Influence of gauge width on rail side wear on track curves.
Keywords: railway, rail, rail wear, gauge width, curve, rail wearing intensity, wheel flange.
Rail is the main bearing element of the upper track construction. Loads, caused by the rolling-stock wheels, act on the element. Therefore, rails have to withstand large dynamic loads in vertical, longitudinal and lateral directions. In order to ensure safe railroad traffic, rails must be strong enough and wearproof. Rail head vertical wearing intensity is greater in the straight sections and on the curves of a larger radius. On the curves with radius less than 650 m change of rails is generally caused by exceeded permissible rail head side wear.
Intensive rail side wear on curves is one of the most important problems for the companies maintaining the railway infrastructure. Not only economical aspect conditions the relevance of the problem; very important factor is road safety as well. Side wear of rails depends on many factors, such as road plain and longitudinal profile, rolling-stock and road technical conditions, steel quality of rails and wheels, maintenance quality of rails and wheels, freight haulage intensity and axial loads. It is very difficult to find an integrated solution for decrement of rail side wear. Ways and methods for solving the problem differ in various countries. In America the scientists join forces to improve steel quality, create new types of fastenings etc . In Western Europe and Australia it is aimed to change the construction of wagon wheel pairs (trying to implement wagons with wheel pairs turning in line with track curve radius [2, 3]). In the developing countries rail wear is decreased looking for better ways of road maintenance, implementing more progressive means for lubrication of wheels and rails etc, some countries (for example, Russia, USA, Germany, Poland, Czech Republic and France) have railway testing centres where experimental researches are carried out and the most complicated natural maintenance conditions of railway lines are simulated. At present Lithuania does not have proper economical possibilities to carry out large and qualitative experimental research, whereas the USA and Russia develop special experimental programs for solving a particular problem (in the USA) or implement complex research programs ordered by Ministry of Transport (in Russia) [4-7].
According to the maintenance rules of Lithuanian railways , the width of the gauge between rail head internal edges in the straight sections and in the curves with radius [greater than or equal to] 350 m, must be 1520 mm. In the curves with smaller radius, the gauge width must be as follows: if radius is from 349 m till 300 m-1530 mm; if radius is [less than or equal to] 299 m-1535 mm. The gauge width in straight sections and curves must not broaden by more than 6 mm and narrow down by more than 4 mm. In the sections with the speed limitation of 50 km/h and higher, deviations must not exceed 10 mm and 4 mm respectively. These norms are the same as in Russia and they do not change since 1970. Till 1970 the gauge width in straight sections and in curves with radius [greater than or equal to] 350 m was 1524 mm. The width of the gauge in the curves with smaller radius was as follows: radius from 349 m to 300 m-1530 mm; radius [less than or equal to] 299 m-1540 mm. The gauge width in the curves and in straight sections should not increase by 6 mm and decrease by 2 mm. In 1970 deviation of 4 mm decrement of the gauge width was accepted. As can be seen, norms accepted in 1970 have decreased the gauge width and increased permissible deviations of decrement. As far as it is known, influence of these changes on interaction of trains and rails was not investigated .
The larger part of railways maintained in Lithuania has a gauge of 1520 mm (98,7 %), while railways of 1435 mm in gauge make merely 1,3 % of Lithuania's overall railway length. The distribution of curves in Lithuanian railway lines by curve radius R is portrayed in Fig 1. During 2003, 21,09 km of railways were replaced with new rails and 30,72 km were replaced with used rails in Lithuania. This cost 1,793 million Litas (including the purchase and the laying price). Rails replaced due to the exceeded head side wear tolerance for 30-40 % of all replaced rails, ie about 0,717 million Litas.
The purpose of this article is to examine the influence of gauge width on rail head side wear intensity on track curves and to carry out experimental research on rail head side wear intensity on curves in Lithuanian railway lines and to estimate the influence of gauge width on it.
2. Mathematical model of wear formation
Interaction of two surfaces (having in mind that one of the surfaces is moving in respect of another one) causes formation of wear. In the upper layers of a material, mechanical and molecular interaction begins and thus surface micro bindings are broken. Wear of two surfaces is related to a multifold fracture of frictional bindings . There are two types of frictional side wear of rails and wheel flanges: a wear caused by slippage when adjoining surfaces decompose and a wear when abrasive parts act on metal. Two main factors influence frictional side wear. They are: 1) wheel flange slippage on rail head side plain; 2) wheel flange pressure on rail head side plain . Additional factors are: steel quality of rails, its chemical composition and microstructure; toughness and roughness of the upper layers of adjoining surfaces; wheel flange tilt angle in respect to rail head; rolling-stock trolley position on the track, defined as wheel flange rolling angle on rail head; radius of the road curve; friction coefficient; lubrication of rails and wheel flanges; ambient temperature and humidity [12, 13]. As wear does not originate without slippage, the main factor determining the rail and wheel wear is the length of slippage distance of the rail flange on the side plain of the rail head.
