Residual Stress State in Tools Used for Thermo-mechanical Metal Forming Processes

ABSTRACT

In thermo-mechanically coupled forming processes for industrial mass production residual stresses are an unavoidable consequence of the alternating inhomogeneous fields of temperature and mechanical stress developing in tools and components dependent on the process parameters applied. Hence, a considerable interest exists to get reliable information about origin and distribution of the relevant residual stress fields and to understand the basic principles of their formation. By way of example a metal forming process based on a predefined locally and temporally differential temperature profile is described, which leads to a characteristic materials property profile and geometrical shape. The development of residual stress in tools (steel AISI H11) used for the thermo-mechanical forming operation of cylindrical flange shafts (steel SAE 6150) is outlined. To this end residual stress analyses by X-ray as well as by neutron diffraction were carried out. The loading situation of the tool was simulated by isothermal, thermal and contact fatigue tests, providing information about cyclic stress and plastic deformation during the manufacturing process.

Introduction

Thermal fatigue loading is one of the most important factors related to lifetime and damage of tools for hot metal forming processes like extrusion, die-casting or injection moulding as well as for many other hot working tools [1-4]. Thermal stresses caused by alternating temperatures in tools lead to characteristic crack formation, propagation and final failure [5,6]. A further important factor related to the lifetime of tools used for hot metal forming is the formation of harmful residual stress states. This article deals with the results of a research project entitled "Analysis and effects of residual stresses in tools and components as a consequence of thermo-mechanical processes". The project is embedded in a large collaborative research centre "Process integrated manufacturing of functionally graded structures on the basis of thermo-mechanically coupled phenomena" established at the three German Universities of Kassel, Paderborn and Dortmund, funded by the German Research Foundation (DFG) [7]. Residual stresses are of outstanding importance during the manufacturing process as well as for strength, lifetime and reliability of the produced components. This is one of the main reasons for the establishment of the project mentioned above.


As an example of a thermo-mechanically coupled forming process Fig. 1 shows the individual production steps of a functionally graded flange shaft which is envisaged for industrial mass production. The process starts with a cylindrical bar (steel SAE 6150) which is heated up to austenitizing temperature by an induction coil. Due to pressing the bar between two tools, a flange shaft is formed. Besides the aim of an exact geometry it is of fundamental importance to realize characteristic temperature-time-courses in the individual regions of the flange shaft to achieve technologically advantageous gradients of microstructures. It is the challenge of the process to adapt the microstructures produced and, hence, the local strength of the material, at the prospected loading state of the components.

Cyclic heating and cooling of the tool as well as mechanical loading leads to cyclic plastic deformation and, as a consequence, to characteristic residual stress states in the tools used [8]. Therefore a great interest exists to understand and to quantify the basic processes of residual stress evolution in order to reliably predict the lifetime of tools under given process parameters. In this paper the development of residual stress in tools used for the thermo-mechanical forming operations of cylindrical flange shafts is outlined. Results of isothermal, thermal and contact fatigue tests, providing information about cyclic stress and plastic deformation during the manufacturing processes are presented and discussed in detail.

Characteristic steps of the flange shaft forming process [9] Experimental results and discussion

Fig. 1 Characteristic steps of the flange shaft forming process [9] Experimental results and discussion

Residual stresses were determined by conventional X-ray diffraction technique [10] using CrKa-radiation at the {211 }-planes of the annealed martensite. For stress evaluation, the sin2y-method was applied and the elastic constant ^s2 = 6.09 10-6 MPa-1 was used. In order to confirm the residual stress distributions determined by successive layer removal in combination with repeated application of X-ray stress measurements, additional neutron diffraction measurements were carried out at the neutron research reactor FRMII in Munich-Garching. Surface strain scans in reflection mode using a wavelength of the neutron radiation of 0.168 nm were performed. For this wavelength the {211} interference line of the annealed martensite occurs at around 29 = 91.5°. A new method to measure near surface residual stress profiles with neutrons up to 200|im underneath the surface of steel samples without time consuming and laborious surface effect corrections was applied [11]. Aberration peak shifts caused by spurious strains, which arise due to the fact that the gauge volume is partially outside of the sample when scanning the near surface region, were minimized by using an optimized bending radius of a SI (400) monochromator. The gauge volume defined by the primary and secondary optics was 1mm3.

By way of example Fig. 2 shows a tool used for approximately 200 hot metal forming operations as described in Fig. 1. A dark ring around the inner bore hole characterizes the heat-affected zone of the tool. Residual stresses measured in radial and tangential direction at the surface are plotted along the radius of the tool. High tensile residual stresses occur in the cyclically heated and cooled area of the tool, whereas in the other regions compressive residual stresses are measured.

