Skip to main content

Taking powder bed fusion precision to the next level with in-situ laser ablation

Manuel Henn, Matthias Buser and Volkher Onuseit, of the University of Stuttgart’s IFSW, combine additive and subtractive laser processes to unlock new manufacturing possibilities

Laser-based powder bed fusion of metals (PBF-LB/M) is an additive manufacturing process used to create highly complex parts by melting multiple fused beads in consecutive layers of metal powder.

The unrivalled freedom of design is its main selling point.

Today PBF-LB/M is no longer found only in scientific labs, with several industrial applications ranging from prototype work and spare-part production to small quantity batch-produced components.

The surface quality of PBF-LB/M parts is often compared to that of castings, and post processing including support structure removal, cleaning, deburring or sanding, which is mandatory to make parts ready to use. Subsequent machining operations are needed if additional features with an accuracy of a few micrometres are specified, such as threads, precision holes or bearing seats. This is necessary due to the inherent limitations in geometrical accuracy of PBF-LB/M in the lateral direction.

The achievable precision is determined by process and system parameters, primarily by the minimum diameter of the laser beam on the surface of the printed part, which is typically 50 to 500μm. Additionally, feed rate and laser power determine the width of the melt pool and thus the minimum structure size. By leaving a gap between two adjacent scan lines, small structures such as internal slits with a width of about 100μm can be created1. However, the generated slits lack contour sharpness caused by the freeform solidification of the melt and the interactions of the incident laser beam, melt pool and powder2,3. This effect can be compensated to an extent using an even smaller focal diameter and finer powders4,5.

But there are numerous industrial applications that require parts with internal micro-structures on the micrometre scale. Such applications cannot be realised with PBF-LB/M components if the structure is geometrically too sophisticated for common subtractive post-processes, or the structure is inaccessible for post-processing.

The aim

Overcoming these limitations is the goal of a new research project at the Institut für Strahlwerkzeuge (IFSW) at the University of Stuttgart. Together with its partners from the Laser Application Centre (LAZ) at Aalen University, and the Institute for Production Engineering (WBK) of the Karlsruhe Institute of Technology, the IFSW launched the research project ADDSUB as part of Innovation Campus Future Mobility, which is funded by the Ministry of Science, Research and Arts of Baden-Württemberg state.

ADDSUB aims to combine additive and subtractive laser processes in the same machine. An animation of such a process is on IFSW’s YouTube channel6. By laser ablating precise micro features in each layer, the team aims to increase achievable accuracy by at least an order of magnitude compared to the conventional PBF-LB/M process.

The application for this technology within ADDSUB is printing top-performing soft magnetic components for electric motors. Such parts have been made from stacked sheet metal to combine the superior flux-conducting properties of the base material with an eddy-current-prohibiting design, i.e. the high-resistance interface between the sheets. With PBF-LB/M, the design constraints by stacking sheets can be overcome, and the geometry of, for example, an optimised stator of a transverse flux machine, or an electric motor for wheel hub drives, can be produced as a single printed part. However, the thin slits required for eddy-current prevention pose a challenge to conventional PBF-LB/M.

To exceed current limitations on narrow slits in printed parts, ultrafast laser ablation of single lines within every newly built-up layer is going to be used to produce vertical slits across multiple layers, mimicking the effect of sheet stacking.

The experiments

To the best of the authors’ knowledge, there is currently no laser beam source of combined continuous-wave and ultrafast capability with the needed high average power available on the market. Therefore, the experimental system incorporates two separate beam sources.

A schematic illustration of such a system is shown in figure 1, consisting of a custom-built powder bed, a galvanometer scanner with an f-theta lens, and two laser sources. A continuous-wave laser is used for the additive process of melting the metal powder for each new layer. The ultrafast laser is used to locally ablate material after each melting process.

Figure 1: Schematic illustration of the experimental setup for the combined additive and subtractive laser processes. (Image: Henn et al.)

Both laser sources operate at the same wavelength of 1,030nm and are guided along the same optical path to the scanning and focusing optics, and finally to the workpiece. Fast and reproducible switching between the two laser sources is enabled by a pneumatically switchable flipping mirror, allowing a machine-intrinsic manufacturing process consisting of sequential additive and subtractive laser processes. The realised focal diameters of the continuous-wave and ultrafast lasers are 190±5μm and 50±5μm, respectively.

