APEM jewuial Advances in Production Engineering & Management Volume 9 | Number 3 | September 2014 | pp 128-138 http://dx.doi.Org/10.14743/apem2014.3.182 ISSN 1854-625G Journal home: apem-journal.org Original scientific paper Influence of welding speed on the melting efficiency of Nd:YAG laser welding Tadamalle, A.P.a*, Reddy, Y.P.a, Ramjee, E.b, Reddy, V.K.b aDepartment of Mechanical Engineering, Sinhgad College of Engineering, Vadgaon, Pune, India bDepartment of Mechanical Engineering, Jawaharlal Nehru Technological University, Kukatpally, Hyderabad, India A B S T R A C T A R T I C L E I N F O Melting efficiency is one of the more important measurable parameters in laser welding when assessing the performance of a process. This paper aims to study the effects of speed on the melting efficiency and energy transfer efficiency of Nd:YAG laser welding process. The weld bead on a 304L austen-itic stainless steel sheet is created by varying the welding speed. The weld samples are cut in the transverse direction by using electric discharge machining, and the cross-section is prepared for metallographic inspection. The cross-sectional dimensions and beads length are measured by using an optical microscope and image analyzer. A methodology is proposed for estimating the weld pool volume from experimental data and a generalised equation for predicting the melting efficiency and energy transfer efficiency is developed. The results obtained by the proposed method have reasonably good agreement with the models proposed by various researchers. The outcome of the result shows the significant influence of welding speed on melting efficiency and energy transfer efficiency in welding of austenitic stainless steel thin sheets. It will be seen from results that one can select appropriate welding speed and processing conditions to obtain desired melting efficiency. © 2014 PEI, University of Maribor. All rights reserved. Keywords: Nd:YAG laser welding Melting efficiency Weld pool volume Energy transfer efficiency Heat affected zone *Corresponding author: aptadmalle.scoe@sinhgad.edu (Tadamalle, A.P.) Article history: Received 22 May 2014 Revised 7 August 2014 Accepted 15 August 2014 1. Introduction The Nd:YAG laser welding process is extensively used in modern industrial applications due to increase in demand for micro products. It is very essential to understand the behavior of laser material interaction, controlling of process parameters and their effect on melting, solidification and efficiency. The fundamental figures of merits for the laser welding process can be expressed in terms of melting efficiency, coupling efficiency, process efficiency, and energy transfer efficiency. The other figures of merits considered are fusion zone size, tolerance, and changing base metal temperature. The melting efficiency quantifies the fraction of net heat input to the work piece that is used to produce the weld pool. The absorption of heat by the material is affected by many factors like process parameters, type of laser, incident power density, and base metal surface condition. The optimum melting efficiency and weld strength of the joint depend upon vaporization temperature of the work piece material. The functional characteristics of laser welding process depends upon welding speed, laser power, size of the weld pool geometry, thermal and process efficiencies. The welding efficiency can be improved by fully utilizing the power supplied to melt the unit volume of material. In literature it has been reported that the effect of very high welding speed on melting efficiency and depth of penetration has been studied [1-2]. The authors discussed the effect of weld- 128 Influence of welding speed on the melting efficiency of Nd:YAG laser welding ing speed on weld bead geometry, and performance parameters such as variation of weld bead diameter from pulse to pulse, duty cycle, and effective pulse energy [3]. The present research work is the continuation of this research on development of new analytical equations to predict the melting efficiency. The experimental and analytical prediction shows that there is a strong dependence of energy absorption on laser power, beam size, and welding efficiency [4-5]. The dimensionless parameters Rykalin (Ry) and Christensen (Ch) are used for computing the melting efficiency of arc and CO2 laser welding. The melting efficiency of the laser welding process is determined by using experimental results, dimensionless parameter models and energy balance equations. Several researchers have formulated analytical equations to predict the melting efficiency in arc welding process [6-10]. The dimensionless parameter model is very valuable in characterizing partial penetration in laser beam welding process. The material independent model reveals that the variation in laser power and focus spot size is insensitive to the