Residual stress distribution at the surface of the tool after approximately 200 hot metal forming operations

Fig. 2 Residual stress distribution at the surface of the tool after approximately 200 hot metal forming operations

To get further information about the origin of these residual stresses, depth distributions were analysed by X-ray and Neutron diffraction at two locations inside and outside the cyclically heated and cooled zone of the tool (see Fig. 3). Outside the heated zone, compressive residual stresses are only measured in a very thin surface layer (Fig. 3, right), inside the heated zone tensile residual stresses of about 450MPa at the surface decrease only gradually and decay within a depth of approximately 1.5mm of the tool. It is assumed that different reasons for these stresses exist. The compressive residual stresses outside the cyclically heated zone result from the final grinding process of the tool manufacturing process whereas the tensile residual stresses inside the cyclically heated zone are produced during the use of the tool as a consequence of cyclic plastic deformation due to the cyclic heating and cooling process [12]. Fig. 3 shows a good correlation of data gained by X-ray and neutron diffraction.

During the forming process of the flange shafts, time-temperature courses were measured at different positions of the hot forming tool to get an idea of the thermal loading situation. Fig. 4 shows three time-temperature courses. During each cycle temperatures rapidly increase and, after reaching maximum values, slowly decrease. Due to the different temperature courses at several points of the tool complex thermal stress states near the tools surface have to be expected.

Residual stress distributions measured by X-ray and neutron diffraction at different points in a thermo-mechanically loaded hot forming tool

Fig. 3 Residual stress distributions measured by X-ray and neutron diffraction at different points in a thermo-mechanically loaded hot forming tool

To get further information about the materials behaviour under isothermal and thermal loading conditions, isothermal fatigue tests as well as thermal fatigue tests were carried out under axial loading states up to temperatures of 650°C. The specimens investigated have been machined from a forged ingot with the following chemical composition (wt-%.): 0.37 C, 1.2 Si, 0.23 Mn, 4.96 Cr, 1.25 Mo, 0.45 V, 0.003 P, 0.002 S, Fe balanced. Blanks of the specimens were heat treated (austenitisation for 20 min at 1025°C, quenching in oil of 60°C, two times tempering for 2 h at 625°C) to achieve a hardness of 44 HRC. The material investigated had the following mechanical properties: Rp0.2 = 1153MPa, UTS = 1384MPa. Cylindrical specimens with a diameter of 7mm and a gauge length of 10mm were machined out of the heat treated blanks. After heat treatment, all samples were hard turned to realize identical starting conditions regarding surface roughness and residual stress state.

Isothermal tests were carried out under stress control with zero mean stress and triangular stress-time paths with a frequency of 0.5Hz. Out-of-phase thermal fatigue tests were realized under complete suppression of total strain. A triangular temperature-time course with a heating and cooling rate of 10K/s was applied using a computer controlled induction system to heat up the specimens. Compressed air nozzles situated near the gauge length of the specimens were used for cooling. Cyclic heating and cooling under complete suppression of thermal expansion leads to fluctuating thermal stresses during the tests. In all cases investigated, cycling softening of AISI H11 occurs. Plastic strain amplitudes are very important indicators to assess the effects of isothermal or thermal fatigue processes. As an example for isothermal tests plastic strain amplitudes for a stress amplitude of 750MPa and different test temperatures, i.e. 20°C, 400°C and 500°C are shown in Fig. 5 (left). Decreasing total lifetimes to failure with increasing test temperature were observed. Cyclic plastic behaviour during isothermal fatigue is characteristic for quenched and tempered steels. Generally a stable microstructure can be assumed except for coarsening processes of precipitated carbides at temperatures well below the tempering temperature of 625°C. For tests at room temperature, plastic strain amplitudes are very small and a nearly macroscopic elastic behaviour is observed for higher numbers of loading cycles. For temperatures of 400°C and 500°C plastic strain amplitudes start to increase continuously after an incubation period and very pronounced cyclic softening occurs [13]. This is attributed to rearrangements of dislocations associated with greater dislocation mobility. In addition, the formation of cracks plays a role and cyclic plastic deformation is superimposed by the consequences of crack opening and closing processes.

Time-temperature courses during the hot metal forming process [14]

Fig. 4 Time-temperature courses during the hot metal forming process [14]

In Fig. 5 (right) plastic strain amplitudes for thermal fatigue tests using upper temperatures of 550°C, 600°C, 625°C and 650°C and base temperatures of 100°C and 200°C are shown. At first very low values of plastic strain can be observed as already shown in the isothermal case. At higher numbers of thermal cycles, the plastic strain amplitudes increase and the specimens exhibit cyclic softening behaviour. A comparison of tests carried out under identical upper temperatures shows higher plastic strain amplitudes for a base temperature of 100°C than for 200°C. Fig. 6 shows upper stresses resulting during thermal fatigue tests at a base temperature of Tmin=100°C. Already after the first loading cycle, tensile stresses are observed which increase with increasing temperature amplitude. At the end of the lifetime, they drastically decrease due to the formation of cracks. If one assumes that specimens loaded under out-of-phase thermal fatigue conditions are representative of a small volume element in the cyclically heated and cooled zone of the tool, then results of the thermal fatigue tests presented explain the formation of tensile residual stress states in this region of the tool.

