Experimental Study on Drilling MDF with Tools Coated with TiAlN and ZrN

There is increasing use of wood-based composites in industry not only because of the shortage of solid wood, but above all for their better properties such as: strength, aesthetic appearance, etc., compared to wood. Medium density fiberboard (MDF) is a wood-based composite that is widely used in the furniture industry. The goal of the research conducted was to determine the effect of the type of coating on the drill cutting blades on the value of thrust force (Ft), cutting torque (Mc), cutting tool temperature (T) and surface roughness of the hole in drilling MDF panels. In the tests, three types of carbide drills (HW) were used: not coated, TiAlN coated and ZrN coated. The measurement of both the thrust force and the cutting torque was carried out using an industrial piezoelectric sensor. The temperature of the cutting tool in the drilling process was measured using an industrial temperature measurement system using a K-type thermocouple. It was found that the value of the maximum temperature of the tool in the drilling process depends not only on the cutting speed and feed rate, but also on the type of coating of the cutting tool. The value of both the cutting torque and the thrust force is significantly influenced by the value of the feed rate and the type of drill coating. The effect of varying plate density on the surface roughness of the hole and the variation of the value of the thrust force is also discussed. The results of the investigations were statistically analyzed using a multi-factorial analysis of variance (ANOVA).


Introduction
Medium density fiberboard (MDF) is a wood-based product widely used in the furniture industry [1][2][3][4][5]. MDFs are composed of wood fibers, bonded with formaldehyde glue under the influence of heat and pressure. The use of medium density fiberboards in industry is associated with their machining during furniture production. One of the most commonly used operations in the production of MDF furniture is drilling. The MDFs machinability is determined by the quality of the surface [5,6], which largely depends on the degree of tool wear and the mechanism of chip formation [7]. Various studies have been carried out to improve understanding of MDF cutting characteristics [4,[8][9][10]. Most of the studies are mainly focused on measuring the cutting forces and the friction phenomenon based on the theory used in the cutting process of metals [11,12]. Cutting forces, temperature, and surface roughness that reflect susceptibility to material processing are three important issues in the machining of wood-based materials. Cutting forces have a direct effect on energy consumption, tool wear, heat generation and the quality of the surface machined [13][14][15].
In order to maintain the relatively long service life of cutting tools used for MDF machining, it is necessary to use tools with a high wear resistance. Blades made from HM (ISO Code K20) cemented X-ray tomographs permits one to obtain tomographic images of the object examined, and then present its spatial (3D) image from many flat (2D) images taken in various positions. The computer tomographic (CT) images contain information about the location and density of the absorbing features in the object. Any difference in material density inside the object can be measured and visualised. Figure 1 shows a picture of the cross-section of an MDF panel, in which a distinct variation in plate density can be observed. The highest density occurs in the outer layers of the panel (to a depth of about 2.3 mm). However, a lower density appears in the inner layer (around 13.4 mm in length) ( Figure 1). the cutting output is the temperature created between the cutting tool and the workpiece. In this paper, the effect that the type of coating on the drill cutting blades has on the value of thrust force, cutting torque, cutting tool temperature, and surface roughness of the hole in drilling MDF panels is investigated. Three types of carbide drills (HW) were used in the tests: not coated, TiAlN coated and ZrN coated.

Material
A typical industrial MDF panel with a thickness of 18 mm was used as the workpiece. The mechanical and physical properties of the material being processed are listed in Table 1. An MDF panel is characterized by a clear differentiation of material density in the cross-section, which results from its multi-layered structure. To more accurately characterise the workpiece, a laboratory measurement of the density profile through the panel thickness was carried out using a Phoenix vtomome X-ray tomograph (GE Sensing & Inspection Technologies, Wunstorf, Germany). X-ray tomographs permits one to obtain tomographic images of the object examined, and then present its spatial (3D) image from many flat (2D) images taken in various positions. The computer tomographic (CT) images contain information about the location and density of the absorbing features in the object. Any difference in material density inside the object can be measured and visualised. Figure 1 shows a picture of the cross-section of an MDF panel, in which a distinct variation in plate density can be observed. The highest density occurs in the outer layers of the panel (to a depth of about 2.3 mm). However, a lower density appears in the inner layer (around 13.4 mm in length) (Figure 1). Furthermore, the hardness distribution of the material processed was measured using a Shore hardness tester (Hildebrand, Oberboihingen, Germany) using the Shore D scale (Figure 2b). As can be seen, the hardness distribution ( Figure 2a) is closely related to the density profile. The highest hardness value occurs in the outer layers with a thickness of approximately 2.3 mm and is equal to 62 °Sh (D scale). As we move away from the outer layer, the hardness decreases, reaching a value of 43 °Sh (D scale) at a depth in the range between 7 and 11 mm. Furthermore, the hardness distribution of the material processed was measured using a Shore hardness tester (Hildebrand, Oberboihingen, Germany) using the Shore D scale (Figure 2b). As can be seen, the hardness distribution ( Figure 2a) is closely related to the density profile. The highest hardness value occurs in the outer layers with a thickness of approximately 2.3 mm and is equal to 62 • Sh (D scale). As we move away from the outer layer, the hardness decreases, reaching a value of 43 • Sh (D scale) at a depth in the range between 7 and 11 mm.

