A R C H I V E S o f

F O U N D R Y E N G I N E E R I N G

Published quarterly as the organ of the Foundry Commission of the Polish Academy of Sciences

ISSN (1897-3310)Volume 8

Issue 2/2008

17 – 20

4/2

A R C H I V E S o f F O U N D R Y E N G I N E E R I N G V o l u m e 8 , I s s u e 2 / 2 0 0 8 , 1 7 - 2 0 17

Computer aided casting methoding of

railway system

St. M. Dobosza, *, A. Chojeckia , **, R. Skoczylasb, *** a Faculty of Foundry Engineering, University of Sciences and Technology AGH,

Reymonta 23, 30-059 Kraków, Poland b KOM-ODLEW, Bluszczowa 25F, 30-439 Kraków, Poland

Corresponding author. E-mail address: *dobosz@agh.edu.pl, **ach@agh.edu.pl, ***Ryszard.Skoczylas@kom-odlew.pl

Received 03.032008; accepted in revised form 31.03.2008

Abstract Authors prepared the casting technology for the frog. Casting method has been verified by simulation, using MAGMASOFT

technique. In multistep simulation it was found that the positioning of casting in the mould feeding and some details of construction must be changed. Finally authors have presented the optimised solution of the technology which eliminates the porosity of casting. Analyses of the distribution of the stress during solidification and cooling in the mould proves that the deformation of casting is acceptable. Keywords: Monobloc frog; Simulation; Solidification; Distribution of Stress.

1. Introduction

Modernization of the Polish railways is absolutely necessary to join the transeuropean system. It means the adaptation to high speed trains at least 200 km/h and axle load not less than 230 kN [1]. For this aim the carbon steel rails are successively replaced by modern, produced in Poland low alloy bainitic steel, much more durable, resistant for the friction and dynamic tension. [2,3]. The modern rails are until now joined with traditional, carbon steel frogs produced by mechanical treatment and welding of elements. Several producers of railways use the monobloc cast high manganese frogs imported from other countries of EU. In 2006 authors began the project, sponsored by the Ministry of the Sciences an Higher Education to study the possibility of the production the bainitic steel frogs easy to join with the rails - in Polish foundries. The aims of the project are as follow: • Selection of a chemical composition of bainitic steel fulfilling

all demands of buyer: UTS minimum 1350 MPa, elongation 12%, hardness 330-400 HB and ductile-brittle transition

temperature below –30° C. Cast steel has to be easily welded with the rail material.

• Elaboration of a casting methoding assuring the II group of overall casting quality and the I group in the region of the edge. (ISO standard 1690). It means that steel must be very clean and the casting very well fed during the solidification process. The foundry sand has to guarantee the clean surface of the casting and be easy to regenerate.

• Establishing parameters of heat treatment and procedures of quality control.

Numerical modelling as a tool for analyzing such processes like mould cavity filling, casting solidification or thermal stresses in castings is known since years. Initially it was applied rather to simple shaped castings [4, 5], but with growing computer power the limitation no more exists. Computer systems like MAGMASOFT enable analyzing real castings regardless casting technique applied [6 – 9]. In the paper authors present some results of optimization of casting methoding by computer simulation using MAGMASOFT system to analyze mould cavity filling, casting

solidification as well as thermal stresses. An optimum casting methoding has been prepared to fulfil the condition of the best feeding of casting during solidification and minimal deformation when cooling.

2. Optimization of casting methoding

Initial shape of the frog, designed by producer of frogs is presented on the Figure 1. This five meter long thin-walled casting with varying cross-section is very difficult to feed. As a first approach, horizontal casting position with the rail head laying down has been selected. Such casting position is often used for high manganese cast steel frog since it ensures good quality of working surfaces of the casting. On each cross-section of the bottom ribs one top feeder has been placed as well as two adjacent side risers, see the Figure 2. At this early stage, time consuming filling cavity simulation was skipped and casting solidification simulation was performed only. An example of results of the simulation is presented in the Figure 3. It shows feeding of the casting. White means sound casting, while colours denote shrinkage defects. The reason of such poor quality is explained in the Figure 4. a)

b)

Fig. 1. Design of the frog: a) view from the top, b) view from

the bottom Distribution of solidification modulus values show that the side risers are not able to feed the hot spot in the middle of casting, since they solidify earlier than the hot spot. A numerous simulation variants with bigger side risers was has been performed but still with the same result. The side risers were not able to feed the hot spot through relative thin ribs. As a next step, the side risers have been removed while the top risers have been enlarged. Additionally, cylindrical material allowances below the risers have been applied to obtain better

riser necks. However, still without good results, since the thin rail neck rail breaks feeding process.

