Construction and Building Materials 133 (2017) 154–162
Contents lists available at ScienceDirect
Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat
Experimental investigation on the application of quick-hardening mortar for converting railway ballasted track to concrete track on operating line Il-Wha Lee, Sukhoon Pyo ⇑ Korea Railroad Research Institute, 176 Railroad Museum Road, Uiwang-si, Gyeonggi-do 16105, South Korea
h i g h l i g h t s Converting technology developed to transform the existing ballasted tracks into concrete tracks. A quick-hardening cement mortar was developed that meets the functional requirements. Quick-hardening mortar was applied to a 52 m-long existing ballasted track of an operating line. Running stability of the train on the converted track was evaluated.
a r t i c l e
i n f o
Article history: Received 9 June 2016 Received in revised form 23 October 2016 Accepted 14 December 2016
Keywords: Quick-hardening mortar Railway track Concrete track Field application Track performance Pre-packed concrete
a b s t r a c t The ballasted track system has been widely adopted on conventional railways due to its ease of construction, but its key shortcoming is its need for frequent maintenance. In recent years, with the growing demand for railway vehicles capable of higher speeds and capacities, the concrete track is becoming more popular due to its high rigidity and relatively low maintenance. A fast converting technology was developed in this study to transform the existing ballasted tracks into concrete tracks using the developed quick-hardening mortar, which has high fluidity and strength development at an early stage, such that it could minimize the interruption of train operation. To assess the applicability of the developed mortar in the field, durability, fluidity and strength development of the mortar were characterized. The developed quick-hardening mortar was applied to a 52 m-long existing ballasted track of an operating line, and structural stabilities under moving train loads were evaluated. Experimental results showed that the track conversion was successful with the quick-hardening mortar, and the concrete structures proved to have excellent performance compared to the existing ballasted track. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction A ballasted track system is the most widely used track system, and transports vehicle loads to sleepers and ballasts beneath the sleepers to reach the trackbed. The track system maintains its stability through the friction created in between gravel, while offering the sufficient rigidity needed for the smooth operation of the train. As the oldest type of track system in the history of railways, the ballasted track is beneficial in terms of having low investment costs in its early stage, and high elasticity. On the other hand, the use of crushed gravel causes dust and contamination, requiring regular cleansing [1], as well as track irregularity due to the repetitive loading of running trains [2]. In recent years, as train speeds have increased, with high-speed trains capable of traveling over ⇑ Corresponding author. E-mail address:
[email protected] (S. Pyo). http://dx.doi.org/10.1016/j.conbuildmat.2016.12.049 0950-0618/Ó 2016 Elsevier Ltd. All rights reserved.
300 km/h, so has the burden of maintaining and repairing the ballasted track [3]. A new track system made of concrete that addresses these weaknesses of the ballasted track has become very popular in recent years [4–6]. Although the construction of the concrete track is expensive in terms of early investment and is a more highly sophisticated process, the concrete track is highly durable and resistant to track destruction, reducing the need for maintenance [4]. In addition, it will improve track regularity and reduce deformation while enhancing ride comfort for passengers [6]. The key measure to take to reduce maintenance with the ballasted track is to get rid of the ballast layer and construct new concrete tracks. However, changing an existing ballasted track into a concrete track requires a lengthy interruption of train operations. To tackle this problem, researchers have pursued a technology that transforms the track system by pouring fast-curing cement paste or mortar deep into the ballast layer during the interruption of
155
I.-W. Lee, S. Pyo / Construction and Building Materials 133 (2017) 154–162