Cutting Tools
In the drilling of MDFs, HW sintered carbide cutting tools were used with different types of cutting blade coating. Two types of coating for cutting blades, i.e., TiAlN and ZrN were used. The selection of coatings was dictated by the fact that they belong to those most commonly used in the machining of composite materials [40]. In the case of cutting wood-based materials, there is a general lack of commercially available cutting tools (especially drills) with additional protective coatings. In the research, it was necessary to measure the temperature of the cutting blade during the drilling process. To measure the temperature between the cutting edge and the workpiece using thermocouples, it was necessary for the tool to have coolant channels. However, no cooling liquids are used in the treatment of MDFs. Therefore, it was necessary to fabricate drills. The geometry of the drills was based on the geometry of the Leitz ® HW/D10/NL35/S10x24/GL70 drill (Leitz GmbH & Co. KG, Oberkochen, Baden-Württemberg, Germany), which is widely used in the drilling of throughholes in MDFs.
A coordinate measuring machine is used to measure the geometrical dimensions of the reference tool in order to make drills with coolant channels identical to the standard ones but made of cemented carbide monolith, which will be covered with two different coatings. The Zoller ® Genius 3 coordinate measuring machine (Zoller, Pleidelsheim, Germany) ( Figure 4) was used to measure and control the geometry of the cutting tools. This machine is used by tool manufacturers to enable one to take The spectral analysis of the elements constituting the material ( Figure 3) was carried out using a scanning electron microscope (TESCAN, MIRA3, Brno, Czech Republic). The spectral analysis of the elements constituting the material ( Figure 3) was carried out using a scanning electron microscope (TESCAN, MIRA3, Brno, Czech Republic).

Cutting Tools
In the drilling of MDFs, HW sintered carbide cutting tools were used with different types of cutting blade coating. Two types of coating for cutting blades, i.e., TiAlN and ZrN were used. The selection of coatings was dictated by the fact that they belong to those most commonly used in the machining of composite materials [40]. In the case of cutting wood-based materials, there is a general lack of commercially available cutting tools (especially drills) with additional protective coatings. In the research, it was necessary to measure the temperature of the cutting blade during the drilling process. To measure the temperature between the cutting edge and the workpiece using thermocouples, it was necessary for the tool to have coolant channels. However, no cooling liquids are used in the treatment of MDFs. Therefore, it was necessary to fabricate drills. The geometry of the drills was based on the geometry of the Leitz ® HW/D10/NL35/S10x24/GL70 drill (Leitz GmbH & Co. KG, Oberkochen, Baden-Württemberg, Germany), which is widely used in the drilling of throughholes in MDFs.
A coordinate measuring machine is used to measure the geometrical dimensions of the reference tool in order to make drills with coolant channels identical to the standard ones but made of cemented carbide monolith, which will be covered with two different coatings. The Zoller ® Genius 3 coordinate measuring machine (Zoller, Pleidelsheim, Germany) ( Figure 4) was used to measure and control the geometry of the cutting tools. This machine is used by tool manufacturers to enable one to take

