Determination of material strength and cutting loads in chipboard milling ? -D. Bouzakis1, G. Koutoupas1, A. Papapanagiotou2, P. ? ikolakakis2, G. ? alamaras3, Y. Venieris3 1. Laboratory for Machine Tools and Manufacturing Engineering, Mehanical Engineering Department, Aristoteles University of Thessaloniki, 54006, Greece 2. DROMEAS S.A., Industrial Area of Serres, 62121, Greece 3. SHELMAN S.A., Industrial area of Vasiliko, Halkis, Greece

ABSTRACT. An alternative method to determine the strength of chipboards has been developed. The developed method provides an accurate estimation of the fracture stress of chipboards through evaluation of measured cutting forces in chipboard milling. The variation in the material properties across the chipboard was taken in to account, so that the method would provide information regarding not only the whole of the structure, but also the individual chipboard layers chipboards consist of. The chip formation/removal cycle in chipboard milling was investigated, by means of a FEM simulation of the machining process. A series of experiments including measurements of cutting forces in chip board milling and chipboard material properties determination by standardised bending tests, was carried out, in order to optimise the FEM simulation model and to verify the results obtained by the developed methodology. In this way, the chip flow and the occurring cutting force fluctuations were elucidated, and herewith the chipboard rapture stress was correlated to stress distributions in the contact area between the tool and the workpiece material, at distinguished cutting edge revolving positions. Finally, using the methodology described in the present paper, chipboard mechanical properties for various chipboard layers and the distribution of cutting loads on the tool rake were deduced. With the aid of easy to determine through experiments parameters, the chipboard strength can be identified. KEYWORDS: Chipboard milling, cutting forces, chip formation, chipboard material properties

1. Introduction Chipboards are widely used in furniture industry, being a cheap solution in applications of lightweight panels. One of the most common types of machining for chipboards is milling. Cutting forces in chipboard milling are relatively lower than the forces in metal cutting. Nevertheless, due to the fact that these forces are dynamic, can be critical for the milling machine and tool life, especially when coated cutting inserts that enhance the performance of the tool are used. [TOE 99] [KLO 99][BOU 99a] [BOU 98c] Standardised tests to determine the mechanical properties of chipboards consist mainly of tensile and bending tests [DIN 93]. These tests however, do not take account of chipboard’s high inhomogeneity and fail to provide informations regarding technological data of the material such as expected cutting forces, chip formation etc. An experimental – numerical procedure for the determination of the size and the distribution of the cutting forces on the cutting tool is presented in the present paper. Moreover, it is possible to accurately estimate the chipboard strength and obtain valuable information regarding the machinability of such materials, by evaluating measured cutting forces, using simple mathematical formulas (see figure 1).

Figure 1. Flow chart of the followed procedure

2. Cutting force measurements Chipboards are not homogenous along their thickness. They consist of three easily identifiable layers: top, bottom, and middle. Top and bottom layers are dense, small grained and stronger than the middle layer, which is large grained, porous and

weak. Usually, chipboards are coated with a thin hard layer of melamine for decorative purposes (figure 2). This feature of chipboards affects the occurring cutting forces in a manner qualitatively shown at the top part of the figure. The shape of the cutting forces distribution across the chipboard explains why chipboard milling cutting inserts wear more where the dense layer and the bulk of melamine cutting takes place [BOU 00] [BOU 99b].

Figure 2. Typical chipboard structure and cutting forces Cutting force measurements where carried out by means of a piezoelectric dynamometer, while milling chipboard specimens of full thickness and individual layers, so that the special material characteristics of each layer could be taken into consideration (figure 3). Thus it was possible to quantify the effect of the different material properties of each layer on the cutting loads and determine the cutting force distribution along the tool cutting edge. As shown in the same figure, the measured cutting forces were analyzed to components used to optimize and verify the FEM model of the milling process simulation. This is described in the following paragraph.

Figure 3. Cutting force measurement and transformation in c and t directions A fluctuation of the registered cutting force signal can be observed [BOU 98a] [BOU 98b]. This is due to the mechanism of chip formation and not due to the machine’s dynamic response, because a fluctuation of size that has been observed

would mean that the tool vibrates with an amplitude of a few hundred microns, which is obviously not the case. The shape of the fluctuation regarding its width and height, depends on material properties of the workpiece and the cutting conditions (rake angle, feedrate, cutting speed, etc.).

3. FEM simulation of the chip flow mechanism The model used in the FEM simulation of the milling process is illustrated in figure 4 [BOU 99b]. Due to symmetry conditions a 2-D and plain stress element formulation was used. Furthermore, in order to keep computing times low, the development of the original workpiece region ABCDEF was meshed, so that the circular tool path BE could be simulated by the straight line B’E’. In this way tool and workpiece intersection forms a chip with variable thickness, just as in the actual milling process.