train operation on a revenue service line [7,8]. This can be considered as a pre-packed concrete technique; however, since the mortar-filling work has to be done within a short period of time so that trains can resume operation thereafter, the material needs to have the property of curing at ultra-high speed, and at the same time a certain amount of working hours for the job are required. The converted concrete track using quick-hardening mortar is an application method using pre-packed concrete, in which preplaced aggregates are filled first, followed by the mixture of water and cement; while to form general concrete aggregates, water and cement are mixed together simultaneously. The suggested construction process is as follows: the existing gravel is replaced by the clean-washed gravel on top of sheets of geotextiles, and quick-hardening mortar is poured into the upper layer of ballasts beneath sleepers, which then becomes a solid slab structure. Fig. 1 illustrates the concept of converting a railway track using quick-hardening cement mortar. Accordingly, the ballast layer within the area from sleepers to where geotextiles are installed is converted into a single concrete slab layer by casting with prepacked material. Furthermore, the existing crushed stone base could also work as a drainage layer in the converted concrete track system. Additional crossing culverts and/or a side water drainpipe could be installed for harsh environments. This experimental research focused on the development of a quick-hardening cement mortar which meets the aforementioned requirements, along with tests to ensure the material’s durability on a lab scale. Prior to its field application, tests on injection and strength were performed in an environment created that is similar to field conditions. In the final stage, field applicability of the new material was assessed using different types of measurements and analyses after the field test. 2. Development of quick-hardening mortar The converted concrete track made of quick-hardening mortar is an engineering method of converting ballasted track into concrete track, and the construction usually is carried out at night during the hours in which there are no trains being operated, necessitating a speedy construction process. A quick-hardening
mortar is developed in this section to come up with the most optimized mixture that meets chemical and physical requirements. The size of the aggregates used throughout this study was maintained at 22.4–63 mm, which is exactly the same as that of the gravel used for conventional tracks. 2.1. Raw materials and proportions of their mixture Table 1 shows the proportions of the mixture used in this filling method that meet the requirements of high fluidity, fluiditymaintainability for over a certain period of time, quick hardening and high durability. Chemical proportions used for the mixture of rapid-hardening cement specially developed in this research are illustrated in Table 2. 2.2. Fluidity of the developed mortar In terms of fluidity, it is important to secure substantial time in the injection process so that the material can flow deep down into the bottom of the structure. If the material becomes hardened within just a few minutes, it will have difficulty reaching the bottom of the ballast layer, causing the problem of insufficient filling. This approach also has the aim of offering methods of maintaining fluidity for the safe and efficient performance of mortar casting, while securing enough time for the work in the field. To satisfy the aforementioned requirement, the developed quick-hardening mortar is designed to maintain fluidity for over 20 min by using set-control agents, as can be seen in Table 1. Fluidity was evaluated by J14-funnel efflux time, based on JSCE F 541 [9]. Fig. 2 shows the results of the travel time of the developed mortar over time after the first chemical reaction starts. As can be seen from the graph, fluidity was maintained for more than 20 min, which was the target. 3. Durability tests of developed quick-hardening mortar The durability of concrete railway tracks could be degraded by biological, chemical and physical factors, and for this reason it is critical to ensure durability to maintain long-term stability in
Fig. 1. Concept of converting railway track using quick-hardening cement mortar.
Table 1 Details of mix proportions (by mass).
§ y à
Cement
Water
HRWRA§
Set-control agent
Sand Ay
Sand Bà
1
0.40
0.027
0.0014
0.33
0.34
HRWRA: High range water reducing agent. Maximum grain size = 1.2 mm. Maximum grain size = 0.42 mm.
Table 2 Chemical composition of the cement. Oxide
SiO2
Al2O3
Fe2O3
CaO
MgO
SO3
K2O
Na2O
LOI
Content (wt.%)
13.4
15.0
1.9
51.2
1.79
12.9
0.43
0.13
3.25
156
I.-W. Lee, S. Pyo / Construction and Building Materials 133 (2017) 154–162
structures. To rank the chloride penetration resistance of the developed mortar, the rapid chloride permeability test (RCPT), a procedure for measuring the electrical conductivity of concrete, was carried out according to AASHTO T277 [13] (ASTM C1202 [14]). The cylindrical concrete specimen used in the test was 150 mm in diameter and 50 mm thick. More than two specimens were used to measure RCPT for two different curing ages. The averaged passed charges were 552 C (Coulombs) in 7 days of curing and 442 C in 28 days, which is very low chloride ion penetrability (100–1000 C) according to ASTM C 1202. The results showed that the values of curing for 7 days and 28 days each turned out to be very close, suggesting that the characteristic of permeability is not affected by the duration of curing time. 3.3. Chemical environmental resistance Fig. 2. Flow of quick-hardening mortar.