Cutting Tools
In the drilling of MDFs, HW sintered carbide cutting tools were used with different types of cutting blade coating. Two types of coating for cutting blades, i.e., TiAlN and ZrN were used. The selection of coatings was dictated by the fact that they belong to those most commonly used in the machining of composite materials [40]. In the case of cutting wood-based materials, there is a general lack of commercially available cutting tools (especially drills) with additional protective coatings. In the research, it was necessary to measure the temperature of the cutting blade during the drilling process. To measure the temperature between the cutting edge and the workpiece using thermocouples, it was necessary for the tool to have coolant channels. However, no cooling liquids are used in the treatment of MDFs. Therefore, it was necessary to fabricate drills. The geometry of the drills was based on the geometry of the Leitz ® HW/D10/NL35/S10x24/GL70 drill (Leitz GmbH & Co. KG, Oberkochen, Baden-Württemberg, Germany), which is widely used in the drilling of through-holes in MDFs.
A coordinate measuring machine is used to measure the geometrical dimensions of the reference tool in order to make drills with coolant channels identical to the standard ones but made of cemented carbide monolith, which will be covered with two different coatings. The Zoller ® Genius 3 coordinate measuring machine (Zoller, Pleidelsheim, Germany) ( Figure 4) was used to measure and control the geometry of the cutting tools. This machine is used by tool manufacturers to enable one to take measurements in a fully automatic way. Fully automatic precise measurements are assured by five numerically controlled axes (X, Y, Z, C, B). numerically controlled axes (X, Y, Z, C, B).
The Genius 3 device is equipped with an optical system of two cameras: a main camera for measurement in transmitted and reflected light with a magnification of 50× and a tilt camera for measuring only in reflected light with a lens capable of focusing in 3D mode with a magnification of 200×. To accurately capture the details of the cutting blade, the device is equipped with eightsegment automatically tuned LED lighting.  The computer-aided design of the tool was developed in the Numroto ® Plus program (NUM AG, Teufen, Switzerland). Numroto Plus software is commonly used to design rotary cutting tools and to generate a 3D model (Figure 5c) after prior definition of all the geometrical values of the cutting tool, the selection of the type of grinding wheels and workpiece size, and the setting of machining parameters on a numerically controlled grinder. The grinding wheels used in the fabrication of drills are Toolgal ® diamond grinding wheels (TOOLGAL Industrial Diamonds Ltd., Degania, Israel). Two 1A1 grinding wheels and two types 11V9 and 12V9 were used for the fabrication of drills.
The finished project made in the Numroto ® Plus program was exported to the 5-axes grinder from a Saacke ® model UW IF (SAACKE GmbH & Co. KG, Pforzheim, Germany), shown in Figure  5a,b. An effective way to increase the durability of cutting tools is to apply a coating on their blades. The most commonly used method is Physical Vapour Deposition (PVD) in which the solid metal is vaporised in a high vacuum environment and deposited in the form of a thin layer on the tool surface. The deposited layer, usually with a thickness of 3-5 μm, has a very high hardness, usually in the range of 2000-3000 HV, which significantly increases the resistance of the tool blades to abrasive wear. The machined tools ( Figure 6b) were PVD coated with zirconium nitride (ZrN) and aluminum titanium nitride (TiAlN) using an EIFELER VACOTEC PVD Alpha 400 vacuum reactor (Vacotec GmbH, Düsseldorf, Germany) ( Figure 6a). The Genius 3 device is equipped with an optical system of two cameras: a main camera for measurement in transmitted and reflected light with a magnification of 50× and a tilt camera for measuring only in reflected light with a lens capable of focusing in 3D mode with a magnification of 200×. To accurately capture the details of the cutting blade, the device is equipped with eight-segment automatically tuned LED lighting.
The computer-aided design of the tool was developed in the Numroto ® Plus program (NUM AG, Teufen, Switzerland). Numroto Plus software is commonly used to design rotary cutting tools and to generate a 3D model ( Figure 5c) after prior definition of all the geometrical values of the cutting tool, the selection of the type of grinding wheels and workpiece size, and the setting of machining parameters on a numerically controlled grinder. The grinding wheels used in the fabrication of drills are Toolgal ® diamond grinding wheels (TOOLGAL Industrial Diamonds Ltd., Degania, Israel). Two 1A1 grinding wheels and two types 11V9 and 12V9 were used for the fabrication of drills. measurements in a fully automatic way. Fully automatic precise measurements are assured by five numerically controlled axes (X, Y, Z, C, B).
The Genius 3 device is equipped with an optical system of two cameras: a main camera for measurement in transmitted and reflected light with a magnification of 50× and a tilt camera for measuring only in reflected light with a lens capable of focusing in 3D mode with a magnification of 200×. To accurately capture the details of the cutting blade, the device is equipped with eightsegment automatically tuned LED lighting.  The computer-aided design of the tool was developed in the Numroto ® Plus program (NUM AG, Teufen, Switzerland). Numroto Plus software is commonly used to design rotary cutting tools and to generate a 3D model ( Figure 5c) after prior definition of all the geometrical values of the cutting tool, the selection of the type of grinding wheels and workpiece size, and the setting of machining parameters on a numerically controlled grinder. The grinding wheels used in the fabrication of drills are Toolgal ® diamond grinding wheels (TOOLGAL Industrial Diamonds Ltd., Degania, Israel). Two 1A1 grinding wheels and two types 11V9 and 12V9 were used for the fabrication of drills.
The finished project made in the Numroto ® Plus program was exported to the 5-axes grinder from a Saacke ® model UW IF (SAACKE GmbH & Co. KG, Pforzheim, Germany), shown in Figure  5a,b. An effective way to increase the durability of cutting tools is to apply a coating on their blades. The most commonly used method is Physical Vapour Deposition (PVD) in which the solid metal is vaporised in a high vacuum environment and deposited in the form of a thin layer on the tool surface. The deposited layer, usually with a thickness of 3-5 μm, has a very high hardness, usually in the range of 2000-3000 HV, which significantly increases the resistance of the tool blades to abrasive wear. The machined tools ( Figure 6b) were PVD coated with zirconium nitride (ZrN) and aluminum titanium nitride (TiAlN) using an EIFELER VACOTEC PVD Alpha 400 vacuum reactor (Vacotec GmbH, Düsseldorf, Germany) ( Figure 6a). An effective way to increase the durability of cutting tools is to apply a coating on their blades. The most commonly used method is Physical Vapour Deposition (PVD) in which the solid metal is vaporised in a high vacuum environment and deposited in the form of a thin layer on the tool surface. The deposited layer, usually with a thickness of 3-5 µm, has a very high hardness, usually in the range of 2000-3000 HV, which significantly increases the resistance of the tool blades to abrasive wear. The machined tools ( Figure 6b     The coatings were made on a substrate previously sprayed with high energy ions, free of oxides and enriched with elements forming a strong adhesion-diffusion bond. The selected properties of the coatings used are listed in Table 2. Three types of drills with coolant channels were fabricated: an HW carbide drill with a ZrN coating, an HW carbide drill without a coating and an HW carbide drill with a TiAlN coating ( Figure 6b). The spectral analysis of the elements included in the coating applied ( Figure 7) was carried out using a TESCAN ® scanning electron microscope (TESCAN, MIRA3, Brno, Czech Republic). The coatings were made on a substrate previously sprayed with high energy ions, free of oxides and enriched with elements forming a strong adhesion-diffusion bond. The selected properties of the coatings used are listed in Table 2. Three types of drills with coolant channels were fabricated: an HW carbide drill with a ZrN coating, an HW carbide drill without a coating and an HW carbide drill with a TiAlN coating ( Figure  6b). The spectral analysis of the elements included in the coating applied ( Figure 7) was carried out using a TESCAN ® scanning electron microscope (TESCAN, MIRA3, Brno, Czech Republic).

