Minimizing Heat Deformation Caused in Metal Cutting

MINIMIZING HEAT DEFORMATION CAUSED IN METAL CUTTING

Jong-Yun Jung

Hengbo Cui

Xiangyu Hou

Dug-Hee Moon

Industrial & Systems Engineering

ChangwonNationalUniversity

Changwon, Gyeongnam, Korea

Choon-Man Lee

Mechanical Design & Manufacturing Engineering

ChangwonNationalUniversity

Changwon, Gyeongnam, Korea

Abstract.An experimental design and performance analysis of heat deformation in machining processes are given in this paper. High-speed machining processes generate severe heat, deformation is generated especially for thin parts. Deformed parts are detected distinctively. The amount of deformation of a thin part is mainly dependent on heat and the shape of a component. The selection of optimal parameters decreases the heat that rises during the cutting period. This paper describes the selection of machining parameters to minimize the heat-deformation. The experimental design of using the Taguchi method is employed to optimize the cutting parameters. Three factors and three levels for each factor are designed using Taguchi method. The effects of machining parameters to minimize the heat deformation are analyzed using analysis of variance (ANOVA) and the signal to noise ratio (S/N).

1INTRODUCTION

For the need of aircraft components, weight and strength are the basis features for the materials. Hardened and heat treated aluminum or titanium alloy is good for the component because they are weight and strength. Most of the Aircraft products are manufactured using machining processes. High-speed end milling is one of the frequently used processes in metal working areas. During machining processes, work-pieces especially for thin parts are easily deformed because heat generated is high and the melting point of the material is relatively low. Under these conditions, some problems rise in cutting thin aluminum alloys. The most serious problem is that parts are deformed severely by heat[1].

Due to the problem above, the selection of optimal parameters in machining is important because they can decrease the temperature. However, the desired machining parameters are determined based on experience or by use of a hand book in machining shop, but this does not mean that the selected machining parameters are optimal or near optimal for the machining performance.

This research presents the selection of optimal machining parameters using Taguchi method that reduces the number of experiments. Two types of specimen are designed for the experiments. This paper draws conclusions by analyzing the experiments and comparing the specimens.

2MACHINING CONDITIONS

For this study, in order to obtain a good deformation results in end-milling operations with flat end mills, cutting tool, and cutting conditions are well selected. The setup of the process is important. Figure 1 shows thespecimen which is carried nine different experiments and Figure 2 showsspecimen for the second set of experiments. In each experiment, different sets of parameters are selected to test the deformed level. The material used as work-piece is AL 7050/T7451 that is widely used for aircraft components. The chemical compositions of specimen material are shown on Table 1 and the mechanical properties of specimen material are shown on Table 2. The specimens are machined in a five axis high-speed vertical machining center manufactured by European company. The specification of the machine tool is shown in the following Table 3. The cutting tool is a flat end-mill and TiN coated carbide with four flutes. Its diameter is 16mm. A Coordinate Measuring Machine (CMM) is used to measure the specimens, with the specification: X stroke 1.5m, Y stroke 2.2m, Z stroke 1.0m; error: 2.8+L/333m, which is shown in Figure 4.Also, surface roughness tester is used to measure the surface roughness of the specimens. The tester machine is shown in Figure 5 [2].

Table 1.Chemicalcompositions of specimen material.

C / Wt. % / C / Wt. %
AL / 87.3-90.3 / Mg / 1.9-2.6
Cr / Max 0.04 / Mn / Max 0.1
Cu / 2-2.6 / Si / Max 0.12
Fe / Max 0.15 / Ti / Max 0.06
Zn / 5.7-6.7 / Zr / 0.08-0.15
Other / Max 0.1

Table 2.Mechancial properties of specimen material.

Density(lb / cu. in.) / 0.102
Hardness Brinell / 140
Electrical Resistance(ohm-cm ) / 4.3e-006
Modulus of Elasticity / 10400ksi
Ultimate Tensile Strength / 76000psi

Table 3. Specification of the machine tool.

Work Area (mm) / X: 750
Y: 600
Z: 520
Work Spindle / SK 40
SpeedRange (rpm) / 30,000
Tool Changer, number of tools / 32
FeedRange (mm/min) / 20,000
Rapid traverse,linear axes(m/min) / 50

Fig. 1. Specimen for first design.Fig. 2. Specimen for second design.

Fig. 3. Machine tool for the experiments.

