessitating careful selection of the forming material and intensive maintenance. The sub-ballast layer consists of fine gravel, medium gravel, or coarse sand. This layer functions as a filter layer between the subgrade and the ballast layer. The subgrade layer comprises the soil layer in all its forms, including original, repaired, and artificial. The load is distributed through the ballast or sub-ballast layer, and the subgrade is responsible for carrying it [1,2,3].

Following the regulations in the Indonesian railway system, the subgrade layer must be solid and have a minimum strength or a California bearing ratio value of 8% [4]. The subgrade layer’s resilience and load-bearing capacity are critical to its performance in its function. Its supporting subgrade layer must bear the load of the ballast and sub-ballast layers. Moreover, the stress it undergoes is the result of the force transmitted from the sleeper to those layers, which are, in turn, transmitted and distributed to the subgrade layer by the ballast and sub-ballast [5].

One of the most crucial parts of conventional railway lines – the ones that bear the greatest stress from train traffic – ballast structures continue to be prevalent in Indonesia’s railway system. Unfortunately, the ballast in Indonesia is weak and prone to various forms of structural damage, such as deformation and loss of ballast, causing the relatively low speed of passenger and freight trains in the country [2,6]. The maximum permitted operating speed is around 90 km/h, even though some passenger train series in Indonesia can reach a speed of up to 120 km/h. Additionally, the highest permitted operating speed is only about 70 km/h, even though the Babaranjang train in South Sumatra can reach a speed of about 90 km/h [7]. However, conventional railway lines in Indonesia have a maximum axle load capacity of only 18 tonnes [4]. Meanwhile, the maximum train axle load that tracks in many European countries can support has risen to 20 to 23 tonnes [8]. Trains can carry over 30 tonnes of axle load, even in the US [9]. Therefore, it is rather evident that train derailments will become more common in Indonesia if trains keep going faster than the allowed speed and carrying capacities. In a similar vein, the safety of Indonesia’s passenger and freight trains heavily depends on the functionality of conventional railway track construction, particularly ballast components.

Moreover, Priambodo [10] stated that asphalt is a viscoelastic material with various characteristics and properties. To begin with, thixotropy, which is the tendency for asphalt to harden when not subjected to stress, is a crucial factor. In addition, there is the rheologic aspect, which pertains to the asphalt’s condition and reveals the time-dependent relationship between stress and strain. Asphalt can retain its elasticity even when subjected to brief loads. Nonetheless, asphalt can undergo plasticity (viscosity) if the loading period is sufficiently prolonged. Finally, there is thermoplastic asphalt, of which the viscosity can vary in response to variations in temperature. As the temperature rises, the viscosity of asphalt decreases, causing it to thin down, and the reverse is also true.

Furthermore, Rianung [11] defined asphalt aging as a factor that determines the durability of the asphalt mixture. As asphalt ages, it hardens and becomes stiffer. It, in turn, makes the asphalt mixture more resistant to permanent deformation and better able to distribute loads. On the flip side, if the asphalt mixture is too brittle, it will crack more readily and be less resistant to repeated loads. Table 1 demonstrates that according to American society for testing and materials standards, asphalt cement can be categorized according to its penetration value at 25°C.

60/70 penetration-grade asphalt binder is a material commonly used in the flexible pavement system in Indonesia. However, asphalt of any type and classification has never been employed as a structural material for railway lines in the Indonesian railway system. It is expected that the use of asphalt can strengthen the ballast structure and be a solution to problems in conventional railway line structures in Indonesia [12,13,14].

Latex is a liquid milky sap obtained from stripping the bark of rubber trees. Indonesia is one of the largest producers of natural rubber in the world, making it easy to find high-quality latex. There has to be an effort to create and employ latex processing products to raise the commodity prices and utilization of latex. Nevertheless, at the moment, the selling price of latex commodities on the Indonesian market is still inexpensive. The use of natural rubber or latex as a hot mix asphalt (HMA) mixture has been proven to boost quality, extend service life, and reduce flexible pavement maintenance costs [15]. In addition, Setiawan et al. [16] also discovered that using tire rubber with varying sizes as a mixture could escalate the durability of the ballast material, as indicated by the smaller percentage of ballast wear after a compression test.

However, roving fiber is a polyester or epoxy fiber typically utilized as a medium for the middle layer of fiberglass plates. Roving fiber, once molded into a fiberglass shape, could be drawn taut enough to resemble a web of interconnected strands. Table 2 displays the fiberglass characteristics. As many researchers asserted, the compressive and split tensile strengths of non-sand concrete could be enhanced using roving fiber in place of sand. With the addition of 5% roving fiber concentration, the non-sand concrete obtained a compressive strength of 1.81 MPa and a split tensile strength of 0.29 MPa [17,18].

Hence, this study emphasized the utilization of a binder and stabilizer material mixture of latex and roving fiber in constructing the ballast layer using a 60/70 penetration-grade asphalt binder. The mechanical qualities of chopped strand mat (CSM/M) fiber roving with S-glass material were considered the most appropriate for application to asphalt mixtures. Thus, this type of roving was adopted to support several research parameters.

