Improved model for estimating the expected roughness in dry turning of magnesium UNS M11311.

De Pipaon, Jose Saenz ; Rubio, Eva ; Villeta, Maria 等

1. INTRODUCTION

A problem with which is encounter in different fields of industry such as transportation is the need to reduce energy consumption by economic and environmental reasons (Ballerini et al., 2001). A way of achieving this is reducing the weight by the use of lighter components made of light alloys (mainly aluminium, titanium and magnesium) because they have an excellent weight resistance ratio. Of the light alloys mentioned, the magnesium is the one that presented a lower density (1740 kg/[m.sup.3]). However, magnesium presents problems with the heat generated in the machining process, since it has a tendency to be flammable. In process of machining pieces of magnesium and in particular in pieces with inserts of steel or stainless steel is especially critical due to both the heat generated during the machining process that can make the magnesium burn, and the sparks generated in the machining of steels at cutting speeds up to 200-300m/min that can cause the ignition of chips or dust of the magnesium.

In industries, such as the aeronautical and motor, the differents machining pieces must fulfill strict requirements for surface finish (usually 0.8[micro]m<Ra<L6[micro]m). The surface roughness obtained in a machining operation can be considered as the sum of independent effects. First of all is the result of the geometry of the tool and the feed used, but also depends on other variables such as cutting speed, depth of cut or rake angle (Woldman & Gibbons, 1951) and the irregularities appeared during the process (mainly, the built-up-edge (BUE) but, as well, vibrations of the machine tool; inaccuracy in the movements of the machine due to irregularities and defaults in the achievements, heat expansions and so on; plastic deformation of the material; friction; formation of discontinuity chips; failure of the work material; or surface damage caused by the chip flow) (Boothroyd, 1998).

The widespread expression to estimate the surface roughness in turning processes using tools with round nose comes represented by equation 1.

[Ra.sub.ideal] = 0.032[f.sup.2]/r (1)

[Ra.sub.ideal] is the surface roughness ideal, f is the feed and r is the nose radius of the used tool in the mechanized.

This experimental study is focused on establishing a model for estimating the surface roughness, (in terms of Ra) of bars of magnesium UNS M11311 obtained by dry turning that collects the influence of others cutting parameters (besides of the feed, f such as the cutting speed, v, and depth of cut, d) and type of tools, T, used.

2. METHODOLOGY

The work follows the general methodology developed for the whole project whose main steps are describe in the next paragraphs (Saenz de Pipaon et al., 2008):

* Previous activities to machining process. These activities consist of the design of experiments, the preparation of the test material and the protocols to calculate the cutting parameters values and to registry data and observations.

* Turning tests. In these tests workpieces of different light alloys are mechanized under certain conditions of feed, cutting speed, depth of cut and types of tools.

* Monitoring processes. In order to get graphics documents that can be analyzed after the process, all the turning tests described before have been photographed and recorded by video and both the chips obtained and tools used have been photographed with a camera of high resolution.

* Roughness measurement. Measurements of the surface roughness have been made using a surface roughness tester in three generatrices separated one from each other 120 and denoted by G1, G2, G3. In each one of them, the roughness, in terms of Ra, has been measured in four different sections of the length of the workpiece denoted by L1, L2, L3, L4.

* Data processing and analysis of results. The data thus obtained have been treated with statistical techniques according to the design of experiments used (Montgomery, 1991) (Taguchi, 1987). Specifically by fractional factorial orthogonal designs (Taguchi L27).

Three potential factors of design that influence the process have been identified, two are quantitative, cutting speed, v, and feed speed, f, and one qualitative, the type of tool coating, T. It is suspected that these factors interact with each other.

Previous studies suggest that the surface roughness can vary depending on the length of the workpiece, L, (Rubio et al., 2009) and the type of cutting tool used. Therefore, the factor L also should be included in the design. Besides, as the roughness is measured in the direction of the generatrices of the piece, G, the design also should be take into account this factor. Table 1 shows the factors and their levels fixed for them.

3. APPLICATIONS

For this study, the workpieces used in the turning tests were cylindrical bars with a diameter of 40mm and length of 125mm (useful 100mm) of magnesium alloy UNS M11311.

The machining tests have been made for a depth of cut d = 0.25 mm. Three different types of tool, from SECO manufacturer, with identical geometry and different coatings have been used. Concretely, one specifically for non ferrous metals and two for steels with a coating of Ti(C,N)+[Al.sub.2][O.sub.3]+TiN. The manufacturer references are: HX, TP200 and TK2000 respectively.

The cylindrical bars were dry turned on an EMCO Turn 120 CNC lathe equipped with an EMCO Tronic T1 numerical control.

