- Email: [email protected]
Temperature Profile and Imaging Analysis of Laser Additive Manufacturing of Stainless Steel
Available online at www.sciencedirect.com
Physics Procedia 41 (2013) 828 – 835
Lasers in Manufacturing Conference 2013
Temperature profile and imaging analysis of laser additive manufacturing of stainless steel M. Islama , T. Purtonena*, H. Piilia, A. Salminena,b, O. Nyrhiläc a
Lappeenranta University of Technology, Research group of laser processing, Tuotantokatu 2, FII -53850 Lappeenranta, Finland b Machine Technology Center Turku Ltd,Lemminkäisenkatu 28, FII 20520 Turku, Finland c EOS Finland,Lemminkäisenkatu 36, 6 FII 20520 Turku, Finland
Abstract Powder bed fusion is a laser additive manufacturing (LAM) technology which is used to manufacture parts layer-wise from powdered metallic materials. The technology has advanced vastly in the recent years and current systems can be used to manufacture functional parts for e.g. aerospace industry. The performance and accuracy of the systems have improved also, but certain difficulties in the powder fusion process are reducing the final quality of the parts. One of these is commonly known as the balling phenomenon. The aim of this study was to define some of the process characteristics in powder bed fusion by performing comparative studies with two different test setups. This was done by comparing measured temperature profiles and on-line photography of the process. The material used during the research was EOS PH1 stainless steel. Both of the test systems were equipped with 200 W single mode fiber lasers. The main result of the research was that some of the process instabilities are resulting from the energy input during the process. © 2013 The Published byPublished Elsevier B.V. © Authors. 2013 The Authors. by Elsevier B.V. Selection and/or peer-review under responsibility the German Scientific Laser Society (WLTLaser e.V.) Society (WLT e.V.) Selection and/or peer-review under of responsibility of the German Scientific Keywords: laser additive manufacturing; powder bed fusion; laser; additive manufacturing; stainless steel; monitoring; balling phenomenon; rapid manufacturing; dmls; slm
1. Introduction Powder bed fusion (PBF) is one of the laser additive manufacturing (LAM) technologies which uses laser beam to solidify powdered metallic materials layer-wise to form functional 3D parts. The process itself is
*
Corresponding author. Tel.: +358405153349 E-mail address: [email protected]
1875-3892 © 2013 The Authors. Published by Elsevier B.V. Selection and/or peer-review under responsibility of the German Scientific Laser Society (WLT e.V.) doi:10.1016/j.phpro.2013.03.156
M. Islam et al. / Physics Procedia 41 (2013) 828 – 835
often called with different names, e.g. selective laser melting (SLM), selective laser sintering (SLS) and direct metal laser sintering (DMLS). Some of the process names are registered trademarks for various companies, and the process nature can be melting even though the name suggests otherwise. [1, 2] In sintering, the metal powder is heated to a temperature slightly below the melting point [3, 4]. The resulting structure with sintering process has high porosity and its mechanical properties are not enough for actual manufacturing. Another process variation is to melt material completely in order to improve the density of the final part. There is no melting if the energy input is not high enough to cause a physical change of the powder material [5, 6]. The energy input is determined by the laser power, scanning speed and the hatch spacing for the laser scan pattern. Hatch spacing has an impact in the heat distribution. If a shorter distance than necessary is used, it can create surface deterioration by generating excessive heat [7]. Balling phenomenon is a typical defect associated with selective laser melting and it is one of the main factors that reduce the quality of the manufacturing. In this phenomenon, the material is not molten to a flat layer, but instead creates relatively large spherically shaped particles. This phenomenon has been widely researched and there are several theories explaining its origins and occurrence [5, 7, 8, 9, 10, 11]. Most of these researches were done with a 0.25 mm layer thickness, which is considered as fairly thick compared to 0.02 mm used in DMLS process. The scanning hatch spacing was 0.1 mm which is typical for SLM. According to Kruth et al., the molten material is forming metal balls on top of the previously solidified layer due to insufficient wettability of the previous layer [10]. There are two kinds of balling phenomena observed during SLM operation. [5, 9] Using low laser power gives an increase to the balling characterized by highly coarsened balls possessing an interrupted dendrite structure in their surface layer. In the second kind balling phenomenon, a large amount of micrometer-scaled are formed on the laser sintered surface at a high scan speed. Kruth et al. have studied that by combining high laser power with high scan speeds, the balling can be reduced [10]. Due to this phenomenon, temperature control of SLM process is critical for obtaining optimum circumstances and to improve quality of final product. For this reason several different approaches to monitor the process have been researched. Most of these are utilizing different types of temperature measurements and photography. [12, 13, 14, 15, 16, 17] Nomenclature LAM
Laser Additive Manufacturing
SLM
Selective Laser Melting
PBF
Powder Bed Fusion
SLS
Selective Laser Sintering
DMLS Direct Metal Laser Sintering 2. Experimental The aim of this study was to characterize powder bed fusion process of stainless steel with two different test setups. First setup (Setup A) used a simple processing chamber with a mechanical powder spreading system and a 200 W IPG Fiber laser with Scanlab scanner optics. The second setup utilized an EOS research system similar to a commercially available EOS M270, also with 200 W IPG fiber laser. The process chamber had a nitrogen atmosphere with both of the setups. Both setups also included monitoring with a Thyssen Laser-Technik TCS pyrometer and a Baumer CMOS camera with Cavitar CAVILUX diode laser illumination. The diameter of measurement range of pyrometer was approximately 15 mm, which is considerably larger than the area of laser beam. Both measurement tools were covering the whole test shape area all the time. The scanner optics and laser were controlled in test setup A with SCAPS SAMLight software, and in test setup B with the EOS PSW software.