Resulting plastic strain amplitudes in case of isothermal loading (left) and thermal loading (right) for different temperatures or temperature ranges

Fig. 5 Resulting plastic strain amplitudes in case of isothermal loading (left) and thermal loading (right) for different temperatures or temperature ranges

Courses of upper stress for different temperature amplitudes at a base temperature of 100°C

Fig. 6 Courses of upper stress for different temperature amplitudes at a base temperature of 100°C

The important role of cyclic thermal stress on residual stress formation is emphasised by the results shown in Fig. 7 Here, tangential and radial residual stress distributions are shown, measured at the surface of tool models which were only cyclically heated (without mechanical load) by an induction coil and quenched by water spraying with time-temperature courses as shown in Fig. 4. Three thermocouples were fixed 1mm beneath the tools surface at distances of 5mm, 15mm and 25mm from the inner borehole. In these tests maximum temperatures of approximately 550°C are measured [14]. As one can see, starting from manufacturing induced compressive stresses of approximately -200MPa in tangential direction and tensile residual stresses of approximately 100MPa in radial direction a tensile residual stress maximum develops close to the inner bore hole and maximum values increase with rising numbers of loading cycles. These distributions are qualitatively very similar to the distributions shown in Fig. 2. These results clearly demonstrate the significance of cyclic thermal loading for the formation of residual stress states.

Tangential (left) and radial residual stress distributions (right) at the surface of a tool which was loaded by thermal cycles only

Fig. 7 Tangential (left) and radial residual stress distributions (right) at the surface of a tool which was loaded by thermal cycles only

Another focus of research is the investigation of residual stress formation induced by cyclic mechanical contact between tool and workpiece. For this purpose, flat specimens (50mm x 50mm x 20mm) of the tool steel AISI H11 were cyclically loaded under contact fatigue conditions using cylinders made of WC-Co hard metal and after that, the resulting residual stress fields were analysed using successive layer removal in combination with repeated application of X-ray stress measurements. Depth distributions of the residual stresses measured in the centre of the contact area in both directions along and perpendicular to the contact line are given in Fig. 8. Residual stresses were measured in longitudinal direction (left) and perpendicular to the contact line (right) after cyclic loading with a maximum force of 202kN for the numbers of loading cycles indicated.

Grinding induced residual stresses altered already after the first loading cycle. For stresses measured parallel to the contact line with increasing numbers of loading cycles, a thin layer of high compressive residual stresses occurs followed by a region with smaller compressive residual stresses with amounts of -50MPa to -260MPa in a depth of approximately 0.1 – 0.25mm. Then amounts of compressive residual stresses increase again, reach maximum values in a distance to surface of approximately 0.4 – 0.6mm and then decrease continuously. For higher numbers of loading cycles, amounts of compressive residual stresses increase slightly and the position of the maximum is shifted to larger distances from the surface. The stress components measured perpendicular to the contact line show a maximum of tensile residual stress about 0.1mm below the surface followed by a maximum of compressive residual stresses which becomes more pronounced and is shifted deeper below the surface with increasing numbers of loading cycles.

Residual stress distributions as a consequence of cyclic contact loading after different numbers of loading cycles

Fig. 8 Residual stress distributions as a consequence of cyclic contact loading after different numbers of loading cycles

Conclusion

The application of modern thermo-mechanically coupled hot metal forming processes allows the economic mass production of components with unique combinations of geometries and local materials microstructures or strength respectively. A characteristic feature of the new processes is that inhomogeneous thermal as well as mechanical loading states occur. Identification and assessment of basic processes responsible for the formation of characteristic residual stress states in tools and components due to thermo-mechanically coupled forming operations was the main purpose of the work carried out in this research project. The importance of thermal cycles and the associated cyclic plasticity on the residual stress state of tools used for hot metal forming of flange shafts was confirmed by the results of isothermal and thermal fatigue tests. In cyclically heated and cooled volumes tensile residual stresses have to be expected when complete suppression of thermal expansion is supposed. This effect occurs already at the beginning of the use of the tool. It has been shown that for the case investigated here, cyclic thermal loading is the most important process controlling the formation of residual stress states in the tool used and that mechanical loading plays a minor role.

Next post:

Previous post: