Advanced Cut-Edge Characterization Methods for Improved Sheared-Edge Damage Evaluation in High-Strength Sheet Steels

Abstract

This study investigates shear cutting of high-strength steel sheets, a process known to negatively impact the forming and fatigue properties of the material. The localized deformation near the cut edges imposes sheared-edge damage, especially in advanced high-strength steels where severe shear deformation occurs in the very vicinity of the cut edge. In this work, an extensive experimental investigation was carried out on punched holes of thin sheets, using light optical microscopy and metallographic techniques for sheared-edge damage assessment. These methods provided detailed insights into the sheared-edge damage and offer a thorough understanding of the deformation behavior in the shear-affected zone. Advanced engineering cut-edge investigation methods have been developed based on 2D and 3D stereo light microscopy for non-destructive panoramic cut-edge parameters and cut-edge profile determination along cut-hole circumference. Such methods provide an efficient evaluation instrument for challenging close-cut holes, with the possibility of industrial in-line monitoring and machine learning applications for Industry 4.0 implementation. Additionally, the study compares grain shear angle measurement and Vickers indentation for deformation assessment of the cut edge. It concludes that grain shear angle offers higher resolution. This parameter is therefore postulated as relevant for assessing the sheared-edge zone. The findings contribute to a deeper understanding of sheared-edge damage and improve evaluation methods, potentially enhancing the use of high-strength steels in automotive and safety-critical applications.

1. Introduction

Shear cutting remains the most common cutting technique in the cold sheet forming process due to automation possibilities and cost-efficiency. This technique involves a shearing process where the sheet metal is separated by a moving punch, pushing the workpiece against the fixed die. This process typically results in the formation of a cut edge with a characteristic profile, consisting of rollover formation, the burnish zone, the fracture zone, and burr formation, as seen in Figure 1. Figure 1 also shows the distinction between primary and secondary burnish and fracture formation, features that might appear at low cutting clearances. While shear cutting offers efficiency and precision, the nature of the operation inevitably causes localized deformation and damage to the material, which can manifest as burrs, micro-cracks, and strain hardening near the cut edge [1]. This region, often referred to as the shear-affected zone (SAZ), is known for affecting the formability of sheared parts and triggering edge cracking, especially for advanced high-strength steels (AHSSs) [2]. This has been experimentally shown for both close cuts [3] and open cuts [4], varying cutting clearance and material grade [5], and with varying punch tool design [6]. The sheared-edge damage is also known to affect the fatigue properties of high-strength sheets since the cutting process induces micro-cracks and notches [7,8,9] and residual stresses to the cut edge [10,11]. Sheared-edge damage is also a factor in hydrogen embrittlement of AHSS [12,13]. For instance, AHSS grades with ultimate tensile strength (UTS) of 1000 MPa and above, now increasingly used in lightweight construction and safety-critical applications, are particularly susceptible to localized sheared-edge damage due to their higher strength and reduced ductility compared to lower strength conventional AHSS sheets. As a result, understanding and controlling the residual state of the SAZ, introduced by shear cutting, is important for ensuring product reliability and optimizing performance during the manufacturing process. Factors influencing the characteristics of the SAZ include the material mechanical properties and cutting conditions, especially the cutting clearance and the cutting speed [14]. However, in addition to the initial punch/die clearance, the punch/die radius and surface quality evolution during forming significantly influence the evolution of edge quality and the occurrence of edge cracking [15,16,17,18]. The high strength of AHSS results in high shearing loads, affecting tool surface integrity [15,19,20]. Adhesive and abrasive wear mechanisms alter the punch and die radius, leading to an increase in the effective cutting clearance. This effect is evident in in situ force-displacement monitoring during trimming operations, which shows that tool wear is associated with increased trimming force, prolonged die penetration, and elevated energy dissipation. Tool wear leads to increased burr height, rougher edge surfaces, and a larger SAZ with more pronounced microstructural damage. Closed-trimming operations, such as hole punching, are highly susceptible to tool wear because of constrained material flow and higher localized stresses. As wear progresses, the quality of the pierced holes declines more rapidly than in open-trimming processes, leading to an increase in rollover, fracture zones, and burrs. Conversely, open-trimming operations experience a more gradual reduction in edge quality, thus permitting a slightly extended tool life before critical degradation is observed [21]. These findings highlight the need for optimized tool maintenance schedules, wear-resistant coatings, and adaptive trimming strategies tailored to the specific needs of shearing AHSS sheets. Incorporating real-time monitoring of tool condition and edge quality metrics can further enhance process reliability and product consistency. Non-destructive techniques have become increasingly vital in monitoring tool wear and assessing part quality in metal forming and trimming operations. Among these, force monitoring and acoustic emission (AE) analysis have emerged as powerful tools due to their sensitivity to subtle changes in process dynamics and material behavior. Some works have shown that AE feature parameters such as amplitude, energy, rise time, and frequency content can be correlated with specific damage mechanisms [22,23]. For example, high-frequency, low-rise-time signals are typically associated with brittle fracture events, while lower-frequency emissions may indicate plastic deformation or tool wear. AE monitoring could successfully detect the onset of tool wear and crack initiation during stamping, highlighting its potential for real-time quality control in trimming processes [24]. The damage induced at the SAZ include the formation of micro-cracks, as in the ferrite/martensite interface in DP steels. AE sensors are capable of capturing transient elastic waves generated by crack initiation and propagation, especially in brittle or semi-brittle interfaces [15,24,25]. Traditionally, sheared-edge damage investigation has been performed at different magnification levels, starting from naked-eye observation to microscopic investigation and using metallographic techniques for detailed analysis. Figure 1 shows conventional cut-edge inspection techniques and presents typical appearance of secondary burnish formation in close-cut hole edges, as well as burr and fracture surface roughness. Figure 2 shows the open-cut secondary burnish formation in a different proportion, from little to strongly manifested secondary burnish formation merging with primary burnish formation. Some irregular burr and rough cut-edge fracture surfaces are also displayed.