829
830
M. Islam et al. / Physics Procedia 41 (2013) 828 – 835
The material used for both of the tests was EOS PH1, which is a precipitation hardening stainless steel. The particle size for the powder was determined with a Beckman Coulter laser diffraction particle size analyzer. The mean particle size for the powder was approximately 42 μm. The chemical composition of the material can be found on table 1. Table 1. Chemical composition of EOS PH1 material. [18]
[Weight -%]
Fe
Cr
Ni
Cu
Mn
Si
Mo
Nb
C
Balance
14-15.5
3.5-5.5
2.5-4.5
Max 1
Max 1
Max 0.5
0.15-0.45
Max 0.07
With test setup A, the laser melting process is performed inside a chamber filled with nitrogen. The laser beam is focused onto the powder bed through a normal float glass window. The pyrometer, illumination and camera systems are also operating through this window. The error created by the window to the pyrometer measurements was compensated during the testing based on information gathered during preliminary test period. The construction of the test setups can be found on the figures below.
Fig. 1. Test setup A on the left and Test setup B on the right.
The test set included some preliminary tests that were done to verify the operation of the pyrometer and to receive some information about the correlation between the pyrometer measured temperatures and both the energy input and the result of the melting process. The tests were performed by melting one single 20 μm layer of the powdered material with two different predefined scanning paths or hatch patterns, one for each test setup. The fusion area was a 5x5 mm square with 100 μm hatch spacing. The hatch patterns are slightly different due to different scanning tactics in the used scanner control software. The PSW software scanned the square by creating the outer border first and then filling it with horizontal lines. The SAMLight was programmed to add a small vertical line between every horizontal line into the filling pattern. The hatch patterns are shown in the figure below.
831
M. Islam et al. / Physics Procedia 41 (2013) 828 – 835
Fig. 2. The hatch patterns used in Test setup A (on the left) and Test setup B (on the right).
One major difference between the test setups was the lack of heated platform in the test setup A. Due to this reason the melting tests were done on top of the powder bed. This can give some differences between the achieved temperatures with the system. And due to this fact the cooling rate of the molten material is different. 2.1. Calculation of energy input and cooling rates Due to the three dimensional nature of the process, volumetric energy input was chosen to describe the amount of energy input during the selective laser melting process. This is commonly applied with the powder bed fusion technologies. The energy inputs were calculated according to the following equation. =
(1) , where
E = Volumetric energy input in J/mm3 P = Laser Power in W v = Scan speed in mm/s h = Hatch spacing in mm d = Layer thickness in mm.
The cooling rates were calculated according to the following equation. This equation is used to give an approximate value of the average cooling rate between two different temperatures. =
1
0
1
0
(2) , where
k = Cooling rate in °C/s T0 = Initial temperature T1 = Final temperature t0 = Initial time t1 = Final time.
3. Results and Discussion The energy inputs for different setups were calculated according to equation 1. The temperature measurements showed the fluctuation of the temperature values according to the hatch shape. The amount of fluctuation is in relation to the scanning speed used during the testing. The tests concluded with test setup A had various different scanning speeds which can be seen as high fluctuation in the measured temperatures.
832
M. Islam et al. / Physics Procedia 41 (2013) 828 – 835
The measurements can be seen on the figure below. Different energy inputs are indicated by different colors in the graphs.
Fig. 3. Temperature measurements with different volumetric energy inputs [J/mm3] with Test setup A (on the left) and Test setup B (on the right).
