对不锈工具钢粉末选择性激光溶解的进一步研究外文文献翻译、中英文翻译
对不锈工具钢粉末选择性激光溶解的进一步研究外文文献翻译、中英文翻译,对于,不锈,工具钢,粉末,选择性,激光,溶解,进一步,研究,钻研,外文,文献,翻译,中英文
International Journal of Machine Tools Selective laser sintering; Absorptance; Simulation materials 2. Laser sintering of metallic materials shows post processing other than surface finishing. In the single track work 3, where M2 and H13 tool steel and 314S-HC stainless steel powders were examined, different track meters 3,4. Since at constant laser power and (for layers) was compared to the incident energy. An explanation was given in terms of absorptivity of the laser energy into the bed increasing with speed. But the observation for single ARTICLE IN PRESS track was not much emphasized as it was thought that the conditions, in which the increase occurred, being associated with the formation of elliptical section (non-flat) tracks, 0890-6955/$-see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.ijmachtools.2006.09.013 C3 Corresponding author. Tel.: +441133432165; fax: +441133432150. E-mail address: t.h.c.childsleeds.ac.uk (T.H.C. Childs). great promise, but significant further research and under- standing are still required for obtaining high-quality parts. This paper is built on previous studies on selective laser melting (SLM) of single tracks 3 and single layers 4 of ferrous alloy powders in the surface of a deep powder bed by a scanning laser beam. SLM is an emerged name for the direct route of SLS when the complete melting of powder occurs rather than the sintering or partial melting 5. The aim is to create a strong part that is usable without further scan spacing, delivered energy density to the bed reduces with increasing speed, the initial expectation was that for any level of laser power, melted mass would reduce as scan speed increased. However, the reverse was observed in both single-track and single-layer works. In fact, there existed scan speed ranges in which the mass of melted track or layer increased (or fluctuated in the case of layer) with scan speed, at constant laser power and scan spacing. To study this, the amount of heat required to melt the observed mass 1. Introduction One current trend in production is the shortening of lead times for product development. New processes, especially those in the field of layer manufacturing, support this trend. In addition, they open new possibilities for manufacturing 1. One group of such processes, selective laser sintering or SLS, has become popular for rapidly manufacturing freeform parts using a wide range of forms were identified, depending on the laser power and scan speed that was used. Explanations for the transitions between these various forms of track were given in terms of the melt pool dimensions, temperatures and temperature gradients that existed. In the single-layer work 4 through studying the topography of melted layers from H13 tool steel, the onset of porosity was addressed. Subsequently the melted masses of single tracks and single layers were investigated at various process para- Further studies in selective tool steel M. Badrossamay, School of Mechanical Engineerin Available online Abstract Previous reported work on the selective laser melting of a H13 tool in which the layer mass increases/fluctuates with increasing speed. stainless steel powders to identify if they reveal similar behaviours. selected laser powers. Furthermore, the masses of the layers have thermal model. It has been found that at a constant laser power, expected from a constant assumed absorptivity into a powder bed. r 2006 Elsevier Ltd. All rights reserved. anufacture 47 (2007) 779784 melting of stainless and T.H.C. Childs C3 University of Leeds, Leeds, UK 2006 powder bed surface has shown that there is a scan speed range paper expands the investigation towards M2 tool steel and 316L ranges of scan spacings and scan speeds have been examined, at compared with those predicted from an existing finite element variation of mass with scan speed is much less than might be ARTICLE IN PRESS of were not practically useful for processing. However, the simulated results of single-layer melting have shown that the variability of layer mass for the speed range is not predictable with a constant absorptivity a value. Indeed, understanding the causes of a variation becomes of central importance to fundamental studies of single-layer forma- tion. This paper expands the investigation towards M2 tool steel and 316L and 314S-HC stainless steel powders to identify these material behaviours in single-layer melting, in conditions in which dense layers have been formed, that might form the foundation for the subsequent layers. The questions that it addresses are the variation of a in the SLM process and effects of scan spacing and thermal history on the melted mass. In an alternative approach for explanation of mass variations, an idea based on the process efficiency is developed later. 