Equipment and Machining Conditions
The drilling process was carried out in two stages: on a CNC vertical milling machine and on an EMCO ® CNC lathe (EMCO GmbH, Hallein, Austria). A schematic diagram of the configuration of the measurement path and the measurement data archiving system is presented in Figure 8. In the first stage of testing, the CNC milling machine recorded the values of thrust force (F t ) and cutting torque (M c ) during MDF machining.

Equipment and Machining Conditions
The drilling process was carried out in two stages: on a CNC vertical milling machine and on an EMCO ® CNC lathe (EMCO GmbH, Hallein, Austria). A schematic diagram of the configuration of the measurement path and the measurement data archiving system is presented in Figure 8. In the first stage of testing, the CNC milling machine recorded the values of thrust force (Ft) and cutting torque (Mc) during MDF machining. In the first stage, three holes (with the same set of cutting parameters) were drilled in an MDF panel with dimensions 130 mm × 30 mm × 18 mm on a CNC milling machine. The value of the thrust force and the cutting torque was measured using the Kistler ® 9345B2 piezoelectric industrial sensor (Kistler, Winterthur, Switzerland). The signals from the sensor were recorded on a personal computer (PC) disk via the National Instruments ® 6034E (National Instruments Corporation, Austin, TX, USA) 16-bit analogue-to-digital card with a sampling rate of 50 Hz. The surface topography of each completed hole was measured in two locations (every 180°) using the Hommel-Etamic T8000RC CNC profilometer (Jenoptik, Jena, Germany) ( Figure 9). In order to measure the surface roughness, the test sample was cut into two parts along the hole axis. One measurement was made for each part separately. In the second stage of the investigations, the holes were drilled (with the same set of cutting parameters) in MDF workpieces with a diameter of 30 mm and a thickness of 18 mm on a CNC lathe. In the first stage, three holes (with the same set of cutting parameters) were drilled in an MDF panel with dimensions 130 mm × 30 mm × 18 mm on a CNC milling machine. The value of the thrust force and the cutting torque was measured using the Kistler ® 9345B2 piezoelectric industrial sensor (Kistler, Winterthur, Switzerland). The signals from the sensor were recorded on a personal computer (PC) disk via the National Instruments ® 6034E (National Instruments Corporation, Austin, TX, USA) 16-bit analogue-to-digital card with a sampling rate of 50 Hz. The surface topography of each completed hole was measured in two locations (every 180 • ) using the Hommel-Etamic T8000RC CNC profilometer (Jenoptik, Jena, Germany) ( Figure 9). In order to measure the surface roughness, the test sample was cut into two parts along the hole axis. One measurement was made for each part separately.