A comparison of the pyrometer measurements with equal energy inputs using different test setups showed similar results. The figure below shows the measured temperature curves from both setups with 2500 J/mm3 energy input. It can be seen that the average temperature during the process behaves similarly. The slight drop in the beginning of temperature curve measured with setup A is created by the laser control software.
Fig. 4. Temperature measurements with 2500 J/mm3 energy inputs with Test setup A and Test setup B.
The different setups were also compared with the photography. Table 2 shows the imaging results with two different energy inputs and maximum temperatures during the process. The results show similar behaviour of the melt during the process. The amount of balling was also determined as equal.
833
M. Islam et al. / Physics Procedia 41 (2013) 828 – 835
Table 2. Temperature and imaging analysis of the two different setups with 2500 and 5000 J/mm3 energy inputs. Setup A Energy input, [J/mm3]
Active illumination image
Setup B Maximum temperature, [°C]
Active illumination image
Maximum temperature, [°C]
2500
1308
1348
5000
1224
1270
The figure below shows how the cooling rates alter between the two different setups. The cooling rate between temperatures of 1000 to 500 °C with a 2500 J/mm3 energy input with setups A and B were approximately 1887 °C/s and 205 °C/s, respectively. The cooling rates were calculated according to equation 2, which in this case gives an average value for the rate. This value can be used to compare different setups. It can be seen that heat conduction to powder material is much faster than to the preheated platform even though the heat conductivity of powder bulk material is much lower than solid material. The difference in cooling rates can have an effect to the final microstructure of the manufactured part with certain materials, and deformations of the part, but it does not seem to have any effect to the balling phenomena.
Fig. 5. Cooling rate of the components manufactured with different setups. Energy input of 2500 J/mm3.
By calculating the maximum and average temperatures with different energy inputs, it could be noted that the energy input has a distinct effect to the temperatures. When the energy input is initially increased, the process temperature is increased rapidly to over 1200 °C which can be reached with less than 500 J/mm3 energy input. When the energy input is increased to approximately 1000 J/mm3, the temperature reaches maximum. This can be concluded as such that the absorption of laser power at this point is highest. Further increase in energy input reduces the process temperatures, which is most likely due to increased reflection of the laser beam from the melt pool. Other possibility would be reflection from solid platform underneath but this was not the case since the photography would have revealed it. The image analysis from the process showed that with too low energy input, the material is not melting properly and when the energy input is increased, the process temperatures are decreasing. The temperature changes according to the energy input can be seen on the figure below.
834
M. Islam et al. / Physics Procedia 41 (2013) 828 – 835
Fig. 6. Calculated maximum and average temperatures from the pyrometer testing with both test setups.
The extent of the balling phenomenon during the process was evaluated by the photography. The results showed that the balling was reduced to minimum with energy input range between 1400-1700 J/mm3. When the energy input was insufficient, balling and porosity of the material was visible throughout the process. By increasing the energy input, the amount of balling was reduced and the porosity was no longer visible. With energy inputs beyond 1700 J/mm3 the balling became again clearly visible. Further increase in energy input resulted in increased balling. 4. Conclusions The two test setups showed similar behavior in the process temperatures according to the changes of energy input. The maximum process temperatures were attained with energy inputs of approximately 1000 J/mm3. The balling phenomenon was observed during these tests with certain energy input ranges. By increasing or decreasing the energy input from the optimum range, the amount of balling was visibly increased. It was shown that the optimum energy input area for these setups was from 1400 to 1700 J/mm3. The effect of scanning speed to the balling phenomena was not registered during these tests. Acknowledgements This research was completed with the help of two different projects, InnoBusiness-ArvoBusiness consortium and Innoliitos. The authors would like to thank TEKES and Päijät-Hämeen Liitto for funding of the projects. References [1]