Fig. 2. Initial feeding system: each section of the casting fed by

three feeders: one top feeder and two side ones

Fig. 3. Results of casting feeding prediction: assumed feeding

system is not able to ensure sound casting

Fig. 4. Solidification modulus distribution confirms that feeders are too small to fed the critical hot spot (light area at the bottom of the casting). Rail foots (upper corners of the casting) cannot

be good fed, as well Simulation of casting solidification without any riser is shows, in the Figure 5; that hot spot is located in the same place regardless casting position. However, the hot spot can be fed better if rail heads are on the top position. To obtain a better

A R C H I V E S o f F O U N D R Y E N G I N E E R I N G V o l u m e 8 , I s s u e 2 / 2 0 0 8 , 1 7 - 2 0 18

feeding pattern, the cross-section of the frog has been corrected to ensure directional solidification, see the Figure 6.

Fig. 6. A cross-section of the frog: initial design (left), and final

design enabling directional solidification (right) Next, the ribs on the bottom side of the frog have been removed. To find a right position, number and size of risers, a number of solidification simulation has been performed. After establishing feeding system, a smaller but sufficient number of bottom ribs has been added again. In the last stage of the casting methoding optimization, an appropriate gating system has been designed, see in the Figure 7.

Fig. 7. Final casting methoding: optimized feeding system and

gating system. Filling simulation was also performed, although not shown in the paper

Metal enters mould cavity from the bottom side, flowing through extensive system of runners, to ensure uniform temperature of liquid metal before solidification. The Figure 8 shows an example of casting solidification on the same cross-section as already shown on the Figure 3 (this time no rib is present there). It is clear, that directional solidification occurs. As a result, sound casting is obtained both in the analyzed cross-section, as well as in other critical cross-sections. Some additional cooling ribs have been added at casting ends, to improve soundness in the regions. A small porosity occurs between some risers, but generally the proposed casting methoding may be accepted. Feeders removing from the working surfaces of the casting will require more efforts, but such feeders position gives more rigidity to the working surfaces. To check distortion of the casting, thermal stresses have also been analyzed. The Figure 10 shows distribution of normal stresses along casting length. Distribution of the stress shows, that top risers located between external rails counteract greater distortion of the top surfaces of the casting. The Figure 11 shows displacement of the casting in the vertical direction using multiplication factor equal to 15. It can be seen that distortion of the upper surface of the frog along its length does not exceed 3 mm.

a)

b)

c)

Fig. 8. An example of solidification of a casting cross-section for optimized layout: directional solidification occurs. Solid

fraction: a) 10%, b) 20%, c) 70%

A R C H I V E S o f F O U N D R Y E N G I N E E R I N G V o l u m e 8 , I s s u e 2 / 2 0 0 8 , 1 7 - 2 0 19

Fig. 9. Distribution of normal stress σxx after cooling down into

room temperature

Fig. 10. Displacement of the frog into vertical direction

(magnification 15x). Distortion of the upper surface of the rails should not exceed 3 mm

3. Conclusions

Optimisation of technology of the frogs proved that for very long and complicated steel casting classical method of the mould technology cannot be adapted. The simulation of the temperature distribution indicates that the porosity free casting cannot be obtained, without change of its positioning in the

mould, and some changes of construction. Such modifications will raises the cost of the mechanical treatment but eliminates the defects of casting. After the multi step simulation one can receive the optimisation of the casting methoding. Simulation is the unique way to study the distribution of stress during solidification and cooling of the casting. It is very important for the long casting of bainitic steel. Proposed technology limits the deformation of 5 m long casting to less then 3 mm.

Acknowledgments Scientific research financed from education funds for 2007-2009 of the project No R 07 007 02

References [1] Decision no 1692/96 of the European Parliament and of the

Council of 23.07.1996. [2] F.C. Robles Hernández, N.G. Demas, D.D. Davis, A.A.

Polycarpou, L. Maal, Mechanical properties and wear performance of premium rail steels, Wear 263 (2007) 766-772.

[3] K.M. Lee, A.A. Polycarpou, Wear of conventional pearlitic and improved bainitic rail steels, Wear 259 (2005) 391-399.

[4] R. Skoczylas, Calculation of Thermal Stresses in Solidify-ing and Cool-ing Cylinder-Shaped Casting, 12th Symposium, ITiMO AGH, Kraków (1986) 35-54 (in Polish).

[5] R. Skoczylas, Experimental and Numerical Modelling of Filling of Mould Cavity. Acta Metallurgica Slovaca 4 (1998), No. 2, 237-241.

[6] L. Baryła at al, Effects of implementation of program MAGMASOFT in Odlewania Żeliwa ZAWIERCIE SA, Przegląd Odlewnictwa 55 (2005) No. 9, 568-571 (in Polish).

[7] M. Dukała at al, Implementation of MAGMASOFT in ZETKAMA SA, Ibidem, 572-576 (in Polish).

[8] W. Targowski, S. Krawczyk, Using program MAGMASOFT to optimize gravity die aluminum caliper, Ibidem, 578-584 (in Polish).

[9] R. Skoczylas, M. Cieplak, Integration of casting simulation results with CAE, Ibidem, 586-597 (in Polish).

A R C H I V E S o f F O U N D R Y E N G I N E E R I N G V o l u m e 8 , I s s u e 2 / 2 0 0 8 , 1 7 - 2 0 20