terms of track conditions. The aim of this durability test was to evaluate freeze-thaw and chemical (e.g. acid rain) resistances of the concrete track. 3.1. Freeze-thaw resistance The freeze-thaw resistance of the developed concrete specimens (100 100 400 mm3) was evaluated according to ASTM C 666 [10] procedure A. After casting, the specimens were covered with plastic sheets and stored at room temperature for 24 h prior to demolding. The specimens were then submerged in tap water at 20 °C for an additional 13 days. Temperature variation for the test was from 4 °C to 18 °C. Two specimens were tested, and relative dynamic modulus of elasticity was measured based on longitudinal mode described in ASTM C 215 [11] every 30 cycles, and is listed in Fig. 3. As can be seen in the figure, the developed material showed a relative dynamic elastic modulus of more than 80 percent. This meets the performance requirement for grade 2 high performance concrete. According to Goodspeed et al. [12], grade 2 performance is a level which enables it to be applied to a region with over 50 freeze-thaw cycles per year. 3.2. Chloride ingress Chloride ingress often leads to the corrosion of the reinforcing steel and a consequent reduction in the strength of concrete
Fig. 3. Freeze-Thaw results of the concrete.
The chemical resistance was determined under aggressive chemical environmental conditions in accordance with ASTM C267 [15] using the following chemicals: 2% hydrochloric acid (HCl), 5% sulfuric acid (H2SO4), 4% calcium chloride (CaCl2). The specimens used for the test were cylindrical in shape, with a diameter of 150 mm and a length of 300 mm. The specimens were cured in the air (20 °C and 50% RH) for seven days after casting and then immersed in the chemicals for 28 days. During 28 days of immersion at room temperature, the chemical resistance was evaluated about every three days. For the evaluation, specimens were cleaned with deionized water and then the weight change of the specimens was determined as follows:
weight change ð%Þ ¼
W1 W2 100 W1
where W1 is the weight of the specimens before immersion and W2 is the weight of cleaned specimens after immersion in the chemicals. Fig. 4 shows the weight changes of the specimens under aggressive chemical environments. The pre-packed concrete specimen with 2% hydrochloric acid showed a weight loss of 0.40% in 9 days, 1.39% in 18 days and 1.82% in 28 days on average; weight loss of 0.83% in 9 days, 1.13% in 18 days and 1.40% in 28 days with 5% sulfuric acid; and weight gain of 1.13% in 9 days, 1.56% in 18 days and 1.45% in 28 days with 4% calcium chloride, respectively. The results show that the developed mortar has a strong chemical resistance property and a tendency of saturation over time. 4. Evaluation of applicability of the developed quick-hardening mortar In order to evaluate the field applicability of the developed quick-hardening mortar, a pilot scale specimen was built to mimic the field conditions. Table 1 shows the mixture proportions used in the test, and a specimen was built at the size of 1.5 1.5 0.45 m to fill in the developed mortar. The entire procedure is illustrated in Fig. 5. After casting, observations were made on the side of the specimen to check on injected status, and on the inside of the specimen structure by coring after curing. The injection performance proved to be quite successful when observed on the side (Fig. 5c) and by the cross section of the cored specimen (Fig. 5d). Furthermore, considering the minimum aggregate size of ballast, 22.4 mm, filling the ballasted structure is much more straightforward compared to conventional prepacked concrete. Table 3 reveals the results of tests of compressive strength of the specimen made in cylindrical molds and cored specimen from the structure: averaged values from three specimens, respectively. The developed mortar proves to be excellent in terms of strength performance, as the results indicate at its early stage.