Gibson, I., Rosen, D.W., Stucker, B. 2010. Additive Manufacturing Technologies - Rapid Prototyping to Direct Digital Manufacturing. Springer Science and Business Media, LLC. New York. ISBN: 978-1-4419-1119-3. [2] Shellabear, M., Nyrhilä, O. 2004. DMLS - Development history and state of the art. Proceedings of the 4th International Conference on Laser Assisted Net Shape Engineering. LANE 2004. Erlangen, p. 393-404. [3] Kruth, J.P., Levy, G., Klocke, F., Childs, T.H.C. 2007. Consolidation phenomena in laser and powder-bed based layered manufacturing, CIRP Annals - Manufacturing Technology, Vol. 56, Iss. 2, 2007, p. 730-759, ISSN 0007-8506, 10.1016/j.cirp.2007.10.004. (http://www.sciencedirect.com/science/article/pii/S0007850607001540) [4] Kruth, J.P., Mercelis, P., Van Vaerenbergh, J., Froyen, L., Rombouts, M. 2005. Binding mechanisms in selective laser sintering and selective laser melting. Rapid Prototyping Journal, Vol. 11 Iss. 1 p. 26 - 36
M. Islam et al. / Physics Procedia 41 (2013) 828 – 835
[5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18]
Yasa, E.; Deckers, J.; Kruth, J.P.: The investigation of the influence of laser re-melting on density, surface quality and microstructure of selective laser melting parts, Rapid Prototyping Journal, vol. 17 iss. 5, p. 312-327 (2011) O'Neill, W.; Sutcliffe, C.J.; Morgan, R.; Landsborough, A.; Hon, K.K.B.: Investigation on Multi-Layer Direct Metal Laser Sintering of 316L Stainless Steel Powder Beds. [online document] Available at http://www.sciencedirect.com/ , (2007),1-2 Gu, D., Shen, Y. 2008. Processing conditions and micro-structural features of porous 316L stainless steel components by DMLS, Applied Surface Science, Vol. 255, Iss. 5, Part 1, 30 December 2008, p. 1880-1887, ISSN 0169-4332. (http://www.sciencedirect.com/science/article/pii/S0169433208015456) Li, R., Liu, J., Shi, Y., Wang, L., Jiang, W. 2012. Balling behavior of stainless steel and nickel powder during selective laser melting process. The International Journal of Advanced Manufacturing Technology. Vol. 59, N 9-12. p. 1025-1035. Springer-Verlag. (http://dx.doi.org/10.1007/s00170-011-3566-1) Niu, H.J., Chang, I.T.H. 1999. Instability of scan tracks of selective laser sintering of high speed steel powder, Scripta Materialia, Vol. 41, Iss. 11, 5 November 1999, p. 1229-1234, ISSN 1359-6462. (http://www.sciencedirect.com/science/article/pii/S1359646299002766) Kruth, J.P., Froyen, L., Van Vaerenbergh, J., Mercelis, P., Rombouts, M., Lauwers, B. 2004. Selective laser melting of iron-based powder. Journal of Materials Processing Technology, Vol. 149, Iss. 1 3, 10 June 2004, p. 616-622, ISSN 0924-0136. (http://www.sciencedirect.com/science/article/pii/S0924013604002201). Gu, D., Shen, Y. 2009. Balling phenomena in direct laser sintering of stainless steel powder: Metallurgical mechanisms and control methods. Materials & Design, Vol. 30, Iss. 8, September 2009, p. 2903-2910, ISSN 0261-3069. (http://www.sciencedirect.com/science/article/pii/S0261306909000181) Craeghs, T., Clijsters, S., Yasa, E., Kruth, J.P. 2011.Online quality control of selective laser melting . Proceedings of the 20th Solid Freeform Fabrication (SFF) symposium, Austin (Texas), 8-10 august. Solid Freeform Fabrication (SFF) Symposium. Austin (Texas), 8-10 August 2011. Lehti, A., Taimisto, L., Piili, H., Salminen, A., Nyrhilä, O. 2011. Evaluation of different monitoring methods of laser assisted additive manufacturing of stainless steel. ECerS XII 12th Conference of the European Ceramic Society. Stockholm Islam, M.E., Lehti, A., Taimisto, L., Piili, H., Salminen, A., Nyrhilä, O. 2012. Evaluation of effect of heat input in laser assisted additive manufacturing of stainless steel. Proceedings of the 37th International MATADOR Conference. Springer. p. 361-368. ISBN 978-1-4471-4479-3 Craeghs, T., Clijsters, S., Yasa, E., Bechmann, F., Brumen, S., Kruth, J.P. 2011. Determination of geometrical factors in Layerwise Laser Melting using optical process monitoring. Optics and Lasers in Engineering. Vol. 49, Iss. 12, December 2011, p. 1440-1446, ISSN 0143-8166. (http://www.sciencedirect.com/science/article/pii/S0143816611001941) Chivel, Y., Smurov, I. 2010. On-line temperature monitoring in selective laser sintering/melting. Physics Procedia, Vol. 5, Part B, 2010, p. 515-521. ISSN 1875-3892. (http://www.sciencedirect.com/science/article/pii/S1875389210005055) Pavlov, M., Doubenskaia, M., Smurov, I. 2010. Pyrometric analysis of thermal processes in SLM technology. Physics Procedia, Vol. 5, Part B, 2010, p. 523-531, ISSN 1875-3892. (http://www.sciencedirect.com/science/article/pii/S1875389210005067) EOS GmbH - Electro Optical Systems. 2008. Material data sheet. EOS StainlessSteel PH1 for EOSINT M 270.