energy transfer efficiency in continuous wave laser beam welding process. A dimensionless parameter model developed for evaluating the melting efficiency in CO2 laser beam welding has been applied in gas tungsten arc welding (GTAW), plasma arc welding (PAW), and autogenous arc welding processes [11-14]. The energy transfer efficiency of arc and laser welding process is estimated by using calori-metric method and thermal expansion measurement techniques for thin sheets and the results obtained are correlated with the reflection method. The process efficiencies of laser engineered net shaping process is estimated for the tool steel and copper powder material deposits [15-19]. Transient 3D numerical model was developed to study the temperature field and molten pool shape during laser welding and volume of fluid method was employed in the calculations [18]. The influence of welding process parameters, preheating and heat absorption on different laser welding efficiencies is investigated [19-21]. It was found that the global efficiency of the laser welding process decreases slightly by varying the welding speed from 10 mm/s to 2 mm/s, but drops significantly below 1 mm/s [22]. The preheating of weld samples has an effect on absorption efficiency and other efficiencies in the beginning of the process and in the formation of the keyhole [23, 24]. The absorption efficiency and keyhole coupling efficiency are determined by using laser energy reflected from molten pool and by considering plasma effect studied under steady state condition [25]. This paper aims to study the effect of process parameters on melting efficiency (nm) of Nd:YAG laser weld joints. A semi empirical equation is proposed to estimate the weld pool volume from experimental data, a dimensionless parametric equation to estimate melting efficiency. The section 2 deals with the models proposed by various researchers to evaluate the melting efficiency in different welding process, and section 3 enumerate the experimental procedure adopted. The equations proposed for prediction of weld pool volume and melting efficiency are presented in section 4 and section 5 respectively. The results and discussions are illustrated in section 6 and the conclusions of present work are given in section 7. 2. Melting efficiency Melting efficiency is defined as the ratio of energy required to create a molten pool from heat energy supplied to the energy absorbed by the work piece. A small percentage of the total energy is used for melting the fusion zone and rest of the energy is dissipated to the surroundings by means of various modes of heat transfer. The literature review reveals that processing parameters, heat flow pattern and thermo-physical properties have significant influence on melting efficiency. A dimensionless parameter model can be used to determine the actual heat input to the metal and the melting efficiency. The dimensionless parameters Ry and Ch are derived from the Rosenthal heat flow solutions. These parameters can be evaluated by using the Eq. 1 developed for the arc welding and CO2 laser welding process [9]. The Rosenthal heat flow solution model ignores mass additions to the melt pool. Hence modifications are required in the heat flow equations that can incorporate mass addition to the Rosenthal heat flow solution model. However, a semi empirical model can provide reasonable estimates of melting efficiency for both the arc welding and CO2 laser welding process. Advances in Production Engineering & Management 9(3) 2014 129 Tadamalle, Reddy, Ramjee, Reddy s A Pj s ,, Ch = —— and Ry = (1) az y azSh where P, is the heat input to the substrate (W), a is the thermal diffusivity of base metal (mm2/s), Sh is the enthalpy of melting (J/mm3), A is the weld cross-sectional area (mm2), s is the welding speed (mm/s). The melting efficiency (nm) for the arc welding process [10] can also be expressed as the ratio of dimensionless parameters Ch and Ry [8] as given in Eq. 2. Ch s A Sh r]m = T= ~W (2) The dimensionless parameter Ry play an important role in estimating the melting efficiency and it is a nonlinear function of heat input and welding speed. Several researchers have developed equations to predict the melting efficiency by using 2D and 3D heat flow conditions. The 2D and 3D heat flow conditions are expressed in terms of thermo-physical properties and processing parameters for predicting the melting efficiency. The mathematical models applied for 2D and 3D heat flow conditions in arc welding process are presented in Eq. 3 and Eq. 4, respectively [10]. /8a\ V5s w) (3) 1.35 I 1 + 11 + 10.4 a2V \ (4) 0 w) where w is the bead width (mm). A model applied for predicting the melting efficiency based on power delivered to the substrate material in arc welding process is given in Eq. 5: Vm= exp[ -( 1 + ^ ^ (5) In present work, the term power input P,, supplied by the laser source to the material is replaced by the product of energy transfer efficiency (ne), voltage (V) and current (I). A semi empirical Eq. 6 is used to predict the melting efficiency based on heat input required to melt the material [7] in GTAW process: v St0cp(t)AT+AHl (6) VetPi where AH/ is the latent heat of fusion (J/mm3), v is the total volume of melted substrate (mm3), T is the temperature of the weld pool (K), To is the initial temperature (K), t is the laser on time (ms), and Cp is the heat capacity (J/mm3K). The experimental procedure adopted in this work for welding of 304L austenitic stainless steel using Nd:YAG laser welding machine is explained in the next section. 