Detailed investigations of sheared-edge damage are conventionally performed using destructive techniques, such as cross-sectional analysis through metallography and microhardness tests. These methods provide valuable insights into the extent of SAZ damage, allowing quantification of the degree of deformation and identify defects like cracks, voids, or brittle phases. However, the destructive nature of these methods limits their applicability, especially in cases where circumferential heterogeneity and spatial variations in sheared-edge damage need to be assessed. Such circumferential heterogeneities typically appear in punched sheet holes in industrial forming lines and reduce the stretch–flangeability of the edge [26] due to unavoidable tool wear or tool misalignment during part production. In contrast, the use of non-destructive cut-edge assessment methods allows for repeated use of the shear cutting specimens and are required for in-line measurements. In this work, the main focus is placed on conventional readily and widely available non-destructive testing methods using light optical microscopy (LOM), which is part of the standard equipment in industrial laboratory environments. Therefore, expensive, time-consuming, punctual, and detailed non-destructive methods such as scanning electron microscopy (SEM) or electron backscatter diffraction (EBSD) are disregarded, although their use in laboratory SAZ damage identification is well proven [27,28,29].

This study investigates the sheared-edge damage in 1.5 mm complex-phase advanced high-strength steel (AHSS) with UTS $\ge 1000$ MPa, which is characterized by highly localized shear deformation and a narrow shear-affected zone (SAZ). Such localization challenges the assessment of edge quality and its influence on subsequent forming and fatigue performance. The experimental work presented in this article is based on hole punching using varying process parameters, where the resulting damage was analyzed across macro-, meso-, and micro-length scales. Conventional SAZ characterization methods, including Vickers hardness (DIN ISO 6507) [30], which is frequently used in conventional cutting [31,32], two-stage shear cutting [33], and post-cutting fatigue studies [10], and grain shear angle measurements [12,34,35,36] are evaluated for their suitability in large-sample investigations. In addition, the study explores the use of 2D panoramic and 3D profile microscopy as efficient, non-destructive techniques for detailed cut-edge assessment, particularly for small, close-cut geometries. The overall objective is to improve the reliability of SAZ characterization and propose practical, scalable methods to address the heterogeneous edges typical of high-strength steels. The methods developed in this work make use of cost-effective microscopy instruments with minimal image processing, providing both research and industry with valuable tools to enhance shear cutting processes and improve the formability and fatigue resistance of AHSS components.

2. Materials and Methods

In the present article, the 1.5 mm CR700Y980T-CH automotive sheet steel from voestalpine Stahl GmbH, Linz, Austria, was investigated in the cold-rolled, pickled, and continuously annealed condition. This grade will be referred in the following as CP1000HD. This is a cold-rolled complex-phase steel with high ductility and an ultimate tensile strength of 1000 MPa. This material is commonly used in automotive applications, including crash management systems and structural components such as side-impact beams, rocker panels, frame-rail reinforcements, tunnel stiffeners, and fender and bumper reinforcements, where high energy absorption and formability are critical. The CP1000HD grade has a fine-grained microstructure consisting of homogeneous bainite, martensite, and retained austenite. The mechanical tensile properties of the CP1000HD grade are shown in

Table 1. Mechanical tensile properties in different material directions relative to the rolling direction (RD) for CP1000HD with a thickness of $t=1.5$ mm: yield strength ($YS$), tensile strength ($UTS$), uniform elongation ($UE$), and total elongation ($TE$) for 80 mm gauge length; n-value at uniform elongation ($n_{UE}$); and r-value determined at 4% elongation ($r_4$).

RD	$\mathit{Y}\mathit{S}$ [MPa]	$\mathit{U}\mathit{T}\mathit{S}$ [MPa]	$\mathit{U}\mathit{E}$ [%]	$\mathit{T}\mathit{E}$ [%]	${\mathit{n}}_{\mathit{U}\mathit{E}}$ [-]	${\mathit{r}}_4$ [-]
$0^{\circ }$	893	1052	7.3	11.1	0.071	0.91
${45}^{\circ }$	905	1052	7.0	10.6	0.068	1.02
${90}^{\circ }$	909	1062	7.1	10.7	0.068	0.96

.

2.1. Sheet Punching

Sheet punching is a shear cutting process commonly applied in the sheet-forming industry prior to hole-collaring operations and for creation of bolt joints. This process produces a hole in the sheet using a fixed die and a moving punch in a close-cut condition according to the schematic image in Figure 3.

