Analysis of Failures of Leaf Springs on the Continuous Casting Machines

František Trebuňa, Jozef Bocko, František Šimčák, Marián Buršák

  Open Access OPEN ACCESS  Peer Reviewed PEER-REVIEWED

Analysis of Failures of Leaf Springs on the Continuous Casting Machines

František Trebuňa1, Jozef Bocko1,, František Šimčák1, Marián Buršák1

1Department of Applied Mechanics and Mechatronics, Technical University of Košice, Faculty of Mechanical Engineering, Košice, Slovakia


At the first look very simple and for several hundred years used elastic element can be critical part in complex equipment where its failure invokes extraordinary high losses. In the paper is given a treatment for the solution of spring failure problem by experimental methods of stress analysis and resulting proposals for its final solution.

At a glance: Figures

Cite this article:

  • Trebuňa, František, et al. "Analysis of Failures of Leaf Springs on the Continuous Casting Machines." American Journal of Mechanical Engineering 1.7 (2013): 304-308.
  • Trebuňa, F. , Bocko, J. , Šimčák, F. , & Buršák, M. (2013). Analysis of Failures of Leaf Springs on the Continuous Casting Machines. American Journal of Mechanical Engineering, 1(7), 304-308.
  • Trebuňa, František, Jozef Bocko, František Šimčák, and Marián Buršák. "Analysis of Failures of Leaf Springs on the Continuous Casting Machines." American Journal of Mechanical Engineering 1, no. 7 (2013): 304-308.

Import into BibTeX Import into EndNote Import into RefMan Import into RefWorks

1. Introduction

During production of slabs by continuous casting the steel is distributed through number of gits in vertical direction into copper form cooled by water. Heat is conducted through the copper into the water and during this process solidification of surface layer of liquid steel occurs. In order to prevent sticking of solidified layer to the copper plates, it is necessary to vibrate crystallizer and this is for various velocities of casting controlled by computer. Crystallizer is positioned in vibration mechanism that ensures vibration movement. The vibration mechanism is equipped by two leaf springs Figure 1 that transfer forces between moving and stationary parts of vibration mechanism. During operation premature failures of leaf springs occurred. The number of spring damages is different on two continuous casting machines that are in operation. Slope angle of leaf springs to the horizontal plane is 4.574° and 5.71°, respectively and the maximal vertical deflection 2.95 mm in the middle of spring is ensured by eccentric shafts.

During solution of this problem the residual stresses in broken leaf spring were determined by the hole-drilling method. Strain gages for the hole-drilling method [1, 4] can be seen in Figure 2. The maximal level of residual stresses (pressure) does not exceed 40 MPa Figure 3.

Strain gage measurements of stresses in leaf spring were realized for fastened spring ends and prescribed vertical deformation of leaf spring in tensile test machine [2, 5, 6]. In order to provide the measurements on leaf springs in the tensile test machine it was necessary to use holding fixture because leaf spring is not in horizontal position and axial component of force should cause shifting of spring from pedestal of tensile test machine. Fastening of holding fixture together with leaf spring is given in Figure 4.

Figure 1. Shape and dimensions of leaf spring
Figure 2. Locations of strain gages RY 61 for the hole-drilling method measurement
Figure 3. Residual stresses determined by ASTM E 837 01
Figure 4. Leaf spring with applied strain gages in tensile test machine

Methodology of the measurement was, with respect to the needs of using different heads and gages for the measurement with pressure and tension, developed only for the tests with pressure force. The authors have supposed that type of joining between spring and holding fixtures as well as prestress in the spring have big influence on relation between deformation and the force needed for creation of prescribed deflection and these assumptions were tested experimentally.

Leaf spring was freely placed on holding fixture and the chucking wedges were completely free. The bolts on the ends of leaf spring were completely free too. After movement of traverse of the tensile test machine to the neighborhood of leaf spring and after equilibrating of strain gage apparatus the measurement began. The records in figures from individual measurements represent stress levels in axial direction of leaf spring in locations 1 to 4 Figure 5. The steps correspond to the change of deflection by 0.5 mm. The measurement was accomplished during the process of loading as well as unloading.