In Fig 2 mechanical scheme of wear formation, when one part is slipping on top of another one is presented. Having the same conditions, the longer the slippage distance l, the greater the wear [DELTA]h of part 1.
[FIGURE 2 OMITTED]
In experimental and theoretical investigations the wear formation of rails and wheels is analysed as the interaction of two steel bodies, taking into consideration mechanical properties of steel. However, the disadvantage of such research is unnatural conditions, ie laboratory conditions are rather different from those of natural ones. Therefore material mass wear volume concept [V.sub.N], [m.sup.3] is used. It is directly proportional to the slippage distance l and normal load N and inversely proportional to the metal wear resistance H [7, 14, 15]:
[V.sub.N] = N x f x l / H (1)
In order to evaluate the influence of gauge width on intensity of rail head side wear, empirical dimensionless parameter--wearing index, is used. It is convenient to have such an empirical index when mathematical model is used to describe the influence of gauge geometrical parameters on rail head side wear. The wearing index evaluates factors causing wear formation. It is directly proportional to the force N acting on rail by wheel, friction coefficient f and distance of wheel slippage on rails [l.sub.s] and inversely proportional to the area A of rail contact with wheel. The wearing index [R.sub.N1] may be determined by formula [16, 17]:
[R.sub.N1] = N x f x [l.sub.s] / A (2)
From Eq (2) it follows that in order to reduce the wearing index [R.sub.N1] (and the intensity of rail head side wearing on curves), it is necessary to diminish the wheel load on the rail and the wheel flange slippage distance on side plain of the rail head.
The slippage distance [l.sub.s]. consists of orbicular [l.sub.a] and longitudinal [l.sub.i] slippage distances. The orbicular slippage distance on rails is not related to geometrical parameters of tracks but only to wheel parameters.
The gauge width impact on rail wearing is expressed through longitudinal slippage distance [l.sub.i] on rails. Reduction of longitudinal wheel slippage distance on rails would diminish the wearing index [R.sub.N1]. Longitudinal wheel slippage occurs due to the fact that the outer stretch of rail on the curve is longer than the inner one. It is practically impossible to avoid wheel slippage on rails ([l.sub.i]= 0), but there is a theoretical possibility to calculate gauge S, under which there would be no longitudinal train wheel slippage on rails :
S = 2 x p x [rho] x c + R x (a + 2b) / R - p[rho] (3)
where p is the flange tilt angle in respect to horizontal, expressed as 1:p. 1:p of new wheels equals to 1:20, and 1:p reaches 1:100, 1:200 and even more as wheels wear; [rho] is the radius of the wheel rolling circle, R--the curve radius, c--distance between side edge of rail head to axis, a--the distance between internal wheel pair flanges at altitude of where gauge width is measured and b--the flange thickness at altitude in the place where gauge width is measured.
The expression (3) allows calculating S needed to equal the longitudinal slippage distance on curve to 0. Calculations are carried out for the train car's wheel pair ([rho] = 475 mm, a = 1440 mm and b = 25 mm) and UIC60-class rails (c = 36 mm). The results analysis is presented in Fig 3. Widening of rail gauge on curves would diminish the longitudinal wheel slippage on rails.
[FIGURE 3 OMITTED]
3. Experimental research of gauge influence on rail head side wear intensity
The aim of the experimental research was to identify the influence of gauge width on external rail head side wearing intensity.
The research was carried out in June 2001-Sept 2002 on 19 curves of Lithuanian railway lines. The upper structure of the railway on curves was as follows: rails UIC60 and R65, reinforced concrete sleepers, and crushed stone ballast. The measurements were made in the length of the orbicular curve near every tenth rail axis.
Every curve contained from 25 to 35 measurement points (taking into account curve radius) . In every fixed curve point, cant, gauge and external rail head side wear were measured and rail head side wear intensity [w'.sub.s] was calculated according to the formula:
[w'.sub.s] = [DELTA]w / I x 10, mm/10 mln t (4)
where [DELTA]w--change of rail head side wear during the research period (mm); I--amount of the carried loads during the research period, mln t.