The most significant process parameter of the hole punching process is the cutting clearance (Cl), which denotes the horizontal space between the punch and the die. Cutting clearance is expressed as a percentage according to Equation (1), where $R_{Die}$ and $R_{Punch}$ denote the die and punch radius and $t_B$ is the blank thickness.

$$\mathrm{Cl}[\%]={\displaystyle \frac{R_{Die}-R_{Punch}}{t_B}}·100$$

(1)

The sheet-punching experiments conducted in this work were performed according to the ISO16630 HET standard procedure [37], which describes a testing method used for evaluating the stretch–flangeability of sheet metals. The experiments were conducted for cutting clearances ranging from 5% to 40% using a servo-hydraulic machine with an ISO16630 dedicated hole-punching tool with 150 kN maximum load. The total press cylinder force was recorded with a calibrated 200 kN strain-gauge load cell and the punch displacement was measured with a 100 mm Linear Variable Differential Transformer (LVDT) inductive displacement sensor. A total force of 5 to 15 kN was required for the blank holder spring device before punching. The punching depth was adjusted so that the bottom dead center corresponds exactly to the full material thickness. The punching piston speed, with a range of 0.05–55 mm/s, was set to 0.05 mm/s, corresponding to quasi-static loading conditions. The testing procedure described in the ISO standard is performed in quasi-static conditions, thus reducing the dynamic processing effects such as strain-rate dependency and adiabatic heating of the material. Such effects are otherwise common in industrial process, but in the present study they are disregarded for a clearer discerning of the material deformation behavior.

The punching of a 10 mm diameter ISO16630 hole in the center of a 100 × 100 mm sheet sample was made with a dedicated 4-column tool (Figure 3a) with tight coaxiality specifications (±5 µm) between upper and lower tool and high fitting accuracy for punch and die inserts (±5 µm). This allowed for an accurate (±1%) cutting clearance compliance around the cut hole perimeter. The punching tools (punch and die) were made of high-performance tool material with titanium carbonitride coating. Sharp tools were used with a circular cutting edge with an estimated 30 µm radii. The cutting processes investigated in this study employed a punch with a diameter of ${\mathrm{D}}_{\mathrm{P}}=10.00$ mm. Die diameters ${\mathrm{D}}_{\mathrm{D}}$ were varied to achieve different cutting clearances, as summarized in

Table 2. The die diameter ${\mathrm{D}}_{\mathrm{D}}$ used in experiments for clearances (Cl) 5–40%.

Cl [%]	5.3	8.5	10.5	12.1	17.1	20.5	23.8	27.0	33.7	40.3
${\mathbf{D}}_{\mathbf{D}}$ [mm]	10.16	10.25	10.31	10.36	10.51	10.61	10.71	10.81	11.01	11.21

. The blank thickness was ${\mathrm{t}}_{\mathrm{b}}=1.5$ mm

The effect of non-coaxiality on the ISO16630 test was previously reported as one of the main causes of the well-documented resulting scatter from this test [38]. Due to this, the coaxiality of the ISO16630 dedicated hole-punching tool used in the present work was controlled by the interrupted punching of a paper sheet. Figure 4 shows the results from paper punching where a slight unevenness in punch intrusion was detected. This caused some circumferential variation in the cutting clearance and consequently, the cut-edge morphology, to be further discussed in Section 4.

2.2. Cut-Edge Investigation and Shear-Affected Zone

Following the hole punching experiments, the cut edges were investigated and the distribution of cut-edge parameters was determined. The cut-edge parameters, schematically shown in Figure 1, consist of rollover formation, burnish surfaces, fracture surfaces, and possible burr formation. Cut-edge investigations are challenging since local formation of cut-edge parameters may vary within narrow ranges of the hole circumference and through the blank thickness. Naked-eye evaluation with a 2–3× magnifying glass is largely insufficient and at best suitable for a quick evaluation of the burnish-to-fracture ratio or the evaluation of burr presence.

A digital portable USB microscope with 50× magnification has proved to be much more useful for daily laboratory qualitative investigation of industrial components or testing samples with cut-edge issues, as shown in Figure 5a. Indirect conclusions can be drawn from the cut edge’s condition regarding cutting tool wear of the punch and die, as well as the coaxiality and clearance of the cutting tools. Quantitative cut-edge investigations require, however, some more advanced methodology. A classic method would be with the metallographic cross-sections shown in Figure 5b, which is particularly useful for determining the local clearance, the fracture angle, or the presence of burrs. This is, however, a destructive and time-consuming testing method, which delivers precise cut-edge information but only for a very specific location.

In this work, a focus has been placed on advanced non-destructive optical methods for cut-edge investigations in close-cut hole conditions. Traditionally, cut-edge characterization relies on destructive techniques such as metallographic cross-sections and microhardness indentation, typically using Vickers hardness testing. While these methods provide valuable information on local work hardening and deformation, they are limited to single cross-sections, are time-consuming, and cannot capture variations along the full hole perimeter. To overcome these limitations, two optical methods have been developed. The first, referred to as the EISYS (Edge Inspection System), provides a 360° panoramic 2D front view of the entire hole perimeter, enabling fast and comprehensive mapping of cut-edge features. The second method utilizes a high-magnification 3D stereo microscope to reconstruct detailed 3D edge profiles at selected positions along the hole circumference. In addition to these optical techniques, this study introduces grain shear angle measurement as a relatively new and physically meaningful way to characterize SAZ deformation. This approach offers higher spatial resolution and better captures localized shear deformation compared to conventional hardness-based methods, providing a valuable extension to established cut-edge assessment techniques.