2. Modelling An existing finite element thermal model for laser sintering of polymers 6 has been developed to predict the mass of melted metallic material by the scanning laser beam. It is a transient heat conduction three-dimensional model in which the temperature rise caused by the travelling laser beam is calculated 7. The model takes into account the latent heat and the influence of both porosity and temperature on the thermal properties of the powder bed. In the temperature calculation, thermal conductivity k of the powder bed is assumed to vary with porosity as observed experimentally in 8: k k s 1C0 C15 1a k C15 0:78 , (1) where k s is the solid conductivity and a k is chosen to fit the measured initial bed conductivity to its measured initial porosity. k s and heat capacity C p are assumed to vary linearly with temperature: k s k 0 k 1 T; C p C 0 C 1 T, (2) where k 0 , k 1 , C 0 and C 1 are material-dependent coefficients. The influence on temperature of phase change with latent heat L between the solidus and liquidus temperatures T S to T L is treated by the temperature recovery method 9. Powder bed density is then assumed to change from its initial value to that of solid material as temperature increases from T S to T L . In the previous work 3, concerned with melting the first track in a powder bed, the temperature of the powder bed at the start of processing was assumed to be uniform, at a pre-determined ambient value. In modelling multi-track (i.e. layer) melting, the temperature existing in the bed at the start of processing a track is the temperature distribution existing from processing all the previous tracks. Since the current model calculates temperatures in a limited volume surrounding the laser beam, it takes into M. Badrossamay, T.H.C. Childs / International Journal780 account this issue (temperature history) approximately. The model presents lower estimates and approximate upper estimates of melted masses for the layer. The lower estimate assumes the bed to be cooled to ambient temperature before each new track is started. The approximate upper estimates are based on calculating melted masses of raster-scanned tracks each 3mm (five laser beam diameters) long, allowed to cool down to the point that all molten material has resolidified before the next track is started. The initial temperature distribution for melting the next track is taken to be that existing at the instant that the previous track has just resolidified. This is felt to give an overestimate to the initial temperature distribution around the region of the next track. In this sense, the calculated melted mass of the next track is likely to be approximately an upper estimate. Thermal conductivity k and absorptivity a of a powder bed are inputs to the thermal model. A calorimetric approach based on theory in 10 was used in 1 to determine these values. It was shown that these obtained data validated the simulation in a low scan speed region 11. An alternative approach to predict the absorptivity, based on heat balance, is used here. Assuming that no evaporation occurs on the surface of the powder bed the overall energy balance equation is expressed as Q abs Q melt Q loss , (3) where Q abs and Q melt are the absorbed energy and required energy to melt a layer mass m per unit area, respectively, and Q loss is the rest of energy which disappears through the powder bed. The heat absorbed per unit area scanned at speed U by a laser of power P is aP=Us, where s is scan spacing. Q melt is defined as mC P ave T m C0T 0 L, where C P ave is the average specific heat from ambient bed temperature T 0 to the metals melting temperature T m and L is the latent heat of fusion. Energy is lost from the melting zone by conduction, convection and radiation. However, the major source of heat loss through the powder bed is the conduction mode and contributions of convec- tion and radiation are so small as to be neglected without any loss of accuracy. By the assumption of negligible heat losses through the powder bed in the heat balance Eq. (3), the minimum estimate of absorptivity a min of laser energy into the powder bed can be expressed as a min mUs=P C0C1 C P ave T m C0T 0 L C2C3 . (4) Although, Eq. (4) gives only a minimum estimate of a in a heat balance sense, it expresses dependency of absorptivity on the process parameters. This issue is discussed later. Consideration of a values and their reliability is of major importance in this paper. 