Equipment and Machining Conditions
The drilling process was carried out in two stages: on a CNC vertical milling machine and on an EMCO ® CNC lathe (EMCO GmbH, Hallein, Austria). A schematic diagram of the configuration of the measurement path and the measurement data archiving system is presented in Figure 8. In the first stage of testing, the CNC milling machine recorded the values of thrust force (Ft) and cutting torque (Mc) during MDF machining. In the first stage, three holes (with the same set of cutting parameters) were drilled in an MDF panel with dimensions 130 mm × 30 mm × 18 mm on a CNC milling machine. The value of the thrust force and the cutting torque was measured using the Kistler ® 9345B2 piezoelectric industrial sensor (Kistler, Winterthur, Switzerland). The signals from the sensor were recorded on a personal computer (PC) disk via the National Instruments ® 6034E (National Instruments Corporation, Austin, TX, USA) 16-bit analogue-to-digital card with a sampling rate of 50 Hz. The surface topography of each completed hole was measured in two locations (every 180°) using the Hommel-Etamic T8000RC CNC profilometer (Jenoptik, Jena, Germany) ( Figure 9). In order to measure the surface roughness, the test sample was cut into two parts along the hole axis. One measurement was made for each part separately. In the second stage of the investigations, the holes were drilled (with the same set of cutting parameters) in MDF workpieces with a diameter of 30 mm and a thickness of 18 mm on a CNC lathe. In the second stage of the investigations, the holes were drilled (with the same set of cutting parameters) in MDF workpieces with a diameter of 30 mm and a thickness of 18 mm on a CNC lathe. The temperature value was measured during machining between the cutting edge and the workpiece using the National Instruments ® 9212 industrial system (National Instruments Corporation, Austin, TX, USA). Two K-type thermocouple wires with a diameter of 0.2 mm were used for temperature measurement. The thermocouple wires were mounted in the cooling liquid drill channels ( Figure 10). Signals from the measurement system were recorded in digital form on a personal computer (PC) disk. The sampling rate of signals during experiments was 5 Hz, and a 24-bit measurement card was used. The temperature value was measured during machining between the cutting edge and the workpiece using the National Instruments ® 9212 industrial system (National Instruments Corporation, Austin, TX, USA). Two K-type thermocouple wires with a diameter of 0.2 mm were used for temperature measurement. The thermocouple wires were mounted in the cooling liquid drill channels ( Figure 10). Signals from the measurement system were recorded in digital form on a personal computer (PC) disk. The sampling rate of signals during experiments was 5 Hz, and a 24-bit measurement card was used. The cutting parameters used during the drilling experiments are listed in Table 3. Three replications were made for each of the sets of cutting parameters. This research methodology was applied to drilling with the following types of drills: not coated, ZrN coated and TiAlN coated.
In order to avoid the accidental influence of the radius of the cutting edge of drills (εr) on the machinability indicators (thrust force, cutting torque, temperature and surface roughness), the drills with the same value of the cutting edge radius were selected. The variation of the the cutting edge radius εr of drills used varied between 5.91 and 5.95 μm. The Zoller Genius 3 measuring machine (Zoller, Pleidelsheim, Germany) ( Figure 4) was used to measure the εr radius. Figure 11 shows the details of measurement of radius of the drill's cutting edge.   The cutting parameters used during the drilling experiments are listed in Table 3. Three replications were made for each of the sets of cutting parameters. This research methodology was applied to drilling with the following types of drills: not coated, ZrN coated and TiAlN coated. In order to avoid the accidental influence of the radius of the cutting edge of drills (ε r ) on the machinability indicators (thrust force, cutting torque, temperature and surface roughness), the drills with the same value of the cutting edge radius were selected. The variation of the the cutting edge radius ε r of drills used varied between 5.91 and 5.95 µm. The Zoller Genius 3 measuring machine (Zoller, Pleidelsheim, Germany) ( Figure 4) was used to measure the ε r radius. Figure 11 shows the details of measurement of radius of the drill's cutting edge.
In order to avoid the accidental influence of the radius of the cutting edge of drills (εr) on the machinability indicators (thrust force, cutting torque, temperature and surface roughness), the drills with the same value of the cutting edge radius were selected. The variation of the the cutting edge radius εr of drills used varied between 5.91 and 5.95 μm. The Zoller Genius 3 measuring machine (Zoller, Pleidelsheim, Germany) (Figure 4) was used to measure the εr radius. Figure 11 shows the details of measurement of radius of the drill's cutting edge.   Figure 11. Measurement of radius of the drill's cutting edge.  Figure 12 shows the variation in thrust force, cutting torque and cutting edge temperature as a function of cutting length when drilling the MDF at a cutting speed of 35 m/min and a feed rate of 167 mm/min using the TiAlN coated drill. It was observed that five major phases of the variation of the thrust force (Ft) value can be identified during drilling (Regions I-V in Figure 12). In the period referred to as Phase I, the chisel edge of the drill penetrates into the material producing a rapid increase in the value of the thrust force. During this phase, instead of cutting, the chisel edge of the drill is pressed into the material. Then, the drill starts to cut the outer layer of the material with the highest density and hardness. When the chisel edge of the drill leaves the area of material with the highest density (depth approx. 2.2 mm), the force value decreases due to the fact that the drill sinks into the middle layer of the panel with lower density and hardness. During Phase II, the drill penetrates to a depth of h = 8.1 mm, which is equal to the height of the cutting blades of the drill (Figure 5c). During this time, there is an increase in the cross-sectional area of the cutting layer, which results in a proportional increase in the thrust force. Phase III corresponds to cutting with a constant cross-section of the cutting layer in the middle layer of the panel. Stabilisation of the thrust force value is then observed. At the beginning of Phase IV, the value of the thrust force starts to increase rapidly due to the fact that the drill begins to penetrate the outer layer of the material (higher density and hardness). The thrust force reaches its maximum value. When the chisel edge of a drill leaves the workpiece (cutting path approx. 18 mm), the thrust force starts to decrease (Phase V). The value of the thrust force drops to zero at the end of Phase V when the cutting edges of the drill have left the workpiece. It was observed that five major phases of the variation of the thrust force (F t ) value can be identified during drilling (Regions I-V in Figure 12). In the period referred to as Phase I, the chisel edge of the drill penetrates into the material producing a rapid increase in the value of the thrust force. During this phase, instead of cutting, the chisel edge of the drill is pressed into the material. Then, the drill starts to cut the outer layer of the material with the highest density and hardness. When the chisel edge of the drill leaves the area of material with the highest density (depth approx. 2.2 mm), the force value decreases due to the fact that the drill sinks into the middle layer of the panel with lower density and hardness. During Phase II, the drill penetrates to a depth of h = 8.1 mm, which is equal to the height of the cutting blades of the drill (Figure 5c). During this time, there is an increase in the cross-sectional area of the cutting layer, which results in a proportional increase in the thrust force. Phase III corresponds to cutting with a constant cross-section of the cutting layer in the middle layer of the panel. Stabilisation of the thrust force value is then observed. At the beginning of Phase IV, the value of the thrust force starts to increase rapidly due to the fact that the drill begins to penetrate the outer layer of the material (higher density and hardness). The thrust force reaches its maximum value. When the chisel edge of a drill leaves the workpiece (cutting path approx. 18 mm), the thrust force starts to decrease (Phase V). The value of the thrust force drops to zero at the end of Phase V when the cutting edges of the drill have left the workpiece.

