PSI - Issue 19

T. Kato et al. / Procedia Structural Integrity 19 (2019) 238–248

239

2

Author name / Structural Integrity Procedia 00 (2019) 000 – 000

wheels [1]. RCF failures are characterized by two crack initiation modes: surface crack initiations and subsurface crack initiations. Internal defects such as voids and non-metallic inclusions are known to be the possible origins of subsurface cracks [2-6]. An internal horizontal crack occurs typically at the depth of the maximum shear stress associated with Hertzian contact. Thus, it is important to reveal the effect of internal defect sizes on subsurface crack initiations to reduce removals of the heavy haul car wheels. Authors have studied the relationship between subsurface crack initiations and internal defect size, and proposed a method of predicting the critical defect sizes for subsurface crack initiations in the service conditions [7]. In this prediction method, the critical defect sizes for the subsurface crack initiations are estimated from the results of finite element analyses and fatigue tests of wheel steel based on the multiaxial fatigue assessment model. The purpose of this study is to clarify the effect of wheel size and tread braking on subsurface crack initiation of heavy haul car wheels by using the proposed method. Finite element analyses were performed to calculate the subsurface stress states of actual wheels using different wheel size models, or applying tread braking conditions. Finally, the critical defect sizes for subsurface crack initiation are estimated, and the critical defect sizes in each wheel size and tread braking condition are discussed.

Nomenclature RCF

rolling contact fatigue

shear stress ( r : radial direction,  : circumferential direction)

 r 

shear stress range

  

 n stress normal to the shear stress plane   eq,max pure shear stress range with an equivalent fatigue damage in the multiaxial stress state  n,max  n at the maximum fatigue damage k material constant

2. Finite element analysis method 2.1. Finite element analysis model

To calculate the stress distributions at the subsurface area of the wheel rim during rolling contact with rails, thermal stress analyses simulating heat treatments in the wheel manufacturing process and contact analyses simulating the rolling contact are performed. These analyses are similar to those of the previous studies [8]. Moreover, thermal analyses simulating tread braking combined with the contact analyses are performed to assess the effect of tread braking on the stress distributions at the subsurface of the wheel rim. Figure 1 illustrates the finite element models. A piece of the wheel and a piece of the rail of length 1 m are employed in these analyses. The diameters of the wheel model are 38, 36 and 33 inches, which correspond to B38 (125 Ton, single wear freight car wheel), H36 (110 Ton, single wear freight car wheel) and J33 (85 Ton, single wear freight car wheel) in accordance with the AAR (the Association of American Railroads) specification, respectively [9]. A hollow wear profile is used for the tread profile of the wheel models, as shown in Figure 2. The worn profile is obtained from the survey of TTCI (Transportation Technology Center, Inc.) [3]. The rim thickness of the model is 25.4 mm. A profile of the rail model is defined based on the worn profile obtained in the field as presented in Figure 2. The element size in the contact area of the wheel model is ~1 mm. This element size is ~1/10th the contact patch length. To verify the mesh size, an elastic analysis is carried out using a model with the same mesh size. The contact pressure and the subsurface stress are almost the same between the finite element analysis results and the Hertzian theory. The material properties of AAR Class-C wheel steel described in the AAR specification [9] are used for the wheel model. An elasto-plastic model for the wheel models and an elastic model for the rail models are employed.