PARTICULARITIES OF MIXING WITHIN MELT POOL DURING LASER ALLOYING OR LASER CLADDING BY DIRECT POWDER INJECTION
B.Medres, L.Shepeleva and M. Bamberger
Surface Technologies Ltd. Israel
ABSTRACT. The mechanism for stirring the liquid within the melt pool is considered. When active evaporation doesn’t occur, a melt is mixed up due to the pressure of gas transporting the injected particles, the gravity and the surface tension. Estimation was done for motion rates of liquid within the melt pool. It was shown that when deep melt pool appeared, a multiple stirring of moving melt (300-350 cm/s) prevailed a plane parallel to the laser beam movement.
Keywords:direct powder injection, a mechanism of the turbulence in the melt pool.
Introduction
LSA and LC are used for increasing the surface wear and corrosion resistance of machine parts. These effects are achieved by to the alloying of the surface layer or the formation a surface cladded layer. This procedure may be realized by at least two ways. The first technique consists of preliminary covering the surface of the treated part with the alloying substance followed by laser melting [1]. Another way is the direct injection of alloying (cladding) elements in a form of powder or wire into the melt pool [2].
Liquid flow within the melt pool is the main mechanism of powder distribution in the melt in both techniques of LSA and LC using CW-CO2 laser. Therefore understanding the melt self-stirring within the melt pool during LSA and LC is of highest technological importance [3].
Different mechanisms for the liquid motion within the melt pool were proposed for laser surface treatments utilizing powerful lasers. But, the treatment conditions corresponded either to the regime of active evaporation [2,3] or to laser melting of pre-coated surface [3,4].
In this work the mechanism of liquid stirring within the melt pool under LSA and LC with direct injection of alloying and cladding elements is considered.
Experimental conditions
The processes taking place within the melt pool under LSA of AISI 1045 steel with Ni2 B powder and LC of copper with NiCrBSi powder were studied. A CW- CO2 laser of maximum power of 3000 W operated at 1750-2500 W was used for heating. The laser beam diameter was 0.2 cm at focal plane. The angle between the powder stream and the treated surface was about 450. The treatments were carried out at low scanning velocity (V) of 1-2 cm/c resulting in a pool depth (Z) which equals the beam radius (reff), and at a high V of 5-10 cm/s that provides Z < reff. The powder carrying gas pressure was (1-2)105
Results and Discussions
Slicing the specimens, in a plan parallel to the laser beam movement direction revealed evidence of intensive turbulent flow in the deep melt pool, as can be seen in Fig1.
Fig.1 The longitudinal-sectional view of a LSA zone
Peculiarities of LSA and LC Performed by Direct Injection of Alloying Elements.
The principle distinction of direct injection LSA (LC) techniques from those using a pre-coating lies in the different forces affecting the melt pool surface. For the former cases the main force operating on the melt pool surface is the pressure of carrier gas. The numerical estimations made by [2], show that
surface tension produces a pressure of about 103 Pa on the melt pool surface. However, the carrier gas delivers a pressure of (1-2)105Pa on the melt pool surface that is several orders of magnitude higher than the pressure provided by surface tension.
Obviously, the force affecting the melt pool surface, which are related to the carrier gas, lies in the plan parallel to the laser beam scan direction. The fact that its value is much higher than that of surface tension suggests that melt motion within the pool will occur in the plane of this force action, i.e. in the plane parallel to the laser beam movement direction. Thus, the carrier gas pressure must dominate the vertical flow of melt, which is most important for LSA processing [3]. Let us discuss the mechanism of processes occurring within the melt pool appearing upon direct injection LSA with using the scheme presented in Fig.2.