835
Physics Procedia 41 (2013) 828 – 835
Lasers in Manufacturing Conference 2013
Temperature profile and imaging analysis of laser additive manufacturing of stainless steel M. Islama , T. Purtonena*, H. Piilia, A. Salminena,b, O. Nyrhiläc a
Lappeenranta University of Technology, Research group of laser processing, Tuotantokatu 2, FII -53850 Lappeenranta, Finland b Machine Technology Center Turku Ltd,Lemminkäisenkatu 28, FII 20520 Turku, Finland c EOS Finland,Lemminkäisenkatu 36, 6 FII 20520 Turku, Finland
Abstract Powder bed fusion is a laser additive manufacturing (LAM) technology which is used to manufacture parts layer-wise from powdered metallic materials. The technology has advanced vastly in the recent years and current systems can be used to manufacture functional parts for e.g. aerospace industry. The performance and accuracy of the systems have improved also, but certain difficulties in the powder fusion process are reducing the final quality of the parts. One of these is commonly known as the balling phenomenon. The aim of this study was to define some of the process characteristics in powder bed fusion by performing comparative studies with two different test setups. This was done by comparing measured temperature profiles and on-line photography of the process. The material used during the research was EOS PH1 stainless steel. Both of the test systems were equipped with 200 W single mode fiber lasers. The main result of the research was that some of the process instabilities are resulting from the energy input during the process. © 2013 The Published byPublished Elsevier B.V. © Authors. 2013 The Authors. by Elsevier B.V. Selection and/or peer-review under responsibility the German Scientific Laser Society (WLTLaser e.V.) Society (WLT e.V.) Selection and/or peer-review under of responsibility of the German Scientific Keywords: laser additive manufacturing; powder bed fusion; laser; additive manufacturing; stainless steel; monitoring; balling phenomenon; rapid manufacturing; dmls; slm
1. Introduction Powder bed fusion (PBF) is one of the laser additive manufacturing (LAM) technologies which uses laser beam to solidify powdered metallic materials layer-wise to form functional 3D parts. The process itself is
*
Corresponding author. Tel.: +358405153349 E-mail address: [email protected]
1875-3892 © 2013 The Authors. Published by Elsevier B.V. Selection and/or peer-review under responsibility of the German Scientific Laser Society (WLT e.V.) doi:10.1016/j.phpro.2013.03.156
M. Islam et al. / Physics Procedia 41 (2013) 828 – 835
often called with different names, e.g. selective laser melting (SLM), selective laser sintering (SLS) and direct metal laser sintering (DMLS). Some of the process names are registered trademarks for various companies, and the process nature can be melting even though the name suggests otherwise. [1, 2] In sintering, the metal powder is heated to a temperature slightly below the melting point [3, 4]. The resulting structure with sintering process has high porosity and its mechanical properties are not enough for actual manufacturing. Another process variation is to melt material completely in order to improve the density of the final part. There is no melting if the energy input is not high enough to cause a physical change of the powder material [5, 6]. The energy input is determined by the laser power, scanning speed and the hatch spacing for the laser scan pattern. Hatch spacing has an impact in the heat distribution. If a shorter distance than necessary is used, it can create surface deterioration by generating excessive heat [7]. Balling phenomenon is a typical defect associated with selective laser melting and it is one of the main factors that reduce the quality of the manufacturing. In this phenomenon, the material is not molten to a flat layer, but instead creates relatively large spherically shaped particles. This phenomenon has been widely researched and there are several theories explaining its origins and occurrence [5, 7, 8, 9, 10, 11]. Most of these researches were done with a 0.25 mm layer thickness, which is considered as fairly thick compared to 0.02 mm used in DMLS process. The scanning hatch spacing was 0.1 mm which is typical for SLM. According to Kruth et al., the molten material is forming metal balls on top of the previously solidified layer due to insufficient wettability of the previous layer [10]. There are two kinds of balling phenomena observed during SLM operation. [5, 9] Using low laser power gives an increase to the balling characterized by highly coarsened balls possessing an interrupted dendrite structure in their surface layer. In the second kind balling phenomenon, a large amount of micrometer-scaled are formed on the laser sintered surface at a high scan speed. Kruth et al. have studied that by combining high laser power with high scan speeds, the balling can be reduced [10]. Due to this phenomenon, temperature control of SLM process is critical for obtaining optimum circumstances and to improve quality of final product. For this reason several different approaches to monitor the process have been researched. Most of these are utilizing different types of temperature measurements and photography. [12, 13, 14, 15, 16, 17] Nomenclature LAM
Laser Additive Manufacturing
SLM
Selective Laser Melting
PBF
Powder Bed Fusion
SLS
Selective Laser Sintering
DMLS Direct Metal Laser Sintering 2. Experimental The aim of this study was to characterize powder bed fusion process of stainless steel with two different test setups. First setup (Setup A) used a simple processing chamber with a mechanical powder spreading system and a 200 W IPG Fiber laser with Scanlab scanner optics. The second setup utilized an EOS research system similar to a commercially available EOS M270, also with 200 W IPG fiber laser. The process chamber had a nitrogen atmosphere with both of the setups. Both setups also included monitoring with a Thyssen Laser-Technik TCS pyrometer and a Baumer CMOS camera with Cavitar CAVILUX diode laser illumination. The diameter of measurement range of pyrometer was approximately 15 mm, which is considerably larger than the area of laser beam. Both measurement tools were covering the whole test shape area all the time. The scanner optics and laser were controlled in test setup A with SCAPS SAMLight software, and in test setup B with the EOS PSW software.