3. Experimentation Single layers have been produced, by a scanning CO 2 laser beam, in the surface of beds made from gas-atomized powders of composition and size fraction listed in Table 1. Machine Tools (a)(c) cases are explained in text. 0 100 125 0.0 0.2 0.4 0.6 0.8 s= 0.06 mm s= 0.18 mm s= 0.30 mm s= 0.48 mm s= 0.60 mm Scan speed (mm/s) 25 50 75 150 Absorptivity min . Machine Tools & Manufacture 47 (2007) 779784 783 the larger the scan spacing, the higher absorptance. As mentioned earlier, this would be because a higher propor- tion of unprocessed powder is being scanned, which has a higher absorptance than the previously processed material. As evidence, a study based on the temperature measure- ments in a stainless steel powder bed at a power that caused bed melting, showed that absoptivity reduces from 0.25 to 0.1 as scan spacing reduced from 75% to 20% of beam diameter 15. Despite the proposition of increasing absorptivity with scan speed, the reverse trend might be expected from a physical point of view. For a given laser power and beam diameter, the lasermaterial intraction time decreases as the laser beam scanning speed increaces. This reduces the deposition of laser energy on the surface and concequently the temperature of the bed surface decreases as the scanning speed increases. Therefore, the absorptivity of the powder bed is expected to be lower at high scanning speeds for the same laser parameters 16. Therefore, it would seem quite logical to consider alternative explana- tions for this. In this paper, an idea based on the process efficiency is presented. The presented model by Eq. (4) ignores the heat loss. This is an acceptable assumption as long as heat losses remain approximately a constant fraction of the absorbed energy. However, some reported works on similar issues (such as welding or laser cutting) suggest that heat losses are influenced by the process parameters. 110 and 140W and s 0.3mm, (b) for 316L at 110W and various scan technologies, International Journal of Machine Tools and Manufac- ture 38 (1997) 12571287. ARTICLE IN PRESS M. Badrossamay, T.H.C. Childs / International Journal of Machine Tools & Manufacture 47 (2007) 779784784 The Rosenthal model for moving line sources is a well- established solution for simulation of either single pass welding in thin materials or a fully penetrating keyhole weld, as well as laser beam cutting process. The model assumes that the energy is absorbed uniformly along a line in the depth direction. As the penetration of a laser beam in the powder bed is much longer than in bulk materials, this model could be applied for SLM as well. Swifthook and Gick 17 presented an analytic solution to the model depending on the scanning speed. From their solution it can be derived that at high speeds (typically 450mm/s) around 52% of the delivered energy is lost through conduction 18, while, for low speeds (typically o5mm/s) the loss could be around 95%. On the other hand, Schulz et al. 19 gave a correlated analytical solution of conduction loss taking into account the cylindrical heat source in laser cutting process as P con 2rC P DT urt P e =2 C0C1 C00:7 where P con is the conduction loss, r the material density, C p the heat capacity, DT the relative elevated temperature, u the cutting speed, r the radius of beam, t the material thickness, and P e is the Peclet number, which equals ru/k (k is the thermal diffusivity). It can be seen that with increasing Peclet number, cutting speed and heat loss decrease. Finally, as discussed so far, it seems the neglecting heat loss assumption cannot be applied in all ranges of process parameters. In fact, in low speed processing heat losses play an important role in the process and can not be neglected. The layer mass variations might be addressed by this route. However, more