130 Advances in Production Engineering & Management 9(3) 2014 Influence of welding speed on the melting efficiency of Nd:YAG laser welding 3. Experimental procedure The welding of 304L austenitic stainless steel sheets is carried out by using pulsed Nd:YAG laser welding machine, TruLaser station 5004 (Trumpf), designed to deliver a maximum laser power of 4 kW. The process parameters, thermo-mechanical properties and chemical composition of 304L stainless steel are given in Table 1 and the setup used for experimentation is shown in Fig. 1. The blank sheets are cleaned before welding by 6-8 % NaOH solution followed by 20 % HNO3 solution. The samples are cut into rectangular specimens of 30 mm by 50 mm with the help of wire cut electric discharge machine to avoid the distortion. The weld quality and aspect ratio of the WBG depends upon welding speed. The threshold value of melting front propagation was found between 1.1-1.4 mm/s and when welding speed is less than 1 mm/s the molten pool irradiates for a longer period of time which results in lower coupling efficiency [22]. The welding speed beyond 10 mm/s does not provide sufficient time to melt the material. Hence, the experiment are planned to conduct on 0.5 mm thick sheet to create a bead on plate by varying the welding speed from 2-10 mm/s in steps of 1 mm/s. Fig. 1 Specimen mounting arrangement and experimental setup Table 1 Process parameters and thermo mechanical properties and chemical composition of 304L stainless steel [3] Process parameters Thermo mechanical properties of the 304L stainless steel Parameters Values Parameters Values Parameters Values Beam diameter 0.4 mm Density 8030 kg/m3 Poisons ratio 0.29 Pulse duration 2 ms Elastic modulus 193 GPa Melting point 1723 K Frequency 25 Hz Mean coefficient of expansion 18.4 nm/m/K Refractive index 3.81 Fe Beam angle 90 ± 5° Thermal conductivity 20 W/mK Enthalpy 8.7 J/mm3 Pulse energy 2.76 J Specific heat 500 J/kgK Diffusivity 5.7 mm2/s C Mn Ni Cr Si V N Proof stress Yield strength Elongation 0.3 2.0 8-12 18-20 0.75 0.07 0.1 170 MPa 485 MPa 40 % Table 2 Weld bead geometry dimensions measured from weld samples Exp. No. Welding speed (mm/s) Bead width (|mf | Bead length (|m) Depth of penetration (|m) 1 2 895 912 500 2 4 845 845 472 3 5 y30 y30 375 4 y 691 691 233 5 8 68y 687 190 6 9 682 682 189 y 10 610 648 100 Advances in Production Engineering & Management 9(3) 2014 1B1 Tadamalle, Reddy, Ramjee, Reddy 4. Estimation of weld pool volume In laser welding process, the selection of preferred levels of welding speed plays an important role to achieve higher melting efficiency. The experiments are conducted by varying welding speed and by keeping all other process parameters constant. The transverse cross-section area, bead length (BL) and bead width (BW) of the weld bead geometry (WBG) are extracted from the samples prepared for metallographic inspection, Fig. 2(a) to Fig. 2(f). The WBG and HAZ profile dimensions are measured by using Metatech (Hitachi) electron microscope. The digitized data obtained from the HAZ profile is used to generate polynomial equations at different welding speeds. This digitized data is best fitted to obtain the second order polynomial equations. A sample second order polynomial equation and corresponding curve obtained from digitized data at a welding speed of 8 mm/s is shown in Fig. 3. Similarly other second order polynomial equations are derived from experiments conducted at different welding speeds. A generalized second order polynomial equation applicable to the geometry of any heat affected zone profile is given in Eq. 7. Fig. 2 Samples prepared for metallographic inspection and measurement of WBG 132 Advances in Production Engineering & Management 9(3) 2014 Influence of welding speed on the melting efficiency of Nd:YAG laser welding Fig. 3 Digitized data fitted to obtain a second order polynomial equation f(x) = Ct + C2x + C3x2 (7) The coefficients C\, C2, and C3 are constants which are evaluated by using the polynomial equations derived from digitized HAZ data and the variable x denotes the bead width. This data is measured from the cross-sectional view of the micrographs obtained at different welding speeds shown in Table 2. The digitised data is used to obtain a HAZ profile of the weld pool shown in Fig. 3 and the corresponding second order polynomial equation. The weld pool volume estimated by revolving half section of the polynomial curve about the axis of the laser beam is given in Eq. 8. d V = J(C1 + C2x + C3X2)2 dx (8) 0 A generalized equation proposed to evaluate the weld pool volume by using best fit curve obtained from the geometric mean of all corresponding coefficients. V = 0.512 + 0.028x - 0.007x2 (9) The weld pool volume and its cross-sectional area is a function of depth of penetration, bead length and bead width. The correct measurement of depth of penetration and bead length is difficult, because it requre metallographic preparation to observe cross-sectional view of the WBG. The bead width can be measured with ease hence it is proposed equation to estimate the weld pool volume in terms of bead width as given in Eq. 10. V = 0.512 + 0.028 w - 0.007w2 (10) The coefficients given in Eq. 10 are valid in the range of weld speed from 2-10 mm/s. This range is selected because it gives better results in terms of pulse overlapping factor, variation in bead diameter from pulse to pulse and other performance parameters [3] at the specified values of process parameter considered for the study. 5. Prediction of melting efficiency The dimensionless parameters Ry and Ch can be evaluated by knowing material properties, weld speed, laser power input and weld pool cross-sectional area. The values Ry and Ch are estimated by varying the welding speed and a best fit curve shown in Fig. 4 is obtained by using linear regression analysis technique. The correlation coefficient between Ch and Ry is greater than 0.943 and the relation obtained from the best fit curve is presented in Eq. 11. Advances in Production Engineering & Management 9(3) 2014 133 Tadamalle, Reddy, Ramjee, Reddy 0.7 0.6 0.2 Ch= 0.065 + 0.016 R, -y 0.1 0 2 4 6 Rykalin S 10 Fig. 4 Relation between Rykalin and Christensen dimensionless parameters Ch = 0.065 + 0.016 R y (11) The relationship for melting efficiency derived from Eq. 11 is given in Eq. 12: 0.065 = 0.016 + —— (12) The semi empirical Eq. 12 can be used for predicting the melting efficiency of Nd:YAG laser welding process. The melting efficiency is a function of Ry. The laser beam reflection method is used in this study to estimate energy transfer efficiency. The accurate prediction of weld size is done by considering different values of energy transfer efficiency. 6. Results and discussions The power absorbed by conduction and melting of the substrate material is more at higher welding speeds. The Fig. 5 reveals that total power absorbed by the substrate material increases with increase in welding speed. This is due to the significant difference in temperature between laser source and work piece material as a result of which more amount of heat is utilized to create and maintain molten weld pool. It is found that the power utilized for melting the material is less than that of power absorbed by conduction up to a welding speed of 7 mm/s. However beyond this limit, power absorbed for creating weld pool is greater than heat carried away by the conduction. This is due to less than 4 ms times is available for transferring heat energy to the substrate material by conduction. The stainless steel material has reflectivity 64 %, whereas for copper, aluminium and nickel it is greater than 74.20 %. Many researchers have determined the melting efficiency in welding by considering energy transfer efficiency constant at 0.37 or 0.48 [7, 10]. In this study, laser beam reflection method is employed to compute energy transfer efficiency. 134 Advances in Production Engineering & Management 9(3) 2014 Influence of welding speed on the melting efficiency of Nd:YAG laser welding 40 35 30 „ 25 I I 20 o cilS 10 5 0 Power absorbed by conduction Total power absorbed Power utilised for melting 5 7 Welding speed (mm/s) 11 Fig. 5 Effect of welding speed on power absorbed by the specimen The Fig. 6 shows that there is a significant variation from 10.1-21.2 % in the energy transfer efficiency with respect to welding speed, therefore different values of energy transfer efficiencies are considered instead of a single value. The heat input to the substrate material is directly proportional to the product of energy supplied, energy transfer efficiency and pulse frequency. The energy transfer efficiency is based on materials properties, laser beam reflection, process parameters, operating conditions and power source. The analytical equations proposed by the researchers to estimate the melting efficiency predict differently for different types of welding processes. A non-linear relationship is observed between the melting efficiency, weld speed, and Ry from the results obtained by different researchers. 25 & 20 c - DO