Available online at www.sciencedirect.com
ScienceDirect
Available online atonline www.sciencedirect.com Available at www.sciencedirect.com
ScienceDirect ScienceDirect
Procedia CIRP 00 (2018) 000–000
Procedia CIRP 00 (2017) 000–000 Procedia CIRP 74 (2018) 719–723
www.elsevier.com/locate/procedia
www.elsevier.com/locate/procedia
10th CIRP 10th CIRPConference Conference on on Photonic Photonic Technologies Technologies [LANE [LANE 2018] 2018]
Influence of processing gases laser cladding simulation analysis 28th CIRP Designin Conference, May 2018,based Nantes, on France and experimental tests A new methodology to analyze the functional and physical architecture of a, a Piotran Koruba *, Krzysztof Walla,product Jacek Reiner existing products for assembly oriented family identification a
Wroclaw University of Science and Technology, Faculty of Mechanical Engineering, Lukasiewicz str. 5, 50-371 Wroclaw, Poland
Paul Stief *, Jean-Yves Dantan, Alain Etienne, Ali Siadat
* Corresponding author. Tel.: +48-71-320-4635. E-mail address:
[email protected]
École Nationale Supérieure d’Arts et Métiers, Arts et Métiers ParisTech, LCFC EA 4495, 4 Rue Augustin Fresnel, Metz 57078, France *Abstract Corresponding author. Tel.: +33 3 87 37 54 30; E-mail address:
[email protected]
Laser cladding is a multiphysical process based on thermal phenomena, powder micrometallurgy and fluid dynamics. The impact of process gases is usually overlooked in experimental studies, since when performing laser cladding on flat surfaces, it can be considered negligible. Abstract However cylindrical surfaces are more sensitive. In this study the influence of process gases on laser deposited coatings was investigated. The preliminary dependencies were determined with two-phase flow model describing the interaction of process gases with powder. It gave initial Inparameters today’s business environment, trend towards product variety and customization is unbroken. characterized Due to this development, the need of range for performed the experimental trials.more Subsequently, obtained results were quantitatively in terms of geometrical agile and reconfigurable production systems emerged torecommendations. cope with various products and product families. To design and optimize production and material properties so as to formulate technological systems well as to choose theby optimal product matches, product analysis needed. Indeed, most of the known methods aim to © 2018 2018 as The Authors. Published Elsevier Ltd. This This is an an open open access articlemethods under the theare CC BY-NC-ND license © The Authors. Published by Elsevier Ltd. is access article under CC BY-NC-ND license analyze a product or one product family on the physical level. Different product families, however, may differ largely in terms of the number and (http://creativecommons.org/licenses/by-nc-nd/3.0/) (https://creativecommons.org/licenses/by-nc-nd/4.0/) nature of components. This fact impedes an efficient comparison GmbH. and choice of appropriate product family combinations for the production Peer-review under responsibility responsibility of the the Bayerisches Bayerisches Laserzentrum GmbH. Peer-review under of Laserzentrum system. A new methodology is proposed to analyze existing products in view of their functional and physical architecture. The aim is to cluster these products in cladding new assembly oriented product families for the optimization of existing assembly substrate; lines andStellite the creation of future reconfigurable Keywords: laser with powder; two-phase flow simulation; carrying and shielding gases; cylindrical 6; assembly systems. Based on Datum Flow Chain, the physical structure of the products is analyzed. Functional subassemblies are identified, and a functional analysis is performed. Moreover, a hybrid functional and physical architecture graph (HyFPAG) is the output which depicts the similarity between product families by providing design support to both, production system planners and product designers. An illustrative 1. Introduction bestudy constant during most of theofexperiments [7]. of example of a nail-clipper is used to explain the proposed methodology. Anassumed industrial to case on two product families steering columns thyssenkrupp Presta France is then carried out to give a first industrial evaluation of the proposed approach. © 2017 The cladding Authors. Published Elsevier B.V. Laser can be byconsidered as a multiphysical Nomenclature Peer-review under responsibility of the scientific committee of the 28th CIRP Design Conference 2018. process, whose input parameters are related to used laser
source, substrate and additional materials, utilized kinematic molar mass of argon MAr system and auxiliary operations such as preheating or postspecific gas constant of argon RAr ρAr process heat treatment [1, 2]. During the process there occur density of argon (kg/m3) many physical phenomena connected to heat transfer, fluid dynamic viscosity of argon (kg/(m s)) µAr 3 dynamics, laser beam absorption, powder micrometallurgy densityrange of powder Stellite 6 (kg/m ) ρ 1. Introduction ofSt6the product and characteristics manufactured and/or and weldability [3]. One of the most important parts of mean diameter of powder Stellite 6 (μm) in D St6 assembled in this system. In this context, the particles main challenge research related the technology of laser according average height of layer havg Due to the to fast development in cladding the domain of modelling and analysis is the nowcladded not only to (μm) cope with single d to [4] is development of a process model based on experiment depth of heat affected zone (μm) HAZ communication and an ongoing trend of digitization and products, a limited product range or existing product families, Rh also toirregularity of coating’s height (μm)products to define (empirical - statistical model)enterprises and simulation (physical based digitalization, manufacturing are facing important but be able to analyze and to compare FD drag force (N) model). The necessary and still being developed element of challenges in today’s market environments: a continuing new families. Itacceleration can be observed g product gravitational (m/s2)that classical existing the laser towards claddingreduction process model is thedevelopment powder stream model. tendency of product times and product families are regrouped u flow velocity (m/s) in function of clients or features. However,product during lifecycles. experimental studies onthere the process of laser up particle velocity (m/s) shortened In addition, is an increasing However, assembly oriented product families are hardly to find. 2 p On thepressure cladding many authors take into account the influence only demand of customization, being at the same time in a of global product (N/m family) level, products differ mainly in two x Cartesian coordinate (m) of components and (ii) the such processwith parameters as laser power, speed and competition competitors all over the cladding world. This trend, main the number i, j characteristics: directions in(i)Cartesian coordinates the amount of additional material supplied per unit time [5, 6]. which is inducing the development from macro to micro type of components (e.g. mechanical, electrical, electronical). It is believed rates oflot carrying and shielding gases markets, resultsthat in flow diminished sizes due to augmenting Classical methodologies considering mainly single products can be considered as secondary parameters, thus they are product varieties (high-volume to low-volume production) [1]. or solitary, already existing product families analyze the To cope with this augmenting variety as well as to be able to product structure on a physical level (components level) which 2212-8271 possible © 2018 Theoptimization Authors. Published by Elsevier is an opencauses access article under theregarding CC BY-NC-ND license identify potentials in Ltd. the This existing difficulties an efficient definition and (http://creativecommons.org/licenses/by-nc-nd/3.0/) production system, it is important to have a precise knowledge comparison of different product families. Addressing this Keywords: Assembly; Design method; Family identification
Peer-review under responsibility of the Bayerisches Laserzentrum GmbH.
2212-8271 © 2018 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) 2212-8271 © 2017 The Authors. Published by Elsevier B.V. Peer-review under responsibility of scientific the Bayerisches Laserzentrum GmbH. Peer-review under responsibility of the committee of the 28th CIRP Design Conference 2018. 10.1016/j.procir.2018.08.025
720 2
Piotr Koruba et al. / Procedia CIRP 74 (2018) 719–723 Author name / Procedia CIRP 00 (2018) 000–000
On the other hand, in case of process simulation, it is stated that the distribution of the powder stream profile during coaxial laser cladding is a key parameter for achieving appropriate quality and efficiency of the process [8]. According to conclusions of simulation studies from [9], the volumetric flows of process gases are significant input parameters of the laser cladding process, because not only do they affect the powder delivery speed but also the obtained powder stream profile. For example, in [10] a twodimensional powder flow simulation was presented, using an axisymmetric model. Also, in [11] authors conducted a twophase model simulation for gas as a continuous phase and powder as a dispersed one. The full 3D model of a coaxial cladding nozzle was prepared to determine the distribution of powder particles as a function of particle diameter. It should be emphasized that in known literature authors do not take into account the geometry of the substrate in the analysis. The lack of extended research in the field of experiment may be explained by the fact that in the case of flat substrate the influence of the flow rate of shielding and carrying gases can be considered negligible. On the contrary, when performing the laser cladding process on more complex geometry, this influence exerted on the powder stream profile should be taken into account. The performance of laser cladding process on cylindrical substrate varies even if parameters such as laser power, cladding speed, laser spot size and powder feed rate are held constant. This indicates the importance of proper selection of volumetric flow rates of process gases in order to obtain desirable shape of powder flux. In turn, proper alignment of the laser beam and powder stream allows obtaining the desired geometrical and mechanical properties of the clad layer. This paper presents the results of two-phase flow simulation with different amount of process gases in case of cylindrical substrate. Moreover performed simulations are verified by experimental tests that were carried out with usage of DoE technique. Finally, the influence of process gases flow on cladding properties is determined. 2. Materials and methods In the study concerning the influence of the process gases on the results of laser cladding on the cylindrical geometry of the substrate, the computer simulations as well as experimental verification was carried out on cylindrical substrate. The experimental setup for this research was shown on Fig. 1 and it originally served as a basis for simulation model development.