Feed Force and Cutting Torque
In the case of cutting torque (M c ), it is clearly visible in Phase 1 that there is no machining, and only the chisel edge of a drill is pressed into the workpiece. In Phase I, the cutting torque value is close to zero. Next, the blades sink into the workpiece (Phase II) which results in a proportional increase in the value of cutting torque. The period of this increase continues until the drill reaches the maximum cross-sectional area of the cutting layer, i.e., the drill is penetrated to a depth of h = 8.1 mm. In Phase III, the cutting of the cutting layer with a constant cross section in the middle layer of the panel proceeds. A stabilisation of the cutting torque value is observed. At the beginning of Phase IV, the value of the cutting torque starts to increase due to the fact that the drill begins to enter into the outer layer of material (higher density and hardness). Furthermore, in this phase, the largest cross-section of the material is cut, and the cutting torque reaches its maximum value. When the drill begins to come out of the workpiece (cutting path approx. 18 mm), the cutting torque starts to decrease (Phase V). In the middle part of Phase V, some stabilisation of the cutting torque is observed. As a result, the drill that comes out of the material cuts only the outer layer of the material characterized by high density and hardness. The cutting torque value drops to zero at the end of Phase V when the cutting edges of the drill have left the workpiece. For the analysis of the acquired signals of F t and M c , an individually designed computer program was prepared in the LabVIEW programming language enabling, at selected time intervals, the mean values of the recorded thrust force and cutting torque signals to be determined. The program was based on the automatic determination of values of F t and M c parameters in a specific time range of the signal. The methodology for determining the average values of recorded signals has been described in detail in [5]. Figure 13a-c shows the effect of feed per revolution (f) on the thrust force (F t ) for the three cutting speeds and the three types of drill coating. The thrust force values presented in the graphs were obtained as the average result of three repetitions. outer layer of material (higher density and hardness). Furthermore, in this phase, the largest crosssection of the material is cut, and the cutting torque reaches its maximum value. When the drill begins to come out of the workpiece (cutting path approx. 18 mm), the cutting torque starts to decrease (Phase V). In the middle part of Phase V, some stabilisation of the cutting torque is observed. As a result, the drill that comes out of the material cuts only the outer layer of the material characterized by high density and hardness. The cutting torque value drops to zero at the end of Phase V when the cutting edges of the drill have left the workpiece. For the analysis of the acquired signals of Ft and Mc, an individually designed computer program was prepared in the LabVIEW programming language enabling, at selected time intervals, the mean values of the recorded thrust force and cutting torque signals to be determined. The program was based on the automatic determination of values of Ft and Mc parameters in a specific time range of the signal. The methodology for determining the average values of recorded signals has been described in detail in [5]. Figure 13a-c shows the effect of feed per revolution (f) on the thrust force (Ft) for the three cutting speeds and the three types of drill coating. The thrust force values presented in the graphs were obtained as the average result of three repetitions.
Multi-factorial analysis of variance (ANOVA) carried out in the STATISTICA program allowed the verification of the significance of the influence of several independent variables on the dependent variable. Furthermore, multivariate analysis makes it possible to take the synergistic effect of the product of many variables into account in the statistical model. Taking into account the adopted level of significance of p = 0.05, the statistical significance of particular groups of variables and individual variables is determined. The results of the analysis (Table 4) allow one to reject, at a significance level p = 0.000, the hypothesis concerning the lack of effect of the factors "coating," feed per revolution (f) and cutting speed (vc) on the value of the thrust force (Ft). There was no statistically significant effect of interactions between the factors analyzed. Figure 13d presents a comparison between the results for the values of the thrust force obtained during the experiment with the values obtained based on the analytical models (1, 2, 3). The correlation coefficients obtained were equal as follows: R 2 = 0.990 for the TiAlN coated drill, R 2 = 0.983 for the ZrN coated drill and R 2 = 0.993 for the uncoated drill.  The smallest value of thrust force (Ft) was obtained in the drilling process using a tool without a coating. However, the highest values of thrust force were obtained for a drill with a TiAlN coating. The increase in the value of the thrust force in comparison to machining without a coating was on average about 39%. In the case of a ZrN coated drill, the value of the thrust force obtained was reduced in comparison to the TiAlN coated drill, but it was still higher on average by about 17% in relation to the value obtained with the use of a drill without a coating. For example: for the cutting speed vc = 105 m/min and feed per revolution value f = 0.2 mm/rev the value of the thrust force for the TiAlN coated drill was 33 N, for the tool with a ZrN coating was 27.8 N and for the tool without a coating it was 23.7 N. This can be explained by the differing values of the coefficient of friction between the tool and the workpiece material resulting from the type of coating used. For all the coatings used, the value of the thrust force increases with an increase in the value of the feed per revolution. The value of the thrust force can be described by the Equations (1) The results of the statistical analysis (Table 5) allow one to reject, at the level of significance p = 0.000, the hypothesis that the coating type, feed per revolution (f) and cutting speed (vc) do not affect Multi-factorial analysis of variance (ANOVA) carried out in the STATISTICA program allowed the verification of the significance of the influence of several independent variables on the dependent variable. Furthermore, multivariate analysis makes it possible to take the synergistic effect of the product of many variables into account in the statistical model. Taking into account the adopted level of significance of p = 0.05, the statistical significance of particular groups of variables and individual variables is determined. The results of the analysis (Table 4) allow one to reject, at a significance level p = 0.000, the hypothesis concerning the lack of effect of the factors "coating", feed per revolution (f) and cutting speed (v c ) on the value of the thrust force (F t ). There was no statistically significant effect of interactions between the factors analyzed. Table 4. Significance level of the effect of cutting parameters on the average thrust force (F t ).