829
830
M. Islam et al. / Physics Procedia 41 (2013) 828 – 835
The material used for both of the tests was EOS PH1, which is a precipitation hardening stainless steel. The particle size for the powder was determined with a Beckman Coulter laser diffraction particle size analyzer. The mean particle size for the powder was approximately 42 μm. The chemical composition of the material can be found on table 1. Table 1. Chemical composition of EOS PH1 material. [18]
[Weight -%]
Fe
Cr
Ni
Cu
Mn
Si
Mo
Nb
C
Balance
14-15.5
3.5-5.5
2.5-4.5
Max 1
Max 1
Max 0.5
0.15-0.45
Max 0.07
With test setup A, the laser melting process is performed inside a chamber filled with nitrogen. The laser beam is focused onto the powder bed through a normal float glass window. The pyrometer, illumination and camera systems are also operating through this window. The error created by the window to the pyrometer measurements was compensated during the testing based on information gathered during preliminary test period. The construction of the test setups can be found on the figures below.
Fig. 1. Test setup A on the left and Test setup B on the right.
The test set included some preliminary tests that were done to verify the operation of the pyrometer and to receive some information about the correlation between the pyrometer measured temperatures and both the energy input and the result of the melting process. The tests were performed by melting one single 20 μm layer of the powdered material with two different predefined scanning paths or hatch patterns, one for each test setup. The fusion area was a 5x5 mm square with 100 μm hatch spacing. The hatch patterns are slightly different due to different scanning tactics in the used scanner control software. The PSW software scanned the square by creating the outer border first and then filling it with horizontal lines. The SAMLight was programmed to add a small vertical line between every horizontal line into the filling pattern. The hatch patterns are shown in the figure below.
831
M. Islam et al. / Physics Procedia 41 (2013) 828 – 835
Fig. 2. The hatch patterns used in Test setup A (on the left) and Test setup B (on the right).
One major difference between the test setups was the lack of heated platform in the test setup A. Due to this reason the melting tests were done on top of the powder bed. This can give some differences between the achieved temperatures with the system. And due to this fact the cooling rate of the molten material is different. 2.1. Calculation of energy input and cooling rates Due to the three dimensional nature of the process, volumetric energy input was chosen to describe the amount of energy input during the selective laser melting process. This is commonly applied with the powder bed fusion technologies. The energy inputs were calculated according to the following equation. =
(1) , where
E = Volumetric energy input in J/mm3 P = Laser Power in W v = Scan speed in mm/s h = Hatch spacing in mm d = Layer thickness in mm.
The cooling rates were calculated according to the following equation. This equation is used to give an approximate value of the average cooling rate between two different temperatures. =
1
0
1
0
(2) , where
k = Cooling rate in °C/s T0 = Initial temperature T1 = Final temperature t0 = Initial time t1 = Final time.
3. Results and Discussion The energy inputs for different setups were calculated according to equation 1. The temperature measurements showed the fluctuation of the temperature values according to the hatch shape. The amount of fluctuation is in relation to the scanning speed used during the testing. The tests concluded with test setup A had various different scanning speeds which can be seen as high fluctuation in the measured temperatures.
832
M. Islam et al. / Physics Procedia 41 (2013) 828 – 835
The measurements can be seen on the figure below. Different energy inputs are indicated by different colors in the graphs.
Fig. 3. Temperature measurements with different volumetric energy inputs [J/mm3] with Test setup A (on the left) and Test setup B (on the right).
A comparison of the pyrometer measurements with equal energy inputs using different test setups showed similar results. The figure below shows the measured temperature curves from both setups with 2500 J/mm3 energy input. It can be seen that the average temperature during the process behaves similarly. The slight drop in the beginning of temperature curve measured with setup A is created by the laser control software.
