Skip to main content

Exploring the potential of diode lasers for additive manufacturing

Tim Lantzsch and his colleagues investigate the feasibility of diode lasers for the laser powder bed fusion of stainless steel

This article was co-authored by Markus Schauerte, Martin Traub, Thomas Westphalen and Christian Tenbrock of Fraunhofer ILT, and Johannes Schleifenbaum of the RWTH Aachen University Chair for Digital Additive Production

In recent years, laser powder bed fusion (LPBF) has become an established manufacturing technique due to it enabling the manufacture of complex part geometries without additional tools1.

State-of-the-art LPBF machines feature a combination of multiple single-mode fibre lasers and galvanometer scanners due to their excellent focusability (BPP < 0.4mm·mrad) and dynamic beam positioning. However, due to the high cost of fibre laser beam sources and galvanometric scanners, LPBF machines still pose significant investment costs.

Highly efficient high-power diode lasers could present a more-affordable alternative to fibre lasers in LPBF machines. However, the lower beam quality (BPP > 8mm·mrad) and spectral width (920 to 1,050nm) of commercial high-power diode lasers results in chromatic aberrations and reduced focusability, thus making modifications to LPBF machines necessary.

Current approaches to address these challenges rely on a combination of fixed focusing optics with a gantrybased positioning system2,3,4, resulting in less dynamic laser beam positioning and thus reduced system productivity. To overcome these productivity restrictions current approaches use diode laser arrays to scale melt pool size2,3,4, which results in reduced part quality, namely higher surface roughness and lower geometrical accuracy.

In this study – conducted as part of the Fraunhofer lighthouse project FutureAM: Next Generation Additive Manufacturing – an optical system, featuring a standard galvanometer scanner and a colour-corrected f-theta lens, was developed and integrated into an LPBF machine. Furthermore, the optical system was used to process stainless steel AISI 316L to demonstrate its feasibility for producing fullydense metallic components.

Machine configuration

For the machine setup, a commercial grade fibre-coupled TruDiode 301 laser from Trumpf, emitting a maximum laser power of PL = 300W at λ = 920 - 1,050nm, was used. It was paired with Trumpf’s QBH-compatible collimator (f = 100mm) and combined with an IntelliScan III 30 galvanometer scanner from Scanlab. In order to limit chromatic aberrations resulting from the spectral width of the high-power diode laser, a colourcorrected f-theta lens was developed and manufactured.

The result of the Zemax simulation for the laser spot deformation due to chromatic aberrations at different deflection angles for different wavelengths, as well as the resulting optical design for the f-theta lens, are demonstrated in figure 1. The beam paths indicated in blue, green and red represent different deflection angles of the scanner mirrors M1 and M2.

Figure 1: a Zemax simulation of spot deformation at different deflection angles of the laser beam resulting from chromatic aberrations (left) and the resulting optical design of the colour-corrected f-theta lens (right)

The developed f-theta lens uses a back focal length of 185mm, which resulted in a focus beam diameter of approximately 235µm. The deflection angle of the laser beam was limited to 6.5° to limit chromatic aberration (see figure 1), which resulted in an effective scan field size of ∅ = 90mm. The laser power characteristic and laser beam caustic of the laser optical system are in figure 2.

Figure 2: laser power characteristic (left), laser beam caustic measured at PL = 300W (middle) and laser intensity distribution at focal position (right) for the developed DL-LPBF optical system

The laser optical system yielded a focus beam diameter of ds  = 237µm at a Rayleigh length of zR = 1.9mm. The optical system was mounted onto a motorised z-axis, which was used during the process setup to precisely align the focus position to the LPBF machine’s substrate plate, and was then kept at a constant position during processing.

An overview of the LPBF machine setup is in figure 3.

Figure 3: DL-LPBF laboratory machine setup

Processing results

The LPBF machine setup described above was used to perform the first trials on the diode laser based LPBF (DLLPBF) of stainless steel AISI 316L. Hatching parameters were determined to ensure manufacturing of functional parts with a relative density of ρ > 99.9 per cent. Due to the limitation of the available laser power and the comparably large focus diameter of ds = 240µm, the powder layer thickness and laser power were kept constant at DS = 30µm and PL = 300W respectively. Hatch distance Δys and laser scanning speed vs were varied accordingly. The resulting relative density was determined optically via microsections and image processing.

The parameter range with the resulting densities and build up rates, as well as an exemplary microsection, are depicted in figure 4.