Figure 5. Time-dependent charts of stresses in locations 1,2,3,4

The stress measurement of springs in the frame of equipment for continuous casting of slabs was realized after calibration in tensile test machine and after positioning of springs into the frame. (See Figure 6)

The measurement was realized on testing stand. For the appraisal of spring behavior during the operation the measurements were realized on testing frame of crystallizer during its vibration. Two leaf springs with strain gages were built into vibration mechanism. After clearing of individual strain gages, there was found out that maximal stress levels registered by all biaxial strain gages during frame vibration are approximately at the same levels and the differences are only in signs of stresses. In Figure 7 are given time-dependent charts of stresses in spring during vibration of crystallizer.

Figure 7. Time-dependent charts of stresses in spring during frame vibration

Horizontal and vertical displacements of crystallizer frame as well as accelerations of points in vertical direction in the center of the springs were measured at the same time. The measurements realized during vibration of crystal1izer in testing frame did not exhibit any phenomenon that could result to the failures. The measurements were accomplished for various casting velocities, various slabs widths, for small and big changes of their dimensions and various materials. Analysis of stress state was also performed by the finite element method. The computations were accomplished for prescribed displacements of the center of leaf spring in axial direction to the value of 0.2 mm and in vertical direction by 2.95 mm. Analyses have confirmed that the spring from the point of view off fatigue loading conforms to all conditions.

From the analysis of results gained from analytical, numerical and experimental methods was found out that leaf springs in vibration mechanism are from the point of view of safe operation under common working regimes made of suitable material and their dimensions and fastening should in full range ensure correct operation of continuous casting machine. Damages of these springs document that the states during which these phenomena arise are not common working states and it is not necessary to change the geometry and material of springs or their fastening. It is necessary to remove phenomena that cause overloading and to this time were not satisfactorily revealed.

2. Mechanical Properties of Material and Metallography

Testing samples (bar 6 x 30) were made from the given material of leaf spring made of Cr-Ni-Mo martensitic stainless steel 1.4418 in accordance with STN EN ISO 377 and OEG 13 1011 [3]. Testing bars were taken from the area of transition part (from the thicker part on boundary to the constant thickness) (sign 1) as well as from flat part of leaf spring (sign 2).

Tensile test was realized on tensile test machine FP 100/1 under temperature +20°C according to STN EN l0002-l+ACl. The results of measurements are given in Table 1.

In Figure 8 and Figure 9 is documented the microstructure from the area of geometrical transition of leaf spring (change of spring thickness). Microstructure in Figure 8 and Figure 9 is acicular and it corresponds to low carbonaceous martensit. On the basis of accomplished experiments can be stated that:

Ÿ  mechanical properties of tested leaf spring are in accordance with Cr-Ni-Mo martensitic stainless steel upgraded to quality QT900,

Ÿ  microstructure corresponds to low carbonaceous annealed martensit.

Microstructure is composed of coarse annealed needles of martensit with visible boundaries of coarse grain that can have influence to the strength of material. During the analysis of results the authors have given their attention to possibilities of increasing of working life of leaf springs by improving of mechanical properties by more suitable technology of heat treatment.

Figure 9. Detail of microstructure of tested material

This orientation was caused by fact that microstructure of material contains coarse annealed needles of martensit. Unfortunately, it was discovered that these changes do not remove the problems. It was the reason for using another material for which can be reached yield point 1700 - 1900 MPa and strength limit 1800 - 2050 MPa by appropriate processing. The material for such elements can be based on steels of maraging type. The steels belong to this group not by their chemical composition but by the strengthen mechanism that is based on precipitation of intermetallic phases in matrices (aging) almost non-carbonaceous martensit.

Maraging steels are produced approximately for 25 years. At the beginning they were low carbonaceous (to 0,03% C) steels in which the main element was nickel in amount 8 - 25%. They contained mainly 20 - 25 % Ni with additions of Ti, Al and Nb. Next stage was alloys from the system Fe-Ni-Co-Mo with addition of Ti. The most spread steels of this type contain 18 % Ni. ‎ American specification expresses (besides of Ni) yield point in units ksi (1 ksi = 7 MPa), russian specification characterizes chemical composition. Specific steels are known also according to various corporate names e.g. steel 18Ni250 grade under names: Vascomax 250, Nimar 110, Almar 18-250, Marvac 250, Republic RSM-250, Murphy 1.

Table 1. Mechanical properties of tested material – measured and prescribed

Figure 10. Schema of locations from which the specimens for fatigue tests were taken

3. Fatigue Tests of Specimens

The fatigue tests were based on using of testing machine PWY from SCHENCK Corporation. Because leaf spring is loaded for correct setting by symmetrical bending, this type loading was chosen also for this machine. The specimens were made according to Figure 10.