157
I.-W. Lee, S. Pyo / Construction and Building Materials 133 (2017) 154–162
Fig. 4. Chloride ion penetrability of the concrete.
Fig. 5. Experimental process of the structural application (a) filling the aggregates in the mold; (b) pouring the mortar; (c) side view of the specimen; (d) cross section of cored specimen.
Table 3 Compressive strength of the structural test. Specimen
Specimen made in cylindrical molds Cored specimen from the structure
Materials
Mortar Concrete Concrete
Compressive strength (MPa) 2h
1 day
3 days
7 days
28 days
12.3 9.0 –
24.1 16.0 11.0
30.4 21.4 22.1
34.6 26.7 28.4
49.4 31.7 33.4
5. Evaluation of field applications The feasibility of the field application was evaluated under moving train loads after converting the existing ballasted track using the developed quick-hardening. A ballasted track that had
been operating for more than 30 years at the time with an annual passage load of more than 5000 tons was selected for the application. Two weeks prior to the infilling, the existing gravel was removed before compacting the trackbed, placing waterproof geotextiles, sleepers and clean gravel (see Fig. 6). The pre-process was
158
I.-W. Lee, S. Pyo / Construction and Building Materials 133 (2017) 154–162
Fig. 6. Field application process of the quick-hardening mortar (a) existing ballasted track before conversion; (b) stabilization of new gravel; (c) pouring the mortar; (d) converted concrete track.
Fig. 7. Schematic of sensor installation (not to scale).
carried out within three hours of train operation interruption. Replaced gravel sat for two weeks for stabilization before the infilling of mortar. A 52-meter-long track was successfully constructed using the developed mortar. Fig. 6(c) and (d) show the infilling procedure and completed converted track, respectively. The geotextile used in the converted concrete track with quickhardening mortar should hold mortar so that it does not leak. Since the developed mortar has high fluidity, leaking is a risk if even a tiny hole exists, which could lead to a serious problem with the product. Many different physical properties and performance metrics were considered to select spun bonded nonwoven geotextiles with PET weight of 300 g/m2, tensile strength of 90 kN/m and permeability coefficient of 1.0 101 cm/s, which had been doublelayered to prevent possible damage of the fabric.
The approaching track needs to gradually bridge the rigidity gap between the converted concrete track and the existing ballasted track. For this purpose, elasticity pads used on the approach track have greater value (75 kN/mm) than the fastener stiffness (22.5 kN/mm) of the converted track so that the track support stiffness could go through a transitional phase. Fig. 7 shows the schematic of the concrete track laid between ballasted tracks. The track and rail behavior were measured to evaluate running stability under actual train loads in the sector after four months of construction. Linear variable displacement transformer (LVDT), strain gauge and piezoelectric accelerometers with a sampling rate of 1000 Hz were used to measure relative displacement, strain at the foot of rail and acceleration, respectively. By measuring 10train car loads at a speed of 65 km/h four times each, the measured
I.-W. Lee, S. Pyo / Construction and Building Materials 133 (2017) 154–162
159
Fig. 8. Deployment of sensors.
results are analyzed and categorized. Fig. 7 shows five locations with installed sensors, whereas Fig. 8 shows examples of sensor installation. It should be noted that relative vertical displacement and acceleration of a track in an approaching track were estimated by measuring the data on a sleeper, not on the gravel. Fig. 9 shows examples of recorded data at C1 spot and Fig. 10 shows the relative vertical displacement of rail at each location. Fig. 11 shows the peak values from four tests at each spot for relative vertical and horizontal displacements of rail, vertical
displacement of track and strain of rail. The measured vertical displacement in the converted concrete track sector stands at less than 0.5 mm, showing lower displacement compared with the approaching track where pad elastic modulus of high rigidity is applied. Due to sudden changes in stiffness between the converted concrete track and the approaching track, higher displacement was observed in B1 location compared to B2 and B3. Although vertical displacement on the converted concrete track is low with high rigidity, it was measured as relatively high displacement in
Fig. 9. Examples of recorded data at C1 spot: (a) relative vertical displacement of rail; (b) relative horizontal displacement of rail; (c) relative vertical displacement of track; (d) strain of rail.