Tests Applied
Level of Significance (p ≤ 0.05) cutting speed (v c ) 0.000 feed per revolution (f) 0.000 coating 0.000 coating × cutting speed 0.997 coating × feed per revolution 0.564 Figure 13d presents a comparison between the results for the values of the thrust force obtained during the experiment with the values obtained based on the analytical models (1, 2, 3). The correlation coefficients obtained were equal as follows: R 2 = 0.990 for the TiAlN coated drill, R 2 = 0.983 for the ZrN coated drill and R 2 = 0.993 for the uncoated drill.
The smallest value of thrust force (F t ) was obtained in the drilling process using a tool without a coating. However, the highest values of thrust force were obtained for a drill with a TiAlN coating. The increase in the value of the thrust force in comparison to machining without a coating was on average about 39%. In the case of a ZrN coated drill, the value of the thrust force obtained was reduced in comparison to the TiAlN coated drill, but it was still higher on average by about 17% in relation to the value obtained with the use of a drill without a coating.  The lowest value of torque was obtained using the drill without a coating in the drilling process ( Figure 14). The highest values of cutting torque (Mc) recorded in the experiments were obtained using a drill with a TiAlN coating. The increase in the value of cutting torque when machining using coated drills in comparison to that using an uncoated drill was approximately 35%. In the case of a ZrN coated drill, the value of the cutting torque (Mc) obtained was less than that using a TiAlN coated drill, but it was still greater by approx. 15% in relation to the value of (Mc) obtained with a drill without a coating.   The results of the statistical analysis (Table 5) allow one to reject, at the level of significance p = 0.000, the hypothesis that the coating type, feed per revolution (f) and cutting speed (v c ) do not affect the value of the cutting torque (M c ). In the case of interactions between the factors analyzed, no statistically significant effect of these factors on the value of (M c ) was observed. The lowest value of torque was obtained using the drill without a coating in the drilling process ( Figure 14). The highest values of cutting torque (M c ) recorded in the experiments were obtained using a drill with a TiAlN coating. The increase in the value of cutting torque when machining using coated drills in comparison to that using an uncoated drill was approximately 35%. In the case of a ZrN coated drill, the value of the cutting torque (M c ) obtained was less than that using a TiAlN coated drill, but it was still greater by approx. 15% in relation to the value of (M c ) obtained with a drill without a coating.
For example: for the cutting speed v c = 35 m/min and feed per revolution value f = 0.2 mm/rev, the cutting torque (M c ) for the TiAlN coated drill was 0.35 Nm, for the ZrN coated drill it was 0.3 Nm and for the drill without coating was 0.26 Nm. In a similar manner to the effect of drilling parameters on the value of the thrust force, this fact can be explained by the differing values of the coefficient of friction between the tool and the workpiece resulting from the type of drill coating. For all the coatings used, the value of the thrust force (F t ) increases with an increase of the feed per revolution (f) value. The value of the thrust force (F t ) can therefore be described by the Equations (4)- (6).
The correlation coefficients obtained between the experimental and predicted values of cutting torque M c (Figure 14d) were equal as follows: R 2 = 0.981 for a TiAlN coated drill, R 2 = 0.997 for a ZrN coated drill and R 2 = 0.993 for an uncoated drill. For all the coatings of tools used, the value of the cutting torque (M c ) increases as the feed per revolution (f) increases.

Analysis of Surface Topography
Surface topography is one of the main features taken into account to evaluate the surface quality in machining processes. The value of the roughness average Ra parameter was measured in the longitudinal direction of the holes machined in MDF panels. The choice of this indicator to assess surface roughness was dictated by its very frequent use in production plants [23]. The value of the roughness average (Ra) was determined on the basis of the surface topography map in selected measurement sections.  Three areas with a different character of surface roughness can clearly be seen on the surface topographies presented. The first and the second area occur in the outer layers of the MDF panel and the third area in the middle layer of the board. This diversity can be explained by the abovementioned multi-layered structure of the MDFs. The resulting surface topography accurately reflects changes in both hardness and panel density.  Three areas with a different character of surface roughness can clearly be seen on the surface topographies presented. The first and the second area occur in the outer layers of the MDF panel and the third area in the middle layer of the board. This diversity can be explained by the abovementioned multi-layered structure of the MDFs. The resulting surface topography accurately reflects changes in both hardness and panel density.
Measurement of the roughness average (Ra) parameter was carried out in accordance with the recommendations of ISO-4288:2011. The test conditions of the surface roughness measurement were adopted in accordance with Table 6. The mean groove spacing (RSm) value ( Figure 16) in the measurements was in the range between 0.13 and 0.4. Table 6. Setup for the roughness measurement (EN ISO 4288).    The value of the Ra roughness average parameter was determined separately for the outer layer and middle layer of the MDF. Therefore, the additional effect of both the density and hardness of the workpiece on the Ra parameter can be analyzed. Figures 17 and 18 show the effect of the feed per revolution value for the three cutting speeds and all types of tool coatings. The values of roughness average (Ra) in Figures 17 and 18 are the average of six measurements. As mentioned, two measurements were made for each hole, and the drilling of each hole was repeated three times. The results presented in Figure 17 refer to the measurement of the surface roughness in the middle layer of the workpiece.