There are several methods used to determine the compressive strength of materials, including ballast layers. The choice of method depends on the material being tested and the standards or protocols in place for that specific type of material. Uniaxial compressive strength test is the most straightforward and commonly used method for determining the compressive strength of materials, where a cylindrical or cubic sample of the material is prepared and the sample is placed in a universal testing machine (UTM). A compressive load is applied at a constant rate until the sample fails and the maximum load at failure is recorded, and compressive strength is calculated [20,21,22]. Triaxial compressive strength test is often used for materials such as soils and rocks, where the interaction between different stress components is important [23,24]. Point load test is commonly used for rock and aggregate materials [25,26]. Rebound Hammer test (Schmidt Hammer) is a non-destructive method and is used primarily for concrete [24,27]. Indirect tensile strength test (Brazilian test) determines the compressive strength through tensile failure [28,29]. Flexural strength test measures the tensile strength of concrete beams and provides insights into the compressive strength characteristics [20]. Split cylinder test is used to determine the tensile strength of concrete, which is then correlated with compressive strength [29,30,31]. Each method has its specific applications and suitability depending on the material and context of the test. Uniaxial compressive strength tests are the most direct and widely used method, particularly for materials such as concrete and rocks. Other methods, such as the rebound Hammer test and the triaxial test, offer additional insights and are useful in different scenarios.

This study has several objectives: (1) analyzing the effect of using latex and roving fiber on the characteristics of the HMA mixture consisting of 60/70 penetration asphalt; (2) evaluating the effect of using a mixture of 60/70 penetration asphalt, latex, and roving fiber as binder and stabilizer on the compressive strength and abrasion of ballast layer structures of conventional railway lines; and (3) investigating the performance of rail tracks with conventional ballast layer structures without a mixture, compared with the performance of rail tracks with ballast layer structures mixed with asphalt, latex, and roving fiber. This research has the potential to shed light on how conventional railway lines can be enhanced in terms of quality and durability by combining asphalt, latex, and roving fiber. It could also inform government policy regarding the application of mixtures of asphalt and latex in conventional rail lines in Indonesia, which might have their structural capacity increased by focusing on the ballast layer structure.

2 Research method

2.1 Equipment and materials

Figure 1 exhibits the equipment employed in this research located at the Transportation and Highway Laboratory of the Civil Engineering Department of Universitas Muhammadiyah Yogyakarta, encompassing a scale (Figure 1a), a sieve (Figure 1b), a cup (Figure 1c), a Los Angeles test (Figure 1d), an oven (Figure 1e), a penetrometer (Figure 1f), a pycnometer (Figure 1g), a ductility test (Figure 1h), a ballast box (Figure 1i), a manual compactor (Figure 1j), a 40 × 20 cm steel plate (Figure 1k), and UTM for compressive strength test (Figure 1l). Figure 2 displays four materials tested in this study: ballast aggregates (Figure 2a), 60/70 penetration-grade asphalt binder (Figure 2b), roving fiber (Figure 2c), and latex (Figure 2d). This research began with physically and mechanically examining the ballast aggregates and 60/70 penetration-grade asphalt binder. Subsequently, the materials were mixed, and the specimens were compacted 25 times per layer. Finally, the specimens were tested for compressive strength, and the results were analyzed.

Figure 1

Equipment: (a) scale, (b) sieve, (c) cup, (d) Los Angeles test, (e) oven, (f) penetrometer, (g) pycnometer, (h) ductility test, (i) ballast box, (j) manual compactor, (k) 40 × 20 cm steel plate, and (l) UTM for compressive strength test.

Figure 2

Materials: (a) ballast aggregates, (b) 60/70 penetration-grade asphalt binder, (c) roving fiber, and (d) latex.

2.3 60/70 penetration-grade asphalt binder testing

The research utilized asphalt with a penetration value of 60/70, which was sourced from the Transportation and Highway Laboratory of the Civil Engineering Department of Universitas Muhammadiyah Yogyakarta (Figure 2b). Initially, the solid asphalt was heated for a minimum of 4 h at a temperature of 155°C. Subsequently, the asphalt underwent testing following the Indonesian National Standard (BSN) for penetration, softening point, specific gravity, ductility, and oil losses; the findings are displayed in Table 4. Figure 3 illustrates the specimens. For every 10 cm thickness of the ballast structural layer, 60/70 penetration-grade asphalt binder was poured onto the ballast layer’s surface, accounting for 2 and 4% of the ballast’s weight, respectively.

Table 4

Test results of the 60/70 penetration-grade asphalt binder

Parameter	Result	Specification	Unit	Standard
Penetration	61.6	60–70	0.1 mm	SNI 2432-2011 [35]
Softening point	53	48 minimum	Celsius	SNI 2434-2011 [36]
Specific gravity	1.05	1.0 minimum	g/cc	SNI 2441-2011 [37]
Ductility	160	≥100	100–200 cm	SNI 06-2432-1991 [38]
Oil losses	0.2	≤0.8	%	SNI 06-2440-1991 [39]

Figure 3

60/70 penetration-grade asphalt binder testing: (a) penetration test using penetrometer, (b) softening point test, (c) specific gravity test, (d) ductility test, and (e) oil losses test using oven.

