THE RESEARCH OF THE RAILWAY RAIL FOR ANALYSIS OF SURFACE INITIATED ROLLING CONTACT FATIGUE CRACKS

Problems . Under the influence of dynamic loading from trains and natural conditions, the railway track deteriorates, characterized by the appearance of defects and residual deformations of the railway track, which lead to accelerated wear and malfunction of other elements of the rail. The appearance of the specified defects, failures and destructions of the elements of the rail grating is associated with the action of additional local stresses arising in the rails of the railway track. On a low-quality track grating with deviations of the parameters of the rail track, when the train is moving, there is an unaccounted interaction of the track and rolling stock with local overload of the interaction elements along the path. As a rule, the unaccounted interaction of track and rolling stock is accompanied by the impact of the wheel and the rail with overloading the contact area. When the impact of the wheel and rail strikes, stresses exceeding the endurance limit occur in the rail head. In this case, the rails are born in the head and begin to develop defects of contact-fatigue origin in the form of horizontal vertical and transverse cracks. The aim of the study . Conducting practical studies of the origin and development of contact-fatigue cracks by wear of the rails in the laboratory. Methods of implementation . An experimental unit was designed and manufactured to investigate the rails for contact damage. The unit is mounted on a mechanical single-cylinder press of the K 2322 model with a nominal force of 16 tons at a nominal stroke frequency of 120 per minute. The bulk of the experimental setup contains a wheel-rail friction unit. The force of clamping the rail to the wheel. Provides a hydraulic system. For testing, a test program was developed, which provides for two-hour tests, after which tests for the presence of defects (micro cracks) on the rail are conducted. Research on the presence of defects is carried out by the method of magnetic powder non-destructive testing. Research results . Studies have shown that contact-fatigue cracks in the rail head occurred after approximately 200 thousand load cycles. This generally coincides with the results presented in other papers, which show the occurrence of similar damage after 190-290 thousand cycles, depending on the properties of rail steel (hardness, surface roughness, etc.). Conclusions. The main cause of the origin and development of contact fatigue defects in the rail head is insufficient contact fatigue strength of the rail steel. For carrying out practical researches in laboratory conditions the methodology was developed, the experimental unit was constructed and made. Experimental studies have shown that the occurrence of defects on the surface of the rail is observed after 200-250 thousand load cycles. With the further increase in the number of load cycles, there is a rapid development of existing contact-fatigue cracks and the formation of new cracks in the contact zone.


Introduction.
In most of the countries railway is the foundation of the transport system, which deals with 50 percent or more of passenger and freight flow.
To ensure the cargo and passengers transportation with necessary reliability and safety, the railway track must always be in good working condition, it also must have the necessary stability, reliability and low operating costs.
Under the influence of dynamic loading from trains and natural conditions, the railway track deteriorates. Rail wear intensity depends on many factors. Rail wear rate is affected by: acting load (wheel contact pressure on rail), temperature (contact), locomotive's type and mode of motion, impact from the environment, physicochemical modification of surfaces during friction and wear, lubricating properties materials and lubrication methods [1].
Often there are defects of contact-fatigue origin in the form of horizontal, vertical and transverse cracks are born and start to develop. Part of such defects with contact-fatigue origin reaches 70 % of total number of rails defects [1].
The development of surface cracks with subsequent peeling of particles of the material is one of the main mechanisms of wear of the rail head during contact interaction with the wheels of the rolling stock [2]. It is known [3] that a rail in rolling contact is subjected to repeated applications of high friction loads (due to traction, braking, curving, etc.), the surface material will deform plastically. Cracks will form when the fracture strain is exceeded. This fracture strain is far above that of tensile tests, the reason being the beneficial influence of the compressive stresses [4]. In the case of alternating directions of frictional loading (for instance due to alternating traction/braking), the material will not ratchet in the same manner since plastic deformations will occur in both directions causing the accumulated plastic strain to be close to zero. Failure will be caused by low-cycle fatigue.
On straight sections of the track, such cracks most often form in the central part of the head of each rail, on curved ones -on the lateral working face of the head of the external rail [2].
Papers [5][6], devoted to this problem, notes that the main cause of contact-fatigue defects appearing and development in the rail head occur due to insufficient contact-fatigue strength of the rail steel. After multiple repeated alternating plastic deformation of the surface layers of the rail head, the plastic properties of the material are exhausted and cracks appear [2]. Thus, studies conducted in this paper are relevant and can be used to investigate the process of rail contact-fatigue defects to increase the rails and railway track life.
The aim of the work is to carry out practical studies of the contact-fatigue cracks origin and development by studying rails wear in the laboratory.