Fig. 1. Schematic (a) and photo (b) of experimental setup for laser cladding
The simulation was performed in ANSYS Fluent environment, wherein two-phase flow model was developed. Argon was modelled as continuous phase and cobalt-based alloy Stellite 6 powder as solid dispersed phase (Tab. 1). Table 1. Properties of continuous and dispersed phases utilized in simulation Material property
MAr
RAr
Value
39.95
208.1
ρAr (kg/m3) 1.633
µAr (10-5 kg/(m s)) 2.286
ρSt6 (kg/m3) 8.75
D St6 (µm) 65
The additional material used during experimental trials was gas-atomized spheroidal powder MetcoClad 6F, whose chemical composition is similar to Stellite 6. Furthermore, a 160 mm long, 84 mm outer diameter AISI 4330 steel tube with a wall thickness of 5 mm was used during the laser cladding experiments as a substrate. 2.1. Simulation model in Ansys Fluent In the simulation of powder particles and process gases flow, the Euler-Lagrange calculation method was used. It involves considering carrying gas as a continuous phase and the powder particles as discrete, dispersed phase. During the simulation of gases flow, the Navier-Stokes equations (Eq. 1) are solved and for dispersed phase the trajectories of particles are determined based on the Lagrange algorithm (Eq. 2). The dynamic viscosity was determined on the basis of the Sutherland formula described in [12]. ∂ ∂xi
�ρ𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴 ui uj �=-
𝑑𝑑𝑑𝑑𝑢𝑢𝑢𝑢𝑝𝑝𝑝𝑝 𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑
∂p ∂xi
+
∂ ∂xj
�μ𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴 ∙(
= 𝐹𝐹𝐹𝐹𝐷𝐷𝐷𝐷 �𝑢𝑢𝑢𝑢𝑖𝑖𝑖𝑖 − 𝑢𝑢𝑢𝑢𝑝𝑝𝑝𝑝 � + 𝑔𝑔𝑔𝑔 �
∂ui ∂xj
𝜌𝜌𝜌𝜌𝑝𝑝𝑝𝑝 −𝜌𝜌𝜌𝜌 𝜌𝜌𝜌𝜌𝑝𝑝𝑝𝑝
+
�
∂uj ∂xi
)� +ρ𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴 𝑔𝑔𝑔𝑔
(1) (2)
In order to calculate the two-phase flow, a geometrical model of cladding nozzle and cylindrical substrate was prepared. The axisymmetric model was used to simplify the geometry to half of the cross-section. This approach significantly accelerates the simulation process while retaining accuracy. Subsequently, characteristic parts of the computational model were extracted and defined (Fig. 2).
Piotr Koruba et al. / Procedia CIRP 74 (2018) 719–723 Author name / Procedia CIRP 00 (2018) 000–000
721 3
of the cladding nozzle increases. This may become a reason for blowing out the powder before it reaches the melt pool. In the case of equal volumetric flow rates of shielding and carrying gases, gas velocity increases only under the powder outlet. This may cause the boost in momentum of powder particles, making them reach the melt pool more easily, hitting it with greater force. This might result in larger penetration depth and dilution and, consequently a smoother clad face.