Figure 4. FEM model for the simulation of chip formation mechanism Due to the fact that the HM tool Elasticity Modulus is significantly higher than that of the chipboard, the tool was assumed to be rigid, consisting of small tool segments. Cutting force calculation could take place in each one of them, thus making it possible to calculate its distribution along the tool rake face. The workpiece material model was assumed bilinear. Its Elasticity Modulus and yield

stress were obtained by standardised bending tests [DIN 93]. The material tangent modulus was determined using an experimental-numerical method [BOU 99c], while the fracture stress of the material was optimised so that the calculated cutting forces converged to the measured ones (figure 5).

Figure 5. Calculated and measured cutting forces and their correlation with the formed chips

Using the FEM model for the different layers of the chipboard, the stresses on the chipboard and the force distribution on tool rake face were ascertained. Typical simulation results of the developed stress distribution and the formed chips when milling different layers of the chipboard are presented in figure 6.

Figure 6. Stress distribution and formed chips in different cases of chipboard milling As mentioned above, a convergence in forces proves an accurate chip formation and flow mechanism simulation. Each peak and valley on the cutting force signal corresponds to a breakage of the flowing chip. The force fluctuation in this milling case depends on the chip formation mechanism. In the beginning of the fluctuation period, the chip is compressed causing an increase in the cutting force until the material exceeds its compressibility limits and breaks. At that moment, the force has reached a maximum value and then decreases until the tool recovers contact with the undeformed material, where the chip formation cycle starts all over again (see figures 7, 8).

Figure 7. Chip formation mechanism in top chipboard layer milling

Figure 8. Chip formation mechanism in middle chipboard layer milling As indicated by the calculated cutting force distribution, at the beginning of the chip formation cycle, the chip thickness is smaller than the determined by the feedrate nominal thickness (fz=0.5 mm). Moving on to the compression phase, the chip regains its nominal dimensions, as the distribution of the cutting force becomes

triangular. When the cutting force reaches its highest value, the distribution gets flatter because of the high compression of the workpiece material, and wider, due to the movement of the chip towards the tool’s axis of revolution. After that, the chip breaks and is being removed, causing a relocation of the force application area (see figures 7, 8). As illustrated in figure 5, the dense layer of the chipboard produces a greater number of chips than the middle porous layer, because it exceeds its compression limits sooner and therefore braking more frequently. For better comprehension of the chip formation timing, the model is presented in its original undeformed state. The blank areas represent elements that failed under loading, and so were excluded from any further calculation [LST 97]. Depending on the material properties, the fluctuation height and width varies, being both wider and higher when the material is more compressible (middle chipboard layer), therefore allowing a grater percentage increase of the cutting force. Moreover, due to high compression, the breakage of the chip is rather irregular, leaving a real undeformed chip thickness often much smaller than the nominal one, thus causing a greater cutting force drop (see figure 9).

Figure 9. Comparison of the cutting force fluctuation between middle and top chipboard layer The results also showed that the load distribution due to cutting forces on the tool rake face verges towards a triangular distribution in the region where melamine

cutting takes place, while it gets slightly flatter in the areas of sparse and dense layer cutting (see figure 10).

Figure 10. Cutting force distribution on the rake face of the tool

4. Determination of chipboard strength through cutting force evaluation An algorithm for the determination of chipboards fracture strength was developed, based on the results obtained by both, FEM simulation of the milling process, and the cutting force measurements. It has been experimentally determined, that cutting forces in chipboard milling can be described by the following formula: Fch = k b h1-mc Where: Fch: characteristic cutting force in the normal to the tool rake direction, approximately at a 90o revolving position, defined as sescribed bellow b: chip width h: chip thickness k, 1-mc : experimentally determined material coefficients In order to determine the chipboard fracture strength, the following procedure was followed. Individual chipboard layers are milled for various cases of maximum chip thickness, while the cutting forces are measured. From the registered signal, the characteristic cutting force Fch is extracted, as the value of the mean line of the cutting force fluctuation, for a tool rotation angle of 90o (see figure 11).

Figure 11. Determination of the characteristic cutting force Fch

The fluctuation mean line is the line that connects the values of the cutting force occurring during the compression phase of the chip formation cycle (phases D2 and S2 in figures 7 and 8 correspondingly). Approximately, this mean value occurs at the middle of the fluctuation height. At that point, as the FEM analysis of the chip formation mechanism showed, the chip has a thickness close to the nominal one, and the workpiece material yields at a shear plane, under the triangularly distributed cutting load. The shear plane is illustrated in figure 12, for various chipboard layers.