2.2.1. Two-Dimensional Optical EISYS Methodology

A cut-edge inspection tool designed for delivering a 360° panoramic view of the cut-edge surface was used for the characterization of 10 mm diameter holes prepared under different punching conditions on 100 × 100 mm ISO16630 samples [39]. The cut-edge parameters were determined with the optical device shown in Figure 6 and Figure 7. A preliminary manual centering device was used with the specimen in burr-down configuration (Figure 6a,b). The whole clamping tool cassette was then transferred to the optical device and rotated 360° in 2° intervals along the hole perimeter (Figure 6c,d). The sample rotation was accurately centered around a 5 × 5 mm 45° reflecting fixed mirror, enabling a frontal orthogonal projected view on the hole cut edge and allowing accordingly for an accurate quantitative panoramic cut-edge parameter analysis for material thickness up to around 3–3.5 mm. Each time, pictures were generated with three lightning configurations, as shown in Figure 7b–d (coaxial, oblique left, oblique right). The red backlight for the upper rollover side and the green backlight on the lower burr side also help identifying the cut-edge boundaries. The local thickness was also measured manually around the hole location. After 360° stitching together of all pictures, the developed cut-edge parameter profiles for rollover, burnish, fracture, and burr height were determined via image analysis according to Figure 8.

Figure 8a–c show the raw panoramic images of the punched hole using 27% cutting clearance and the identified cut-edge parameters. Figure 8d delivers the digitalized distribution of the cut-edge parameters. Additionally, Figure 8e gives a bar chart of the average cut-edge parameter distribution over the blank thickness, and the error bars show the circumferential variation (standard deviation) of each cut-edge parameter. The EISYS procedure was performed for all cutting clearances, presented in Table 2, and the reconstructed panoramic images of the cut edges, plotting the relative proportion of cut-edge parameters, are shown in Appendix A.

2.2.2. Three-Dimensional Stereo Optical Cut Edge Methodology

A Keyence stereo 3D microscope (VHX-7000) device has been used to measure the full 3D cut-edge profiles of the punched edges around the hole perimeter in 90° steps with regard to the rolling direction (0°, 180° cross-section along the rolling direction; 90, 270° cross-section transverse to the rolling direction). A scan was performed with the Keyence microscope in the Z-direction through the focal range of a sample in order to build a fully focused image with a large working distance (depth of field ≈25 mm), similar to an already proven technique [40,41]. The pre-selected X-Y area was then scanned delivering a 2D/3D stitched image, as shown in Figure 9a. Virtual 3D profile cross-sections shown in Figure 10a–c were then generated from recorded images. While rollover and burr-side information were obtained directly from the horizontally lying sample (Figure 10a,c), the front view (Figure 10b) of the actual cut edge 90° to the burnish surface was obtained by a mirror technique. As for the EISYS 2D technique, a small free-moving 5 × 5 mm 45° prismatic glass mirror was introduced manually in front of the cut edge, reflecting the cut-edge vertical image into a flat image in the objective (Figure 9b–d). In this way, a 3D cut-edge picture from a mirror view can still be processed in the 3D stereo microscope, which was particularly advantageous for cut hole edge investigations. Such a mirror reflection technique allows a non-destructive analysis of around 4 mm wide local cut-edge sections at any location along the hole edge.

The cut-edge profiles were then built by combining the individual top, bottom, and front view virtual profiles with simple translation/rotation image analysis operations as displayed in Figure 10d. By following geometric conditions, the whole cut-edge profile can be reconstructed from the individual rollover, front view mirror, and burr profiles: the blank thickness being accurately measured, the upper and lower edges horizontally assumed in the bulk material region far away from the cut edge, and a vertical orthogonal cut along the primary burnish–punch axis for X = 0. The 3D stereo optical cut-edge methodology presented in this work requires the combination of only three recorded images, thereby minimizing image post-processing and ensuring low computational cost. This makes the advanced Keyence stereo 3D microscope mirror technique ideal for rapid cut-edge measurements, utilizing an affordable and minimal equipment set-up.

Some advantages of such advanced experimental optical techniques in comparison to traditional metallographic cross-sections are the broader general overview of the cut edge’s shape, as well as the freedom of virtual cuts, which can be repeated without destruction of the cut-edge samples as in the metallographic sections. From those virtual 3D cut-edge profiles, the evolution and scattering of the cut-edge conditions versus the cutting process parameters along the hole perimeter can be determined at whatever location needed. The optical resolution reached around 1–2 µm at 400× magnification, which was high enough for accurate 3D cut-edge parameter determination, especially useful for local cutting clearance, horizontal rollover, fracture angle, or burr width parameters, which were not delivered by the 2D front view panoramic EISYS device. Both non-destructive optical techniques from the 2D EISYS panoramic views and 3D stereo microscope profiles complement each other. More time-consuming, specific metallographic cross-sections were therefore only mandatory for more in depth microstructure and grain shear angle analysis from etched cut-edge cross-sections.