Fig. 4. Coordinate measuring machine.Fig. 5. Surface roughness tester.

The side walls are used for the measuring the deformation. From the Figure1 and 2, ten points marked with in two rows is shown in the side wall, these are two groups of measuring points for this experimental design. Twenty points (four groups) are defined to be measured, ten points are shown in the front side and the other ten points are on the back side to these ten points.

3TAGUCHI METHOD

A large number of experiments have to be carried out when the number of the process parameters are increased. In the experiments, the parameter design proposed by the Taguchi method is adopted. The use of Taguchi method reduces the number of cutting experiments for design optimization of the machining parameters. Results of the cutting experiments are analyzed using the S/N and ANOVA analysis.

3.1Application of Taguchi Method

Taguchi method uses a special design of orthogonal arrays to study the entire parameter space with only a small number of experiments. Cutting speed, feed rate, and depth of cut are selected as the cutting parameters. Three levels of each factor are shown in the following Table 4, an L9 orthogonal array with four columns and nine rows is used. Each cutting parameter is assigned to a column, nine cutting parameter combinations being available. This array has eight degrees of freedom and it handles three-level design parameters. Each cutting parameter is assigned to a column, nine cutting-parameter combinations being available. The experimental layout using an L9 orthogonal array is listed in Table 5.

Table 4.Factors and selected levels.

Factors / Level1 / Level2 / Level3
Speed(rpm) / 5000 / 10000 / 15000
Feed(mm/tooth) / 0.05 / 0.1 / 0.15
DOC(mm) / 0.3 / 0.65 / 1.0

Table 5. Experimental layout using an L9orthogonal array.

No / Cutting parameter level
A / B / C
Speed / Feed / Depth of cut
1 / 1 / 1 / 1
2 / 1 / 2 / 2
3 / 1 / 3 / 3
4 / 2 / 1 / 2
5 / 2 / 2 / 3
6 / 2 / 3 / 1
7 / 3 / 1 / 3
8 / 3 / 2 / 1
9 / 3 / 3 / 2

3.2Analysis of the S/N ratio

In the Taguchi method, the signal to noise ratio is the ratio of the mean to the standard deviation. Taguchi uses the S/N ratio to measure the quality characteristic deviating from the desired value [3]. Table 6-13 show the experimental results for S/N ratio calculated by Minitab. Table 6, 7, 8 and 9 give the results of S/N ratio in the first experimental design. Table 10, 11, 12 and 13 give the results of S/N ratio in the second experimental design. Delta means the absolute value between Max. and Min. S/N ratio of each machining parameter [4]. The bigger the delta of that parameter is, the bigger the deformation is strongly influenced by that parameter. From the tables, the cutting speed has the most important effect on heat deformation in the first experimental design, but in the second one, all the selected parameters have a significant effect on heat deformation.

Table 6. Response table for average S/N ratio (top-front points).

Cutting Factor / S/N ratio of each level / Delta
1 / 2 / 3 / Max-min
Speed / 36.32 / 30.96 / 32.26 / 5.36
Feed / 33.07 / 31.18 / 35.29 / 4.11
DOC / 31.39 / 32.50 / 35.65 / 4.26

Table 7. Response table for average S/N ratio(bottom-front points).

Cutting Factor / S/N ratio of each level / Delta
1 / 2 / 3 / Max-min
Speed / 33.67 / 29.19 / 24.87 / 8.80
Feed / 26.65 / 27.95 / 33.13 / 6.47
DOC / 32.64 / 26.58 / 28.51 / 6.06

Table 8. Response table for average S/N ratio(top-back points).

Cutting Factor / S/N ratio of each level / Delta
1 / 2 / 3 / Max-min
Speed / 24.09 / 21.31 / 21.00 / 3.09
Feed / 20.71 / 22.53 / 23.16 / 2.44
DOC / 22.54 / 21.39 / 22.46 / 1.15

Table 9. Response table for average S/N ratio(bottom-back points).

Cutting Factor / S/N ratio of each level / Delta
1 / 2 / 3 / Max-min
Speed / 26.73 / 24.22 / 27.30 / 3.08
Feed / 26.02 / 25.29 / 26.94 / 1.65
DOC / 23.39 / 25.76 / 29.12 / 5.73

Table 10. Response table for average S/N ratio(top-front points)

Cutting Factor / S/N ratio of each level / Delta
1 / 2 / 3 / Max-min
Speed / 23.70 / 25.10 / 24.24 / 1.40
Feed / 25.11 / 24.47 / 23.47 / 1.64
DOC / 23.88 / 25.12 / 24.04 / 1.24

Table 11. Response table for average S/N ratio(bottom-front points).