Fig. 2. Schematic of investigated geometry (a) and boundary conditions (b)
To obtain more accurate simulation results, the grid was concentrated in places that were significant for the research. In this case it was the space between the outlet of the nozzle and the surface of the tube to be cladded. Subsequently, the parameters of the two-phase flow analysis such as boundary and initial conditions and parameters of the ambience were determined. Initial velocity values for inlet 1 and inlet 2 were calculated for three sets of process gases flows. It should be also emphasized that the analysis, which was carried out, was a steady-state simulation for two-phase laminar flow. 2.2. Experimental setup The effect of process gases on results of laser cladding on cylindrical surfaces was also determined using experimental tests. The experimental setup consisted of: laser cladding nozzle COAXpowerline (IWS Fraunhofer), powder feeding system H-PF2/2 (GTV), laser unit LDF 4000-30 (Laserline) and for manipulation RV60-40 robot and tilting turntable RDK05 (REIS). During performed laser cladding tests, the main process parameters such as laser power, beam focus, cladding speed and powder feed rate remained constant, see Tab. 2. Table 2. Constant laser cladding process parameters during experiments Process parameter
Laser power (W)
Laser spot diameter (mm)
Value
400
1.5
Cladding speed (mm/s) 10
Powder feed rate (g/min) 1.8
Fig. 3. Simulation results of gas velocity (a) and powder concentration (b)
Similar observations can be made referring to Fig. 3b) showing particle mass concentration. For the initial set of gas parameters, which had been used for deposition of coatings on flat surfaces, it was observed that the area of high powder concentration is relatively small in crucial process zone. Due to increasing the flow of carrying gas the high concentration area became larger. This might result in greater amount of melted powder and smooth and high cladding. On the other hand, with the increase of shielding gas flow the red area is much smaller. The analysis of obtained data allows stating that the highest particle mass concentration was obtained with initial process gases volumetric flows. However, the values presented on a line graph concern only the centerline and for initial case the maximal value is not on the substrate surface. Increasing the flow of carrying gas helps achieve the smallest dispersion of powder concentration, the largest area of high concentration and the maximum value at the substrate surface.
3. Results and discussion
3.2. Results of experimental tests
The outcome of the computational simulation and experimental tests are presented separately in detail. Further, the relationships between the two studies will be described.
The results of laser cladding process were presented on Fig. 4, which shows the cross section images of resulting Stellite 6 coatings deposited on cylindrical steel substrate.
3.1. Results of simulation tests The results of performed simulation were shown on Fig. 3 and can be divided into the results considering gases velocity and describing the behavior of powder concentration. It can be noticed that for a significant amount of shielding gas in relation to the transport gas, the gas velocity under the outlet
Piotr Koruba et al. / Procedia CIRP 74 (2018) 719–723 Author name / Procedia CIRP 00 (2018) 000–000
722 4
Table 4. Exemplary chemical composition of laser deposited coatings Coating No.
Co
Cr
Si
C
Fe
2
68.23
27.44
2.58
1.19
0.56
4
64.27
25.98
2.39
1.17
6.19
The difference in iron content observed in the chemical composition (Tab. 4.) of the coatings occurred due to excessive dilution and diffusion of iron to the coating. Fig. 4. Cross-sections of deposited coatings showing cladding irregularity
The measured coatings properties were gathered in Tab. 3. Coating No. 1 is a reference coating, since the gas parameters for this case were the same as for the flat surfaces. Table 3. Properties of Stellite 6 coatings for different flows of process gases Coating No.
CG (l/min)
SG (l/min)
havg (mm)
dHAZ (mm)
Rh (mm)
HV 0.1
1
4
10
200.7
472.5
43.3
599
2
4
15
176.4
412.1
63.9
634
3
10
15
276.2
647.4
25.6
572
4
10
10
301.8
710.3
15.3
548
The impact of process gases on coatings properties were determined quantitatively by generating Pareto charts for height irregularity, heat-affected zone depth and microhardness of the coatings. Apart from the independent influence of variable process parameters, the charts also included interaction between them, since CG:SG means the interaction between carrying and shielding gas. As can be seen on Fig. 5, the increase of carrying gas flow can smooth the clad face and prevent formation of possible discontinuities. On the contrary, it may cause a considerable deepening of HAZ (Heat Affected Zone). Although carrying gas is more influential factor, shielding gas is still crucial for changing coating properties. The higher shielding gas flow resulted in higher cooling rates and reduction of HAZ, but also caused the discontinuities and gaps in the coating.