With the experimental cut-edge profiles, the fracture angles were measured at positions located at $0^{\circ }$, ${90}^{\circ }$, ${180}^{\circ }$, and ${270}^{\circ }$ around the hole circumference for each cutting clearance. In this work, the fracture angle was determined between a best-fit tangential line in the fracture zone and the vertical burnish surface axis [7]. Figure 11 shows how the fracture angle $\alpha$ was defined for both (a) no-burr and (b) burr conditions, whereas (c) shows the two fracture angles ${\alpha }_1$ and ${\alpha }_2$ associated with secondary burnish formation.

2.2.3. Metallographic Cut-Edge Investigation—Vickers Hardness Measurement

Local shear deformation of the SAZ was evaluated using Vickers micro-hardness tests in accordance with the ISO 6507-1 standard [30] on metallographic cross-sections of cut edges for a separate section per cutting clearance, aligned with the rolling direction. Given the near-isotropic behavior of the material, time-consuming hardness testing along several rolling directions would give insignificant variations of the results and was therefore out of the scope in the present work. Hardness measurements have been performed with LECO LM100/300AT pyramid micro-hardness testers and a Zeiss Axio microscope for hardness track image documentation. According to the Vickers hardness standard, the minimal distance from any indentation point to the edge should be at least 2.5 times the average indentation diagonal width. The minimal distance between neighbor indentations should be at least 3 times the average indentation diagonal width. The average indentation width is a function of the Vickers indentation weight applied (as well as the overall indented material strength).

In agreement with the ISO 6507-1 test standard, the Vickers hardness required a minimum distance between indentation measurement points of 50–70 µm for HV0.1 tests (100 g test load), which resulted in insufficiently detailed measurements for assessing work hardening near the cut edge. To increase testing resolution for this investigation, the test load was reduced to 25 g (HV0.025), which was the smallest calibrated weight possible and significantly smaller than examples found in the literature using HV0.1 [8,13,42] or HV0.05 [43]. A comparable 20 g testing weight (HV0.02) has also been used in [44] for a 1200 MPa AHSS dual-phase (DP) grade with similar SAZ strain localization challenges.

Hardness measurements were extracted along three horizontal paths on the metallographic cross-sections, which were the middle of the burnish zone, at the boundary between the burnish and fracture zones, and in the center of the fracture zone on the punched edge. In the vicinity of the punched edge, two parallel tracks were made with a 30 µm offset, using a 15 µm spacing between points. This method increased the amount of points close to the cut edge, thus increasing the result’s resolution in the cut-edge vicinity, as also performed in [43]. At distances from the cut edge greater than 200 µm, only one track was tested with 50 µm spacing between indentations. With a 25 g load (HV0.025), a 30 µm spacing could be achieved from the punched edge for the initial indentation compared, for example, with a minimum 70 µm distance from the edge using HV0.5 as mentioned in [43] for a DP800 steel grade. Figure 12 shows the three parallel tracks for the cut edge produced by 12% cutting clearance.

2.2.4. Metallographic Cut-Edge Investigation—Grain Shear Angle Analysis

Determining the SAZ width and damage imposed by shear cutting to the cut edge was also performed by measurement of the etched microstructure material grain shear angle, performed on cut-edge cross-sections. First, the investigation consisted of creating metallographic sections longitudinal to the rolling direction and perpendicular to the hole-punched edge. The polished metallographic section was dipped in Nital solution for around 10 s at room temperature. The Nital etching solution consisted of 2% nitric acid (HNO3) and 65% etanol (EtOH) and this solution is commonly used for the etching of metals. A trial-and-error manual etching technique was used until optimal microstructure etching contrast was reached.

In the next step, a high-resolution image analysis of the metallographic cut-edge cross-section was created. The image analysis was performed on the etched samples using a digital stereo microscope set-up (Keyence VHX-7000, Osaka, Japan) for light optical microscope analysis at 1000× maximum magnification. A scan was performed in the Z-direction through the focal range of a sample in order to build a fully focused image of the etched microstructure in the cut-edge vicinity. The pre-selected X-Y area was then scanned delivering a sharp 2D stitched image. The 3D modus in the vertical direction allowed for better resolution and sharper images in contrast with classical 2D metallographic microscopes. The X-Y stitching device enabled a broad field of view by stitching multiple single images with high magnification and resolution. This quality of the image was necessary since the SAZ width is very narrow and steep in the cut-edge region for the investigated steel grade. From the high-quality images of the metallographic sections, the material’s grain shear angle was manually calculated with Image J software (1.53 Version) for six vertical positions along the cut edge and into the SAZ width (burnish start, burnish center, burnish end, fracture start, fracture center, and fracture end). The paths were chosen to include all events of the shear cutting process, thus giving a varied and widespread description of the sheared-edge damage vertically along the cut edge.