Regression analysis of the relationship between external rail side wearing intensity and value of gauge (confidence level--95 %) was carried out.
The curves used to examine the impact of gauge width upon rail head side wearing were divided into five groups according to the curve radius R:
1. R [less than or equal to] 350 m;
2. 350 m < R [less than or equal to] 400 m;
3. 400 m < R [less than or equal to] 650 m;
4. 650 m < R [less than or equal to] 900 m;
5. R > 1500 m.
On the curves with radius under 350 m, the lowest intensity of rail head side wearing was found at a gauge of 1535-1540 mm (according to the effective standards, when R<300 m, S=1535 mm, when the radius R is between 300 and 350 m, the rail head side wearing S is 1530 mm) (Fig 4). When the gauge is widened up to 1535 mm, the rail head side wearing on the curve of such radius reaches an average of 2,1 mm when 10 mln tons of gross weight loads pass. If the gauge width is equal to 1530 mm, the outer rail head side wear reaches 2,9 mm after handling the same amount of loads, which is 1,4 times higher.
[FIGURE 4 OMITTED]
After widening the gauge from 1520 mm (effective standard) up to 1530 mm on curves of radius from 350 m to 400 m and letting 10 mln tons of gross cargos pass the railway section, the rail head side wearing would be 1,5 mm. Meanwhile, the side wearing would be 2,5 mm if the gauge was left unchanged 1520 mm (Fig 5).
[FIGURE 5 OMITTED]
The majority of the curves analysed had a radius from 400 to 650 m. The results of data analysis on these curves corroborated the presumption that widening of the gauge from 1520 to 1524 mm would decrease the intensity of the rail head side wearing by 8-9 % (Fig 6).
[FIGURE 6 OMITTED]
On the curves with radius of over 650 m, rails need to be replaced not because of exceeding the rail head side wear tolerance. Experimental research shows that the intensity of rail head side wearing is not high on such curves.
Comparison of experimental results with the results obtained by other authors has shown that, on curves with radius from 400 to 650 m, the increase of gauge width from 1520 to 1530 mm would lead to decrease in rail head side wearing intensity: 1,72 times, according to the authors' research, and 1,1-2,5 times, according to researches conducted by the other authors (Fig 7) [19-21]. The results of the research conducted by the authors might differ from those of other authors due to the different conditions of experiments (different upper construction of the rails, technical condition of the rolling-stock, train traffic speed etc).
[FIGURE 7 OMITTED]
4. Possibilities to increase the durability of rails due to the gauge widening
Results of theoretical and experimental research showed that the effective standards that regulate the gauge width increase on curves and tolerances do not assure the lowest wearing intensity of the rail head on curves. Therefore, it was concluded that the standards should be specified to reduce the intensity of the rail head side wearing. The following improvements were suggested:
1. On curves with a radius of over 650 m, the rail wearing intensity is not high, thus there is no need to widen the rail gauge (gauge of 1520 mm should be installed).
2. The gauge should be widened when the curve radius is less than 650 m. On curves with such a radius, rails need to be replaced more often due to the exceeding the permissible side wearing rather than other defects. On the curves with radius from 400 to 650 m, the gauge width should be 1524 mm instead of 1520 mm. Such widening would decrease the rail head side wearing intensity from 28 to 40 % (Fig 8).
[FIGURE 8 OMITTED]
3. On the curves with radius from 350 to 400 m, the gauge should be widened from the current 1520 to 1530 mm because the gauge width of 1520 mm would lead to an increase in the rail side wearing intensity by a factor of 2 (Fig 9).
[FIGURE 9 OMITTED]
4. On the curves with the radius of less than 350 m, widening the gauge up to 1535 mm ensures the lowest intensity of the rail head side wearing.
The decrement of rails maintenance expenses was calculated. It is related to the purchase and instalment of new rails after implementing the suggested changes. The calculations were done for the entire Lithuanian railway network and have shown that the number of rails that would normally need to be replaced due to the outer rail head side wearing would be reduced by 1,178 km annually, which is estimated at 30,000 EURO. The amount of money could go up in the wake of the modernisation of the Lithuanian railway infrastructure, increase in the axial load of the railways, formation of heavyweight trains and increase in the cargo flows.
1. Frequent replacement of defective rails due to excessive rail head wearing on track curves increases railway maintenance expenses. The main factor affecting the rail wearing on curves is gauge width. The investigation of influence of gauge width on the rail wearing intensity and the possibilities of track widening to decrease the wearing is of vital importance.