160
I.-W. Lee, S. Pyo / Construction and Building Materials 133 (2017) 154–162
Fig. 10. Relative vertical displacement of rail at each location.
Fig. 11. Peak values of each location for four tests: (a) relative vertical displacement of rail; (b) relative horizontal displacement of rail; (c) relative vertical displacement of track; (d) strain of rail.
I.-W. Lee, S. Pyo / Construction and Building Materials 133 (2017) 154–162
Rail at C1
Track at C1
Rail at C2
Track at C2
Rail at B1
Track at B1
Rail at B2
Track at B2
Rail at B3
Track at B3
Fig. 12. Examples of recorded acceleration of rail and track at each location.
161
162
I.-W. Lee, S. Pyo / Construction and Building Materials 133 (2017) 154–162
approaching track. As there is no applicable standard presented for urban trains, the running judgment criteria of the Japanese Shinkansen high speed train [16,17] could be referenced as a guideline, and here a maximum 4 mm of displacement is suggested. Therefore, it can be concluded from the measurements that running safety of the vehicle is satisfied since the converted track in this study is designed for relatively slow speed trains. As can be seen in Fig. 11(b), there is no major difference in the horizontal displacement of rail from one spot to the others. Similar to rail, the relative vertical displacement of the track showed the highest displacement at B1, the transitional area. In terms of strain, a relatively low amount of strain can be found at point B1 since the rail at the point is less constrained. Fig. 12 shows an example of a set of acceleration measurements on rail and track at each location. It can be seen from the figure that acceleration of rail in the converted concrete track is higher than that in the approaching track, which is a different trend compared with the displacement measurement. In other words, it is considered that in concrete track where both structural rigidity and damping are high, moving loads influence rail more than track. In the ballasted track however, moving loads disperse in the sub ballasts layer rather than on the rail. Therefore, the moving loads affect the track structure less in the concrete track system, which otherwise will take a toll on maintenance. On the other hand, the rail parts might require concern for maintenance. It can be concluded from the findings that the construction method turned out to be a success when applied in the field, with high rigidity of the structure. In this research, the quick-hardening mortar is developed to convert the existing ballasted track into a concrete track within the limited non-operation time (2–3 h). The performances of the converted concrete track were evaluated under actual operating train loads while making relative comparisons with the approaching track. The results show that there is little vertical displacement, which indicates excellent train driving compared with the existing ballasted track while relieving the burden of structural loads. Consequently, this can be considered a very efficient system in terms of maintenance. 6. Conclusions This experimental study investigated the application of quickhardening mortar to convert a ballasted track into a concrete track in order to minimize the maintenance work required for the existing railway tracks. A quick-hardening cement mortar was developed in this study that offers sufficient fluidity, durability and rigidity at an early stage in order to apply it to the existing ballasted track. The new material was applied to a certain segment on a railroad track used for actual train operation to convert it into a concrete track. The performance of the converted track was verified under actual metro train loads. The key observations and findings of this study can be summarized as follows: 1. The quick-hardening mortar has an adequate fluidity that lasts more than 20 min and presented a compressive strength of 9 MPa after two hours of casting. Also, the cured specimen obtained from the casting under field conditions presented a compressive strength over 33 MPa in 28 days. 2. Long-term durability of the quick-hardening mortar has been verified through tests such as freeze-thaw resistance, chloride ingress, and chemical environmental resistance. Based on the literature and criteria guidelines, it was found that the mortar did not have any problems related to its exposure to harsh environments. 3. The new method was implemented successfully on a total length of 52 m ballasted track that was subject to actual train