Fig. 4. Temperature measurements with 2500 J/mm3 energy inputs with Test setup A and Test setup B.
The different setups were also compared with the photography. Table 2 shows the imaging results with two different energy inputs and maximum temperatures during the process. The results show similar behaviour of the melt during the process. The amount of balling was also determined as equal.
833
M. Islam et al. / Physics Procedia 41 (2013) 828 – 835
Table 2. Temperature and imaging analysis of the two different setups with 2500 and 5000 J/mm3 energy inputs. Setup A Energy input, [J/mm3]
Active illumination image
Setup B Maximum temperature, [°C]
Active illumination image
Maximum temperature, [°C]
2500
1308
1348
5000
1224
1270
The figure below shows how the cooling rates alter between the two different setups. The cooling rate between temperatures of 1000 to 500 °C with a 2500 J/mm3 energy input with setups A and B were approximately 1887 °C/s and 205 °C/s, respectively. The cooling rates were calculated according to equation 2, which in this case gives an average value for the rate. This value can be used to compare different setups. It can be seen that heat conduction to powder material is much faster than to the preheated platform even though the heat conductivity of powder bulk material is much lower than solid material. The difference in cooling rates can have an effect to the final microstructure of the manufactured part with certain materials, and deformations of the part, but it does not seem to have any effect to the balling phenomena.
Fig. 5. Cooling rate of the components manufactured with different setups. Energy input of 2500 J/mm3.
By calculating the maximum and average temperatures with different energy inputs, it could be noted that the energy input has a distinct effect to the temperatures. When the energy input is initially increased, the process temperature is increased rapidly to over 1200 °C which can be reached with less than 500 J/mm3 energy input. When the energy input is increased to approximately 1000 J/mm3, the temperature reaches maximum. This can be concluded as such that the absorption of laser power at this point is highest. Further increase in energy input reduces the process temperatures, which is most likely due to increased reflection of the laser beam from the melt pool. Other possibility would be reflection from solid platform underneath but this was not the case since the photography would have revealed it. The image analysis from the process showed that with too low energy input, the material is not melting properly and when the energy input is increased, the process temperatures are decreasing. The temperature changes according to the energy input can be seen on the figure below.
834
M. Islam et al. / Physics Procedia 41 (2013) 828 – 835
Fig. 6. Calculated maximum and average temperatures from the pyrometer testing with both test setups.
The extent of the balling phenomenon during the process was evaluated by the photography. The results showed that the balling was reduced to minimum with energy input range between 1400-1700 J/mm3. When the energy input was insufficient, balling and porosity of the material was visible throughout the process. By increasing the energy input, the amount of balling was reduced and the porosity was no longer visible. With energy inputs beyond 1700 J/mm3 the balling became again clearly visible. Further increase in energy input resulted in increased balling. 4. Conclusions The two test setups showed similar behavior in the process temperatures according to the changes of energy input. The maximum process temperatures were attained with energy inputs of approximately 1000 J/mm3. The balling phenomenon was observed during these tests with certain energy input ranges. By increasing or decreasing the energy input from the optimum range, the amount of balling was visibly increased. It was shown that the optimum energy input area for these setups was from 1400 to 1700 J/mm3. The effect of scanning speed to the balling phenomena was not registered during these tests. Acknowledgements This research was completed with the help of two different projects, InnoBusiness-ArvoBusiness consortium and Innoliitos. The authors would like to thank TEKES and Päijät-Hämeen Liitto for funding of the projects. References [1]