2. Mathematical modelling of wear formation was done using wearing index [R.sub.N1]. Analysis of the mathematical model shows that widening the gauge on curves leads to a decreased longitudinal wheel slippage path on rails and the rail head side wearing intensity.
3. The results of the experimental research carried out on 19 Lithuanian railway lines have shown that widening the gauge on the curves with radius less than 650 m, decreases rail head side wearing intensity up to 1,72 times.
4. Based on theoretical and experimental research, suggestions are presented to revise the effective standards for installing and maintaining the rail track. After implementation of these changes, rail gauge width should be as follows: 1520 mm in the straight lines and curves with radius of more than 650 m, 1524 mm on the curves with radius of 400-600 m, 1530 mm on the curves with radius of 350-400 m and 1535 mm on the curves with radius of less than 350 m. The tolerances of the gauge widening are +6 mm to the upper side and -2 mm to the lower side. In the lines with train speed limits of under 50 km/h, the respective deviations are +10 mm and -2 mm.
5. Implementation of proposed gauge widening standards should be gradual through overhaul repairs of the upper track structure repairs, saving up to 30,000 EURO in all Lithuanian railway lines annually. The money could be successfully used for the general repairs of the railroad lines.
[1.] Clayton, P. Predicting the wear of rails on curves from laboratory data. Wear, No 181-183. Paris: Elsevier, 1995, p.11-19.
[2.] Zakharov, S. and Zharov, I. Simulation of mutual wheel/rail wear. Wear, 253(1-2). Paris: Elsevier, 2002, p. 100-106.
[3.] Updhay, R. Decrease of wheel and rail wear. International Railway Journal, 7. London: Europa House, 2000, p. 33-34.
[4.] Bogdanov, V. M.; Zacharov, S. M. Contemporary problems of the system rail-wheel. International Railways ([TEXT NOT REPRODUCIBLE IN ASCII.]), 1. Moscow, 2001. 8 p. (in Russian).
[5.] Lementjev, L. P. Wheels and rails interface and their wearing. Railway and Railway Economy ([TEXT NOT REPRODUCIBLE IN ASCII.]), 11. Moscow: MK-Poligraf, 1999, p. 6-15 (in Russian).
[6.] Podagelis I. Factors acting on points and inner edges of rails wear. In: Proceedings of Civil Engineering and Environment, Vol 387. Vilnius: Technika, 1998, p. 91-94 (in Russian).
[7.] Jendel, T. Prediction of wheel profile wear comparison with field measurements. Wear, 253. Paris: Elsevier, 2002, p. 89-99.
[8.] The rules of railway service (Gelezinkelio kelio prieziuros taisykles). Vilnius: SPAB ,Lietuvos gelezinkeliai", 2000. 213 p. (in Lithuanian).
[9.] Povilaitiene, I. and Podagelis, I. Research into rail side wearing on curves. Transport, 18(3), 2003, p. 124-129.
[10.] Andreev, A. I.; Komarov, K. L.; Karpuschenko, N. I. Wearing of rail and wheels. Railway Transport ([TEXT NOT REPRODUCIBLE IN ASCII.]), 7. Iiscow: Zeleznodoroznoje delo, 1997, p. 31-36 (in Russian).
[11.] Clayton, P. Tribological aspects of wheel-rail contact: a review of recent experimental research. Wear, 191. Paris: Elsevier, 1996, p. 170-183.
[12.] Ilsiuk, V. S. The reasons and mechanics of rail side wearing of rail and wheels. Railway and Railway Economy ([TEXT NOT REPRODUCIBLE IN ASCII.]), No 5. Moscow: IE-Poligraf, 1996, p. 15-20 (in Russian).
[13.] Markov, D. Laboratory tests for wear of rail and wheel steels. Wear, No 181-183. Paris: Elsevier, 1995, p. 678-686.
[14.] Olofssom, U. and Telliskivi, T. Wear, plastic deformation and friction of two rail steels--a full-scale test and a laboratory study. Wear, 254. Paris: Elsevier, 2003, p. 80-93.
[15.] Meng, H. C. and Ludema, K. C. Wear models and predictive equations: their form and content. Wear, No 181-183. Paris: Elsevier, 1995, p. 443-457.
[16.] Karpuschenko, R. I. Lubrication--the only way to reduce the wear. Railway and Railway Economy ([TEXT NOT REPRODUCIBLE IN ASCII.]), No 5. Moscow: IE-Poligraf, 2000, p. 15-18 (in Russian).