work is needed to confirm this issue in a quantitative way. 5. Conclusion Experimental studies on melting single layers of M2 high speed steel and 316L and 314S-HC stainless steels in the surface of powder beds, by a raster-scanning laser beam have confirmed that variations of layer mass with scan speed are not consistent with what is expected from the delivered energy to the powder bed. Two explanations are suggested. The first one, which was supported by the simulation as well, suggests that absorptivity may increase with increasing scan speed. In the second one the heat loss decreasing with scan speed is proposed. In addition, effects of scan spacing on the layer mass and absorptivity was studied. Experimental and simulation results also showed that the thermal history of processing is influential in determining the amount of melt under a laser beam irradiation. References 1 G.N. Levy, R. Schindel, J.P. Kruth, Rapid manufacturing and rapid tooling with layer manufacturing (LM) technologies, state of the art and future perspectives, Annals of the CIRP 52/2 (2003). 525540. 3 T.H.C. Childs, C. Hauser, M. Badrossamay, Selective laser sintering (melting) of stainless and tool steel powders: experiments and modelling, Proceedings of the Institution. of Mechanical Engineers, Part B: Journal of Engineering Manufacture 219B (2005) 339358. 4 T.H.C. Childs, C. Hauser, Raster scan selective laser melting of the surface layer of a tool steel powder bed, Proceedingsof the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufac- ture 219B (2005) 379384. 5 F. Abe, K. Osakada, M. Shiomi, K. Uematsu, M. Matsumoto, The manufacturing of hard tools from metallic powders by selective laser melting, Journal of Materials Processing Technology 111 (2001) 210213. 6 T.H.C. Childs, A.E. Tontowi, Selective laser sintering of a crystalline and a glass-filled crystalline polymer: experiments and simulations, Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture 215B (2001) 14811495. 7 C.M. Taylor, T.H.C. Childs, Morphology of direct SLS-processed stainless steel layers, in: proceedings of the solid freeform fabrication symposium 2002, University of Texas, Austin, Texas, USA, pp. 530537. 8 M. Shiomi, A. Yoshidome, F. Abe, K. Osakada, Finite element analysis of melting and solidifying processes in laser rapid prototyp- ing of metallic powders, International Journal of Machine Tools and Manufacture 39 (1999) 237252. 9 T.C. Tszeng, Y.T. Im, S. Kobayashi, Thermal analysis of solidifica- tion by the temperature recovery method, International Journal of Machine Tools and Manufacture 29 (1989) 107120. 10 H.S. Carslaw, J.C. Jaeger, Conduction of Heat in Solids, second ed., Clarendon Press, Oxford, 1959. 11 T.H.C. Childs, C. Hauser, M. Badrossamay, Mapping and modeling single scan track formation in direct metal selective laser melting, Annals of the CIRP, STC E, 53/1 (2004) 191194. 12 C. Hauser, T.H.C. Childs, C.M. Taylor, M. Badrossamay, S. Akhtar, C.S. Wright, M. Youseffi, J. Xie, P. Fox, W. ONeill, Direct selective laser sintering of tool steel powders to high density: part A- affective of laser beam width and scan strategy, in: proceedings of the solid freeform fabrication symposium 2003, University of Texas, Austin, Texas, USA, pp. 644655. 13 M. Dewidar, K.W. Dalgarno, C.S. Wright, Processing conditions and mechanical properties of high-speed steel parts fabricated using direct selective laser sintering, Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture 217B (2003) 16511663. 14 A.M. Prokhorov, V.I. Konov, I. Ursu, I.N. Mihailescu: Laser Heating of Metals, Adam Hilger, 1990, ISBN: 0-7503-0040-X. 15 C.M. Taylor, Direct laser sintering of stainless steel: thermal experiments and numerical modelling, Ph.D. Thesis, University of Leeds, 2004. 16 U. Tanriver, J. Longobardi, W.P. Latham, A. Kar, Effects of absorptivity, shielding gas speed, and contact media on sheet metal laser welding, Science and Technology of Welding and Joining 5 (5) (2000) 310316. 17 D.T. Swiethook, E.E.F. Gick, Welding Research Supplement (1973) 492498. 18 W.M. Steen, Laser Material Processing, third ed., Springer, London, ISBN 1-85233-698-6, 2003. 19 W. Schulz, D. Becker, J. Franke, R. Kemmerling, G. Herziger, Heat conduction losses in laser cutting of metals, Journal of Physics D: Applied Physics 26 (1993) 13571363. 2 D.T. Pham, R.S. Gault, A comparison of rapid prototyping
收藏