This generated two-dimensional paths into the SAZ width, where the grain shear angle was extracted with even steps. Figure 13a shows exemplarily the paths for the 12% cutting clearance case and the points of grain shear angle measurement. Figure 13b illustrates conceptually how the grain shear angle value $\alpha$ was measured in Image J software (1.53 Version) at the measurement points along the grain lines. The grain shear angle measurements were performed for cutting clearances 5%, 12% 20%, and 27%. Contrary to the automated horizontal Vickers hardness measurements lines in Figure 12, the grain shear angle measurements are performed in a more physically meaningful manner following the material flow lines, starting from defined locations at the cut edge up to the undeformed base material, as illustrated in Figure 13. This is a particular advantage of grain shear angle analysis compared to Vickers hardness testing, which usually requires predefined horizontal or vertical automated hardness tracks.

The high strength level of the investigated CP1000HD steel grade caused an abrupt strain localization tendency; thus, a general analytical exponential fit could not be achieved with enough accuracy. Therefore, a piece-wise linear analysis was preferred, subdividing the total SAZ into a localized bending SAZ versus the localized shear SAZ contribution. A schematic image of grain shear angle results and the piece-wise linear curves is shown in Figure 14, and from this kind of plot it was possible to determine the critical fracture shear angle at $F_{AB}{\left(x\right)|}_{x=0}$, the width of the localized shear SAZ at $x_{localization}$, and the width of the total SAZ at $F_{BC}{\left(x\right)|}_{y=0}$.

3. Results

Using the 3D stereo optical cut-edge methodology presented in Section 2.2.2, the cut-edge profiles for at least four different positions along the hole perimeter were extracted. These profiles are shown in Figure 15. Due to the appearance of a partial burr at around 0° and 180° rotation for the 27% cutting clearance case, several additional profiles were extracted.

The 5% cutting clearance case involved significant circumferential heterogeneity, causing the appearance of secondary burnish islands within the fracture zone and uneven distribution of burnish/fracture surface heights. This heterogeneity is displayed in Figure 16, showing how the cut-edge profile varies locally for different positions along the hole perimeter.

Figure 17 shows a superposition of 3D optical profiles from Figure 15 and Figure 16 in all directions along the hole perimeter versus the metallographic cross-sections sampled longitudinal to the rolling direction. For 5% cutting clearance, only the 3D optical profiles longitudinal to the rolling direction from Figure 16a are shown for clarity. For 27% cutting clearance the 0° metallographic section direction also coincided with the progressive burr/no-burr transition zone along the hole perimeter, leading to some more ambiguous comparison results with optical profiles.

Similarly, the 2D optical EISYS methodology was compared to the metallographic cut-edge cross-section, as presented in Figure 18. Figure 18 shows the cut-edge cross-section extracted longitudinal to the rolling direction and the EISYS results at 0 $\pm {15}^{\circ }$ to the rolling direction, where the outer blank edges and the characteristic cut-edge parameters are compared with red lines.

The final cut-edge results were also defined in terms of rollover, burnish, fracture, and burr height, i.e., the cut-edge parameters and their respective distribution over the blank thickness, in accordance with Figure 8. The results were obtained for the entire range of cutting clearances presented in Table 2 using the 2D optical EISYS method described in Section 2.2.1. The results shown in Figure 19 describe the average cut-edge parameters as black crosses and the circumferential variations as error bars (standard deviation).

Similarly, the fracture angles (defined in Figure 11) are displayed in Figure 20, measured from the cut-edge cross-section profiles in Figure 15 and Figure 16. Due to the slight and unavoidable non-coaxiality of the tools, marginal clearance deviation occurred, which caused variations in the cut-edge morphology and, consequently, the fracture angles. Here, the results showed an increase in fracture angle until a burr appeared at the 27% cutting clearance, which caused the fracture angle to decrease. At higher clearances ≥27%, it can be assumed that the burr width increases to some extent at the expense of the fracture angle. For 27% cutting clearance, partial burr formation occurred, which caused one part of the hole to have a burr, while the other was left without a burr. Therefore, the no-burr/burr fracture angles are displayed distinctly in Figure 20.

The hardness indentation results are shown in Figure 21 for each cutting clearance, combining the three paths in the burnish, burnish/fracture, and fracture zones, as illustrated in Figure 12.

By investigating the material orientation of the SAZ, it was possible to determine the SAZ width and level of deformation in the cut-edge vicinity. The grain shear angle results were obtained by the method described in Section 2.2.4 according to the six paths shown in Figure 13. The combined grain shear angle results from each path were gathered and plotted for each cutting clearance case, as shown in Figure 22. As for the indentation results, the dimensionless normalized cut-edge distance with respect to the blank thickness was introduced. In this figure, the piece-wise linear fitted lines determining the point of localized shear deformation and total SAZ width are shown.

For comparison between the Vickers hardness measurements and the grain shear angle results, the respective results are shown in Figure 23 for 12% cutting clearance.

Figure 24 shows the grain shear angle measurements at 12% cutting clearance for each of the paths mentioned in Section 2.2.4. These results combined form the plots shown in Figure 22b and Figure 23.