Tasks performed in the work:  Analysis of relevance and current state of research of the problem wear and durability of rail railway tracks.  Development of method and creation of an experimental setup for practical studies of contact damage to rails railway track.  Conducting studies of the resistance of the rail to cracking from the influence of the wheel load.  Development of general recommendations for studies of surface initiated rolling contact fatigue cracks of railway rails.
Core material and results. For research of rails for contact damage, the experimental setup has been designed and manufactured. Experimental setup 3D model is shown in Fig. 1. The experimental setup contains the following main parts: the running head, which is shown in fig. 2; hydraulic system, the scheme of which is shown in Fig. 3; table with different guides and fixing plates (Fig. 1). The setup is mounted on a mechanical single-cylinder press of the K 2322 model with a nominal force of 16 tons at a nominal stroke frequency of 120 per minute. The running head is fixed to the slide of the press KD 2322. Fig. 4 presents a general view of the experimental setup that mounted on press of the K 2322 model. Fig. 5 shows a "rail-wheel" friction unit. Fig. 6 presents a general view of the hydraulic system that provides clamping force of the rail to the wheel. 6 -guide roll; 7 -set of base plates for the rail; 8 -guide for the movable plate 3; 9 -guide axis for the movable plate 3; 10 -stud-bolt for greater rigidity setup It is known that the local contact stresses at the contact place is calculated using the following dependence [7]: where R max is the greater of the two radiuses (R w is the wheel radius or R 1 is the radius of the rail head) of the two surfaces in contact; m is the coefficient depending on the ratio R w / R 1 ; E eqv is the equivalent modulus of elasticity of the contact surfaces wheel and rail; F w is the load on the wheel. Since in the setup must have the same conditions of the shaft-wheel and rail contact as the actual contact of the train wheel with the rail, then the force to be generated by the hydraulic system can be determined from the following expression: where train max R is the greater of two radiuses (wheel or rail head) of two contact surfaces in real contact a rail with a train wheel; setup max R is the greater of two radiuses (shaft-wheel or rail head) of two contact surfaces in the setup; m train is coefficient that depends on the ratio of radiuses of the contact surfaces (wheel and rail head) in the real contact the train wheel and the rail; m setup is coefficient that depends on the ratio of radiuses of contact surfaces (shaft-wheel and rail head) in the setup; train eqv Е is equivalent module of elasticity of contact surfaces (wheel and rail) in the real contact the train wheel and the rail; setup eqv Е is equivalent modulus of elasticity of contact surfaces (shaft-wheel and rail) in the setup; train w F is the load transmitted from the train wheel to the rail. Knowing the required force that must be generated by the piston, you can calculate the required pressure in the hydraulic system: where S p is piston area; n is the number of hydraulic cylinders (the setup contains two hydraulic cylinders, n = 2). It is known that the weight of a loaded train car can be 70 tons [8]. The train car has four wheelsets. Then the load on one train wheel is be train 3 70 10 9.8 / 4 2 85750 The radius of the train wheel in the center of the riding circle is R w = 475 mm. The radius of the shaft-wheel in the center of the riding circle of the setup is R setup = 40 mm. The radius of the rail head is R r = 500 mm. Then coefficients are be: m train = 0.395, m setup = 1.12 [7]. Since mechanical properties of the wheel and rails are the same as parts (shaft-wheel and rail) in the setup then train Using the formula (2), the force was calculated to be generated by the hydraulic system: setup 3761.6 .
The pistons of the hydraulic cylinders have a diameter D 2 = 35 мм. Then to create needed force setup w F the pressure in the hydraulic system must be: The test program, which provides for two-hour tests, followed by defects (micro cracks) detecting on the rail body has been developed. Thus, studies are carried out for every 31920 load cycles. Research on the defects presence is carried out by fluorescent magnetic particle inspection method.
The fluorescent magnetic particle inspection method is based on the phenomenon of come together particles of magnetic powder under the influence of magnetic scattering fluxes arising over defects in magnetized objects that are tested [9]. The presence and length of indicator patterns caused by defect scattering fields are recorded visually or automatically by image processing devices.
The following technological operations were applied during the fluorescent magnetic particle inspection of the rail surface:  surface preparation for inspection (degreasing with organic solvents);  magnetization of the controlled object in two perpendicular directions by the applied field method (magnetization occurred simultaneously with the application of a luminescent suspension);  inspection of the controlled surface when irradiated with ultraviolet rays with a wavelength of 360 nm and registration of indicator patterns of defects;  assessment of inspection results. Before testing the rail in the setup ( fig. 7), its surface was examined for cracks using the fluorescent magnetic particle inspection. In fig. 7 shows the image of the original rail during testing for the presence of cracks. As can be seen from fig. 7, there are no cracks on the working surface of the rail. The image can be distinguished areas of the surface that deviate from the norm, but are not surface defects that affect the performance of the rail and do not reflect in UV light.