To observe the machining tests carried out, videos and photographs of the tools were systematically taken during the tests using a Sony Cibershot DSC-P100 digital camera of high resolution.

To measure the roughness of the workpiece were used a measuring surface roughness tester Mitutoyo Surftest SJ-401. The roughness was measured on three generatrices separated 120[degrees] in four sections [L.sub.1]=0-25mm, [L.sub.2]=25-50mm, [L.sub.3]=50-75mm and [L.sub.4]=75-100mm.

4. RESULTS

After making the analysis of variance of the data, the variability in the roughness expected ([Ra.sub.expected]) is modelled by equation (2). Expression that allows establishing a ranking for the best combinations of cutting parameters and tool coatings (Rubio et al., 2009).

[y.sub.i,j,k] = [mu] + [v.sub.i] + [f.sub.j] + [T.sub.k] + [(vf).sub.i,j] + [(vT).sub.ik] + [(fT).sub.j,k] + error (2)

From the analysis of the variance it is possible to affirm that: The feed is the cutting parameter that affects more strongly (75%) to the roughness expected of the variability of the equation 2. The cutting speed, v, the tool, T, and their interaction v-T affect but less. L and G are not statistically influential factors.

Model shown in this work take into account, in addition to the feed, f, others variables such as cutting speed, v, and tool coating, T.

The expected surface roughness given by the new model, [Ra.sub.expected], will be, in general, equal or higher than the [Ra.sub.ideal] defined before, since it is considered a great number of parameters. Figure 1 shows the [Ra.sub.expected] given by the equation 2 versus the [Ra.sub.ideal] given by the equation 1 for the tool TP200.

5. CONCLUSIONS

The work proposes an improved statistical model for estimating the expected surface roughness in dry turning of bars of magnesium UNS M11311.

Such model is coherent with others found in the classical references about the theme but considers, in addition to the influence of the feed, f, others variables such as cutting speed, v, and tool coating, T, and their interaction v-T.

Besides, it can provide a ranking for the best combinations of cutting parameters and tool coatings based on the surface roughness expected. This ranking allows selecting the best cutting conditions and tools to achieve a certain range of surface roughness in terms of Ra.

The main limitations of this research are the values of the cutting conditions used; specially low in comparison with the usually values of these parameters used in the production of the magnesium but adequated for repair operations of hybrid structures such as magnesium-steel or magnesium-aluminium; usually used in aeronautical or automotive industries.

[FIGURE 1 OMITTED]

6. ACKNOWLEDGMENTS

Funding for this work was provided in part by the Spanish Ministry of Science and Innovation (Directorate General of Research), Project DPI2008-06771-C04-02.

7. REFERENCES

Ballerini, G.; Bardi, U.; Lavacchi, A. & Migliorini D. (2001). Magnesium alloys for Structural automotive applications, Proceedings of the 7th International Conference HTCES, Modena (Italy), May-June 2001

Boothroyd, G. (1978). Fundamentos del corte de metalesy de las Maquinas-Herramienta, McGraw-Hill Latinoamericana, ISBN: 968604658-5, Mexico

Montgomery, D. C. (1991). Design and analysis of experiments, John Wiley & Sons, Inc., ISBN: 0471-520004, New York

Rubio, E.M., Saenz de Pipaon, J.M., Villeta, M., Sebastian, M.A., (2009). Experimental study for improving repair operations of pieces of magnesium UNS M11311 obtained by dry turning, Proceedings of the 12th CIRP Conference on Modelling of Machining Operations, Universidad de Mondragon (Ed), pp. , Mayo 2009, San Sebastian, Spain

Saenz de Pipaon, J. M.; Rubio, E. M.; Villeta, M. & Sebastian, M. A. (2008). Influence of cutting conditions and tool coatings on the surface finish of workpieces of magnesium obtained by dry turning, Proceedings of 19th International DAAAM Symposium, Katalinic, B. (Ed.), pp. 609-610, ISBN: 978-3-901509-68-1, Trnava, October 2008, DAAAM Int. Vienna, Vienna

Taguchi, G. (1987). System of experimental design, American Supplier Institute, Vol. 2, ISBN: 0-527-91621-8, New York

Woldman, N. E. & Gibbons, R. C. (1951). Machinability and machining of metals, McGraw-Hill, New York

Tab. 1. Factors and levels for the experimental design

Factors Levels

 1 2 3 4

Tool coating HX TP200 TK2000
v (m/min) 75 150 225
f (mm/rev) 0.05 0.10 0.15
L (mm) 0-25 25-50 50-75 75-100
G ([degrees]) 0 120 240