operation and was over 30 years old. Two weeks prior to the casting, there had been a series of stabilization work after replacing ballast aggregates, and the converting job was performed within the short time that train operation was interrupted. The entire process of this research illustrates that the new material and construction method can efficiently convert tracks without any additional interruption to train operation. 4. Running stability of the train on the converted track was evaluated with measurements taken from track and rail under actual train loads. The highest rail displacement on the concrete track was relatively low compared with the approaching track. Analysis of acceleration results proved that the improved track functioned sufficiently well to form a solid structure with high rigidity, which will reduce maintenance. As an extension of this research, the authors attempt to monitor and analyze the long term behavior of the converted structural behavior, including the monitoring of incidents such as possible cracks. Acknowledgments This research was supported by a grant (16RTRP-B065581-04) from Railroad Technology Research Program funded by Ministry of Land, Infrastructure and Transport of Korean Government. The opinions expressed in this paper are those of the authors, and do not necessarily reflect the views of the sponsors. References [1] R. Jack, P. Jackson, Imaging attributes of railway track formation and ballast using ground probing radar, NDT & E Int. 32 (8) (1999) 457–462. [2] R.N. Iyengar, O.R. Jaiswal, Random field modeling of railway track irregularities, J. Transp. Eng. 121 (4) (1995) 303–308. [3] I.W. Lee, S. Pyo, Y.H. Jung, Development of quick-hardening infilling materials for composite railroad tracks to strengthen existing ballasted track, Composites B Eng. 92 (2016) 37–45. [4] M.J.M.M. Steenbergen, A.V. Metrikine, C. Esveld, Assessment of design parameters of a slab track railway system from a dynamic viewpoint, J. Sound Vib. 306 (1) (2007) 361–371. [5] X. Lei, B. Zhang, Analysis of dynamic behavior for slab track of high-speed railway based on vehicle and track elements, J. Transp. Eng. 137 (4) (2010) 227–240. [6] S.D. Tayabji, D. Bilow, Concrete slab track state of the practice, Transp. Res. Rec. 1742 (2001) 87–96. [7] I.W. Lee, Optimal design of cement mortar pouring type paved track, J. Korean Soc. Railway 9 (2) (2006) 222–229 (in Korean). [8] T. Takahashi, Y. Momoya, K. Itou, H. Naganuma, Y. Oikawa, M. Suzuki, H. Suzuki, Development of ballastless track for existing Shinkansen-line in cold region, Railway Tech. Res. Inst. Q. Rep. 28 (6) (2014) 11–16 (in Japanese). [9] Japanese Society of Civil Engineering standard. Test Method of Flowability for Filling Mortar, JSCE-F 541, 2010. [10] ASTM C666/C666M-15, Standard Test Method for Resistance of Concrete to Rapid Freezing and Thawing, ASTM International, West Conshohocken, PA, 2015. [11] ASTM C215-14, Standard Test Method for Fundamental Transverse, Longitudinal, and Torsional Resonant Frequencies of Concrete Specimens, ASTM International, West Conshohocken, PA, 2014. [12] C.H. Goodspeed, S. Vanikar, R. Cook, High-performance concrete defined for highway structures, Concr. Int. 18 (2) (1996) 62–67. [13] AASHTO T277-86, Rapid Determination of the Chloride Permeability of Concrete, American Association of States Highway and Transportation Officials, Standard Specifications—Part II, Tests, Washington, D.C., 1990. [14] ASTM C1202-12, Standard Test Method for Electrical Indication of Concrete’s Ability to Resist Chloride Ion Penetration, ASTM International, West Conshohocken, PA, 2012. [15] ASTM C267-01, Standard Test Method for Chemical Resistance of Mortars, Grouts, and Monolithic Surfacings and Polymer Concretes, ASTM International, West Conshohocken, PA, 2012. [16] A. Namura, K. Matsuo, S. Miura, Introduction of buffers into a transitional track stiffness region, Railway Tech. Res. Inst. Q. Rep. 11 (2) (1997) 39–42 (in Japanese). [17] M. Ishida, S. Miura, A. Kono, Track deforming characteristics and vehicle running characteristics due to the settlement of embankment behind the abutment of bridges, Railway Tech. Res. Inst. Q. Rep. 12 (3) (1998) 41–46 (in Japanese).