Characterization of damage in forward rod extruded parts

Abstract

In addition to strain hardening and residual stress, damage influences the product performance of forward rod extruded parts. Damage is usually neglected and difficult to quantify. The evolution of ductile damage in metal forming is closely correlated to the load path. An experimental approach using automated energy dispersive X-ray spectroscopy (EDX) particle analysis in scanning electron microscopy (SEM) is used to successfully quantify the void area fraction and obtain information on ductile damage. The method is performed on forward rod extruded 16MnCrS5 workpieces with varying extrusion strains and shoulder opening angles (and thus varying underlying load paths). The quantified damage is directly correlated to the load path, which can be described by the stress triaxiality evolution during forming. Density measurements were performed to further validate the results. By observing the change of strain-weighted stress triaxiality and maximum stress triaxiality, it is shown, that the maximum stress triaxiality is the decisive parameter enabling void nucleation.

Appendix

In order to evaluate the influence of the process parameters on the damage evolution the influence of the load path during forward rod extrusion on an ideal void was studied numerically. For this an axisymmetric simulation model was set up in Abaqus CAE, representing a material point on the center line which travels through the forming zone during forward extrusion (Fig. 20).

Fig. 20

Simulation model

The reference model has a diameter and a height of 2a. A spherical void was positioned in the center of the model. The evolution of the axial and radial stress components σ_z and σ_r occurring along the central axis during forward rod extrusion, determined in Finite-Element-Simulations performed with the software Simufact 13 (previous chapter), were applied to the boundaries of the reference model (Fig. 21).

Fig. 21

Evolution of of the axial and radial stress components σ_z and σ_r occurring along the central axis during forward rod extrusion (16MnCrS5, 2α = 90°, ε = 0.5; μ = 0.04)

Since the stress components are prescribed directly, the behavior of the reference model is not affected by the length a. In order to assure, that the void does not alter the resulting strain along the central axis, which is prescribed by the extrusion strain, its size was decreased incrementally, yielding an initial void radius of b = a/5. This way, the difference between the resulting strain in forward extrusion and the resulting strain in the reference model was less than 1%. It was studied how the shape and the volume of the void changes.

The shape of the ideal spherical void alongside with its shape change in dependence of the extrusion parameters is shown in Fig. 22. Higher extrusion strains lead to voids being increasingly stretched in extrusion direction. The direction of the distortion of the voids is the same for all extrusion strain due to the Lode parameter being identical (Fig. 22).

Fig. 22

Numerically determined distortion of a void due to different extrusion strains (16MnCrS5, 2α = 90°, μ = 0.04)

The change in volume of the voids caused by different extrusion strains and the corresponding triaxiality values are shown in Fig. 23. For comparison, the initial void volume is shown additionally. It can be observed that a small extrusion strain of ε = 0.1 leads to an increase of the void volume due to a positive triaxiality. For ε = 0.3 the void volume is similar to the initial void volume. With extrusion strains higher or equal than ε = 0.5 the mean triaxiality is negative. This causes a reduction of void volume after forming. For extrusion strains higher than ε = 1.2 a void volume of nearly 0 is reached because of highly negative triaxiality values.

Fig. 23

Numerically determined change in volume due to different extrusion strains (16MnCrS5, 2α = 90°, μ = 0.04)

The above mentioned method was repeated for different shoulder opening angles. This yields the advantage that the same strain is reached on the central axis of the extrudate and thus, there is only a change in the hydrostatic part. In Fig. 24 it is shown that only the volume of the void changes. The shape is the same due to identical Lode parameters and extrusion strains.

Fig. 24

Numerically determined distortion of a void due to different shoulder opening angles (16MnCrS5, μ = 0.04, ε = 0.5)

The change in volume of the voids with varying shoulder opening angles compared to the initial void volume is shown in Fig. 25. A correlation between triaxiality and void volume can be seen.

Fig. 25

Numerically determined change in volume due to different shoulder opening angles (16MnCrS5, μ = 0.04, ε = 0.5)

It is noted, that a real material shows no round spherical void. The simulation results merely show a general tendency on how an ideal spherical voids changes its shape and volume under the load paths of forward rod extrusion with different process parameters. The semi-finished products used for forward rod extrusion contain inclusions. Void nucleation occurs at these inclusions due to their breakage or due to the separation of the matrix/inclusion interface, resulting in variously shaped voids with different volumes. Nonetheless, in addition to this new voids will nucleate due to the interaction of plastic strain, hydrostatic pressure, matrix and inclusion.