[17.] Andreev, A. I.; Komarov, K. L.; Karpuschenko, R. I. Wear of rails and wheels. Raiway Transport ([TEXT NOT REPRODUCIBLE IN ASCII.]), 7. Iiscow, 1997, p. 31-36 (in Russian).
[18.] Povilaitiene I. and Laurinavieius A. Reduction of external rail wearing on road curves. Journal of Civil Engineering and Management, Vol X, No 2, 2004, p. 123-130.
[19.] Plotkin V. S. and Kuzmich L. D. About the increased wearing of wheels and rails. Railway Transport ([TEXT NOT REPRODUCIBLE IN ASCII.]), No 8, 1997, p. 51-54 (in Russian).
[20.] Nikiforov B. D. Reasons for rail and wheel wear and ways to decrease it. Railway Transport ([TEXT NOT REPRODUCIBLE IN ASCII.]), No 10, 1995, p. 36-40 (in Russian).
[21.] Karpuschenko N. I. Track parameters and rail wear. Railway and Railway Economy ([TEXT NOT REPRODUCIBLE IN ASCII.]), No 8. Moscow: IE-Poligraf, 1996, p. 6-7 (in Russian).
GELEZINKELIO VEZES PLOEIO ITAKA BEGIU SONINIAM DILIMUI KREIVESE
I. Povilaitiene, I. Z. Kamaitis, I. Podagelis
Pagrindinis gelezinkelio sistemos turtas yra infrastrukturos objektai ir riedmenys. Jiems islaikyti reikia dideliu islaidu, o renovacijai--investiciju. Straipsnyje nagrinejama nuolat aktuali gelezinkelio problema--tai begiu dilimo mazinimo galimybe kreivese. Autoriu tyrimo tikslas yra nustatyti, kaip vezes platinimas ir leistinieji nuokrypiai daro itaka begiu dilimui kreivese, kuriu spindulys mazesnis nei 650 m. Apskaieiuota, kiek sumazetu gelezinkelio prieziuros islaidos, susijusios su nudilusiu begiu keitimu kreivese, kai kelio veze kreivese, kuriu spindulys mazesnis nei 650 m, platinama pagal pateiktus siulymus. Tam tikslui straipsnyje pateikiamas matematinis modelis. Siuo modeliu ivertinama vezes ploeio itaka begiu dilimui kreivese. Eksperimentiniai tyrimai buvo atlikti Lietuvos gelezinkelio liniju kreivese. Pateikiama eksperimentiniu tyrimu rezultatu analize.
Reiksminiai zodziai: gelezinkelis, begis, begio nuodyla, vezes plotis, kreive, begio dilimo intensyvumas, ratu antbriaunis.
Inesa Povilaitiene, Ipolitas Zenonas Kamaitis, Igoris Podagelis Vilnius Gediminas Technical University (VGTU), Sauletekio al. 11, LT-10223 Vilnius, Lithuania. E-mail: firstname.lastname@example.org, email@example.com, firstname.lastname@example.org
Inesa POVILAITIENE. Doctor, Associate Professor at the Dept of Roads since 2005. Research interests: upper railway construction, its strength and durability, railway traffic organisation.
Ipolitas Zenonas KAMAITIS. Dr Habil, Prof Emeritus at the Dept of Bridges and Special Structures. Member of IABSE since 1999. Author and co-author of more than 150 publications, including 6 books. Research interests: concrete structures and bridges, materials, durability, monitoring, and refurbishment.
Igoris PODAGELIS. Doctor, Associate Professor at the Dept of Roads. Author and co-author of more than 40 publications, including 1 book. Research interests: upper railway construction, its strength and durability, technology of repairs.
Received 17 Jun 2005; accepted 01 Feb 2006
Fig. 1 Distribution of curves according to radius R in Lithuanian railways (900<R [less than or equal to] 1500 m) 37,0 % (650 m<R [less than or equal to] 900 m) 13,5 % (400<R [less than or equal to] 650 m) 15,5 % (R [less than or equal to] 400 m) 0,4 % (R > 1500 m) 33,6 % Note: Table made from pie chart.
|Printer friendly Cite/link Email Feedback|
|Author:||Povilaitiene, Inesa; Kamaitis, Ipolitas Zenonas; Podagelis, Igoris|
|Publication:||Journal of Civil Engineering and Management|
|Date:||Jul 1, 2006|
|Previous Article:||Study of sorptivity of self-compacting concrete with mineral additives.|
|Next Article:||Integrated agent-based construction equipment management: conceptual design.|
|Letter: On the rails.|