Abstract
In this paper, the optimization of computer numerical control (CNC) machining parameters were conducted using the Taguchi design method on the surface quality of massive wooden edge glued panels (EGP) made of European larch (Larix decidua Mill). Three machining parameters and their effects on surface roughness were evaluated. These parameters included tool clearance strategy, spindle speed, and feed rate. An analysis of variance (ANOVA) was performed to identify the significant factors affecting the surface roughness (Ra and Rz). Optimum machining parameter combinations were acquired by conducting an analysis of the signal-to-noise (S/N) ratio. Optimal cutting performance for the Ra and Rz was obtained for the cutter at a tool clearance strategy of an offset 16000 rpm spindle speed and 1000 mm/min feed rate. The surface roughness, both the Ra and Rz, increased with increasing feed rate. Optimal cutting performance for Ra and Rz was obtained for a tool clearance strategy of an offset 16000 rpm spindle speed, and 1000 mm/min feed rate cutting settings. Based on the confirmation tests, Ra decreased 2.2 times and Rz 1.8 times compared to the starting cutting parameters.
Download PDF
Full Article
Determination of Optimal Machining Parameters of Massive Wooden Edge-Glued Panels Made of European larch (Larix decidua Mill.) using Taguchi Design Method
Sait Dundar Sofuoglu
In this paper, the optimization of computer numerical control (CNC) machining parameters were conducted using the Taguchi design method on the surface quality of massive wooden edge glued panels (EGP) made of European larch (Larix decidua Mill). Three machining parameters and their effects on surface roughness were evaluated. These parameters included tool clearance strategy, spindle speed, and feed rate. An analysis of variance (ANOVA) was performed to identify the significant factors affecting the surface roughness (Ra and Rz). Optimum machining parameter combinations were acquired by conducting an analysis of the signal-to-noise (S/N) ratio. Optimal cutting performance for the Ra and Rz was obtained for the cutter at a tool clearance strategy of an offset 16000 rpm spindle speed and 1000 mm/min feed rate. The surface roughness, both the Ra and Rz, increased with increasing feed rate. Optimal cutting performance for Ra and Rz was obtained for a tool clearance strategy of an offset 16000 rpm spindle speed, and 1000 mm/min feed rate cutting settings. Based on the confirmation tests, Ra decreased 2.2 times and Rz 1.8 times compared to the starting cutting parameters.
Keywords: Wood machining; European larch; Taguchi design method; Surface roughness
Contact information: Dumlupinar University, Faculty of Simav Technology, Department of Wood Works Industrial Engineering, 43500 Simav/Kutahya, Turkey;
* Corresponding author: sdundar.sofuoglu@dpu.edu.tr
INTRODUCTION
Wood and wood-based materials have been used as a construction material for years, mostly because of their low cost, renewability, strength, and low processing energy requirements. In recent years, the machining of wood products has acquired great importance because of the short supply of wood, as well as increasing environmental awareness among users and manufacturers. Consequently, the machining process is being optimized in terms of tool wear, work piece surface quality, crack initiation, and the utilization of different types of wood (Davim 2011).
The surface quality of solid wood and wood-based panels is one of the most important properties characterizing the wood-machining process and its conditions, as well as manufacturing processes, such as finishing or adhesive strength properties. The surface roughness of wood can be affected by different factors, such as annual ring variation, wood density, cell structure, earlywood ratio and latewood ratio, and humidity (Sandak and Tanaka 2003; Zhong et al. 2013). The stylus method is commonly used in laboratories and for off-line roughness measurements of wood and wood-based panel surfaces (Hiziroglu 1996; Hiziroglu and Graham 1998; Sandak and Tanaka 2003; Zhong et al.2013).
European larch is one of the most important commercial species in Europe. Its wood is highly valued for lumber and particleboard (Muhcu et al. 2015). Wood edge-glued panels (EGP) have a more heterogeneous structure than solid wood (Sutcu 2013). Many furniture companies utilize EGP in products such as table tops, bed or chest panels, and doors (Mitchell and Lemaster 2002).
Computer numerical control (CNC) wood working machinery has been widely introduced in wood industries, especially in the furniture industry. Computer numerical control has been used for grooving, milling, and patterning of furniture materials. The use of CNC has good advantages with respect to the speed, surface quality, and precision machining of work pieces (Costes and Larricq 2002; Ohuchi and Murase 2005; Karagoz et al. 2011; Alves et al. 2015).
The Taguchi method (TM) combines engineering and statistical methods to attain rapid improvements in quality and cost by optimizing the product design and manufacturing processes. Taguchi method is a unique statistical experimental design approach that greatly improves the engineering productivity (Verma et al. 2012). The TM can be effectively used to determine the optimal combinations of machining process parameters. The number of experiments is reduced to minimize the cost and time using an orthogonal array (OA) (Valarmathi et al. 2013). The method explores the concept of quadratic quality loss function and uses a statistical measure of performance called Signal-to-Noise (S/N) ratio. The S/N ratio takes both the mean and the variability into account. The S/N ratio is the ratio of the mean (Signal) to the standard deviation (Noise). The ratio depends on the quality characteristics of the product/process that was optimized. The standard S/N ratios generally used are as follows: Nominal is Best (NB), Lower the Better (LB), and Higher the Better (HB). The optimal setting is depends on combination of the parameters, which has the highest S/N ratio (Patel and Patni 2014). In addition, an analysis of variance (ANOVA) determines the statistically significant process parameters. After the analyses (S/N and ANOVA) have been conducted, a confirmation experiment is performed to verify the optimal process parameters acquired from the parameter design (Taguchi et al. 1989; Nalbant et al.2007; Gologlu and Sakarya 2008). Some researchers have reported on the effect of cutting speed and feed rate on surface roughness in wood and wood-based materials by milling and drilling of a CNC router (Davim et al. 2009; Sutcu and Karagoz 2012; Sutcu 2013; Valarmathi et al 2013). Benotmane and Zirouk (2013) studied the woodturning performance and efficiency by using the Taguchi methodology for dynamic systems. Wilkowski et al. (2011, 2013) used TM in order to investigate machinability and surface roughness of wood materials. Kacal and Gulesin (2011) studied determination of optimal cutting conditions in finish turning of austempered ductile iron using TM. Another study investigated optimization of the combined modification process of thermo-mechanical densification and heat treatment on Chinese fir wood (Li et al. 2013). Sofuoglu (2015) reported on using Artificial Neural Networks (ANN) to model the surface roughness of EGP made of Scotch pine in a machining process with CNC.
In this study, three important process variables (tool clearance strategy, spindle speed, and feed rate) with an experimental set up made by the Taguchi experimental design were investigated with a CNC router. Tool clearance strategy, which is area clearance strategy can be selected: raster or offset by CAM software (Fig. 2). On the machined surface, the average roughness (Ra) and highest mean peak to-valley height (Rz) values were measured with a surface roughness measuring device.
EXPERIMENTAL
In this study, the material used was 30-mm-thick EGP; the wood was from European larch (Larix decidua Mill.) trees. The EGP panels were supplied by ERPAN Massive Panel from Turkey. EGP panels were selected at random from the manufacturer. During the preparation of the EGP panels and manufacturing of the finger joints, laths were not expected to have the same properties (radial + radial, tangential + tangential, knotless, etc). The density of the EGP was measured as 0.62 g/cm3 at 12% moisture content ISO 3130 (1975); ISO 3131 (1975). The experiments were carried out on a Skilled CNC milling machine (Beysantaş A.Ş., Turkey) with a maximum spindle speed of 18000 rpm and a maximum feed rate of 2000 mm/min. The experiments were carried out with Netmak – 0450-07 (Ø10 x 30 x 72) (Netmak group, Turkey) and 10-mm-diameter router cutters made of a solid carbide. New and sharp cutters were used in each cutting test.
Three machining parameters (tool clearance strategy, spindle speed, and feed rate) were evaluated, as indicated in Table 1. A total of 18 pieces with dimensions of 50 x 50 mm were machined on EGP panels by a CNC router (Fig. 1). Depth of cuts were 4 mm (Fig. 2). Surface roughness was measured on every piece, in the radial direction, parallel to the grain, seven times. The measuring parameters (Raand Rz) are described in ISO 468 (2009). ISO 468 defines Ra and Rz all measured relative to a straight mean line: Ra (centerline average) value is the arithmetic mean of the departures of a profile from the mean line. It is normally determined as the mean results of several consecutive sample lengths. Rz (ten-point height) is the average distance between the five height peaks and the five deepest valleys within the sampling length and measured perpendicular to it.
Fig. 2. Parameters of CNC process
Table 1. Assignment of Levels to Factors (Machining Parameters used in the Face Milling of EGP Made of European Larch)
The measurement of surface roughness was conducted according to the protocols in ISO 468 (2009), ISO 3274 (2005), and ISO 4287 (1997). The Surface Roughness Tester (TR200, TIME, China) surface roughness measurement device, which has a diamond stylus, was used. Measurements were taken along the grain. The sampling length was taken as 2.5 mm; the evaluation length was chosen as Lt= 12.5 mm. Surface roughness values were measured with a sensitivity of ± 0.01 µm. The tool measurement speed was chosen as 10 mm/min, the diameter of the measurement needle was 4 µm, and the needle tip orientation was 90°.
Wood machining parameters were used as control factors, where 1 parameter was designed to have 2 levels and 2 parameters were designed to have 3 levels (Table 1). In accordance with TM, a L18 OA table with 18 rows (corresponding to the number of experiments) was selected for the experiments.
Statistical analyses (S/N, ANOVA) were performed by using MINITAB R17 software for a confidence level of 95% (e.g., significance level of 0.05).
To obtain an optimal cutting performance, the-smaller-the-better quality characteristics for Ra and RzS/N ratio (η) are defined as,
(1)
where y is the observed data and n is the number of observations.
RESULTS AND DISCUSSION
Average surface roughness (Ra) values were obtained as 1.89 μm and highest mean peak to-valley height (Rz) values were obtained as 8.30 μm. The measured data gives the normal distribution at the 95% confidence level. The experimental design of L18 and the experimental results and S/N ratios are given in Table 2. The optimal level of the process parameters is the level with the greatest η value. According to η value, main effects plots were prepared (Figs. 3b and 4b).
Table 2. Experimental Design using an L18 Orthogonal Array and the Results and S/N Ratio for Ra and Rz
The optimal cutting performance for Ra was obtained for tool clearance strategy of an offset (Level 2), 16,000 rpm spindle speed (Level 3), and 1000 mm/min feed rate (Level 1) setting (A2B3C1). Table 3 shows the response table mean S/N ratio (signal refers to the desired real value, whereas noise refers to the undesired factors in measured values) for Ra and Rz.
Table 3. Response Table Mean S/N Ratio for Ra and Rz
When analyzing the main effect plot for Rz (Figs. 3a and 4a), the raster tool clearance strategy obtained a smoother surface than the offset tool clearance strategy. Surface roughness values decreased with the decreasing feed rates. In the same way, many researchers have reported the effect of feed rate on surface roughness in wood and wood-based materials machining. According to the obtained results, the surface roughness decreases with increasing spindle speed and increases with the feed rate (Iskra and Tanaka 2005; Salca et al. 2008; Davim et al. 2009; Sutcu and Karagoz 2012; Sutcu 2013). Additionally, some researchers have explained that cutting speed has no effect on surface properties for large ranges of speed. Others reported that cutting speed has a slight effect depending on the process conditions (Costes and Larricq 2002). The value of surface roughness is higher at 12000 rpm than 8000 rpm. In contrast, at 16000 rpm, the surface roughness increased, even though it was expected to decrease.
Analysis of Variance (ANOVA)
ANOVA was conducted for a 95% confidence level. The factors did not exhibit statistical significance at a 95% confidence level. Effects of the tool clearance strategy, spindle speed, and feed rate on surface roughness were minimal (Table 4).
Table 4. ANOVA Results for Ra and Rz
a
b
Fig. 3 (a) Main effect plot for means (surface roughness for Ra); (b) mean S/N ratio for surface roughness for Ra
a
b
Fig. 4. (a) Main effect plot for means (surface roughness for Rz); (b) mean S/N ratio for surface roughness for Rz
Confirmation Tests
Once the optimal levels of design parameters were chosen, the final step was to predict and verify the improvement of the performance characteristics using an optimal level of machining parameters at TM18. An optimum agreement was observed between the predicted and real machining performance (Table 5). An increase in the S/N ratio from the initial machining parameters to the optimal cutting parameters (tool clearance strategy of an offset, 16,000 rpm spindle speed, 1000 mm/min feed rate) for Ra was 6.79 dB and for Rz was 5.06 dB. Based on the confirmation tests, Ra decreased 2.2 times and Rz 1.8 times compared to the starting cutting parameters.
Table 5. Confirmation Tests for Ra and Rz
CONCLUSIONS
The following conclusions can be drawn from the surface roughness (for Ra and Rz) aspects, based on the experimental results obtained for the wooden edge glued panels (EGP) made of European larch (Larix decidua Mill.):
- The surface roughness, both Ra and Rz, increased with increasing feed rate.
- Both of the Ra and Rz surface roughness values for 12000 rpm is higher than 8000 rpm and 16000 rpm spindle speed. Smoothest surface was obtained 16000 rpm spindle speed.
- The Taguchi design analyze method is suitable for solving the wood machining problems by CNC router machining which they were described in the present works. Better results have been obtained in terms of techniques such as Taguchi design method using MINITAB software.
- The surface roughness; parameters for Rz for an offset cutting strategy is smoother than that of a raster cutter strategy. Parameters for Ra approximate values were obtained.
- Optimal cutting performance for Ra and Rz was obtained for at a tool clearance strategy of an offset 16000 rpm spindle speed and 1000 mm/min feed rate cutting settings.
- Based on the confirmation tests, Ra decreased 2.2 times and Rz 1.8 times compared to the starting cutting parameters.
REFERENCES CITED
Alves, P. R. G., Goncalves, M. T. T., and Alves, M. C. D. (2015). “Surface quality of wood in face milling, techniques up-milling and down-milling in CNC machine,” 22nd Proceedings of International Wood Machining Seminar (IWMS 22), Quebec City, Canada, pp. 215-222.
Benotmane, B., and Zirouk, S. (2013). “Analyzing the woodturning process using Taguchi methodology for dynamic systems,” Scient. Res. Essay 8(4), 2046-2057. DOI: 10.5897/SRE2013.5538
Costes, J. P., and Larricq, P. (2002). “Towards high cutting speed in wood milling,” Ann. Forest Sci., 59(8), 856-865. DOI: 10.1051/forest:2002084
Davim, J. P. (2011). Wood Machining, John Wiley and Sons, Hoboken, NJ.
Davim, J. P., Clemente, V. C., and Silva, S. (2009). “Surface roughness aspects in milling MDF (medium density fibreboard),” Int. J. Adv. Manuf. Tech. 40(1-2), 49-55. DOI: 49-55.10.1007/s00170-007-1318-z
Gologlu, C., and Sakarya, N. (2008). “The effects of cutter path strategies on surface roughness of pocket milling of 1.2738 steel based on Taguchi method,” J. Mater. Process. Tech. 206(1-3), 7-15. DOI: 7-5.10.1016/j.jmatprotec.2007.11.300
Hiziroglu, S. (1996). “Surface roughness analysis of wood composites: A stylus method,” Forest Prod. J. 46(7-8), 67-72.
Hiziroglu, S., and Graham, M. (1998). “Effect of press closing time and target thickness on surface roughness of particleboard,” Forest Prod. J. 48(3), 50-54.
ISO 468 (2009). “Surface roughness-parameters, their values and general rules for specifying requirements,” International Organization for Standardization, Geneva, Switzerland.
ISO 3130 (1975). “Wood determination of moisture content for physical and mechanical tests,” International Organization for Standardization, Geneva, Switzerland.
ISO 3131 (1975). “Wood determination of density for physical and mechanical tests,” International Organization for Standardization, Geneva, Switzerland.
ISO 3274 (2005). “Geometrical product specifications (GPS) – Surface texture: Profile method – Nominal characteristics of contact (stylus) instruments,” International Organization for Standardization, Geneva, Switzerland.
ISO 4287 (1997). “Geometrical product specifications surface texture profile method terms, definitions and surface texture parameters,” International Organization for Standardization, Geneva, Switzerland.
Kacal, A., and Gulesin, M. (2011). “Determination of optimal cutting conditions in finish turning of austempered ductile iron using Taguchi design method,” J. Sci. Ind. Res. India 70(4), 278-283.
Karagoz, U., Akyildiz, M., and Isleyen, O. (2011). “Effect of heat treatment on surface roughness of thermal wood machined by CNC,” Proligno 7(4), 50-58.
Li, T., Cai, J.-b., Zhou, D.-G. (2013). “Optimization of the combined modification process of thermo-mechanical densification and heat treatment on Chinese fir wood,” BioResources 8(4), 5279-5288. DOI: 10.15376/biores.8.4.5279-5288
Mitchell, P. H., and Lemaster, R. L. (2002). “Investigation of machine parameters on the surface quality in routing soft maple,” Forest Prod. J. 52(6), 85-90.
Muhcu, S., Nemli, G., Ayrilmis, N., Bardak, S., Baharoğlu, M., Sarı, B., and Gerçek, Z. (2015). “Effect of log position in European larch (Larix decidua Mill.) tree on the technological properties of particleboard,” Scand. J. Forest Res. 30(4), 357-362.
Nalbant, M., Gokkaya, H., and Sur, G. (2007). “Application of Taguchi method in the optimization of cutting parameters for surface roughness in turning,” Mater. Design 28(4), 1379-1385. DOI: 10.1016/j.matdes.2006.01.008
Ohuchi, T., and Murase, Y. (2005). “Milling of wood and wood-based materials with a computerized numerically controlled router IV: Development of automatic measurement system for cutting edge profile of throw-away type straight bit,” J. Wood Sci. 51(3), 278-281. DOI: 10.1007/s10086-004-0663-x
Sandak, J., and Tanaka, C. (2003). “Evaluation of surface smoothness by laser displacement sensor – 1: Effect of wood species,” J. Wood Sci. 49(4), 305-311. DOI: 10.1007/s10086-002-0486-6
Patel, D. H., and Patni, V. N. (2014). “An investigation effect of machining parameters on CNC router,” International Journal of Engineering Development and Research 2, 1583-1587.
Sofuoglu, S. D. (2015). “Using artificial neural networks to model the surface roughness of massive wooden edge-glued panels made of scotch pine (Pinus sylvestris L.) in a machining process with computer numerical control,” BioResources 10(4), 6797-6808. DOI: 10.15376/biores.10.4.6797-6808
Sutcu, A. (2013). “Investigation of parameters affecting surface roughness in CNC routing operation on wooden EGP,” BioResources 8(1), 795-805. DOI: 10.15376/biores.8.1.795-805
Sutcu, A., and Karagoz, U. (2012). “Effect of machining parameters on surface quality after face milling of MDF,” Wood Research 57(2), 231-240.
Taguchi, G., Elsayed, E., and Hsiang, T. (1989). Quality Engineering in Production System Taguchi,McGraw-Hill, New York.
Valarmathi, T. N., Palanikumar, K., and Sekar, S. (2013). “Prediction of parametric influence on thrust force in drilling of wood composite panels,” IJMMME 1(1), 71-74.
Verma, A. S., Suri, N. M., and Kant. S. (2012). “Effect of process parameter of AL-6063 based fly ash composites using Taguchi,” International Journal of Applied Engineering Research 7(11), 1856-1859.
Wilkowski, J., Czarniak, P., and Grześkiewicz, M. (2011). “Machinability evaluation of thermally modified wood using the Taguchi technique,” COST Action FP0904 Workshop “Mechano-chemical transformations of wood during Thermo-Hydro-Mechanical (THM) processing, pp. 109-111.
Wilkowski, J., Rousek, M., Svoboda, E., Kopecky, Z., and Czarniak, P. (2013). “Analysis of the influence of cutting parameters on surface roughness of milled wood based on Taguchi techniques,” Ann. Warsaw Univ. Life Sci. – SGGW, Forestry and Wood Technol. 84, 321-325.
Zhong, Z.W., Hiziroglu, S., and Chan, C. T. M. (2013). “Measurement of the surface roughness of wood based materials used in furniture manufacture,” Measurement 46(4), 1482-1487. DOI: 10.1016/j.measurement.2012.11.041
Article submitted: July 14, 2015; Peer review completed: September 19, 2015; Revised version received and accepted: September 29, 2015; Published: October 2, 2015.
DOI: 10.15376/biores.10.4.7772-7781