In Fig. 8, a shows an image of the rail wear zone in UV light after 31920 load cycles. In Fig. 8, b presents a photograph of the wear zone of the rail. There are no visible cracks on the surface of the rail. The crack initiation began after approximately 200-210 thousand load cycles. The first crack was fixed at 218120 load cycles ( fig.10, a). Its length was 20 mm. In fig. 10, b shows the photograph of the wear surface of the rail after 255360 load cycles. Several cracks are clearly distinguished and tend to develop along the contact zone of the rail and shaft-wheel with indicated numbers of the load cycles (255360). This is broadly in line with the results presented in [10], which shows the occurrence of similar damage after 190-290 thousand load cycles, depending on the properties of rail steel (hardness, surface roughness, etc.).
The rough grinding (Rz = 80 μm) can reduce the crack resistance of the rails by 70-80 %, which can be considered as a result of the influence of residual tensile stresses induced in the surface by strong local heating and increased roughness [10]. Under certain grinding conditions, the hardening formed during the grinding process to a certain extent neutralizes the effect of stress concentrators from the resulting microroughnesses, thereby increasing the crack resistance parameters. A decrease in roughness by 20 μm allows one to increase the wear resistance by 20-25 % due to an increase in compressive residual stresses and an increase in the microhardness of the surface layer [10].
In papers [11][12] also confirms the influence of grinding modes on the wear resistance of rails and their surfaces hardness. During the investigation of contact damage rail on the setup, in addition to detecting fatigue cracks and other surface defects, the shape and size of the wear zone were also determined. As can be seen from fig. 8, b and fig. 9, b, the shape of the worn zone of the rail is close to a rectangular trapezoid. This can be explained by the peculiarities of the experimental setup, when the force of pressing the rail to the wheel increased while the slider moves down. During the upward stroke, the pressing force decreased and its value was in the range of 3-5 kN. In the process of moving the slider, the pressure also varied in the hydraulic system from 1.5 to 3 MPa. The maximum pressure was in the lowest position of the slider. The relationship between the number of the load cycle with the worn area of the rail is presented as the graphical dependence in fig. 11. Table 1 presents the results of studies, in particular, the geometric dimensions (length and width) of the wear zone, as well as the characteristics of defects that occur on the surface. Table 1 Research results the development of two longitudinal cracks almost throughout the wear zone, the occurrence of two cracks up to 10 mm long Fig. 11. The dependence of the number of load cycles on the worn area of the rail As can be seen from table 1, the average crack growth rate is 0.0005 mm / load cycle after its occurrence. In addition, the number of cracks doubles approximately every 30 thousand load cycles since the first crack General recommendations for the implementation of research of the railway rail for analysis of surface initiated rolling contact fatigue cracks 1. Fatigue research can be performed on a mechanical crank press with the use of specially designed setup. Each turn of the crank provides two cycles of loading on a rail. The calculation of the load on the rail to be created by the hydraulic system of the setup is calculated according to the dependence (2).
2. According to the results of the experiment, in order to more accurately record the moment of occurrence of the first crack on the rail surface, it is advisable to divide it into stages, each of which corresponds to 25 thousand load cycles.
3. After each load stage, it is recommended to use the fluorescent magnetic particle inspection method to detect cracks, which is highly sensitive and can detect surface microcracks with a width of 0.001 mm and a depth of 0.01 mm.

1.
To carry out experimental studies of the initiation and development of contact-fatigue defects on the surface of the rail in laboratory conditions, the test program was developed, the experimental setup was designed and manufactured. For the detection of defects of the surface of the rail was used the fluorescent magnetic particle 2. Experimental studies have shown that the occurrence of cracks on the rail surface is observed after 200-210 thousand load cycles, which generally coincides with previously known data.
3. With a further increase in the number of load cycles, the existing contact-fatigue cracks rapidly develop and new cracks form in the contact zone. After occurrence, the crack propagates at an average speed of 0.0005 mm / load cycle. In addition, the number of cracks doubles approximately every 30 thousand load cycles since the first crack.
4. The worn area appears and increases on the surface of the rail during the wheel and rail come in contact. The worn area on the rail was already formed after the first test stage (31920 load cycles for 2 hours testing) with an area of about 241 mm 2 . The shape of the worn area of the rail is close to a rectangular trapezoid. This can be explained by the peculiarities of the experimental setup, when the force of pressing the rail to the wheel increased while the slider moves down. During the upward stroke, the pressing force decreased and its value was in the range of 3-5 kN. As the number of load cycles increased from 31920 to 287280, the worn area of the rail increased from 241 to 800 mm 2 .