367
The value of the Ra roughness average parameter was determined separately for the outer layer 368 and middle layer of the MDF. Therefore, the additional effect of both the density and hardness of the 369 workpiece on the Ra parameter can be analyzed.

401
The results of the analysis (Table 7) allow one to reject, at a significance level p = 0.000, the 402 hypothesis that the coating type and feed per revolution (f) do not affect the Ra parameter value.

403
There was no statistically significant impact of the cutting speed (vc) on the Ra parameter value.

404
Similarly, in the case of interactions between the factors analyzed, no statistically significant impact 405 was observed.

422
The ZrN coating has a much lower heat transfer coefficient compared to the TiAlN coating (Table 2).

423
This causes the ZrN coating to be a barrier to removing heat from the cutting zone. In addition, the 424 MDF panel is characterized by a relatively low value of thermal conductivity.  Figure 18 presents the results regarding the measurement of surface roughness in the outer layer of the workpiece. In both cases, as the cutting speed (v c ) increases and simultaneously the feed per revolution (f) decreases, the value of the roughness average (Ra) decreases. For the inner layer of panel at a cutting speed of 35 m/min and a feed per revolution of 0.2 mm/rev, the roughness average parameter was Ra = 10.3 µm for a TiAlN coated drill, for a cutting speed of 105 m/min and a feed per revolution of 0.1 mm/rev, the roughness average parameter was Ra = 6.6 µm. However, for the outer layer of the panel at a cutting speed of 35 m/min and a feed per revolution of 0.2 mm/rev, the roughness average parameter was Ra = 5.1 µm, while for a cutting speed of 105 m/min and a feed per revolution of 0.1 mm/rev the roughness average parameter was Ra = 3.3 µm for a drill with TiAlN coating. This can be explained by the fact that the accumulation of chips in the chip spaces decreased with increasing cutting speed (v c ). In addition, the very pronounced impact of the type of tool coating on the surface roughness value was noted. It has been found that machining with a ZrN coated tool allows one to achieve the lowest surface roughness value compared to a TiAlN coated tool and an uncoated tool. This may be caused by a different value of friction coefficient as well as of thermal conductivity coefficient depending on the type of coating (Table 2). A higher value of friction coefficient and a lower value of coefficient of thermal conductivity causes an increase in the temperature value in the tool-workpiece contact area. The increase in heat generated in the area of contact between the cutting tool and the workpiece, in the case of MDF, significantly improves the connection of wood fibers and formaldehyde adhesive. It causes the compaction of the bonds between fibers.
The results of the analysis (Table 7) allow one to reject, at a significance level p = 0.000, the hypothesis that the coating type and feed per revolution (f) do not affect the Ra parameter value. There was no statistically significant impact of the cutting speed (v c ) on the Ra parameter value. Similarly, in the case of interactions between the factors analyzed, no statistically significant impact was observed. The experiments conducted showed that the feed per revolution (f) value, cutting speed (v c ) and the type of tool coating have a significant influence on the roughness average (Ra) parameter. However, the Ra parameter values measured in the outer layers are significantly lower than those measured in the inner layer.

Temperature of Cutting Tools
The highest values of tool temperature were observed in the case of the ZrN coated drill, and the lowest in the case of the uncoated drill (Figure 19a-c). The increase in the temperature value for a ZrN coated drill compared to a drill without a coating was about 20%. In the case of the TiAlN coated drill, the value of cutting torque obtained (M c ) is lower compared to the ZrN coated drill, but it was still higher by approximately 13% in relation to the temperature value obtained with the use of an uncoated drill. and R 2 = 0.941 for an uncoated drill (Figure 19d). Table 8. Significance level of the effect of cutting parameters on temperature (T).

Tests Applied
Level of Significance (p ≤ 0.05) cutting speed (vc) 0.006 feed per revolution (f) 0.000 coating 0.000 coating × cutting speed 0.998 coating × feed per revolution 0.943 The temperature value of the cutting edge depending on the coating applied is expressed in the form of the Equations (7) In the case of the ZrN drill, the maximum temperature was 56.3 • C. The smallest value of temperature obtained in the drilling process relating to an uncoated drill was 38.5 • C. This fact can be explained by differentiation in both the coefficient of friction between the tool and the workpiece as well as in the value of the thermal conductivity coefficient resulting from the type of tool coating. The ZrN coating has a much lower heat transfer coefficient compared to the TiAlN coating (Table 2). This causes the ZrN coating to be a barrier to removing heat from the cutting zone. In addition, the MDF panel is characterized by a relatively low value of thermal conductivity.

5.
The feed per revolution (f) and the type of drill coating had a significant influence on the value of the roughness average (Ra) parameter. It has been observed that in the outer layers of the panel, the (Ra) parameter value has a lower value compared to that measured in the middle layer. The lowest value of roughness average parameter (Ra) was observed in the case of the ZrN coated drill, and the highest in the case of the uncoated drill. This fact can be explained by a lower value of friction coefficient and a lower value of thermal conductivity coefficient of ZrN coating compared to the uncoated tool. The lower value both of thermal conductivity coefficient and friction coefficient causes an increase in heat generated in the area of contact between the cutting tool and the workpiece. In the case of MDF, this significantly improves the connection of wood fibers and formaldehyde adhesive. It causes the compaction of the bonds between fibers.
In conclusion, the feed per revolution (f) and the type of tool coating are the dominant parameters that significantly affect the drilling process of the MDF board.