Fig.2 The schematic diagram of the LSA zone creation
This scheme presents the extreme case when tg = 1, where is the angle between the laser beam direction and the normal of liquid/solid interface [3]. Laser beam melts the surface to a specific depth Z. Upon this heating the melted layer affected by the pressure of the transporting gas stream begins to move countercurrent to the scanning direction. When the accelerated melt riches the opposite solidified side of pool it is rejected from it (combined effect of rejection force and gravity) and returns to the center of melted zone. Here the melt is mixed with a fresh melt and a turbulence flow in the plane parallel to the laser beam scan direction, shown schematically in Fig.2, is formed. The holes, which can be seen in the Fig.3 are situated on the plane parallel to the laser beam scanning direction.
Fig.3. The holes on the laser cladding zone formed during LC processes
Evidently, these holes resulted from the absorption of gas with the powder particles and turbulence movement of the liquid under the gas jet action. It is very interesting both in scientific and technological aspects to try to estimate the velocity of the liquid within the melt pool. For this purpose the following equation can be used.[3]
l =
where l - is the velocity of the liquid melt motion, cs -is sound speed in the metal, P is the power of heating source, S is the immediate value of effective heated area, is the density, -is the specific enthalpy of evaporation. The following values were used cssteel = 4.3 10 5 cm/s, = 7gr/cm 3 , steel = 10 4 kJ/gr [3]. Thus, the melt flow velocity was found to be 300-350 cm/s, which is higher than the value of 200 – 260 cm/s obtained in [3] for developed evaporation region. These velocity values seem to be reasonable, because maximum pressure of steam rejection doesn’t exceed 105 Pa, [2], while for direct injection LSA pressure of carrier gas is (1-2)105Pa. Additionally, the power density used in direct injection LSA was about 7.5104 W/cm2 while authors [3] performed thei
their experiments at 5104 W/cm2.
Let us now estimate the intensity of melt stirring in the following way. When the effective diameter of melt pool (deff) is about 0.2 cm, the longest perimeter (L) in the melt pool is equal to the melt/substrate interface, e.g. about 0.628 cm. Melt pool interaction time depends on the values of deff and scanning velocity, i.e. t = deff /V, all other factor being constant. For Z reff the interaction time is 0.1 - 0.2 s, and 0.02 - 0.04 s for Z<reff . Then, at the lowest moving speed of 300 cm/s the melt will cross a distance of 30 – 70 cm, during the interaction time of 0.1 – 0.2 s, which will provide at least 50 – 100 revolutions. Presence of such intensive stirring is clearly seen on Fig.1. Because all main forces affecting the liquid melt pool lie in the plane parallel to laser beam scan direction the melt stirring will also occur in this plane. Under the high scanning rates, when melt pool dimensions fit the condition of Z<reff , the melt pool exists t = 0.02 –0.04 s in this case the number of revolutions of the liquid is very low. We can say that for such high scanning rates the temperature gradient at the edges of melt pool will be approximately the same at all radial directions, that provides mainly the horizontal flow of liquid within melt pool. Obviously, that two considered cases are both the extreme ones and the real process within the considered range.
Conclusions
1.The mechanism of LSA and LC zones formation upon direct injection of alloying powders is considered. It is shown that direct injection LSA and LC zone is characterized by intensive liquid mixing within the melt pool, resulting from the pressure of the carrier gas.
2.When deep melt pool is created, the melt is stirred in the plane parallel to the laser beam scanning, that provides vertical motion of liquid within LSA and LC zone.
REFERENCES
- Rycalin N., Uglov A. and Kokora A.: Laser machining and welding. Moscow,USSR, MIR Publishers,1978,p.308
- William M.Steen: Laser material processing. Springer-Verlag, 1991, p.266
- Vedenov A.,. Gladush G.: Physical processes in laser treatment of materials. Moscow, Energoatomizdat, 1985, p.205
- Uglov A., Soloviov A. and Medres B.: Principles of modern treatment of materials. USSR, Omsk, OHTES, 1988, p.54
- Uglov A., Medres B. and Soloviov A.: “ Laser treatment of tool steels.” Magn. Phys. Chem. Treatment Mater., 1987, (3), 3-10
2 - 1