Cutting Factor / S/N ratio of each level / Delta
1 / 2 / 3 / Max-min
Speed / 23.69 / 24.84 / 24.40 / 1.15
Feed / 25.32 / 24.40 / 23.20 / 2.12
DOC / 23.87 / 25.23 / 23.83 / 1.40

Table 12. Response table for average S/N ratio(top-back points).

Cutting Factor / S/N ratio of each level / Delta
1 / 2 / 3 / Max-min
Speed / 25.86 / 27.02 / 25.78 / 1.23
Feed / 27.66 / 26.13 / 24.87 / 2.79
DOC / 25.56 / 27.63 / 25.46 / 2.17

Table 13. Response table for average S/N ratio(bottom-back points).

Cutting Factor / S/N ratio of each level / Delta
1 / 2 / 3 / Max-min
Speed / 36.85 / 38.41 / 41.13 / 4.28
Feed / 38.38 / 39.62 / 38.39 / 1.25
DOC / 39.26 / 37.94 / 39.19 / 1.32

3.3Analysis of variance (ANOVA)

One of the methods to analyze data for process optimization is the use of the analysis of variance (ANOVA) [5]. Table 14 to 21 show the results of ANOVA for heat deformation which are analyzed by the Minitab software. The first four tables are analyzed based on the first specimen, the last four are based on the second specimen. The Seq SS of cutting speed is the highest in the analysis tables except for the ANOVA of bottom-back points in the first research. In the second research, feed rate is the most important factor for the heat deformation in front surface and cutting speed has a significant effect on the heat deformation in back surface.

Figure 6 to 13 show the mean effects plots for the heat deformation. The dots in the figures express the results of the deformation. It is also found that the dots in cutting speed columns have the fluctuations of the deformation. Therefore, in the first study, cutting speed is the most significant machining parameters affecting the heat deformation. The change of the feed rate has an insignificant effect on heat deformation. In the second study, there is an optimal parameters can be selected, that is cutting speed 10,000 rpm, feed rate 0.05mm/tooth, depth of cut 0.65mm.

Fig. 6. The mean effects plot for deformation (top-front). Fig. 7. The mean effects plot for deformation (bottom-front).

Fig. 8. The mean effects plot for deformation (top-back).Fig. 9. The mean effects plot for deformation (bottom-back).

Fig. 10. The mean effects plot for deformation (top-front). Fig. 11. The mean effects plot for deformation (bottom-front).

Fig. 12. The mean effects plot for deformation (top-back). Fig. 13. The mean effects plot for deformation (bottom-back).

Table 14. Results of ANOVA for heat deformation (top-front points).

Source DF Seq SS Adj SS Adj MS F P
Speed 2 0.000208 0.000208 0.000104 48.08 0.000
Feed 2 0.000104 0.000104 0.000052 23.92 0.040
DOC 2 0.000112 0.000112 0.000056 25.81 0.037
Error 2 0.000004 0.000004 0.000002
Total 8 0.000428

Table 15. Results of ANOVA for heat deformation (bottom-front points).

Source DF Seq SS Adj SS Adj MS F P
Speed 2 0.002410 0.002410 0.001205 219.1 0.000
Feed 2 0.000776 0.000776 0.000388 70.55 0.001
DOC 2 0.001578 0.001578 0.000789 143.5 0.000
Error 2 0.000011 0.000011 0.000006
Total 8 0.004775

Table 16. Results of ANOVA for heat deformation (top-back points).

Source DF Seq SS Adj SS Adj MS F P
Speed 2 0.001225 0.001225 0.000613 72.06 0.000
Feed 2 0.000739 0.000739 0.000369 43.47 0.003
DOC 2 0.000164 0.000164 0.000082 9.65 0.091
Error 2 0.000017 0.000009 0.000005
Total 8 0.002145

Table 17. Results of ANOVA for heat deformation (bottom-back points).

Source DF Seq SS Adj SS Adj MS F P
Speed 2 0.000602 0.000602 0.000301 25.08 0.018
Feed 2 0.000029 0.000029 0.000014 1.21 0.290
DOC 2 0.001621 0.001621 0.000810 67.54 0.000
Error 2 0.000024 0.000024 0.000012
Total 8 0.002276

Table 18. Results of ANOVA for heat deformation (top-front points).