2.7 Material preparation and specimen compaction

This research primarily focused on a ballast-layer structure arranged in a 40 × 20 × 30 cm ballast box. The ballast was arranged in the ballast box until it had a thickness of 30 cm divided into three layers, each with a thickness of 10 cm. A manual compactor with a pounder was employed for compaction; the pounder’s rectangular flat surface measured 40 × 20 cm, and the compactor weighed 4.5 kg. Its free-fall height was 45.7 cm. The relationship between stress (σ) and strain (ε) caused by loading from the UTM was determined by conducting a compressive strength test on each specimen once they were prepared.

Figure 4a depicts the compaction of ballast within the ballast box with a thickness of 10 cm for each layer. As demonstrated in Figure 4b, the latex and 60/70 penetration-grade asphalt binder were mixed. Figure 4c illustrates the pouring of the mixture into the ballast structure while putting the roving fiber on top of the ballast and asphalt–latex mixture appears in Figure 4d. Furthermore, Figure 4e exhibits a ballast specimen with a total thickness of 30 cm. Figure 4f displays the final process of pouring the asphalt–latex mixture into the ballast structure.

Figure 4

(a) compaction of ballast every 10 cm of layer thickness, (b) mixing of 60/70 penetration-grade asphalt binder and latex, (c) pouring of asphalt–latex mixture into ballast structure, (d) roving fiber placement on top of ballast and asphalt–latex mixture, (e) ballast specimen with total thickness of 30 cm, and (f) the last pouring of asphalt–latex mixture.

2.8 Compressive strength test using UTM

Compressive load refers to the amount of load affecting a specimen based on the area of its surface. As displayed in Table 5, compressive strength testing was performed on specimens using a UTM with a maximum load of 6 tonnes. This test aims to obtain several performance parameters of the ballast layer, including stress, strain, compressive strength, and percentage of ballast aggregate abrasion. The initial step was inputting the dimensions of the specimens, the surface area of the loading plate, and the amount of load to be tested. As depicted in Figure 5, a compressive strength test was carried out after the specimens, and the loading plate was placed symmetrically.

Figure 5

A specimen ready for compressive strength test using micro-computer UTM.

3 Results and discussion

3.2 Ballast abrasion

Aggregate abrasion or wear refers to the degradation or breaking of aggregate due to mechanical processes, such as forces during the railway line construction and servicing traffic loads, as well as chemical processes, such as the influence of humidity and temperature changes throughout the day. The aggregate abrasion analysis was performed using filter analysis. The abrasion value indicates the resistance of the coarse aggregate to crushing due to mechanical loads. The abrasion test was run after the compressive strength test by comparing the weight of the material crushed or passed the sieve no. ¾” to the total weight of the specimen, yielding the abrasion percentage of the specimen. Figure 6 exhibits the ballast abrasion test.

Figure 6

Ballast abrasion test.

Several factors, including material preparation, manufacturing compaction, and compression test, affected the abrasion percentage of the specimens. As stated by Setiawan [14], the new-ballast-based sample with the addition of 4% of the 60/70-grade bitumen had the abrasion value around 0.46–0.50% because ballast materials were protected from abrasion during the process of compaction and compressive testing by the presence of the well-distributed bitumen on three layers.

Figure 7 displays the various abrasion percentages of the specimens. The B1 specimen yielded the greatest abrasion value. In contrast, the specimens with the asphalt, latex, and roving fiber added generated much lower abrasion values. Protecting the ballast from abrasion caused by compaction and compression testing could be accomplished with specimens that employed additives such as asphalt, latex, and roving fiber. On top of that, specimen B3, containing 4% asphalt, depicted a lower abrasion value than specimen B4, containing 2% asphalt and 1% latex. In other words, asphalt was more effective than latex in protecting the ballast material from abrasion. According to several studies, the addition of the elastic materials such as scrap rubber on the ballast layer can reduce the degradation of ballast materials since it can minimize the movement of the ballast particles and reduce the friction between aggregates so that material durability increases and material degradation decreases [42,46,47,48].

Figure 7

Ballast abrasion.

4 Conclusion

Ballast material, commonly used in railway tracks, resists abrasion through several factors related to its physical properties and the quality of the materials used, such as material hardness, shape and size, durability, testing and standards, regular maintenance, and environmental factors. The following conclusions were derived from the results of the analysis.

1. Adding asphalt, latex, and roving fiber could enhance the compressive strength of the ballast structure.

2. Comparing the compressive strength of two ballast structures, one with 2% asphalt and 3% latex and the other with 4% asphalt and 1% latex, the former achieved a greater outcome.

3. Asphalt, latex, and roving fiber could be utilized to prevent the ballast from being abrasive during compression and compaction tests.

4. Asphalt was more effective than latex in protecting ballast from abrasion.