4. Conclusions Laser cladding of cylindrical substrate is strongly dependent on the flow rates of carrying and shielding gases. It was confirmed by numerical analysis and experimental trials. The protection from oxidation is improved with increase of the shielding gas flow but the risk of irregularities occurring in the clad increases. Although increasing the carrying gas flow may result in deposition of higher and more uniform coating, it causes higher dilution, deeper HAZ and lower hardness. On the contrary, shielding gas can be considered as additional cooling resulting in higher purity and hardness of the coating. The coating’s irregularity was explained by two-phase flow simulation. The high ratio of shielding to carrying gas flow creates a high velocity field under the central part of the cladding nozzle. This causes a possibility of blowing away the powder by gas blasts. Therefore the powder stream cannot be focused precisely on the substrate and the deposited coating may be lower and contain geometrical defects. Summarizing, the research confirmed that coating properties may be modified by the appropriate selection of process gases flows. Acknowledgements This research was undertaken as part of the Additive Manufacturing Processes and Hybrid Operations Research for Innovative Aircraft Technology Development project funded by the National Centre for Research and Development within INNOLOT program. References
Fig. 5. Pareto charts for impact estimation on: dHAZ (a), Rh (b), HV0.1 (c)
The analysis of influence of gases on microhardness of the coating indicates that their impact is opposite. Microhardness declined with the increase of the carrying gas flow, and with decrease of the shielding gas flow. However, the magnitude of the carrying gas factor is about 2 times greater, which leads to the conclusion that coating microhardness might be reduced by excessive carrying gas flow. Results of chemical composition examination of two deposited coatings (No. 2 and No. 4), support this conclusion.
[1] Poprawe R (Ed.). Tailored Light 2 Laser Application Technology. Berlin Heidelberg: Springer-Verlag, 2011. [2] Baraniecki T., Chlebus E., Dziatkiewicz M., et al. System for laser microsurfacing of metal powder. Weld Int 2016:30(2) p. 98-102. [3] Toyserkani E., Khajepour A., Corbin S. Laser Cladding. Boca Raton: CRC Press, 2005. [4] Pinkerton A.J. Advances in the modeling of laser direct metal deposition. J. Laser Appl 2015;27:S15001. [5] Goodarzi DM., Pekarinnen J., Salminen A. Effect of process parameters in laser cladding on substrate melted areas and the substrate melted shape. J Laser Appl 2015; 27:S29201. [6] Graf B., Ammer S., Gumenyuk A., et al. Design of experiments for laser metal deposition in maintenance, repair and overhaul applications. Procedia CIRP 2013;11:245-248. [7] Song L., Bagavath-Singh V., Dutta B., et al. Control of melt pool temperature and deposition height during direct metal deposition process. Int J Adv Manuf Technol 2012;58(1):247–256. [8] Kheloufi K., Amara E. Numerical modelling of gas/particles diphasic jet in laser cladding by coaxial nozzle. Phys. Procedia 2010;5B:347-352.
Piotr Koruba et al. / Procedia CIRP 74 (2018) 719–723 Author name / Procedia CIRP 00 (2018) 000–000
[9] Wen SY., Shin YC., Murthy JY., et al. Modeling of coaxial powder flow for the laser direct deposition process. Int J Heat Mass Trans 2009;52(2526):5867–5877. [10] Tabernero I., Lamikiz A., Ukar E., et al. Numerical simulation and experimental validation of powder flux distribution in coaxial laser cladding. J Mater Process Tech 2010;210(15):2125-2134.
723 5
[11] Li L., Huang Y. Numerical and Experimental Study on Powder Stream Characteristics in Coaxial Laser Cladding Process. Proceedings of 6th International Conference on Welding Science and Engineering. Beijing, 2015. [12] Badyda J. Corrected Sutherland’s formula for calculating gas thermal conductivity. J Pow Tech 1997;85:45-54.