Source DF Seq SS Adj SS Adj MS F P
Speed 2 0.0001165 0.0001165 0.0000583 32.36 0.003
Feed 2 0.0001799 0.0001799 0.0000899 49.97 0.000
DOC 2 0.0001065 0.0001065 0.0000532 29.58 0.002
Error 2 0.0000036 0.0000036 0.0000018
Total 8 0.0004065

Table 19. Results of ANOVA for heat deformation (bottom-front points).

Source DF Seq SS Adj SS Adj MS F P
Speed 2 0.0000763 0.0000763 0.0000381 12.51 0.056
Feed 2 0.0003124 0.0003124 0.0001562 51.21 0.000
DOC 2 0.0001524 0.0001524 0.0000762 24.98 0.039
Error 2 0.0000061 0.0000061 0.0000032
Total 8 0.0005472

Table 20. Results of ANOVA for heat deformation (top-back points).

Source DF Seq SS Adj SS Adj MS F P
Speed 2 0.0001059 0.0001059 0.0000529 81.46 0.000
Feed 2 0.0000159 0.0000159 0.0000079 12.23 0.054
DOC 2 0.0000051 0.0000051 0.0000025 3.92 0.272
Error 2 0.0000013 0.0000013 0.0000007
Total 8 0.0001282

Table 21. Results of ANOVA for heat deformation (bottom-back points).

Source DF Seq SS Adj SS Adj MS F P
Speed 2 0.0001177 0.0001177 0.0000588 130.8 0.000
Feed 2 0.0000442 0.0000442 0.0000221 49.11 0.000
DOC 2 0.0000061 0.0000061 0.0000031 6.78 0.085
Error 2 0.0000009 0.0000009 0.0000005
Total 8 0.0001689

4CONCLUSION

This research presents the experiments on the heat deformation in the high-speed end milling of the material AL 7050/T7451.

In the first experimental design, from the measurements, the specimens are deformed during the milling practice. Based on the S/N ratio and ANOVA created from analyzing data collected, the factors which are the most effective to the degree of deformation are the cutting speed, the depth of cut, and the feed rate in order. From the main effect plots it can be observed that the lower the cutting speed is the less deformation is attained. The least deformation is derived by the level of cutting speed in 5,000rpm. Also, the higher the DOC is the less the deformation is. But it is displayed not as notable as the cutting speed. The feed rate does a less influence on the deformation.

In the second experimental design, the specimens are also deformed during milling the operation. Based on the S/N ratio and ANOVA, the factors have the similar effectiveness to the degree of cut. From the main effect plots, it is observed that it has a least deformation when the cutting speed is 10,000 rpm, the feed rate is 0.05 mm/tooth and the depth of cut is 0.65mm. From the analysis of variance, it is found that feed rate has a significant effect on the heat deformation in front surface and cutting speed has a significant effect on the heat deformation in back surface.

Therefore, it is clear that different optimal parameters are selected when different shapes of specimen are cut, so experimental design is useful to solve the problem.

The optimal parameters depend on the shape and they are to be selected in accordance with the shape.

ACKNOWLEDGEMENT

This work was supported in part by the Machine Tool Research Center and the MIRAE center of NURI at ChangwonNationalUniversity.

REFERENCES

[1] Zhang, G.M., C. Yu, S.R. Chen, and L.B. An, “Experimental study on the milling of then parts of titanium alloy (TC4)”, Journal of Materials Processing Technology 138(2003) 489-493.

[2] Cui, H.B., J.Y. Jung, and D.H. Moon, “The Selection of Machining Parameters to Minimize Deformation caused by Heat”, Proceedings of the Conference of Society of Korea Industrial and Systems Engineering, Daejon, Korea, (2005).

[3] Lin, T.R., “Experimental design and performance analysis of TiN-coated carbide tool in face milling stainless steel”, Journal of Materials Processing Technology 127 (2002) 1-7.

[4] Ghani, J.A., I.A. Choudhury, and H.H. Hassan, “Application of Taguchi method in the optimization of end milling parameters”, Journal of Materials Processing Technology 145 (2004) 84-92.

[5] Yang, W.H., and Y.S. Tarng, “Design optimization of cutting parameters for turning operations based on the Taguchi method”, Journal of Materials Processing Technology 84 (1984) 122-129.