4. Discussion

Figure 15 shows how the non-destructive 3D Keyence cut-edge assessment technique developed within the present work could produce reliable results for cut-edge profiles around the hole perimeter. The use of such a technique is straightforward as it enables identification and measurement of local profiles, possibly involving unusual features such as secondary burnish formation and partial burrs. Similarly, the 2D panoramic EISYS cut-edge investigation technique enabled an effective separation of each characteristic cut-edge feature, as shown in Figure 19. The present cutting clearance results for the CP1000HD grade fit well with the body of work with AHSS sheets found in the literature [14]. With increasing clearance a steady rollover increase was seen, along with a moderate burnish formation decrease and strong fracture height decrease. Also, a burr formation increase was expected with increasing clearance. There seems to be a minimum value for the burnish zone and the burnish-to-fracture ratio around 15–20% clearance for AHSS grades, as also observed for other 1000–1200 MPa tensile strength steel grades by [39]. Burr formation was triggered around 20–30% clearance, as also observed in [14]. The present CP1000HD punching clearance investigation showed partial burr formation starting at 27% and homogeneous burr formation at 34% and 40% cutting clearance around cut hole edges. Partial burr formation was possibly caused by the slight tool misalignment presented in Section 2.1. Specific cut-edge features such as uneven burrs along the cut-hole perimeter have been reported in similar investigations as a relevant issue impacting further material cut-edge stretch–flangeability [26,41,45].

On the opposite lower side of the cutting clearance spectra, the experimental cut-edge results show islands of secondary burnish formation. The secondary burnish formations are shown in the circumferential representation of the cut-edge parameter distribution in Figure A1, whereas Figure 16 shows the various resulting shapes of the experimental cut-edge profiles along the hole perimeter. As for the case of the 27% cutting clearance, the significant circumferential variation in the experimental cut-edge morphology was caused by the slight tool non-coaxiality shown in Figure 4. The secondary burnish zones were formed due to mismatching fracture angles from the punch and the die edges. As shown in Figure 20 in the low 5% clearance part, the fracture angles are higher compared to the 8–10% clearance level. This is due to a higher total punch penetration depth up to secondary burnish zone end. This is true especially when considering the significantly higher secondary fracture angle ${\alpha }_2$ as compared to the lower primary fracture angle ${\alpha }_1$, as defined in Figure 11c.

The comparison between the metallographic cut-edge cross-section and the 3D Keyence profile technique in Figure 17 clearly validates the use of the non-destructive edge profile method. Similarly, comparison between the cut-edge cross-sections and the 2D EISYS technique also confirms this approach, as shown in Figure 18. The strength of the non-destructive 2D/3D high-resolution stereo optical methods were apparent as the circumferential variations in the cut edge were detectable.

When investigating the SAZ in more detail by means of the grain shear angle analysis method, a few observations can be made from Figure 22, Figure 23 and Figure 24. Firstly, the grain shear angles reached approximately 90° in the cut-edge vicinity, which implies that the material undergoes extreme deformations in mixed compressive/shear stress states without fracturing (Figure 22 and Figure 23). Secondly, the total SAZ width appeared to be non-uniform in thickness direction and steadily decreasing close to the die. As an example, Figure 24 shows the extent of the SAZ width derived from the grain shear angle at different locations across the cut edge from the start of the burnish zone to the fracture end zone. It becomes then apparent that the width of the plastically deformed overall SAZ zone decreases from the cutting punch to the die side. The localized deformation, however, remains confined around a 0.1 normalized thickness level, regardless of the cut-edge zone considered. This behavior is also confirmed in a similar grain shear angle experimental investigation for DP/CP800 grades [34]. The width of the total SAZ also had an obvious clearance dependency (Figure 22) as the bending of the blank structure increased with larger clearances, while the work hardening of the material was also considered to affect the total SAZ width. Accordingly, shear angle measurements may provide additional and relevant information to better understand sheared-edge damage.

By comparison between the Vickers hardness measurements and grain shear angle results in Figure 23, it is apparent that the grain shear angle measurements can better capture the sharp shear-deformation localization tendency of the CP1000HD grade. Even though a small indentation weight (HV0.025) and parallel offset indentation tracks were used for increased resolution, it was still insufficient and resulted in significant scatter (Figure 21 and Figure 23). Compared to the grain shear angle measurement, the point of $x_{localization}$ is rather vague from the hardness measurement. For open-cut configurations, digital image correlation (DIC) may serve as a useful tool in determining the SAZ strain field [46,47,48]. Such a technique enables a step-by-step tracing of the SAZ strain field along the shear cutting process and investigation of individual strain components. However, there is no feasible way of performing DIC analysis of close-cut shear cutting due to the obvious reason that the deformation zone is within the material. In this case, assessment of the SAZ deformation needs to be conducted on post-cutting cross-sections. In [1,34,49], different approaches to calculating the effective and shear strain based on grain shear angles are presented, providing possibilities for the strain mapping of the SAZ without the use of stepwise DIC monitoring. This grain shear angle technique can even be used postmortem for industrial component sheared-edge analysis, whereas DIC online punch monitoring is also not feasible in a press shop environment. Additionally, nano-indentation is a powerful tool for local hardness evaluation [50] and could give more accurate results near the edge [51]. The use of nano-indentation is out of the scope of the present paper but should be considered in future works. One noticeable experimental method consists of a probe arm with two needles [43], with such a mechanical scanning device being able to record 3D profiles of the whole shear cut edge at once with even better accuracy (0.5 µm) than the optical Keyence system (1–2 µm). This testing method is non-destructive but would still require multiple scans around the hole perimeter and is rather used for more accessible open-cut trimmed edges [43].