Gibson, I., Rosen, D.W., Stucker, B. 2010. Additive Manufacturing Technologies - Rapid Prototyping to Direct Digital Manufacturing. Springer Science and Business Media, LLC. New York. ISBN: 978-1-4419-1119-3. [2] Shellabear, M., Nyrhilä, O. 2004. DMLS - Development history and state of the art. Proceedings of the 4th International Conference on Laser Assisted Net Shape Engineering. LANE 2004. Erlangen, p. 393-404. [3] Kruth, J.P., Levy, G., Klocke, F., Childs, T.H.C. 2007. Consolidation phenomena in laser and powder-bed based layered manufacturing, CIRP Annals - Manufacturing Technology, Vol. 56, Iss. 2, 2007, p. 730-759, ISSN 0007-8506, 10.1016/j.cirp.2007.10.004. (http://www.sciencedirect.com/science/article/pii/S0007850607001540) [4] Kruth, J.P., Mercelis, P., Van Vaerenbergh, J., Froyen, L., Rombouts, M. 2005. Binding mechanisms in selective laser sintering and selective laser melting. Rapid Prototyping Journal, Vol. 11 Iss. 1 p. 26 - 36
M. Islam et al. / Physics Procedia 41 (2013) 828 – 835
[5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18]
Yasa, E.; Deckers, J.; Kruth, J.P.: The investigation of the influence of laser re-melting on density, surface quality and microstructure of selective laser melting parts, Rapid Prototyping Journal, vol. 17 iss. 5, p. 312-327 (2011) O'Neill, W.; Sutcliffe, C.J.; Morgan, R.; Landsborough, A.; Hon, K.K.B.: Investigation on Multi-Layer Direct Metal Laser Sintering of 316L Stainless Steel Powder Beds. [online document] Available at http://www.sciencedirect.com/ , (2007),1-2 Gu, D., Shen, Y. 2008. Processing conditions and micro-structural features of porous 316L stainless steel components by DMLS, Applied Surface Science, Vol. 255, Iss. 5, Part 1, 30 December 2008, p. 1880-1887, ISSN 0169-4332. (http://www.sciencedirect.com/science/article/pii/S0169433208015456) Li, R., Liu, J., Shi, Y., Wang, L., Jiang, W. 2012. Balling behavior of stainless steel and nickel powder during selective laser melting process. The International Journal of Advanced Manufacturing Technology. Vol. 59, N 9-12. p. 1025-1035. Springer-Verlag. (http://dx.doi.org/10.1007/s00170-011-3566-1) Niu, H.J., Chang, I.T.H. 1999. Instability of scan tracks of selective laser sintering of high speed steel powder, Scripta Materialia, Vol. 41, Iss. 11, 5 November 1999, p. 1229-1234, ISSN 1359-6462. (http://www.sciencedirect.com/science/article/pii/S1359646299002766) Kruth, J.P., Froyen, L., Van Vaerenbergh, J., Mercelis, P., Rombouts, M., Lauwers, B. 2004. Selective laser melting of iron-based powder. Journal of Materials Processing Technology, Vol. 149, Iss. 1 3, 10 June 2004, p. 616-622, ISSN 0924-0136. (http://www.sciencedirect.com/science/article/pii/S0924013604002201). Gu, D., Shen, Y. 2009. Balling phenomena in direct laser sintering of stainless steel powder: Metallurgical mechanisms and control methods. Materials & Design, Vol. 30, Iss. 8, September 2009, p. 2903-2910, ISSN 0261-3069. (http://www.sciencedirect.com/science/article/pii/S0261306909000181) Craeghs, T., Clijsters, S., Yasa, E., Kruth, J.P. 2011.Online quality control of selective laser melting . Proceedings of the 20th Solid Freeform Fabrication (SFF) symposium, Austin (Texas), 8-10 august. Solid Freeform Fabrication (SFF) Symposium. Austin (Texas), 8-10 August 2011. Lehti, A., Taimisto, L., Piili, H., Salminen, A., Nyrhilä, O. 2011. Evaluation of different monitoring methods of laser assisted additive manufacturing of stainless steel. ECerS XII 12th Conference of the European Ceramic Society. Stockholm Islam, M.E., Lehti, A., Taimisto, L., Piili, H., Salminen, A., Nyrhilä, O. 2012. Evaluation of effect of heat input in laser assisted additive manufacturing of stainless steel. Proceedings of the 37th International MATADOR Conference. Springer. p. 361-368. ISBN 978-1-4471-4479-3 Craeghs, T., Clijsters, S., Yasa, E., Bechmann, F., Brumen, S., Kruth, J.P. 2011. Determination of geometrical factors in Layerwise Laser Melting using optical process monitoring. Optics and Lasers in Engineering. Vol. 49, Iss. 12, December 2011, p. 1440-1446, ISSN 0143-8166. (http://www.sciencedirect.com/science/article/pii/S0143816611001941) Chivel, Y., Smurov, I. 2010. On-line temperature monitoring in selective laser sintering/melting. Physics Procedia, Vol. 5, Part B, 2010, p. 515-521. ISSN 1875-3892. (http://www.sciencedirect.com/science/article/pii/S1875389210005055) Pavlov, M., Doubenskaia, M., Smurov, I. 2010. Pyrometric analysis of thermal processes in SLM technology. Physics Procedia, Vol. 5, Part B, 2010, p. 523-531, ISSN 1875-3892. (http://www.sciencedirect.com/science/article/pii/S1875389210005067) EOS GmbH - Electro Optical Systems. 2008. Material data sheet. EOS StainlessSteel PH1 for EOSINT M 270.
835