On the flat surfaces of specimens 1 to 3 were applied in the most narrow cross sections the strain gages 1-XY91-10/120 with gage factor k=2.05 and resistance 120 The grid of the strain gage perpendicular to the axis of the specimen is used for thermal compensation and at the same time in the sense of Hooke’s law it comprises also inf1uence of transversal deformation. Specimens 2 and 3 were chosen such a way that they have stress concentrator in the narrowest part of the specimen’s cross section. In Figure 11 is shown one specimen taken from leaf spring according to Figure 10.

Fastening of test specimen in clamping jaw of the testing machine PWY is shown in Figure 12. Fatigue tests of leaf spring were based on the thesis that fatigue limit of leaf spring material is according to State Research Institute of Materials in Prague for symmetrical bending

According to the producer, the mechanical properties of material of leaf springs are: strength limit MPa and accordingly.MPa.

Fatigue limit mentioned above is given for the specimen with polished surface. That is the reason why boundary specimen was after its fixation in testing machine Figure 12 loaded by symmetrical bending with amplitude of stress that corresponds to the fatigue limit for symmetrical bending. After 2.6786.106 cycles was clearly demonstrated that fatigue limit of this material exceeds this value. Further were increased the stress amplitudes and after certain number of cycles for the defined amplitudes was the specimen damaged.

Figure 12. Fixation of test specimen in test machine PWY

In Figure 13, Figure 14 and Figure 15 photographs of fractured specimens 1 to 3 are given. In Figure 13 is shown fractured surface of specimen 1, in which is in its most narrow location realized transition to the reinforced part of boundary specimen. From the structure of the fracture surface is seen that it contains segregates inclusions visible without microscope. These were probably introduced into the material by incorrect technology of leaf spring production. Inhomogeneous structure of material is located in several points of the cross section (see Figure 13). In the Figure 14 is shown fractured surface of the specimen 2 that was taken from the middle part of spring. This part is without any segregates inclusions or visible defects. In Fig. 15 is fractured surface of the specimen 3 in the area of transition from thin to thicker middle part of spring. In this part of cross section are again visible conspicuous inclusions in the fracture surface.

4. Conclusions

The analysis of results gained from analytical, numerical and experimental methods demonstrates that leaf springs in the vibration mechanism are from the point of view of safe operation under common working regimes made of appropriate material and their dimensions and fastening should in full range fulfil all requirements needed for proper operation of continuous casting machine.

Failures - fractures of these springs document that the states under which arise these phenomena are not common operation states and accordingly it is not necessary to change material of springs, their dimensions or fastening. However, it is necessary to remove the phenomena that invoke extreme overloading, especially at the beginning of operation.

For the solution of problem of spring breaking, at least to the time of determination of extraordinary load combina-tions, we suggest to use material with significantly higher strength properties.


The work has been accomplished under the research projects VEGA 1/1205/12 - Numerical modelling of mechatronic systems and 1/0289/11 - The use of experimental methods in the development of identification methodology and breakdown predictions of supporting members of mechanical systems.


[1]  ASTM E 837-01 Standard Test Method for Determining Residual Stresses by the Hole Drilling Strain-Gage Method, New York, 2001.
In article      
[2]  Kobayashi A. S., Handbook and Experimental Mechanics, VCH Publishers Cambridge, 1993.
In article      
[3]  Fürbacher L. a kol, Lexikon technických materiálů se zahraničními ekvivalenty. Verlag Dashofer, 1999.
In article      
[4]  TECH NOTE 503-6., Measurement of Residual Stresses by the Hole-Drilling Strain Gage. Method, Measurement Group, Inc., 2003.
In article      
[5]  Trebuňa F. et al., Riešenie problematiky praskania plochých pružín ramov kmitania ZPO. Final report. TU Košice, 2007.
In article      
[6]  Trebuňa F., Šimčák F., Kvantifikácia zvyškových napätí tenzometrickými metódami. Grafotlac Prešov, 2005.
In article      
  • CiteULikeCiteULike
  • Digg ThisDigg
  • MendeleyMendeley
  • RedditReddit
  • Google+Google+
  • StumbleUponStumbleUpon
  • Add to DeliciousDelicious
  • FacebookFacebook
  • TwitterTwitter
  • LinkedInLinkedIn