Figure 12. Stress distribution and shear plane during the chip compression phase

Once the characteristic cutting force Fch has been calculated for a number of different maximum chip thicknesses (e.g. for different feedrates), the coefficient 1mC is determined as shown for the dense layer of an investigated hcipboard, in figure 13. k = Fch / b h1-mc

Figure 13. Determination of the coefficient 1-m c

The coefficient k can then be obtained as the cutting force devided by a chip cross section of 1x1 (mm2). It has been experimentally ascertained that coefficient k provides an accurate approximation of the material’s rapture strength, as that was determined by means of standardised bending tests. Coefficient k represents herewith the compressive load that causes shearing of the material in the simplified consideration of an elementary cubic volume. In order to achieve production monitoring it is possible to simplify the experimental arrangement bypassing expensive instruments such as a dynamometer, by measuring the cutting forces indirectly, by means of an amperometer that registers the power consumption of the spindle motor. Using the registered force signal, material characteristics such as strength can be monitored in real time under production conditions.

Typical results of the analysis in comparison to results obtained by standardised bending tests are listed in the following table:

Chipboard 1

2

Layer Full thickness Top / Bottom Middle Full thickness Top / Bottom Middle

Tensile strength [daN/mm 2] Developed Standardized Method bending tests 2,38 2,1 2,9 2,7 1,3 1,33 2,62 2,32 3,53 3,43 1,02 0,96

Table 1. Typical results of the proposed methodology in comparison to results obtained by standardised bending tests

5. Conclusions An experimental –numerical analysis involving cutting force measurements and a FEM simulation of the chip formation mechanism in chipboard milling was conducted. The obtained results indicated that the form and size of the cutting force signal depend amongst others on the material properties of the workpiece, and elucidated the chip flow mechanism and the cutting force fluctuations. It has been also shown that the chipboard strength can be accurately estimated by evaluation of measured cutting force in chipboard milling. Based on the results of the analysis, a new methodology to determine the rapture strength of chipboards, taking into account the different characteristics of the individual chipboard layers, has been proposed in this paper. The proposed methodology can be utilized in monitoring the production and machining parameters of chipboards. Finally, the conducted analysis made it possible to determine the cutting force distribution on the cutting tool, and correlate the it to the occurring stress field in the tool-workpiece contact area.

6. References [BOU 00] BOUZAKIS, K-D., KOUTOUPAS, G., SIGANOS, A., LEYENDECKER, T., ERKENS, G., PAPAPANAGIOTOU, A., NIKOLAKAKIS, P. “Improvement of cutting performance of PVD coated cemented carbide inserts in chipboard milling considering chip formation analytically described through a FEM simulation”, [BOU 99a] BOUZAKIS, K-D., VIDAKIS, N., DAVID, K., “The concept of an advanced impact tester supported by evaluation software in characterization of hard layer media”, Thin Solid Films, vol 355/356, p 322-329, 1999 [TOE 99] TOENSCHOFF, H.-K., MOHLFELD, A., SPENGLER, C., PODOLSKY, C., “PVDcoated tools for metal cutting applications”, Proceedings of 1 st int. conf. “THE coatings”, October 1999, Thessaloniki, p.1, 1999 [KLO 99] KLOCKE, F., “Coated tools for metal cutting – features and applications”, Key note paper, Annals of the CIRP, vol.48, p.515, 1999 [BOU 99b] BOUZAKIS, K-D., KOUTOUPAS, G., SIGANOS, NIKOLAKAKIS, P., ““Improvement of cutting performance of PVD coated cemented carbide inserts in chipboard milling considering the chip formation”, Proceedings of 1 st int. conf. “THE coatings”, October 1999, Thessaloniki, p.191, 1999 [BOU 99c] BOUZAKIS, K.-D., VIDAKIS, N., “Superficial plastic response determination of hard isotropic materials using ball identations and a FEM optimization technique”, Int. Journalof Materials Characterization, vol.42, p.1-12, 1999 [BOU 98a] BOUZAKIS, K-D., KOUTOUPAS, G., “Determination of chipboard strength through cutting force evaluation”, Proceedings of DAAM International ’98, Cluj-Napoca, Romania, 1998 [BOU 98b] BOUZAKIS, K-D., KOUTOUPAS, G., “Determination of material properties of chipboard layers evaluating cutting force conponents”, Proceedings of WCTE ’98, Laussane, Switzerland, 1998 [BOU 98c] BOUZAKIS, K-D., VIDAKIS, N., KALINIKIDIS, D., LEYENDECKER, T., LEMMER, O., FUSS, H.G., ERKENS, G., “Fatigue failure mechanisms of multi- and monolayer physically vapour deposited coatings in interrupted cutting processes”, Surface and Coatings Technology, vol.108/109, p.526-534, 1998 [LST 97] LIVERMORE SOFTWARE TECHNOLOGY CORPORATION, “LS-Dyna, User’s Manual, Version 940”, 1997 [DIN 93] DIN EN 310, “Holzwerkstoffe (Bestimung des Biege-Elastizitatsmodulus und der Biegefestigkeit)”, 1993