In addition, grain shear angle results may be translated to physically relevant failure strain parameters [1,34,49], while conversion of Vickers hardness values into material properties is less direct [34,52]. These results stated that grain shear angle measurement should be the preferred choice when it comes to metallographic SAZ investigations of materials with a high tendency for strain localization, such as AHSS with UTS > 1000 MPa. An additional benefit of this method is the automation possibility using image analysis and measurements with similar methodologies, as used in [53].

To meet Industry 4.0 demands, there is a growing need for stringent in-line monitoring of cut-edge parameters and overall edge quality, especially for materials with tensile strengths of 800 MPa and above. For such high-strength materials, the relationship between cut-edge quality and forming characteristics, such as stretch–flangeability, sensitivity to hydrogen embrittlement, as well as low- and high-cycle fatigue performance, is crucial. The optical testing methods presented are already part of a comprehensive database linking edge crack sensitivity to cut-edge conditions. For this topic, so-called 2.5D optical photometric stereo technology is on the rise for industrial tool wear monitoring, for example [54]. This is an extension of the 3D Keyence technique presented in this paper. Internal work is also pursued in this direction based on a patented own-shape from shading reconstruction technique with multiple sequential lightning at different known angles of a fixed sample and camera [55]. Additionally, artificial intelligence and machine learning techniques are increasingly used to model hole expansion ability in relation to cut-edge quality, addressing the numerous parameters and complex interactions between cut-edge and material properties [56,57]. When combined with ML algorithms or other AI applications, this in-line monitoring can effectively prevent downstream production issues, thus reducing scrap in production of lightweight AHSS components.

5. Conclusions

Concludingly, for investigation of cut edges produced by shear cutting, the combination of cut-edge analysis tools, ranging from naked-eye observation to micro-mechanical analysis, all lead to better understanding of sheared-edge damage. Based on such experimental findings, the following conclusions can be drawn:

• A thorough understanding of cut-edge parameters (such as nominal vs. effective clearance, hole perimeter coaxiality, and punch tool wear) is essential for accurately interpreting laboratory material test results, particularly for assessments of edge crack sensitivity, hydrogen embrittlement, and fatigue involving cut-edge conditions. Laboratory testing under these conditions is meaningful only when preceded by comprehensive 2D and 3D punch edge characterization as an integral part of the testing protocol. This approach is necessary to distinguish intrinsic material properties from cutting process effects, preventing misinterpretations of material behavior. Also, in industrial forming lines, where circumferential variation in the cut hole edge may occur, cut-edge investigation considering the cut-edge’s perimeter is required. This can be achieved with the 2D panoramic optical EISYS or by the non-destructive 3D optical cut-edge profile determination testing methods developed within this work.
• Investigating the effect of varying the cutting clearance on cut-edge morphology showed that both optical cut-edge investigation methods were useful for detecting the characteristic cut-edge parameters. The results show the appearance of secondary burnish formation at ≈5% and burr formation at ≥ 27% cutting clearance (for sharp tools). In between, smooth fracture zones were detected, with a minimum burnish-to-fracture ratio of 12–15% cutting clearance.
• The 3D optical cut-edge profiling technique is still quite time-consuming. The feasibility has been proven in this investigation for a significantly high number of profiles. A mechanical device for tilting and rotating of the sample could allow for an automatization of an optical high resolution 3D cut-edge profile determination technique in the future. However, the Keyence stereo 3D microscope mirror technique employs simple equipment set-up and minimal post-processing requirements, making it comparably affordable and computationally efficient.
• Due to the narrow shear localization tendency of the CP1000HD grade evaluated in this work, it is concluded that grain shear angle measurement is the preferred method over hardness indentation for AHSS grades since grain shear angle measurements are able to give high-resolution results in the immediate 5 to 10 µm cut-edge vicinity. This is of particularly acute relevance for 1200 MPa strength AHSS with an even sharper strain localization tendency.
• The SAZ grain orientation technique was manually implemented in a time-consuming and tedious manner. It should be further developed according to state-of-the-art image analysis tools. Two-dimensional high-resolution mapping of cut-edge SAZ shear angle would be a valuable addition to Vickers hardness mapping. The shear angle from microstructural measurements allows for a direct physical comparison of experimentally derived shear and Von Mises effective strain results with finite element simulations, as well as material flow motion investigations in shear cutting operations.
• The presented optical techniques provide the data foundation needed for AI and ML to reliably monitor shear processes in-line, enabling real-time edge damage assessment and supporting process optimization in Industry 4.0 environments.

In conclusion, it is worth noting that optical methods have gained significant momentum due to recent advancements in optical hardware, IT computing power, and image analysis software. These new trends suggest that further research and development should focus on fully automated mechanical and optical devices. Systematic cut-edge characterization cannot be achieved on a regular basis with conventional destructive cross-section techniques, especially not if the same sample is then to be tested later on. The advanced optical testing methods presented in this work can contribute to more widespread and scientific cut-edge characterization results involving mass testing. Moreover, this type of 2D/3D high-resolution stereo optical testing device can be easily adapted from the originally targeted ISO 16630 hole with a 10 mm diameter to any trimmed component shape in an open-cut configuration.