For the first experimental phase, pure iron powder with an average particle size of 35μm was used for the additive process in the following example. To melt the metal powder, the continuous-wave laser was operated at an average power of 400W in combination with a scanning speed of 1m/s. The height of each powder layer was set constant to 50μm. A cross jet was used to prevent spatter from reaching the f-theta lens. Nitrogen was used as shielding gas during the process. The smallest possible weld bead width was approximately 200μm.

The subtractive process was performed as follows. With an average power of 30W and a scanning speed of 3m/s, lines with a length of 4mm were ablated after each consecutive layer of powder had been welded onto the part. The pulse energy was 100μJ with a pulse duration of approximately 8ps. With a repetition rate of 300kHz, the resulting pulse overlap was approximately 80 per cent. The number of repetitions for each line was set to 500 after each added layer. The consecutive process steps, PBF-LB/M and ablation, described above were repeated for 100 layers, resulting in cube-shaped samples with an edge length of 5mm. The samples are shown in figure 2. The orientation of the slits created in the right sample is marked with red rectangles.

Figure 2: Samples of pure iron produced by 100 cycles of consecutive additive and subtractive laser processes to reach a height of 5mm. The orientation of the slits created in the right sample is marked with red rectangles. (Image: Henn et al.)

In figure 3 a cross-section is shown perpendicular to the scanning direction of the ablated lines. The porosity visible in this picture could be explained by a less than optimal inert gas atmosphere and the high reactivity of the pure iron with oxygen.

Apart from some minor transversal bulges, the average slit width is 40±10μm. With a built height of 5mm, this results in an aspect ratio of height-to-width of about 125:1. No principal limits for the achievable aspect ratio were found at this stage of the investigations.


This is a major achievement because it pushes the boundaries of what is possible with post-processing, even if the geometry is of great simplicity and accessibility. A second important finding of the initial experiments is that although the slit width is minimally larger than the average particle size of the powder, no residual particles were observed within the slits.

This is a remarkable discovery which is to be focused on during upcoming investigations. It indicates that to some extent, thin empty voids can be produced with this combined process and open ports for later powder removal are dispensable.

Beyond the slits, future investigations will aim to incorporate the controlled ablation process of horizontal or skewed surfaces into the system.

Figure 3: Cross-section through a sample of pure iron. 100 cycles of consecutive additive and subtractive laser processes were performed. The average width of the slits produced by the subtractive laser process is 40±10μm.

As shown by the IFSW7, this has already proven beneficial in a post-processing step to increase the accuracy and surface quality of PBF-LB/M printed parts. The in-situ application of this technology, based on what has been shown previously, will increase the possibilities even further.

It can already be said that combining additive and subtractive laser applications in a consecutive process is a promising approach to achieving a new level of precision in printed parts. Bringing this technology to market requires little more than a flexible laser source within currently available PBF-LB/M machines to cover both process regimes.

The initial results8 of the ADDSUB project indicate a most probable future demand. It demonstrates the enormous potential of future laser materials processing through process combinations on the way to a universal laser machine that expands the horizon for previously unimagined applications.


  1. Yasa E, Kruth J-P, Deckers J. Manufacturing by combining Selective Laser Melting and Selective Laser Erosion/laser re-melting. CIRP Annals 2011; 60(1):263–6.
  2. Goll D, Schuller D, Martinek G, Kunert T, Schurr J, Sinz C et al. Additive manufacturing of soft magnetic materials and components. Additive Manufacturing 2019; 27:428–39.
  3. Goll D, Schurr J, Trauter F, Schanz J, Bernthaler T, Riegel H et al. Additive manufacturing of soft and hard magnetic materials. Procedia CIRP 2020; 94:248–53.
  4.  Li R, Liu J, Shi Y, Wang L, Jiang W. Balling behavior of stainless steel and nickel powder during selective laser melting process. The International Journal of Advanced Manufacturing Technology 2012; 59(9-12):1025–35.
  5. Ziri S, Hor A, Mabru C. Effect of powder size and processing parameters on surface, density and mechanical properties of 316L elaborated by Laser Powder Bed Fusion. ESAFORM 2021.
  7. Holder D, Leis A, Buser M, Weber R, Graf T. High-quality net shape geometries from additively manufactured parts using closed-loop controlled ablation with ultrashort laser pulses. Advanced Optical Technologies 2020; 9(1-2):101–10.
  8. Henn M, Buser M, Onuseit V, Weber R, Graf T. Combining LPBF and ultrafast laser processing to produce parts with deep microstructures. In: Wissenschaftliche Gesellschaft Lasertechnik e.V., editor. Proceedings of the Lasers in Manufacturing Conference 2021.

Editor's picks

Media Partners