Figure 4: identified processing parameter range for the DL-LPBF of stainless steel AISI 316L (left) and corresponding microsection for the chosen process parameter set (right), highlighted in red

The chosen parameter set (highlighted in red) yields a reproducible relative density of ρrel = 99.94 per cent and a theoretical build up rate of 4.5mm³/s.

To improve the surface quality, contour parameters were introduced. The line energy during the contour exposure was varied through adjustment of the laser power and scan speed. The resulting surface roughnesses of side-skin surfaces in as-built condition and after sandblasting are displayed in figure 5. The measured surface of the test specimens is highlighted. A contour line energy of EL,C = 0.43J/mm yielded a minimum average surface roughness of SA = 25.6µm in as-built condition, which corresponded to a minimum average surface roughness of SA = 13.6µm after sandblasting.

Figure 5: resulting average surface roughness SA depending on the contour line energy EL,C in as-built condition and after sandblasting

Furthermore, geometrical restrictions during DL-LPBF, such as maximum free overhang angle and minimum detail resolution, were investigated by manufacturing test specimens using the above-mentioned processing parameters. The results are demonstrated in figure 6.

Figure 6: overview of geometrical restrictions (highlighted red) and manufacturable geometrical features (highlighted green) for DL-LPBF processing of stainless steel AISI 316L

The scanner-based DL-LPBF approach presented within this article enables the manufacturing of free overhang angles of up to α = 40°. A minimum wall thickness of twall = 250µm and a minimum gap with of tgap = 200µm can also be achieved. Horizontal bores with diameters of dbore = 1 - 5mm can be manufactured without internal support structures. Larger bore diameters require internal supports for defect-free manufacturing, whereas smaller bore diameters cannot be manufactured due to powder sintering.

The findings obtained from the process parameter study and the investigation on geometrical limitations were transferred onto a turbocharger impeller demonstrator, which was chosen to show the capabilities of scanner-based DL-LPBF. The impeller had a diameter of 85mm and a height of 50mm. An image of the manufactured and sandblasted impeller, as well as the 3D scan of its geometrical accuracy, using striped light projection, are shown in figure 7.

Figure 7: turbocharger demonstrator manufactured using DL-LPBF (left) and corresponding 3D-scan after sandblasting using striped light projection (right)

The 3D scan of the demonstrator geometry indicated an average geometrical deviation below 0.2mm. However, some of the impeller’s blade tips exhibited significantly larger deformations. These only occured on blade tips with overhang areas oriented parallel to the recoater and facing in the recoating direction. Due to the filigree blade geometry, the overhangs deformed and hindered the recoating of subsequently manufactured layers, resulting in geometrical defects. A possible solution to this problem is the use of dedicated down-skin parameters for the manufacturing of filigree overhang structures.

Conclusion

In summary, scannerbased DL-LPBF enables the manufacturing of fully-dense metallic components with similar properties to those manufactured with state-of-theart fibre laser LPBF machines. This demonstrates the feasibility of diode lasers for additive manufacturing. However, the spectral width and beam quality of high-power diode lasers result in a complex and costly optical design and restrictions regarding the achievable maximum scan field size.

Nevertheless, DL-LPBF, especially when combined with other emerging technologies such as direct blue diode lasers and dense wavelength combining, will be well suited to future LPBF applications.

Tim Lantzsch is team leader of LPBF machine technology at Fraunhofer ILT

References

[1] Terry Wohlers, Robert I Campbell, and Olaf Diegel. Wohlers report 2020. ‘3D printing and additive manufacturing state of the industry’ 2020. 978- 0-9913332-6-4.

[2] Florian Eibl. ‘Laser powder bed fusion of stainless steel with high power multi-diode-laser-array’. Dissertation. 1. Auflage. Edition Wissenschaft Apprimus. 978-3-86359-587-6.

[3] Florian Eibl, Christian Tenbrock, Tobias Pichler, Tobias Schmithüsen, Daniel Heussen, and Johannes H. Schleifenbaum. ‘Alternative beam sources and machine concepts for laser powder bed fusion’. Piscataway, NJ: IEEE, 2017. Proceedings of the 2017 High Power Diode Lasers and Systems Conference (HDP). 9781538632642.

[4] Miguel Zavala-Arredondo, Nicholas Boone, Jon Willmott, David TD Childs, Pavlo Ivanov, Kristian M. Groom, and Kamran Mumtaz. ‘Laser diode area melting for high speed additive manufacturing of metallic components’. Materials & Design. 2017, 117, 305-315. Available from: 10.1016/j. matdes.2016.12.095.

 

Media Partners