Abstract
Due to the width of the laser melt pool and the sintering effect on the surrounding powder, the experimental size of the selective laser melting (SLM) sample will be larger than the design size, which will greatly affect the dimensional accuracy and surface quality of the thin-walled sample. In order to obtain SLM thin-walled TC11 specimens with precise dimensions, an orthogonal experiment was designed to investigate the effects of laser power, scanning speed, and hatch space on the relative density, wall thickness (IWT) increase, and surface roughness of the specimens.
Unlike previous studies, the results showed that IWT and roughness decreased with increasing scanning speed. It was found that due to laser beam drift under high scanning speed conditions, the IWT and roughness of the sample decreased with increasing scanning speed and then increased. The laser beam drift increases with the increase of scanning speed. When the scanning speed is ≤ 400 mm/s, the laser beam drift is almost 0 mm. At a scanning speed of 5000 mm/s, the laser beam drifts 1.51 mm in the x direction and 1.28 mm in the y direction. The drift of the laser beam causes an increase in the cross-sectional area of the sample, resulting in the same energy distribution over a larger area and a decrease in relative density; Meanwhile, the drift of the laser beam causes the scanning paths of different layers to not completely overlap, which increases the roughness of the sample. When the laser power is 350 W and the scanning speed is 1000 mm/s, the cross-sectional area of the sample increases by 1.71 mm2 due to the width of the melt pool, which is greater than the value caused by laser beam drift of 1.70 mm2. In this case, the width of the melt pool is the main factor affecting the size of the specimen. Therefore, in order to reduce the negative impact of laser drift on sample density, the scanning speed should be ≤ 1000 mm/s when the laser power is 350 W.
Introduction
The TC11 alloy with a nominal composition of Ti-6.5Al-3.5Mo-1.5Zr-0.3Si is an α+β biphasic titanium alloy. Due to its excellent comprehensive mechanical properties at room temperature and high temperature, it is widely used in key components of aerospace engines such as pressure plates. Selective Laser Melting (SLM) technology utilizes laser to melt metal powder layer by layer, achieving additive manufacturing of metals. One major advantage is that it can produce near net size and complex shaped parts without the need for mechanical processing, improving material utilization and reducing production costs.
However, process parameters such as laser power and scanning speed can affect the flow and geometry of the melt pool, resulting in experimental dimensions being larger than the design dimensions; In addition, the sintering effect of the molten pool on adjacent powders results in a large amount of partially melted powder on the surface of the sample, increasing the sample size and surface roughness, which once again affects manufacturing accuracy. Currently, some scholars are studying the differences between SLM design and experimental dimensions under fixed process parameters, and proposing compensation strategies to ensure high-precision SLM samples are obtained. Accurate mathematical models can predict the experimental dimensions of SLM samples under different parameters, which is an important guide for optimizing process parameters and sample size design.
However, the working principle of a laser oscillator is to direct the laser beam onto two mirrors and use a remote pendulum controlled by a computer with two or more micro motors at different angles and positions to scan the mirrors along the X and Y axes to deflect the laser beam. The laser scanning speed of SLM equipment can reach 5000 mm/s or even higher. With a significant increase in scanning speed, the delay in mirror movement response may lead to significant laser drift, thereby affecting the dimensional accuracy and quality of SLM samples. The scanning speed in previous studies did not exceed 2500mm/s, which did not fully demonstrate the laser drift caused by the delayed response of the vibrating mirror motion.
Therefore, an orthogonal experiment was designed to investigate the effects of laser power, scanning speed, and scanning spacing on the density, wall thickness increase (IWT), and surface roughness of thin-walled TC11 alloy parts. Discovered laser beam drift caused by high scanning speed. Three quadratic regression mathematical models were established to predict the relative density, IWT, and surface roughness of SLM TC11 alloy specimens under different laser powers, scanning speeds, and hatch spaces. Validate the mathematical model by manufacturing samples of different thicknesses using different laser beam incidence angles. Finally, a model was established to explain the mechanism of the influence of laser beam drift on the size and roughness of SLM specimens.
Materials and methods
This study selected laser power, scanning speed, and scanning spacing as variables, and designed a three factor four level orthogonal experiment, as shown in Table 1.
Fig. 2. Schematic diagram of scanning strategy.
Fig. 3. Effect of SLM process parameters on relative density; (a) laser power, (b) scanning speed, (c) hatch space.
Fig. 4. Comparison of experimental and predicted relative density values for orthogonal specimens.
Fig. 5. Effect of SLM process parameters on IWT; (a) laser power, (b) scanning speed, (c) hatch space.
Fig. 6. Comparison of experimental and predicted IWT values for orthogonal specimens.
Fig. 7. Effect of SLM process parameters on roughness; (a) laser power, (b) scanning speed, (c) hatch space.
Fig. 8. Comparison of experimental and predicted roughness values for orthogonal specimens.
Fig. 9. Diagram of different laser incidence angles.
Fig. 10. Drift of the laser scanning path; (a) 5000 mm/s, (b) 4000 mm/s, (c) 3000 mm/s, (d) 2000 mm/s, (e) 1800 mm/s, (f) 1600 mm/s, (g) 1400 mm/s, (h) 1200 mm/s, (i) 1000 mm/s, (j) 800 mm/s, (k) 600 mm/s, (l) 400 mm/s, (m) 200 mm/s, (n) 100 mm/s.
Fig. 11. Drift value of the laser scanning path.
Summary
The drift of the laser beam causes a decrease in IWT and surface roughness with increasing scanning speed, followed by an increase. The incident angle of the laser beam and the design thickness of the specimen have no significant effect on the relative density, IWT, and roughness of SLM specimens. The differences between experimental values and predicted values were 2.6%, 0.08 mm, and 7.3 μ m, respectively, further demonstrating that the mathematical model can accurately predict.
At a laser power of 350 W and a scanning speed of 1000 mm/s, the cross-sectional area of the sample increased by 1.71 mm2 due to the width of the melt pool, which is greater than the 1.70 mm2 caused by laser beam drift. In this case, the width of the melt pool is the main factor affecting the size of the specimen. Therefore, in order to reduce the negative impact of laser beam drift on relative density, the scanning speed should be ≤ 1000 mm/s when the laser power is 350 W.
Source: Yangtze River Delta Laser Alliance
